The Smartphone Wars: The market share scramble and Apple’s long con

Mobile phone carriers have a crappy record of strategic planning – the history of the industry is rife with massive overinvestment in services consumers didn’t actually want, partly redeemed by massive unanticipated revenue from accidents of technology (I’m looking at you, SMS!). I’ve explained elsewhere that inflation-adjusted carrier ROI is negative.

Even so, the latest news from the analysts is pretty mind-boggling. Remember all those carrier execs rhapsodizing about how iPhone is the awesomest invention since sex? Well, it seems Apple is sucking all the profits out of the carriers that went for it. That has interesting implications for the future. Like, what happens when the carriers decide they’re done being conned?

The story begins:

The price of Apple’s iconic smartphone is heavily discounted by carriers. Those subsidies almost single-handedly devastate profit margins for Verizon, AT&T and Sprint.

The money quote is probably this:

“When we look at the direct and indirect economics that Apple has managed to extract from the carriers, the carrier-level value destruction is quite evident.”

Yeesh…”Value destruction”. The implication, which is backed up by the rest of the article, is that the iPhone is wreaking enough havoc on carrier margins to be seriously damaging. While this neatly explains Apple’s record quarter in 4Q2011, it also makes harder to understand why the carriers are standing still to be milked.

We know what the carriers think they’re buying – increased profits through increased market share. The problem with this theory is that it no longer makes any sense at all. Three of the four major carriers (AT&T, Verizon, Sprint) now carry the Apple product. How can the iPhone be a winning differentiator when almost all of your competitors have it too?

OK, so there might be a second level to the argument: each carrier might be thinking that if it doesn’t carry the iPhone it will have its marketshare eaten by others that do. The trouble with this theory is that Android is still growing userbase and marketshare faster than the iPhone. And though T-Mobile (the one carrier without iPhone) ain’t doing so well, nobody in the industry thinks lack of iPhone – as opposed to, say, weak execution and lack of the capital mass to pursue its buildout – is its problem.

From any carrier’s point of view, the case for dumping iPhone, or at least threatening to do so in order to renegotiate Apple’s subsidy requirement away, seems pretty open and shut. Apple has things all its own way right now – skimming the lion’s share of the profits off the carriers’ business without having to shoulder their risks. But this is an unstable situation, because the carriers’ investors won’t tolerate it indefinitely. What happens when they revolt?

That’s not hard to figure, actually. Best case, Apple is going to have to give up the carrier subsidies, taking a serious hit on volume and profits. Because the alternative is that one or more of the majors will stop carrying the iPhone entirely, which would be much worse. Likely Apple’s market share would actually drop noticeably if that happened. And that would be disastrous, because the other carriers would probably run for the exits.

The bottom line is that Apple’s current performance isn’t sustainable. The losses the carriers are presently eating on the iPhone are going to get squeezed out one way or another, almost certainly re-manifesting as significantly higher unit prices to the consumer. This, of course, will increase Android’s competitive advantage.

643 thoughts on “The Smartphone Wars: The market share scramble and Apple’s long con

  1. I don’t see any of the carriers actually dropping the iPhone, but I could easily see one or more of them start to deemphasize it. Like, one small iPhone display, and maybe fewer unit purchases. That sort of thing. Otherwise, MAYBE increased rates to cover it, though of course they would apply to to all plans to hide the real reason for the hike.

    On a blog note, it was actually very difficult to read this post; maybe too much whitespace, I’m not sure, but I had to re-read at least two paragraphs before my brain would register.

  2. I’m not sure if it’s the whitespace but for me (Chrome and IE9 on Windows) the serif font isn’t easy on the eyes.

  3. By the way, do you plan on introducing other categories than “General” and “Administrative”?

  4. Yeah, that’s what I’d blame, too, whether it’s true or not. It will get repeated often enough to be true.

  5. The question is, can the carriers actually renegotiate a deal?

    AT&T is selling old phones like nobody’s business. One would think they must have gotten some sort of long-term contract to do that.

    Verizon obviously creamed in their pants to get the iPhone. Who knows what they signed?

    All we really know is that Sprint reportedly offered Apple $15.5 BILLION (cue three pinkies) over 4 years…

    My guess is that the carriers are stuck. If T Mobile takes AT&T’s money and does something useful with it rather than giving it to Apple, and stops making some of their other mistakes like the Lightsquared clusterfuck, then maybe this will give them a chance to start making a profit.

    Because the other carriers are going to have to start raising data rates, and whatever else is in the contract, you can bet your bottom dollar they aren’t allowed to charge iPhone users extra.

  6. When all the villagers get tired of the vampire sucking their blood, they’ll band together and drive a stake in its heart.

  7. So the iPhone kills margins, but what does it do to net profits? If my margin drops 5% but my sales volume increases by 10%, I’m a happy camper. Now, the interesting thing is that the iPhone, model 4S in particular, hogs bandwidth like crazy (whether it’s Siri or iCloud is immaterial). So, just how much additional profit does having an iPhone on your network give you?

  8. Its is the users. Iphone users use more bandwidth, because they tend to use their phones a lot more for ordinary things (browsing, downloading music, videos, etc.

  9. It’s basic math. Every smartphone manufacturer has 1 or a few models with a carrier subsidy in the $300-400 range. Every other manufacturer also has more models with lower to no subsidy. Apple may be alone in offering only either un-subsidized phones or phones the carriers are willing to pay a $350-500 dollar subsidy to carry (even HTC and RIMM have models that merit little to no subsidy), but this doesn’t mean the carrier woes are solely due to Apple. All of the high end phone users/data consumers (requiring high subsidies and infrastructure investment from the carriers) are problematic for the carriers (as are the low end users not adopting new data services).

    Suggesting that Apple is uniquely responsible for the carrier problems because they cost $50-100 more in acquisition costs than what even most readers here would consider as “comparable or better than the iPhone” over, on average, 2 years only tells me that Apple is uniquely more desirable and profitable (or as profitable) than the lower-priced competition.

    Attempting to bite back on the subsidy isn’t going to improve the carriers’s situation; it will merely stall the necessary market transition.

  10. Every smartphone manufacturer has 1 or a few models with a carrier subsidy in the $300-400 range.

    [Citation needed.]

  11. The carriers are stuck. Even customers who don’t want an iPhone will look crosseyed at a carrier that doesn’t offer it. (If you don’t have the iPhone, you must not be a ‘real’ wireless company.)
    What I think will happen is that, rather than confront Apple, the carriers will hold a secret (illegal) conclave, and agree to raise their prices for the iPhone simultaneously. After all, it worked when the tuna canners all went from 6 ounce to 5 ounce cans at the same time. (What a miracle of the free market!)

  12. The question is not whether or not customers will buy iPhones. The question is why do carriers subsidize the iPhone purchase to the point where there is no profit remaining for them on that iPhone customer. Why do I, as a wireless telco, want an iPhone customer whose phone has cost the telco so much, and who’s data useage will be such a burden on the network.

  13. I would argue that the premise that the iPhone is uniquely damaging to the carriers is false.

  14. What we *want* is for the carriers to stop subsidizing phones, stop adding crap to the phones, and just sit back and be a dumb pipe like we want them to.

    I know the focus here is the sustainability of Apple vs. Android, but you’re also failing to take into consideration the fact that Microsoft is practically giving away Vista Phones at this point; the subsidies are likely to be flowing in the opposite direction. Apple won’t be able to keep up with a competitor who’s willing to throw billions down the hole for years at a time to buy market share.

  15. “Otherwise, MAYBE increased rates to cover it, though of course they would apply to to all plans to hide the real reason for the hike.”

    Of course, this has already happened several times in the last 5 years, but not solely due to Apple. Mostly in the form of restrictions and caps to existing data rate plans because of any and all high cost, network-intensive consumers. The fact is: the benefits and deficiencies of any one model or type of consumer can be distributed across, extracted from, or made up from, in large measure, a carrier’s entire pool of subscribers. Goldman’s and McCormick’s analysis isn’t necessarily well-supported. We do not have very transparent data for individual models, individual manufacturers, or individual platforms across all significant factors for the carriers (sub growth, churn, ARPU, EBITDA, sub termination fees, data usage and infrastructure investment, etc…) in an industry measuring ROI on the multi-decade timeframe.

  16. With firefox on kubuntu, the font is large enough. To make me set it a couple of clicks smaller, still a bit on the large side, which then will require me to set it back up when I leave. And makes the fort for this note a bit on the small side.

    Jim

  17. > What we *want* is for the carriers to stop subsidizing phones, stop adding crap to the phones, and just sit back and be a dumb pipe like we want them to.

    Yeah the problem with that is as ESR has pointed out (in the link in the OP) that that is a negative ROI business. They need to get the value add ons to leverage their infrastructure.

  18. “Apple won’t be able to keep up with a competitor who’s willing to throw billions down the hole for years at a time to buy market share.”

    Microsoft is basically making up the subsidy gap between what the carriers will offer Apple for the iPhone and what they will provide for all other comparable smartphones ($200-250). But this is a Microsoft cost simply to spread the platform; they do not gain in device or carrier or license revenue. I don’t see how Apple will not be able to keep up with Microsoft when they have more market and enterprise value, cash, revenue, profit, and market share.

  19. Given that Apple doesn’t care about market share all that much as much as profit share why CAN’T Apple keep up with a competitor that dumps billions down the whole for years whether that’s Google or Microsoft?

  20. Yes, it’s usually safe for the primary assumption to be that the marketplace (for smartphone pricing, not carrier operations) is broken.

  21. I misunderstood. I thought you were saying the subsidy was in the $300-400 price range, just like the Apple subsidy.

    Oh, yeah, that’s because we were talking about subsidies:

    Suggesting that Apple is uniquely responsible for the carrier problems because they cost $50-100 more in acquisition costs than what even most readers here would consider as “comparable or better than the iPhone” over, on average, 2 years only tells me that Apple is uniquely more desirable and profitable (or as profitable) than the lower-priced competition.

    Everything I’ve read says that Android subsidies are typically $150 and iPhone subsidies are typically $400. $400 – $150 != $50 or evey $100.

  22. I don’t think google’s dumping billions — everything I’ve seen suggests that they’re getting some sort of positive ROI, even if not huge.

    I’m sure Apple will make money, at least for awhile, even if relegated to a boutique business, and you know that a lot of us here don’t really care about its profit.

    However, if Microsoft wraps $400 around each handset, I think Apple’s (and HTC’s and everybody else’s) profits will certainly shrink at least a little.

  23. Well, it seems Apple is sucking all the profits out of the carriers that went for it. That has interesting implications for the future. Like, what happens when the carriers decide they’re done being conned?

    Given that two of the major carriers pushed hard to GET the iPhone your assumption is that

    a) they’re getting conned
    b) they’re so stupid as to not realize it

    Where the other option is that the iPhone was such a huge competitive advantage that Verizon had to get it even after investing $100M+ in its Droid ad campaigns. Which means they aren’t getting conned and they aren’t revolting any time soon. They will make money later in the contract, the data plans bring in more revenue and profit and they are increasing prices to offset the subsidies.

    But no, Apple is simply conning the stupid carriers and is ripe for revolt.

    What’s funny is that your analysis from a year ago was wrong on many points. Take for example this assertion:

    Third: Google’s strategic direction looks smarter every day. They’ve put themselves in a position where they profit from the buildout without having to eat negative returns. Same goes for Apple, but Google’s advertising profits will probably scale better with network usage volume than Apple’s hardware sales.

    Evidently not so much given how well Apple’s profits scaled this year.

    The big one is why the carriers have fallen into Android’s honey trap. They’ve traded cost-cutting in present time (by the amount of NRE they no longer have to use on in-house software development) for future loss of control over their customer base.

    And this…which is the exact opposite of what happened in 2011. Carriers didn’t emphasize cost-cutting at all but did the opposite by agreeing to a high Apple subsidy despite having a distinct Android handset advantage over AT&T. Fear that they’d lose of control of their customer base to AT&T was one oft repeated reasons.

    So the whole analysis on the important parts from a year ago is suspect…and crowing about the lack of skinning doesn’t obscure that. Especially since the abandonment of skinning might easily be because of the realization that investing in the iPhone was a much better idea and betting on Android alone, skin or no skin that was a losing strategy.

    The bottom line is that Apple’s current performance isn’t sustainable. The losses the carriers are presently eating on the iPhone are going to get squeezed out one way or another, almost certainly re-manifesting as significantly higher unit prices to the consumer. This, of course, will increase Android’s competitive advantage.

    The losses the carriers are presently eating will be recouped by higher contract prices and tiered data services, not from higher unit prices on handsets. Android has no consumer price advantage and in fact subsidizes Apple by helping recoup carrier expenses. That makes carriers happy but not so much Android device makers.

    Apple’s current performance isn’t infinitely sustainable but it’s not ending in 2012 either.

  24. You may be right but there’s no need for an illegal conclave for it to happen. The normal way for industries to operate is for everyone to follow the market leader. For example, Coke publicly announces it will decrease its can size by .5 oz. Then over the next week, other companies one by one announce they’ll follow suit.

  25. Is there a breakout for how the iphone subsidy compares to similar smartphones? The article only compares iPhone to non-iPhone customers, such as when it writes:

    The cost of adding an iPhone customer is about 40% higher than the cost for the average non-iPhone customer, according to Sprint.

    Given that Android is on models all over the price and feature spectrum, it seems unfair to compare iPhone subsidies to even low spec Android models.

  26. Yes, subsidy. Each of those are subsidized in the $300-400 range (mostly $350). (The crappy HTC WP7 Trophy is cheating a little on my part: Verizon is also discounting $150 on top of the $250 subsidy probably to clear inventory that isn’t moving; “subsidy” or “discount” — same cost to the carrier, same revenue for the manufacturer.)

    I don’t know what you’ve read, but your reading is close on “average” — not any and all Android devices. Android devices occupy the full range of 0-$400+ subsidies (I’ve seen some estimates for the SGS2 north of $450); there is usually 1-3 models from each leading manufacturer per upgrade cycle (that I’d guess is in the 8-10 month range) that are receiving a $300-400 subsidy, and each of those manufacturers usually has several other lesser models (with HTC and RIMM having the fewest models/highest subsidies next to Apple). And, as with the Trophy, many models that receive low to no subsidy also often lead to higher rates of “discounts” to clear inventory. Most “true” smartphones are averaging about $250 subsidy.

  27. It becomes most relevant if the carriers can upsell to a higher bandwith tier and the simple act of buying an iPhone (vs. Android) is what turns a customer into a bandwith-consuming fiend looking for the next fix.

    The question is whether this happens often enough to make up for the customers who don’t pay a lot for data, and who would have been happy with Android at a $150 subsidy, but accepted an iPhone at a $400 subsidy because it was a great deal.

  28. Again, citation needed. The price list you show doesn’t show the cost to the carriers. Apple squeezes the carriers coming and going. Whatever the price Apple charges in the Apple store is the most that a carrier can get away with for an unsubsidized phone, but that’s zero margin.

    You can’t show a price list that says you can buy phone ‘x’ from a carrier outright for one price or with a contract for another price, and claim the difference is a subsidy.

  29. You may not care about profit but LG, HTC, Sony and all the android device makers being pummeled by Samsung do. For them Android is not a competitive advantage. If one of them doesn’t do an amazon like closed fork I’d surprised. Sony is best positioned for it but not executing all that well.

    As for whether Google is dumping billions into Android I guess that depends on if you think investing 12.5B on Motorola is a good idea or not. My impression was that if anyone got played in the 2011 smartphone arena it was Google by Moto and not the carriers by Apple. Jha’s threat to turn Moto’s patent portfolio against other Android device makers and collect licensing fees was masterful. The dude went all in on a moderately good hand and made Google fold.

  30. So the real question is which way it will go. Will it follow the “ESR path”:

    “The losses the carriers are presently eating on the iPhone are going to get squeezed out one way or another, almost certainly re-manifesting as significantly higher unit prices to the consumer. This, of course, will increase Android’s competitive advantage.”

    …which would benefit Android, or the “Patrick path”:

    “My guess is that the carriers are stuck…the other carriers are going to have to start raising data rates, and whatever else is in the contract, you can bet your bottom dollar they aren’t allowed to charge iPhone users extra.”

    …which would be really bad for Android, seeing as Android customers would then be directly subsidizing iPhone users. Android can’t directly compete with iPhone if the lower price of the Android hardware can’t somehow be passed along to Android purchasers. This would feel like a major market failure.

    OTOH, I agree with Ignatius:
    “What we *want* is for the carriers to stop subsidizing phones, stop adding crap to the phones, and just sit back and be a dumb pipe like we want them to.”

    Arguments that the carriers cannot make money that way would seem to be wrong, since there is nothing in the above statement that sets a maximum price that can be charged for the dumb pipe. There should be an equilibrium point somewhere that lets the carriers maximize revenue; if they are showing losses, their prices are too high or their (unnecessary) expenses are too high.

    Of course, it’s possible to get into a game-theory trap where the first to raise prices loses, so no one raises prices and everyone loses. See “airline industry”. There must be some way to break out of this, but I don’t know how. The difference here is that the airline industry has excess capacity, while the carriers are still short of capacity (hence the recently-dropping data/month quotas, like Virgin Mobile’s drop from 5 GB to 2.5 GB).

  31. When I click through that first link (and give a zip code) it tells me that in my area, the Motorola DROID RAZR MAXX from Verizon (android 2.3.5) is $650 unsubsidized or $300 subsidized, implying a subsidy of $350. The second link tells me the Samsung Galaxy Nexus from Verizon (android 4.0) has the exact same pricing ($650-$300=$350 subsidy). The third link tells me the Blackberry Curve 9370 is $410 unsubsidized or $100 with a 2-year contract, so the subsidy there is $310. The third link says the HTC Trophy (Windows phone) has a $400 subsidy ($430 unsubsidized, $30 subsidized).

    How do those not constitute “subsidies in the $300-$400 range?”

  32. Actually, given the rapid rate at which new android models are pushed out I wonder if the carriers are really subsidizing to the degree you think or the handset makers are taking it in the shorts.

    I don’t think discount and subsidy counts the same to the carrier.

    Given pretty much only Samsung is making a ton of profit I have to think that the carriers aren’t the ones being shafted with excess inventory they have to subsidize hugely to move. Instead they have a deal where after a certain date they discount the price and it comes out of the handset maker’s pockets.

    This isn’t much different than Best Buy being able to send unsold Touchpads back to HP if they wanted.

    For the iPhone the carrier is taking little risk that they’ll move. Even Sprint with their bulk buy.

  33. The best starting point for a manufacturer comparison begins at Average Sales Price data.

    (The most useful source may be Horace Dedieu’s compiled data as presented using Google’s Motion Charts. Remember that this is all mobile phones, not just smartphones. Select the vendor chart tab at the bottom if it’s displaying the table of data. Select the line chart tab in the upper right, and select “ASP” for the verticle axis. Of course, there are other good data sets and visualizations there too:

    http://www.asymco.com/hire-me/vendor-data/)

    You’ll then have to do your own analysis from two perspectives: (1) analyze each vendor’s product line, pricing of each model, likely sales share per each model, margin per model, etc. They do not have to provide much transparency here; there isn’t much penetrating, accurate analysis outside of Apple. Maybe RIM the most. The rest are Asian manufacturers who are very vague and have a diverse product mix or depressing stories like Motorola that don’t inspire much analysis. and (2) look at the carrier’s reported figures across their product lines, pricing of each model, subsidized and unsubsidized, likely sales share per model, margin per model on a device and subscription basis, and other value factors for the carriers — subscriber additions, churn, ARPU on a model basis, etc… Again, they don’t need to be very transparent here.

    There are thousands of little hints here and there. But not much solid reporting. We now even have to consider new factors like direct platform to manufacturer subsidies such as Microsoft to Nokia (but I do not believe HTC and other partners, or at least not as much) and likely Google to its own MMI unit in the future. I don’t think my admittedly rough numbers in earlier comments are that unreasonable.

  34. To put that another way: nobody but the carrier cares if you get a 2-year contract, so nobody but the carrier is going to pay you $300-$400 to lock yourself into that contract. Microsoft and Apple and Google surely want you to buy their phone and do stuff that drives costs down, but that should to a first approximation affect both the subsidized *and* the unsubsidized price about the same. If you (Patrick) think somebody else besides the carrier is helping to pay customers $350 to buy the latest droid specifically on a contract but not helping pay for the same phone without one: Who would do that? And why? And what’s your evidence for this?

  35. When I click through that first link (and give a zip code) it tells me that in my area, the Motorola DROID RAZR MAXX from Verizon (android 2.3.5) is $650 unsubsidized or $300 subsidized, implying a subsidy of $350.

    When people say the iPhone has a $400 subsidy, they mean that the carrier collects the $200 or $300 from the customer, adds $400 to it, and gives it to Apple. (Apple fanboys are always crowing about Apple’s ASP; this is why it’s high.)

    Apple sets the price on unsubsidized iPhones by what it charges in their stores. Carriers can hardly charge more, but it’s my understanding they don’t get much per handset, if anything. It’s a privilege to be able to sell an Apple phone, you see.

    By contrast, I expect that carriers make a small profit on most unsubsidized Android phones they sell. I can’t prove it, but black market and third party prices seem a lot lower than the carriers, and you know that when somebody’s selling an unlocked phone with a warranty, they’re not getting a carrier subsidy.

    Also, with the current warped economics, the carriers don’t really sell that many unsubsidized phones, so they have a certain incentive to sell unsubsidized phones at as high a price as possible, to make the subsidized ones look like a really great deal. After all, if an unsubsidized iPhone is $650 and a contract iPhone is $200, and an unsubsidized Android phone is $300, the discriminating customer might expect that Android phone for free on contract.

  36. “…who would have been happy with Android at a $150 subsidy”

    [Citation needed.]

  37. Google’s net price for Mot is a lot less than $12.5 billion and it remains to be seen how much of that is “down the hole” since they haven’t even spent it yet.

  38. “What we *want* is for the carriers to stop subsidizing phones, stop adding crap to the phones, and just sit back and be a dumb pipe like we want them to.”

    *I* want the carriers to subsidize phones (because I’m going to have a mid to high-range data plan anyway — I want a $500-800 super device in my hand, not a $100 piece of crap), stop adding crap to the phones (accomplished via Apple), and just sit back and be a dumb pipe… (closest with Apple).

  39. What do you want a citation for? That Android phone subsidies can be a lot smaller than Apple subsidies? Here, try this:

    “Cusick estimates that the average iPhone subsidy is $350 at Verizon, $375 at AT&T and close to $400 at Sprint. This compares to about $150-250 on similar Google (GOOG) Android devices.”

    http://tech.fortune.cnn.com/2011/12/04/j-p-morgan-verizon-is-getting-its-iphone-reward-this-quarter/?source=yahoo_quote

    BTW, I’m still waiting on your citation…

  40. “Carriers can hardly charge more, but it’s my understanding they don’t get much per handset, if anything.”

    [Citation needed.]

    “By contrast, I expect that carriers make a small profit on most unsubsidized Android phones they sell. I can’t prove it, but black market and third party prices seem a lot lower than the carriers, and you know that when somebody’s selling an unlocked phone with a warranty, they’re not getting a carrier subsidy.”

    [Citation needed.]

    “Also, with the current warped economics, the carriers don’t really sell that many unsubsidized phones, so they have a certain incentive to sell unsubsidized phones at as high a price as possible, to make the subsidized ones look like a really great deal.”

    [Citation needed.]

    All analysis, data, and statements I’ve ever seen show a very direct correlation between carrier subsidy and on-/off-contract device price.

  41. Look, dickwad, I posted a citation (actually a couple of posts ago, but I’ve reposted it in multiple comments on this post, now, as well) claiming that iPhone subsidies are higher than Android. Here it is one more time:

    http://tech.fortune.cnn.com/2011/12/04/j-p-morgan-verizon-is-getting-its-iphone-reward-this-quarter/?source=yahoo_quote

    Without any such citation, you made an assertion to the contrary. I asked you for a citation to back up this supposed fact, and instead, you keep asking me for citations to back up my opinions.

    I’m entitled to my opinions, and you’re entitled to your own little reality, but don’t think this behavior will help you foist it off on the rest of us.

  42. And instead of providing said citation, I noticed you just doubled down:

    All analysis, data, and statements I’ve ever seen show a very direct correlation between carrier subsidy and on-/off-contract device price.

    This is fine. I love new data points. But you gotta show your homework. I’d actually be thrilled to see this, because then it would indicate that everybody posting about how the carriers are fucking up and giving all their profits to Apple for the next two years are simply wrong.

  43. I disagree with Cusick’s assessment of “comparable” based on the data to me.

    If you somehow want to pretend that off-contract pricing is not reflective of subsidy pricing, or that Samsung is charging carriers less than $450 for SGS2s, that’s fine by me.

    And, no, I don’t really want a citation. I find that a petulant, useless copout that’s not useful to the conversation.

  44. “…which would be really bad for Android, seeing as Android customers would then be directly subsidizing iPhone users. Android can’t directly compete with iPhone if the lower price of the Android hardware can’t somehow be passed along to Android purchasers. This would feel like a major market failure.”

    This has been the case for a while. And the carriers are platform agnostic up to the point that one causes them to gain, keep, or lose users so they do not view it as one subsidizing another; they view it as present revenue to fund necessary and inevitable investment in infrastructure or conversion of their pricing plans to the new data demands.

    Apple has achieved price parity with comparable competition (while still only occupying a narrow, high-end product band) by passing on their premium to an intermediary consumer, the carriers, and both Apple customers are happy, insofar as the carriers are either gaining or not losing higher range ARPU subscribers.

  45. You don’t seem to understand that I find the notion that when all data points to on-/off-contract pricing directly correlating to carrier subsidies, direct knowledge that Apple’s carrier subsidies do correlate directly to its on-/off-contract pricing, manufacturing ASP data indicating far higher ASP of leading Android phones than you claim is plausible… all of this makes your statements just as unsupported. I find it incomprehensible that you can look at ASP data and say that comparable Android devices aren’t getting subsidized by $300 or more. It seems like there are virtually zero true smartphones subsidized less than $200.

  46. Don’t forget that wireless carriers, even in this modern age of fancy phones, are still American telcos, so they only ever improve their ROI by screwing the customer down ever-tighter. They never do anything that is more about making the customer happy than about locking her up in a box. Sure they’ll sell Androids if Apple forces them to (that whole episode was far more about Apple pimp-slappin’ its hoes to make ‘em think twice than about telco execs falling in love with Google; even so the telcos are smarter than Hollywood so they still own the customer relationship), and they’ll hold their noses at the faint whiff of “freedom” they catch now and again. But if Apple is running a con they’re only too happy to sign up: each of the three likes getting in a fight that might kill somebody, because after all it might not be me! (OK maybe Sprint is a bit delusional here but her top execs know they’ll get a soft landing come takeover time.) Duopoly is much better for the long term than triopoly, and every child of Ma Bell thinks mostly about the long term. Here we are coming up on thirty years since the “divestiture” and they still haven’t wrapped their little game yet, although they have convinced most customers that the Princess Phone is worth a two-year contract and they have unloaded all those crappy rural ILECs (see “FairPoint Communications”, “Hawaiian Telcom”, etc.). Maybe that’s “market failure” or maybe a superior competitor could make more money by giving the customer what she wants, but that shit doesn’t matter. This is what they are, and this is what they do.

    Short of ending the FCC’s tyranny over radio transmission (which option I wholeheartedly support), there is no hero coming to save us. You have capitalism because nothing better is on offer.

    So, we’ve seen it with the music labels, Hollywood, and the telcos. If you want to know to which industry Apple is likely next to lay waste, think about other stodgy old boys’ clubs with counter-intuitive business models and a bilious hatred for their customers. I would say the airlines but I can’t figure out how a tiny electronic gizmo would dominate that business.

  47. @Cathy:
    “…if they are showing losses, their prices are too high…”

    Of course, that should have said priced too low.

  48. And I find a list of subsidized vs. unsubsidized prices a petulant, apparently deliberately confusing cop-out to a real question about why you believe what you hold to be a fact. Hence [Citation still needed]

    If you somehow want to pretend that off-contract pricing is not reflective of subsidy pricing, or that Samsung is charging carriers less than $450 for SGS2s, that’s fine by me.

    You haven’t showed anything that indicates that off-contract price – contract price == subsidy price, so I’m not sure who’s “pretending” here, and the question is not how much Samsung charges carriers for the Galaxy S2, it’s how much Samsung charges carriers, MINUS the amount the carrier gets from the customer.

  49. “…and the question is not how much Samsung charges carriers for the Galaxy S2, it’s how much Samsung charges carriers, MINUS the amount the carrier gets from the customer.”

    Jesus Christ, do you have to have a citation for what people are paying for SGS2s too?! I didn’t know that was a [Citation needed.]-worthy.

  50. A few minutes ago you said “Everything I’ve read says that Android subsidies are typically $150 and iPhone subsidies are typically $400.”

    Now you link to an article which claims the typical iPhone subsidy is $350 or $375 at Verizon or AT&T (it’s only $400 at sprint), but adds “This compares to about $150-$250 on similar Google (GOOG) Android devices.”. So: you compared the high end of the range given for Apple with the low end of the range given for Android. Whereas doing the reverse (which seems indicated if one is comparing the high end most-comparable-to-apple Androids to the most-commonly-sold iPhones) would have shown, indeed, a difference of $100.

    On reflection I think I know where some of the funny numbers are coming from. If Droid has a list price of $650, there’s some actual wholesale price the carriers can buy them at – let’s call it $550. If you buy an unsubsidized phone from Verizon, they buy the phone for $550 and sell it to you for $650, making an immediate $100 in profit. If you buy a phone with a two-year plan, they again buy the phone for $550 but sell it to you for $300, which equals an immediate short-term loss of $250. The difference between the price discount and their loss amount – two different measurements of the “subsidy” – is that one includes their normal profit margin on the retail sale and the other does not. For accounting purposes it’s conceivable they count that phone as “worth” $650 when they sell it (since that’s the retail value – what they could have sold it to somebody else for without a contract), in which case the two numbers would be the same, but that’s probably not the way it’s being accounted for here.

  51. Given that two of the major carriers pushed hard to GET the iPhone your assumption is that

    a) they’re getting conned
    b) they’re so stupid as to not realize it

    I don’t think that necessarily follows. Lots of people and businesses do dumb things all the time, that they know are dumb at the time, because they know they’ll be punished if they don’t do them. Apple has apparently very cleverly negotiated a really good deal for itself, and the negotiators on the other side — do they have better job security when Apple brings in more users at a high cost, or would they have had better job security when they explained why they decided not to bring in Apple?

    The thing is, just like every lost T-Mobile customer will be blamed on Apple, every AT&T or Verizon new customer or customer not leaving will be credited to Apple. (At least Sprint had the balls to print their numbers showing percentage of iPhones going to new customers.)

  52. Now you link to an article…

    WHich I read when it first came out and misremembered. But the basics are right — more subsidy for Apple than Android.

    On reflection I think I know where some of the funny numbers are coming from. If Droid has a list price of $650, there’s some actual wholesale price the carriers can buy them at

    Which is exactly what I was trying to explain, apparently not very well.

    For accounting purposes it’s conceivable they count that phone as “worth” $650 when they sell it…

    You have to keep your eye on the ball. If you look at carrier acquisition costs (which I haven’t done this quarter, but you can find it in their quarterlies) you can essentially see how much the subsidies cost them. The supposed retail price has nothing to do with how much it costs to acquire a customer.

    The retail price only matters if they sell a phone at retail, and I don’t think that happens nearly as often (on postpaid) as the subsidies. I was fairly sure of this when I calculated this out for AT&T a couple of quarters ago, but a really quick sanity check will show you that (for postpaid services) AT&T and Verizon don’t have plan discounts for when you’re bringing your own device, unless the situation changed while I wasn’t looking. The thing is, they’re so keen to have data be worth ‘x’ that they will do anything (including give extra money to Apple) to not have to reduce ‘x’.

  53. No, the question is how much the carrier pays Samsung for the handset, duh.

  54. The update (about the 3%) was at the bottom of the original article, but thanks for reinforcing that Apple manages to squeeze retail margins. Points deducted for not noticing that the same thing that lets them do that (the Apple store) also operates in the phone space.

  55. So all the analysts who think Android subsidies are lower are wrong? So Verizon is lying when they said that iPhone subsidies impacted their bottom line? Why did they never say this about Droid?

  56. I think what will happen is that we’ll see an uptick in prepaid and in MNVOs. Sprint is playing heavily in both these arenas, and the other carriers are dabbling there, too. I sincerely doubt that Apple’s contract allows them any control over prepaid pricing…

  57. > the simple act of buying an iPhone (vs. Android) is what turns a customer into a bandwith-consuming fiend looking for the next fix.

    Inadvertently, possibly, and *bandwidth* as such may not be (all of) the issue.

    I heard from one of the network guys at the carrier where I work that iOS sets up and tears down a lot of small session connections for these little activities, whereas Android tries to keep connections open and multiplexes activities from many apps through one session on the network. Which doesn’t necessarily mean more or less aggregate data volume, but there are network costs associated with the setup and administration of many small sessions that aren’t correlated with the volume of data used in a session.

    (This was a couple of years ago and may have changed in newer versions of iOS)

  58. “and the question is not how much Samsung charges carriers for the Galaxy S2…”

    “No, the question is how much the carrier pays Samsung for the handset, duh.”

    Huh?

  59. Yes. Does that seem so shocking that analysts aren’t doing a good job at understanding this market. That they have a distorted view of “comparable” — how many tech reviews claim a low-end device is comparable and then it ultimately proves to be a dog in the market and not comparable?

    iPhone subsidies were a drag because all iPhones have a high-end subsidy that is 25-30% higher than comparable high-end Android devices. A lot of Android devices are not at the high-end.

    You seem really confused: I have said over and over again there is an iPhone premium in carrier subsidies. I have argued that it is only 25-30% higher for comparable devices, not 267% higher. In aggregate, yes, it is likely to be 200-235% higher. That is where we disagree.

  60. Points deducted for citing an article with the ludicrous and erroneous theory that Apple was accounting for a $500 versus $800 price disparity by recouping 50% retail margins in its own stores versus other retail partners rather than Apple’s massive scale and Motorola’s complete incompetence and thinking that this idiotic theory has anything to do with carrier subsidies.

  61. For example, the recent Sprint quarterly that prompted the most recent gut-wrenching discussion about whether the subsidies are worth the net adds / minimized losses to Sprint:

    http://newsroom.sprint.com/article_display.cfm?article_id=2179

    Note that for the “Sprint Platform” (the CDMA, not the dying Nextel iden), they added 539 K postpaid customers but added 899K prepaid customers and 954K affiliate/wholesale customers.

    They’re not going to be as ecstatic about any of those customers, because the ARPU for prepaid customers is a lot smaller than for postpaid, but if 1.8 million new customers provide ARPU of $25, that’s actually a lot more money than 540K new customers with an ARPU of $60, presumably at a much smaller acquisition cost.

  62. You certainly didn’t present that argument very well, especially at the start, when you just handed me a price list, and seemed to disagree that iPhone was subsidized more than Android.

    Now even you admit that it’s subsidized more even for “comparable” phones, so you’re basically agreeing with my argument.

  63. The only thing it has to do with carrier subsidies is to show that Apple’s retail stores let them force retailer margins to be much smaller than industry standard. You’ve been one of the ones crowing about Apple’s handset ASPs — doesn’t this data point jibe with that?

  64. Hmmm. I bought my iPhone 4s from an on line store, with no carrier involved. Then I just walked to the carriers’ own shop and had my SIM card exchanged for a microsim.

    Here the carriers will sell you a phone without a contract. So will about every brick and mortar or on line electronics store. If you commit to a contract you get a subsidy on your phone purchase. Phone and contract are effectively unbundled, but bundling is still an option.
    So since consumers have the choice of not committing to a carrier when buying a phone, even if they want a particular phone one could assume the subsidy carriers give when you do commit reflects the value of that commitment to the carrier.
    Incidentally the maximum subsidy given on a Samsung Galaxy IIS purchase by Swisscom is 650, CHF, on an iPhone it’s 500CHF… So this carrier actually appears to have a higher margin on the iPhone than on competing Android phones…

  65. In the Netherlands, I can buy a Motorola RazR (unspecified model) for 475 euro mail order, no SIM, no plan. That is tax (19% VAT) included.

    That would be $530 without tax (assuming they use real exchange rates).

    This is the site:
    http://www.wehkamp.nl/computer-telecom/mobiele-telefoons/simlockvrije-telefoons/motorola-razr-mobiele-telefoon/C31_6F7_F75_744807/?BC=GOA&BAC=CET&cm_mmc_o=7BBTkwCjC2Ha C 4B_bwkw fwkwuBBEl C mbFkBgLMybXw fwkwuBBEl C (a-K)CjCmbFkBgLMybXw fwkwuBBEl C 4BfByBkz qzOq C oxISSoiI (a-K)CjCPyBzpj%2b4BfByBkz %2bqzOq&cm_guid=1-_-100000000000004628814-_-11495522584&cm_mmca1=12921-556F38A8-52D1-4A70-AFC4-8882C2F8C921S11495522584

  66. For this discussion, it would be nice to know whether Apple realizes more profit in the USA with their captured phone market than in other places where customers are used to buy phone and data plan separate?

  67. If they know it’s stupid at the time it’s not a con, now is it?

    And if multiply carriers knowingly choose the “stupid” option then is it stupid? Or another business decision made rationally and shows the value of the iPhone vs Android?

  68. “Serif font”? In FF, it’s a large, widely spaced sans-serif, which is one reason I have been turning off the page style, to get a more readable font, and better density (less whitespace).

  69. A while back we were talking about carrier earnings reports. You argued that you couldn’t trust carriers to tell the clear truth; they’d slant reports to make themselves look good.

    That article quotes an anonymous carrier customer service rep. You’re willing to trust that.

    I am sincerely not trying to be snarky here, Patrick: think about your own confirmation bias.

  70. By the way, this also makes even the most deeply threaded comments quite readable.

  71. Moreover, it’s mentioned in T-Mobile’s most-recent 10Q

    Discipline on the cost side contributed to year-on- year margin improvement, while postpay churn, in particular related to the iPhone 4S launches by competitors, will continue to be an area of concern.”

    and

    The quarter-over-quarter improvement in net customer additions was driven by improvements in both contract and prepaid gross additions resulting from the introduction of unlimited Value plans discussed above and growth of prepaid unlimited Monthly 4G plans. This growth may be impacted in the fourth quarter of 2011 due to competitor launches of the iPhone 4S.

    T-Mobile has *never* mentioned the iPhone before.

    Source: http://www.t-mobile.com/Company/pdf/English/TMUS_Q3_2011_Press_Release.pdf

  72. T-Mobile will use that money (and spectrum, let’s not forget the spectrum) to build out LTE, so they can get on the iPhone 5.

  73. Even when android phones are heavily subsidized, it doesn’t last long. Android phones tend to drop in price like a rock

  74. I still haven’t seen any proof that the iPhone is leading the carriers to have negative earnings, or 0% profits. “…no profit remaining for them on that iPhone customer.” Hah.

    Margins shrinking? Sure. Transfer of power in a market the carriers once ruthlessly held. Yep.

    Thinking that all the carriers woes are due to the iPhone, silly at best.

  75. No, they do not. US sales are increasingly decreasing as international market entries increase. This quarter was an anomaly of sorts because US availability was much higher as rollouts were added outside of the U.S. This increasing international % with occasional re-upticks in the US because of new releases has had little impact on Apple’s ASP.

  76. Reminds me of the Japanese saying:

    Those who eat Fugu are stupid.
    Those who do not eat Fugu are also stupid.

    Even though Verizon and AT&T already had the iPhone, there were a lot of unknowns for T-Mobile going into the deal. How many new customers would it attract? How much would cost on the existing customer side? How much more ARPU would it drive? How much more investment in congested areas to support the new phones?

    They could have good guesses for all these. They might even have been accurate on the side of what happened. They will never know exactly how it turned out if they took the other path. Maybe it’s all just buyer’s remorse, shared very publicly so they don’t get beat up too badly.

  77. It’s another data point. I haven’t made up my mind whether to believe it or not. Usually if I have made up my mind, I say so.

  78. It says that Eric spent a lot of time, two threads ago, telling me he didn’t have to address my arguments because he’s categorized me as an Apple partisan. But I shouldn’t stick you in the same bucket.

  79. Unless you know what the wholesale price of the phone is to the carrier, you don’t know what the subsidy is. This is because you don’t know what margin the carrier achieves for the unsubsidized phone.

  80. Point of order: the carriers aren’t sharing complaints about Apple publicly. This thread is predicated on a bunch of analysts deciding that the decline in EBITDA is due to Apple.

    Engadget has the cost of the LTE buildout, for Sprint, estimated at 4 to 5 billion over five years. Peg that at a billion a year for the sake of the back of the envelope. Given the lower than expected rate of LTE adoption, it’s a possible factor.

    And hey — Sprint’s CEO tags those costs as significant:

    “Hesse said that investors would have to wait for 2014 to see a meaningful acceleration in profit margins. Sprint expects to have completed its network upgrade by then. By the fourth quarter of 2014 he promised an increase of 12 to 16 percentage points from the fourth quarter 2011 number.

    “‘You have to get through (the upgrade) to see the margin expansion,’ he told Reuters. ‘That’s just how long it takes to completely revamp your network.'”

    Maybe he doesn’t want to piss Apple off, who knows?

    And maybe the timing’s wrong for network build out to be part of this. But I’d be more interested in the original link if it at least addressed possible counter-cases.

  81. But we don’t know what subsidies are in other countries — we can only look at overall profit percentage. I don’t think we have an answer to Winter’s question, unfortunately. Wish we did, it’s a good one.

  82. Google/Motorola speculation thread:

    What is google going to do with mot? (note: starter ideas not necessarily mutually exclusive)

    Leave it alone?
    Sell off parts?
    Sell/trade/swap strategic patents to Android partners?
    Partner with MNVOs?
    Become an MNVO?
    Buy T Mobile?
    Add features that help the users break the back of the telcos?
    – Multiple SIM cards (and possibly IMEI numbers) to cherry pick different plans?
    – Embedded always available hotspot not under carrier control?
    – Seamless WiFi VOIP/carrier voice/carrier VOIP switching with Google Voice?
    – Any-carrier software defined radio?
    – Unlocked boot loader, pre-rooting, easy recovery to factory defaults
    Add features that help the telcos break the back of the users?
    – Make impossible to root
    – No WiFi, except maybe through AT&T’s public network
    Define the next gen flagship Android handset?
    Define the next gen throwaway Android handset?
    Open Android further and encourage serious hacking on Mot handsets?
    Close Android further and support other Android OEMs a lot less?
    Create open-source hardware designs that other manufacturers can build to get component costs down?
    Switch to whatever the latest incarnation of Moblin/Maemo/Meego/Tizen is?
    Close up shop and become a pure play patent troll?

  83. The heavy cost of subsidizing the Apple iPhone and its other smartphones also pressured margins, CFO Fran Shammo said in a conference call with investors. Shammo warned earlier this month that strong sales of the iPhone would result in a 500 to 600 basis point decrease in wireless margins.

    “pressured margins” and “warned earlier” sure sounds to me like at least Verizon expects some investors to react unhappily to the results of Verizon carrying Apple products.

    Also, I don’t think you understand EBITDA — the capital expenditure for the LTE rollout will not be included in EBITDA.

  84. No, I don’t see what has been unclear to you at all. My first statement was every manufacturer has 1 or more models with a subsidy int he $300-400 range. What are you so confused by? What have you been arguing with me about?

  85. You’re oversimplifying. The capital expenditure won’t be included, but cost of labor, R&D, and so forth are. Hm, I don’t know if cost of spectrum gets depreciated or not, but I assume it does.

  86. BTW, here’s what Warren Buffett has to say about EBITDA:

    “It amazes me how widespread the use of EBITDA has become. People try to dress up financial statements with it.”

    “We won’t buy into companies where someone’s talking about EBITDA. If you look at all companies, and split them into companies that use EBITDA as a metric and those that don’t, I suspect you’ll find a lot more fraud in the former group. Look at companies like Wal-Mart, GE and Microsoft — they’ll never use EBITDA in their annual report.”

    “People who use EBITDA are either trying to con you or they’re conning themselves. Telecoms, for example, spend every dime that’s coming in. Interest and taxes are real costs.”

    So, EBITDA is designed to hide a lot of negative stuff. Unfortunately, as a metric, it doesn’t hide the phone subsidy. Perhaps the telcos need to start renting handsets instead of selling them, to make EBITDA go back up?

  87. Oh god yes. I spent 10 years in Silicon Valley riding the bubble up and then down. Afterwards I was very interested in how people were shading financial statements.

  88. Let me get this straight:

    You said that the LTE rollout was “x” and that was obviously weighing on EBITDA, and I specifically said that LTE capex wouldn’t be included in EBITDA (without saying that all LTE expenses were EBITDA), and I’m the one who’s oversimplifying?

  89. I would argue that the stability of the ASP does answer the question: if subsidies in other countries are notably more or less than in the US, it would be indicated in a changing ASP as the domestic/international mix shifts.

  90. Given that Android is on models all over the price and feature spectrum, it seems unfair to compare iPhone subsidies to even low spec Android models.

    The entire point of the subsidy is to provide the telco with a benefit. Not the customer. One would think the telco is happiest when the customer is paying for the most expensive plan and using the least amount of data, and not generating any support calls or trouble tickets. I can certainly believe that Apple phones (especially with the Apple store) are the best at that latter one, but if a customer is going to go for the expensive plan, it’s worth it to the carrier to give him a hefty subsidy even for a crappy handset.

    Having said that, there is a lot to be said, both from an implementation point of view and from a “don’t confuse the customer” point of view, for not changing the subsidy based on the selected plan.

    Also, if (as I have posited before) the very act of buying an iPhone causes the customer to become a dataholic, then it actually makes sense to give the highest subsidy to an iPhone even on the cheapest plan, because the customer will need to pay more to get his fix later.

  91. @Tim F.:

    No, I don’t see what has been unclear to you at all. My first statement was every manufacturer has 1 or more models with a subsidy int he $300-400 range. What are you so confused by? What have you been arguing with me about?

    I still don’t believe that Android subsidies are that high for high-end handsets in general, although as others point out, they may be for certain handsets when they first come out. In any case, this is somewhat contradicted by the statement I was replying to — if the Apple subsidy is 25-30% higher than comparable Android handsets, then Android handset subsidies might barely eke into the $300 range, but nowhere near $400.

    iPhone subsidies were a drag because all iPhones have a high-end subsidy that is 25-30% higher than comparable high-end Android devices. A lot of Android devices are not at the high-end.

  92. @Tim F.:

    I would argue that the stability of the ASP does answer the question: if subsidies in other countries are notably more or less than in the US, it would be indicated in a changing ASP as the domestic/international mix shifts.

    There aren’t enough variables in that equation. ASP = Subsidy PLUS up-front customer acquisition cost. If up-front customer acquisition costs are higher in other countries (and I’m sure they are in some), then the subsidies are smaller in those countries if ASP is constant.

  93. T-Mobile is doing a limited experiment with Apple-sized high Android subsidies (but you gotta do the sucky mail-in rebate thingy):

    http://www.tmonews.com/2012/02/t-mobile-pushes-out-statement-valentines-day-sale-website-now-live/

    Also note in that release, that they are, on some plans, now splitting out the subsidy — with the “Unlimited Value” plan you can still get the shiny new smartphone, but pay an extra $20/month, so the pricing gets a bit more transparent.

    More info on the value plans here:

    http://explore.t-mobile.com/value-plans

  94. I’m not that sure that ‘subsidy’ it the correct concept here.
    On a sample of one, I bought the nexus S a year and a half ago. I paid full price for it, 520 I think, and put it on my kid’s family plan with t-mobile.

    I could have bought it for 199 on a contract, but a quick look at the numbers said that I would be paying about 500 more over two years for the cheap phone and contract vs my share of the family plan and the full price phone.

    Present value of that 500 would have to be discounted — time value of money, + risk that I wouldn’t pay up for the full contract time + probably some cut for BestBuy, but even so it looks like T-mobile would have been around 400 to the good at the end of the two years if I had gone with the contract.

    I’ll admit that I haven’t looked at what the numbers would be for different phones and plans, but like the rent to own places, for the most part the contracts and subsidized phones would probably be less than optimum for a customer that can put up the money up front.

  95. @Jim Hurlburt:

    > I’m not that sure that ‘subsidy’ it the correct concept here.

    Exactly. Except that, unless you shop around intelligently like you did, you might get hit with the same monthly whether you buy the smartphone outright or not, because the big two carriers really don’t want people to get used to smaller monthly bills.

    BTW, I just noticed in that tmonews article I gave a link to (quoting a press release from T Mobile):

    A recent online Omnibus survey4 with Harris Interactive found that four-in-ten (44%) of first time smartphone buyers in 2012 say the cost of the data plan associated with the smartphone is the most influential reason they have not purchased a smartphone before. The value for customers extends beyond the “Valentine’s Day Sale” as T-Mobile offers the best rate plan pricing on America’s Largest 4G Network.

    In other words, you and me aren’t the only people who would rather pay up front for hardware than buy it from Aaron’s Rent To Own…

  96. “if the Apple subsidy is 25-30% higher than comparable Android handsets, then Android handset subsidies might barely eke into the $300 range, but nowhere near $400.”

    Last I checked, $450 is ~28% greater than $350. I’m unclear on what you think is out of the ballpark.

    “If up-front customer acquisition costs are higher in other countries (and I’m sure they are in some), then the subsidies are smaller in those countries if ASP is constant.”

    I am not following: Winter asked about Apple’s profits. If the theory is that the change in device price is perfectly offset by the carrier subsidy, which is preserving the ASP, then Apple’s profits remain the same (which they are).

    If the theory is that Apple is getting more or less subsidy (in a manner directly related to end user cost (?) but not specifically tied to the American sales price, then we would ultimately see some regional impact on the ASP.

  97. The bottom line is that Apple’s current performance isn’t sustainable.

    The profits from iPods weren’t sustainable, because eventually everyone who wanted one would have one, and people would use iPhones as iPods. And the tremendous lead the iPad has over other tablets is probably not sustainable, as others jump on the tablet bandwagon. But Apple clearly thinks long-term, and has no problem cannibalizing their own sales if it means moving forward. So while Apple’s current performance (in the narrow sense of its relation to the carriers) may not be sustainable, I would bet that this does not indicate a coming collapse, or even a stumble.

    And let’s give Apple credit for disrupting the previous carrier model, which involved a bunch of undistinguished phones filled with crapware and tied to carriers. This sudden concern for the welfare of carriers reminds me of all the geek hatred of Flash that suddenly diminished as soon as it became clear that iOS wouldn’t support it.

  98. @Patrick:
    “Unless you shop around intelligently like you did, you might get hit with the same monthly whether you buy the smartphone outright or not, because the big two carriers really don’t want people to get used to smaller monthly bills.”

    Of course, this means that they are completely losing revenue from people like me, who simply will not sign up for a large monthly fee no matter what service is providing or how cheap the up-front cost was.

    If Boost Mobile or Virgin Mobile didn’t exist with their $25 (now $35) plan, or something similar from T-Mobile, I simply wouldn’t have a smart phone. And if affordable prepaid options go away at some point, I’ll discontinue service and revert to a dumbphone. The service simply isn’t worth $100 per month to me, period, and there is nothing the carriers casn do to change that.

    @T-Mobile press release, quoted by Patrick:
    “A recent online Omnibus survey4 with Harris Interactive found that four-in-ten (44%) of first time smartphone buyers in 2012 say the cost of the data plan associated with the smartphone is the most influential reason they have not purchased a smartphone before.”

    This doesn’t surprise me at all. I think it’s a myth that everyone wants a small up-front cost and don’t care about the monthly fee. If that were the case, no one would buy cars with cash, but the last time I bought a car I was told by the dealer that it’s pretty common.

  99. @PapayaSF

    But Apple clearly thinks long-term, and has no problem cannibalizing their own sales if it means moving forward.

    That’s correct. Apple is self-disruptive, which is why they have been so successful.

    With regard to the argument around whether carriers are profiting from the iPhone, I have to say that I don’t feel like I have a good enough understanding of how the carrier industry works to comment intelligently on this.

    However, I have an observation. The iPhone, in various guises, has been around for almost five years now, and carriers worldwide still seem hell-bent on being able to offer the device.

    So, it seems to me that if they *were* hurting from the iPhone that they have had ample time to drop it. Instead we are seeing the opposite. *More* carriers are offering the iPhone than ever before. Also, here in the UK, every carrier offers the phone, and it has been that way almost since the start, (although O2 had a brief period of exclusivity at the beginning).

    So, my question is, has something changed recently that makes you guys think that the carriers are doing worse now from the iPhone than they were in the past?

  100. @Tim F:

    Last I checked, $450 is ~28% greater than $350. I’m unclear on what you think is out of the ballpark.

    Okaaay, what’s $350, and what’s $450, and where’s your evidence that either of those are true subsidy numbers?

    If the theory is that Apple is getting more or less subsidy (in a manner directly related to end user cost (?) but not specifically tied to the American sales price, then we would ultimately see some regional impact on the ASP.

    No, the subsidy is the extra that the carrier adds to what the customer pays for the phone. You cannot directly relate ASP to subsidy without having that third variable — what is the cost to the consumer of the handset in the different markets.

    @PapayaSF:

    And let’s give Apple credit for disrupting the previous carrier model, which involved a bunch of undistinguished phones filled with crapware and tied to carriers.

    We’ve given them that credit multiple times over the last couple of years. I’m more than happy to reiterate it.

    This sudden concern for the welfare of carriers reminds me of all the geek hatred of Flash that suddenly diminished as soon as it became clear that iOS wouldn’t support it.

    The concern is not for the welfare of the carriers per se; rather that if they are, in fact, subsidizing Apple users and then later making it up (partly) on Android users, that distorts the market significantly. People complain bitterly about the “Microsoft tax” where you buy a computer and throw away the software you didn’t use. If the analysts are right and Android users wind up getting taxed monthly to support the carriers giving extra money to Apple, it’s yet more evidence that Apple has learned entirely too well from the master.

  101. @Tom:

    So, my question is, has something changed recently that makes you guys think that the carriers are doing worse now from the iPhone than they were in the past?

    Sure. Having the iPhone is a competitive advantage, until it’s not (e.g. when all the carriers have it). Now, all the carriers have a competitive advantage over TMobile…

    But seriously, it might be worth an extra subsidy for the reduced support calls and higher ARPU. Hard to know from the publicly available data, but some of the analysts who seem to think the iPhone is a problem for Sprint (for example) probably have access to more in-depth data.

  102. We know what the carriers think they’re buying – increased profits through increased market share. The problem with this theory is that it no longer makes any sense at all.

    Hey, isn’t “increased profits through increased market share” the Android strategy? How’s that working so far?

    Three of the four major carriers (AT&T, Verizon, Sprint) now carry the Apple product. How can the iPhone be a winning differentiator when almost all of your competitors have it too?

    The iPhone has moved from a differentiator to a business necessity. A wireless carrier who doesn’t have one is like a 7-11 without cold beer.

  103. @Patrick

    Sure. Having the iPhone is a competitive advantage, until it’s not (e.g. when all the carriers have it). Now, all the carriers have a competitive advantage over TMobile…

    That argument might sound quasi-plausible in the US, but if you look at markets like the UK, where every carrier (or ‘network’ as we call them here) has had the iPhone for years, it doesn’t stand up.

    As PapayaSF says, the iPhone is a necessity for any carrier. Yes, it is not a differentiator any more, but that doesn’t mean that it isn’t making them money.

    There are more than 200 carriers worldwide offering the iPhone. Are they all just stupid fools being conned by Apple?

    analysts who seem to think the iPhone is a problem for Sprint (for example) probably have access to more in-depth data

    If the argument is based on what ‘analysts’ are saying about undisclosed data I am very far from convinced.

  104. This is largely an American problem; in many places outside the Land of the Free with Two-Year Contract, the phone market is largely decoupled from the service market and the iPhone does quite well there.

    If anything, the iPhone sucking profits out of the carriers will force the service and phone markets to decouple here; and the carriers will become more like dumb bit-haulers. America may finally join the 21st century in terms of cell network economics.

    And you can add that to the list of things Apple revolutionized.

  105. @Tom:

    There are more than 200 carriers worldwide offering the iPhone. Are they all just stupid fools being conned by Apple?

    The question is not whether a carrier offers the iPhone. The question is whether the carrier (a) subsidizes (subsidises, for you) the iPhone handset to a much greater extent than other handsets, and (b) hides the subsidy by artificially keeping service rates high for all handsets — the same for iPhone and others, and not raising them when you buy the handset or, more importantly, lowering them after the contract period is over.

    Most of the rest of the world is different from the US for several reasons:

    – Lower carrier phone subsidies in most countries
    – Multiple competing compatible GSM networks mean that there has been a healthy market for unlocked phones for a long time
    – Minimal iPhone penetration

    On that last one, note that last quarter, Apple’s best iPhone quarter ever, AT&T sold 20% of their iPhone output. And Verizon and Sprint together sold another 15% of Apple’s iPhone output. So, as Jeff Read notes, this is largely an American problem — Americans are enticed to accept cheap iPhones, but pay really inflated data rates.

    On the compatible networks, note that earlier iPhones were effectively locked out of Verizon and Sprint just by the hardware capabilities.

  106. I needed a good laugh, thanks Patrick!

    “- Minimal iPhone penetration”

    As of 10/11, the iPhone had 5% of the entire worldwide handset market. It’s share of the smartphone market is far higher than that.

    And interestingly,

    “The iPhone is now the top smartphone in the enterprise with 45 percent marketshare among mobile workers, up from 31 percent in 2010.” (http://mobile-workforce-project.ipass.com/cpwp/wp-content/uploads/2011/11/ipass_mobileworkforcereport_q4_2011.pdf)

  107. Spent 20+ minutes on a post that my browser just ate. Grrr. So I’ll try again but may be terse.

    The subsidy is the difference between the on-contract price and the WHOLESALE price, not on-contract and retail. The relationship between wholesale and retail prices is… complicated. And very dependent on the distribution model in any given market. If you think otherwise, I guarantee you paid WAY too much for your car.

    Problem is, the distribution system for phones in the US is broken. Nasty, inefficient, opaque, poorly serving the needs of nearly everyone. But it’s what we have, and we’re stuck with it because the retailers think it works in their faver, and they also happen to be the gatekeepers to the services that phones need to be useful. This system is so messed up, there is definitely incentive (to *someone*) to break it.

    But to do so would require an alternate retail source, as well as viable alternte sources of service. This is one reason I was so excited by things like the TMo $30 monthly 4G. To break the current system we’re going to need to see a huge increase in MVNO’s, discount plans, decent prepaid options and the like. Which requires viable carrier(s) who are not happy with how the current system is working for them. But Sprint and TMo right now are too small, too poor, too inept, too determined to play Verizon mini-me, you name it.

    Who has lots of money, a physical (if not B&M) distribution system for goods and motivation to remove the strangehold of the current wireless service gatekeepers? I predict it’s only a matter of time until Amazon buys TMo, and I just laid out the short version of why.

  108. @pinhead

    As of 10/11, the iPhone had 5% of the entire worldwide handset market. It’s share of the smartphone market is far higher than that.

    He meant that in countries other than the US the penetration was minimal. The total world marketshare doesn’t actually tell us anything about that.

    However, it is also wrong to say that in countries outside the US there is minimal penetration. In the UK, Germany, and France, share of the smartphone market is about 20%. In Japan it is about the same as in the US, and in Australia it is quite a bit higher than the US.

    I am also not sure that there are “lower carrier phone subsidies” in non-US countries. On my network an iPhone 4S currently costs £99 with a £35/month contract. The price direct from Apple is £499. Not sure if that is representative, or how it compares to subsidies in the US, but it’s one data point.

  109. Several years ago the talk was about Apple buying a carrier and breaking the current system. They found a way to twist the system to their benefit, instead. Brilliant for Apple, not so good for the rest of us.

    The combination of an Amazon and a properly-capitalized, competently-run TMo would be such an enormous win to nearly everyone who isn’t Appple, Verizon or AT&T that it could serve as the lever to help move the US more to the European decoupled phone seller/network provider model.

  110. @pinhead:

    I was comparing and contrasting the US with the rest of the world. But you knew that, right? Please tell me you knew that.

    Speaking of US vs. rest of world, you do realize that 49% of the survey respondents in that survey you quote from were from North America, right? Not to belittle Canada or Mexico or anything, but I think that and some other issues in the report might skew the results. But Cathy can speak better to that.

    Finally, you realize that the report shows a microcosm of Apple vs. Android, but delayed due to corporate decision making, right? As in Blackberry market share hasn’t yet hit a steep decline, Apple market share grew by around 50%, and Android, from a much lower starting point, almost doubled its share, which (according to the report) “is reflective of more choice in the enterprise.” In other words, more companies are finally allowing employees to bring Androids to work.

    @Tom:

    Good data point — similar to US pricing. How much would, e.g. a Samsung Galaxy S2 be?

  111. @Greg:

    > I predict it’s only a matter of time until Amazon buys TMo, and I just laid out the short version of why.

    Interestingly, the value of TMobile went up (a bit) after the failed acquisition. Not only the breakup money fees, which will soon be frittered away, but the spectrum AT&T had to fork over.

    But Amazon’s not the only one that could benefit from picking up TMobile. Perhaps a consortium of Amazon, Google, and Dish Network.

  112. @Patrick Maupin

    Good data point — similar to US pricing. How much would, e.g. a Samsung Galaxy S2 be?

    Same network, same £35/month contract, price for the Samsung Galaxy S II is £1.95! Wow!

    Unlocked price for that phone with no contract is £429.99.

  113. @Tom:

    > However, it is also wrong to say that in countries outside the US there is minimal penetration.

    I didn’t mean to say that in _every_ country outside the US, just in aggregate. The US is a large single market. UK, Germany, and France together smaller than US, and Japan and Australia together much smaller than US. I sincerely doubt that all 200 carriers you mentioned offer really good iPhone subsidies, but I would be interested in further data points.

  114. Looks like it is experimentation time again! I like this one much more than the last ‘new’ theme.

  115. WTF?

    Is it me or did the site theme just change twice in as many minutes?

  116. Needs less whitespace on the right side. It doesn’t expand properly.

  117. Ok, I’m sorry. That Samsung galaxy S2 actually is free with a £35/month contract. It just appears they are charging £1.95 of it upfront for some weird reason.

  118. @Tom:

    > Same network, same £35/month contract, price for the Samsung Galaxy S II is £1.95! Wow!

    See, I don’t think this sort of deal is available in the US, except for special sales such as T Mobile’s Valentine’s Day Sale. For example, AT&T brought it out at $200 (but I think you can get it for $90 now at amazon) and I think it’s $230 on TMobile.

  119. If you bring your own device, can you get the same service for a lower price? (And, if so, can you get them to drop the price on the contract phone after a certain amount of time?)

  120. Yes, I clearly understood your bogus point. To say that the iPhone has minimal penetration outside of the US is completely bogus.

  121. >Looks like it is experimentation time again! I like this one much more than the last ‘new’ theme.

    Yeah, the one you saw flash by is called “Admire”. It does look better than TwentyEleven. But it has nested comments.

  122. @Patrick Maupin

    If you bring your own device, can you get the same service for a lower price? (And, if so, can you get them to drop the price on the contract phone after a certain amount of time?)

    If you bring your own phone you can get the same service for £25/month. I don’t really know about your second question, but I assume that after the £35/month contract ends you could switch to the £25/month product.

  123. After reading this thread with an open mind, I have come to the (unsupported by hard facts) that the subsidies even out (Apple v Android). Not mentioned are the BOGO specials frequently found on Android phones, and I got my Fascinate for $0 with a contract extension at a time when VZW was charging $200. I believe that Apple is doing a better job at keeping a greater percentage of what the carrier+consumer spends on each phone.

    That being said, the current US marketing system (with its emphasis on subsidies) is broken for sure. The current status quo has its roots in the Ma Bell days and also reminds me of the way car dealerships used to operate. There’s no doubt the will change, the question is how and who the agents of change will be and what the new (and more profitable) business model will be. Because of capital and operating costs, wireless is a tough business. The incumbents are not going to reinvent themselves except in the face of new competition that renders the status quo obsolete. Where is that going to come from?

    Greg, your comments about T-Mo and Amazon are interesting but I just don’t see that Amazon has the financial muscle to pull it off ($48b revenue, $1b or so profit, $9.5b cash on hand, p/e 134). They might be a player since they’d obviously like to sell phones but someone else will be putting up the real money.

    So what is the business model that renders the current carrier model obsolete and how does it get introduced? I don’t know.

    I’d like it to be a carrier that offers bits in a pipe for reasonable cost with minimal overhead, but I don’t know if that can actually be profitable. If I had 20 or 30 billion dollars to kick it off, I don’t know that I’d be willing to take the gamble. I’ve been looking over VZ’s financials to get a picture. VZ as a bitpipe would have lower revenue. That would be offset by lower SG&A expenses, and cost of revenue might drop some. Depreciation isn’t likely to change. Without knowing a whole lot more about the wireless business than I do, I’d speculate that Bitpipe VZ won’t be any more profitable than Current VZ. So unless Sprint or T-Mo adopts a bitpipe strategy as a desperation/survival tactic, I don’t see change coming from within the industry, and I don’t see anyone outside the industry injecting enough cash to change the landscape. Hell, in another quarter Apple will have enough cash on hand to buy VZ outright, but I wouldn’t bet on them doing it.

    Something else besides the profit motive will be the agent of change.

    One possibility is a regulatory change regarding how spectrum is allocated. Not likely, since any change that made spectrum allocation more efficient would devalue the incumbents’ current spectrum holdings, and that is political suicide.

    Another possibility that comes to mind would be a Carterfone style decision pertaining to the wireless carriers. Skype tried this a few years ago and failed. If anyone else tries again, it will have to be someone with a lot to gain by it and the willingness to make mortal enemies of Verizon and AT&T at least. Google/Motorola is a remote possibility here. The risk would be substantial, and a failed outcome could destroy the Motorola brand and quite possibly Android as well in the US market. If they did it just to blaze the way for others to follow, the upside is limited. If they had IP that made it difficult for others to follow, then it might be worthwhile. In fact quite worthwhile. The other possibility could be Microsoft. They bought Skype, and if they felt that WP was becoming a losing cause, they might think it worthwhile to shoot for the moon. Especially if they had IP that made it difficult for others to follow. And, they have Nokia who would go along for the ride since they have no US market share to lose.

    But both of those ideas are speculative. One is politically impossible and the other is just plain unlikely.

    We can only hope.

  124. One other thought crosses my mind…If Google/Motorola or Microsoft/Nokia were to partner with either Sprint or T-Mo to mount a Carterfone challenge, that would go a long way toward defusing the inevitable arguments that VZ and AT&T would advance why it wouldn’t work/would be inefficient/would harm the consumer etc.

    But now three companies are throwing in their lot together on a bet the farm strategy. Still unlikely.

  125. Problem is, the distribution system for phones in the US is broken. Nasty, inefficient, opaque, poorly serving the needs of nearly everyone. But it’s what we have, and we’re stuck with it because the retailers think it works in their faver, and they also happen to be the gatekeepers to the services that phones need to be useful. This system is so messed up, there is definitely incentive (to *someone*) to break it.

    Mmmm…Apple tried with the original iPhone price model and had to go to a subsidized model (can’t recall why).

    Then Google failed with the original nexus big time trying to change the model.

  126. look at the cash flows, not just the GAAP numbers.

    most Cable TV companies have never made any profit either,
    but they keep playing…..

  127. So new IDC report out.

    According to IDC, Samsung and Apple are basically neck and neck for smartphone position with Apple getting marginally(1Mil/0.1% of market) ahead over the christmas period. Apparantly Samsung was the overall leader for the year due to the bumper result in Q3.

    Two Caveats. Firstly, it’s based on shipment data so insert allegations of channel stuffing here. Secondly, Samsung’s numbers represent all smartphones not just Android ones .

    Also, something that was linked on Groklaw recently(well I went through to the actual Forbes article not the rehash article on tech2) . Study says iOS apps crash more often than android ones. While I wouldn’t even begin to suggest this “proves” anything, I do think it’s one data point to suggest that the Appleigentsia need to be careful about fragmentation accusations, the kettle may not be as pristine as they believe.

  128. @Tom:

    > If you bring your own phone you can get the same service for £25/month.

    This is where TMobile seems to be headed with some of their new offerings. AFAIK, all the other carriers (e.g. all the ones carrying iPhones) segregate their services into prepaid (where the phone subsidy is minimal or nonexistent) and postpaid (where they don’t want to lower the monthly even if you BYOD, because that would set a precedent they would hate.)

    But TMobile now seems to have a postpaid offering where there’s essentially a 2-year adder for financing the smartphone.

  129. @Nigel:

    Mmmm…Apple tried with the original iPhone price model and had to go to a subsidized model (can’t recall why).

    There was speculation at the time that even at the $499 or $599 price, AT&T had to fork over a bit each month to Apple, so it might have been subsidized even then (and AT&T had a $175 early termination fee on phones at those prices, too…)

  130. It seems likely that someone will buy T-Mobile and try to blow it out. Whether they will focus on prepaid, try to turn it into a cheap bit-hauler, or attempt to convert it into a carbon copy of the Big 3 is an unknown.

    Given all the years that investors have continued to pour money into airlines despite it the industry as a whole showing long-term losses, I conclude that there’s always someone who thinks they can build a better mousetrap and make money doing so.

  131. “So what is the business model that renders the current carrier model obsolete and how does it get introduced? I don’t know. I’d like it to be a carrier that offers bits in a pipe for reasonable cost with minimal overhead, but I don’t know if that can actually be profitable.”

    Can actually be profitable at what price? Has anyone done the calculation to see what the monthly fee would need to be for a profitable no-frills bit-hauler that cut SG&A and operational costs to a minimum? Given the huge amount of money that the carriers are pouring into subsidies and attempts to lock people into long-term contracts, I have to suspect that that calculation would show a fee less than $100/month, but I don’t have the data to make a real estimate.

  132. At one point (before AT&T) they were talking about floating a T Mobile IPO.

    I think they should do that, but offer some shareholder perks up front. You know, like maybe if you buy and hold $10K worth of shares, you get 75% off the price of any family plan. Maybe 20% off for $1K. Maybe cut-rate ebook and movie transit to Kindles or Nooks if you put in $5 billion.

    And they should use a dutch auction IPO like google and kick the banks out.

    I’d invest. It’s not like Americans don’t bleed money to cellphone companies anyway. Might as well do it directly if it means that I can bleed a little less in the monthly bill.

  133. Just as an aside, I was told sometime about a decade ago that AT&T had done a study on usage and costs for local and long distance land lines, which concluded that accounting and billing for minutes was a lose, and that they’d be far more profitable charging a flat rate for access to all subscribers.

  134. I see the motivation for breaking the current US wireless industry model for distributing phones coming from someone wanting to make a profit *selling phones*. As cell phones consume all other mobile devices, as they cease to be just phones and become personal computing devices and much else besides… selling/distributing those devices is going to be too big, too important and too potentially profitable to be left in the hands of wireless carriers.

    As it is, selling phones and selling service have been artificially merged, to the detriment of both. The carriers lose huge sum of money “selling” phones, on the bet of making it up elsewhere. Imagine how screwed up things would be if gas stations gave away cars, and what gas stations would have to do to themselves and their customers to try to make that profitable.

    Also, Amazon just recently made a large bet in online content distribution, making them even more likely to be willing to take the step of buying TMo. It’s true though that they might not be big enough to make it work alone. It might take a consortium- don’t leave out WalMart. They’ve already started making tentative steps in that direction.

  135. The Apple business model for anything is “Apple gets all the money.” That’s certainly their model for iPad magazine publishing, where they’ve actually conned print publishers into believing the iPad is a magickal money fountain they just need to tap into.

    (As far as I can tell, the real business model for iPad magazine publishing is “the executive suite all need new iPads to, er, check how our stuff looks, yes that’s right.”)

  136. While I wouldn’t even begin to suggest this “proves” anything, I do think it’s one data point to suggest that the Appleigentsia need to be careful about fragmentation accusations, the kettle may not be as pristine as they believe.

    Code is buggy, even Apple code. Even rocket science code subjected to full up systems engineering, fagan inspections, yadda yadda yadda. That apps are buggy moving from XCode 3 to XCode 4 (that sucked) and from iOS 4.x to iOS 5.0 isn’t all that surprising.

    That has zip to do with fragmentation and the burden on the developer. By November of 2011 iOS 5 adoption rate was 38%. From an app sustainment we elected to drop iOS 3 support in our latest enterprise app and to go ahead and use iOS 5 features if desired. For a indie developer I’d likely support 3.2 for the widest base but not blink an eye if I had to move to iOS 4+. I’d lose all the iPhone/iPod 1 customers but that’s less than 1% by now.

    Here are some numbers from Marco and his app also from Nov 2011:

    http://www.marco.org/2011/11/30/more-ios-device-and-os-version-stats-from-instapaper

    For reference iOS 5 was only 49 days old at the time.

    Looking at the Android SDK version stats today 21% are still using Froyo and only 3.4% using ICS.

    http://www.appbrain.com/stats/top-android-sdk-versions

    We killed our enterprise android app in favor of web apps but if I were going to continue to maintain it I’d stay with using no more than 2.2. I wouldn’t be looking at ICS support for another six months unless I wanted to make a tablet version. Probably not even then as I’d target the Fire.

    As the article states, iOS apps will stabilize as devs get used to changes in both the tool chain and SDK features.

  137. Deja vue all over again. What made the updating of MS Windows such a problem, is now bugging down Android.

    We all know how the seamless integration of hard- and software brought Apple: in distant second place to Windows.

    Now we see the sequel play out the exact same story:

    Moto Exec on Android Upgrade Delays: It’s The Hardware
    http://www.pcmag.com/article2/0,2817,2400023,00.asp

    “When Google does a release of the software … they do a version of the software for whatever phone they just shipped,” she said. “The rest of the ecosystem doesn’t see it until you see it. Hardware is by far the long pole in the tent, with multiple chipsets and multiple radio bands for multiple countries. It’s a big machine to churn.”

    Motorola understands that consumers want their Android upgrades sooner, but the process is complicated, she said. First there’s hardware support, then the layering in of custom software from manufacturers like Motorola, and finally, phones must be re-certified by carriers, taking more time.

  138. We all know how the seamless integration of hard- and software brought Apple: in distant second place to Windows.

    Well, that allowed Apple to survive as a distant second place to Windows, and they are now gaining. And how are those third- and fourth-place players doing? Does the sale of scores of millions of iPads tell you something?

    You are trying to make hardware and software fragmentation a virtue in the marketplace. I don’t think it is, not in smartphones or tablets, at least.

  139. @Greg:
    “I see the motivation for breaking the current US wireless industry model for distributing phones coming from someone wanting to make a profit *selling phones*. As cell phones consume all other mobile devices, as they cease to be just phones and become personal computing devices and much else besides… selling/distributing those devices is going to be too big, too important and too potentially profitable to be left in the hands of wireless carriers.”

    Good thought, Greg. That would point toward a Microsoft or Google/Motorola. Google needs to be careful not to look like they are competing with their own customers, but Microsoft has so little presence in phones that they have little to lose. Surely they have enough cash to either buy T-Mobile outright, or take a controlling stake in it.

    I don’t think Walmart would be interested in tying up a lot of cash investing in this business. I could see them as a retail partner without a significant investment stake, where their contribution would be shelf space and consumer mindshare/promotion/high-traffic shelf space.

    Re: Amazon, I don’t see the benefit to their core business.

  140. @Papayasf
    Sorry, but Apple survived due to active support from MS who were desperately in need of a harmless, but credible competitor.

    Without the investments and Office for Mac, Apple would have folded in the 1990s

  141. From WSJ: it looks like Google is getting into the hardware business.

    “Google Inc. is developing a home-entertainment system that streams music wirelessly throughout the home and would be marketed under the company’s own brand…

    “Google plans to make a branded home-entertainment system, steps up the company’s rivalry with Apple by adding a new platform for competition… The effort marks a sharp shift in strategy for Google, which for the first would time would design and market consumer electronic devices under its name.”

    http://online.wsj.com/article/SB10001424052970203824904577213430617644196.html?mod=WSJ_hp_mostpop_read

  142. @Larry Yelnick:
    “Android is killing desktop linux.”

    This is silly. From the article Larry references:

    “The switch to Android 4.0 is changing thinking around Webtop, Motorola’s very unusual mode which turns a phone into a mini-laptop running a desktop version of Linux. The desktop-Linux version may not be needed post-ICS, Wyatt mused.

    “‘With ICS it’s more interesting because it starts to have some of the full, desktop-like features,’ she said. Chrome for Android may make Webtop’s desktop version of Firefox less necessary, for instance.”

    Well, sure. It’s been clear for some time that mobile devices are going to take over a lot of the user occasions that used to be handled by desktops and laptops. The browser is front and center for this. That applies equally to Linux desktops, Windows desktops, and Macs. If you are just browsing the web, you don’t need the full home computer capabilities.

    But for software development, or running complex analytical software, or even high-powered game software with complex user interfaces, mobile devices just can’t compete. They are limited by the user interface, regardless of how powerful the device may become.

    Of course, if we see the rise of mobile-device-plugged-into-full-size-keyboard-and-monitor, that may change. But that would require a very flexible mobile operating system, and I don’t think current versions of Android, or iOS, or WinPhone, could handle this.

  143. @Winter

    Without the investments and Office for Mac, Apple would have folded in the 1990s

    No doubt Apple was helped by MS’s investments, but you can’t possible substantiate a statement like ‘they would have folded’ without that help. There’s absolutely no evidence for this.

  144. @Cathy

    But for software development, or running complex analytical software, or even high-powered game software with complex user interfaces, mobile devices just can’t compete. They are limited by the user interface, regardless of how powerful the device may become.

    Right. Linux-on-the-desktop is almost exclusively used by hackers and other techies who are not going to be satisfied with Android as a replacement for their PCs. I would say that Linux-on-the-desktop is actually the OS *least* susceptible to disruption from mobile platforms.

  145. Spoilers on blog post linked in above comment: James A. Donald successfully explains why Wikipedia has failed dismally, except the bit about being both useful and popular. He also uses more words with less skill than Robin Hanson, who invented the prediction market, to explain the same neat idea. HTH!

  146. @Tom:
    “Linux-on-the-desktop is almost exclusively used by hackers and other techies who are not going to be satisfied with Android as a replacement for their PCs. I would say that Linux-on-the-desktop is actually the OS *least* susceptible to disruption from mobile platforms.”

    Well said, Tom.

    My post on Google new home entertainment system seems to be stuck in moderation…

  147. Right. Linux-on-the-desktop is almost exclusively used by hackers and other techies who are not going to be satisfied with Android as a replacement for their PCs.

    Desktop Linux is losing even in this demographic. Linux is only free if your time has no value. Hackers with shit to do buy Macs.

  148. Desktop Linux is losing even in this demographic.

    Any evidence?

    Linux is only free if your time has no value. Hackers with shit to do buy Macs.

    I’m a Mac user, and this is an incredibly stupid and arrogant thing to say.

  149. When Microsoft succeeds in locking down the boot loader, where are the linux hackers going to find hardware on which to run their favorite distro?

    Fortunately, that will take a while, (end of the decade), and esr will have started to collect Social Security by the time we would see him appear in public with an Apple product in-tow.

  150. Any evidence?

    Only anecdotal. When I see hackers they are increasingly carrying Macs under their arms rather than Linux laptops. If someone is hacking in a coffee shop, chances are it’s on a MacBook. The ones I’ve talked to tend to justify their decision as follows:

    a) Apple is the only company that makes decent hardware anymore (I would add Lenovo to that list, but really those are the only two especially for laptops)

    b) They want something that Just Works, which Linux doesn’t

    When you think about it, most hackers are not ideologues like Stallman. They are engineers; and paying more money (and compromising complete openness) for Apple’s vertical-integration advantages (the It Just Works effect), the coolest native-code platform in the world (Cocoa), and an audience for their code outside hackerdom is a completely legitimate engineering tradeoff.

    From my observations the trend is clearly in favor of the Mac as the hacker’s platform of choice. I know it’s not much, but I’m not the only one noticing this.

  151. When Microsoft succeeds in locking down the boot loader, where are the linux hackers going to find hardware on which to run their favorite distro?

    You know you can still get 5 1/4″ floppy disks, right? A few specialty shops sell them for those applications that still need them — my eye doctor’s visual field testing machine for instance. They’re harder to get and more expensive, though.

    It’ll be that way with conventional, open-boot-loader PCs. They will be rare and expensive, but they will be there for the few who really need them.

    Commodity hardware will only boot the vendor-supplied signed kernel image, and may even require signed, approved application code as well. Microsoft is actually accelerating this process by blurring the transition from the old peecee model to the new tablet model, referring to Windows 8 ARM tablets as “PCs”.

  152. a) Apple is the only company that makes decent hardware anymore (I would add Lenovo to that list, but really those are the only two especially for laptops)

    b) They want something that Just Works, which Linux doesn’t

    When you think about it, most hackers are not ideologues like Stallman. They are engineers; and paying more money (and compromising complete openness) for Apple’s vertical-integration advantages (the It Just Works effect), the coolest native-code platform in the world (Cocoa), and an audience for their code outside hackerdom is a completely legitimate engineering tradeoff.

    You don’t have to convince me of the merits of the Macintosh, but different platforms work for different people. There is a certain minority of programmers who have decided that, for them, Linux works best, and that minority is unlikely to change that decision because of Android.

    Also, I wouldn’t be so sure that just because you see somebody carrying a Mac that they are necessarily using OS X exclusively. I have Linux installed on mine as a dual boot option, and while *I* don’t use it as my primary OS, many do.

  153. Tom, indeed. The MacBook makes a great Linux platform.

    To be fair, many of those hackers on Macs I mentioned do use Linux. On virtual machines. But as a primary desktop OS, it stinks. I don’t see too many actual dual-booters.

  154. That has zip to do with fragmentation and the burden on the developer.

    Um… Wot? I thought fragmentation was about a shitty experience for the end user which can impact whether they enjoy(and by extension will stay with) the platform. Such as (for example) repeatedly crashing applications.

    If developer burden is the issue then i’ll renew my voice that fragmentation isn’t a problem for android. For good developers it’s not going to matter much either way and crappy developers would rather work in blub language (i.e. java) than something that looks like someone attempted plastic surgery on C++ and ended up with Melanie Griffith.

  155. @Jeff Read:
    “It’ll be that way with conventional, open-boot-loader PCs. They will be rare and expensive…”

    Evidence? Not on tablets or phones, which I believe live in very different parts of the tech spectrum, but actual desktop or laptop machines that have locked boot loaders? I haven’t seen or heard of any, but perhaps I’m not paying sufficient attention.

  156. >actual desktop or laptop machines that have locked boot loaders? I haven’t seen or heard of any, but perhaps I’m not paying sufficient attention.

    There aren’t any. In Jeff Read’s fantasy universe, some nefarious coalition of closed-source software vendors is going to somehow pressure every PC hardware maker into manufacturing only PCs with locked bootloaders in the future.

    This will never happen, because the customers that PC makers ship in volume to – businesses with IT departments and server rooms – would scream bloody murder and defect to manufacturers who weren’t so fucking stupid at the first sign of such an attempt. IT departments (even those not using Linux internally) don’t want to lose the control point of being able to create and boot custom distributions and won’t stand for having it taken away from them. This intransigence has the accidental effect of protecting individual consumers.

  157. “Fragmentation isn’t a problem for Android”? Seriously? As the article says: in three months more than 2/3rds of iOS devices were running iOS 5. When will 2/3rds of Android devices be running ICS? My guess: never. And that’s just software fragmentation.

  158. @Tom
    “Linux-on-the-desktop is almost exclusively used by hackers and other techies who are not going to be satisfied with Android as a replacement for their PCs.”
    @Jeff Read
    “Desktop Linux is losing even in this demographic. Linux is only free if your time has no value. Hackers with shit to do buy Macs.”

    Indeed, this must be true as this has been going on for 15 years now. I know because “reliable” sources are writing about the decline of Linux from a minuscule start since 1995. If you start small and keep losing users in droves, you must be dead really soon.

  159. FWIW, an anecdote point: the Wikimedia Foundation uses a lot of Apple hardware … running Ubuntu. This is for daily office stuff in a small-to-middling charity. (You can run Mac OS if you like, and you only get Windows if you have a clear need, e.g. accounting and HR. They also use a lot of Dell.) I’ve asked someone to make a public blog post or page on this for public interest, but am unsurprised no-one’s gotten to it.

  160. As the article says: in three months more than 2/3rds of iOS devices were running iOS 5. When will 2/3rds of Android devices be running ICS? My guess: never. And that’s just software fragmentation.

    OK great. 2/3rds of devices are having a reliable experience on an old version of Android.
    So we’re agreed, there’s no fragmentation issue.

  161. @David Gerard
    That is not an insightful comment. An Android device is more like a computer than a TV. And I can envision using a netbook/laptop running Android. I do not see. A TV set fulfilling my computational needs looks much less likely.

    Anyhow, a TV running linux runs loo inux whatever you try to call it.

  162. @Winter – The TV is much closer to Unix as we know it than the phone is, unless you go to particular effort to get a bash prompt. The useful aspect is that Android is open source, or at least open core. It’s barely Linux, that it’s Linux in particular isn’t anywhere near being the important thing about it.

  163. @David Gerard
    If people had written KDE or Gnome versus Android, OK.

    But Linux versus Android is silly.

  164. @Winter – that it is. I’m annoyed the industry expects my next netbook to be a tablet. And that I can’t find a new 9″ netbook anywhere unless I order directly from China.

  165. There aren’t any. In Jeff Read’s fantasy universe, some nefarious coalition of closed-source software vendors is going to somehow pressure every PC hardware maker into manufacturing only PCs with locked bootloaders in the future.

    I never said that.

    What they’ll do is split the market into a consumer market and a business market. Consumer market is locked only. Business market pays a hefty premium for general-purpose hardware. Or, PC vendors only sell general-purpose hardware to businesses with a Dun & Bradstreet number and/or who place orders of a certain minimum volume. Or, PC vendors can sell certificates to businesses that let them sign their custom OS and apps, much like Apple’s corporate developer program for the iPhone and iPad. (You don’t get to reflash the iPad’s OS under this program but you do get to approve apps of your own for distribution on your company’s iPads.)

    Also, note well that though servers look more like PCs in terms of hardware loadout than they did in years past, the server market is still quite different from the PC market. So the market may yet be split again into unlocked servers and locked desktop machines. End users in the corporate environment may not even get PCs but Wyse terminals, that connect to VMware instances running Citrix in the company’s cloud.

    Corporate IT has absolutely nothing to do with whether the little guy gets access to general-purpose, unlocked computer hardware.

  166. >Consumer market is locked only. Business market pays a hefty premium for general-purpose hardware.

    The business PC market is as efficient as the consumer market. What they will pay a slight premium for is not completely piss-awful hardware. (Yes, there’s actually something worse than the average business PC.) Windows will be standard on office workers’ desks until the end of time, Win32 on the desktop and Java on the server will be the new COBOL. Microsoft could probably get away with locked desktop PCs if they can effectively subsidise the locked desktops.

  167. @Android

    If people had written KDE or Gnome versus Android, OK.

    But Linux versus Android is silly.

    We were specifically talking about ‘linux-on-the-desktop’, which means Debian, Fedora, Ubuntu, etc etc etc.

  168. @Jeff Read

    Business market pays a hefty premium for general-purpose hardware. Or, PC vendors only sell general-purpose hardware to businesses with a Dun & Bradstreet number and/or who place orders of a certain minimum volume.

    We are well into fantasy-land now.

    You think businesses are going to be happy paying a ‘hefty premium’ to get what they want? Why? They will not accept this, and anybody in the market who tries it will lose business.

    As for the idea of vendors only selling unlocked hardware to businesses, I barely know where to start. Why would they do this? It’s completely crazy. It might open up a nice opportunity though for a business that bought large numbers of unlocked computers and then resold them to anybody who wanted them.

  169. @JonCB

    Um… Wot? I thought fragmentation was about a shitty experience for the end user which can impact whether they enjoy(and by extension will stay with) the platform. Such as (for example) repeatedly crashing applications.

    The burden on the consumer is simply having a device that can’t or wont go to the latest version of Android even if the hardware is capable of supporting it.

    If developer burden is the issue then i’ll renew my voice that fragmentation isn’t a problem for android. For good developers it’s not going to matter much either way and crappy developers would rather work in blub language (i.e. java) than something that looks like someone attempted plastic surgery on C++ and ended up with Melanie Griffith.

    Good developers want to test on as many target platforms as possible. Fragmentation makes this harder. Especially with Sense/TouchWiz/Blur.

    Good developers like using the most powerful/latest API features to make their apps better than the competition. OS rev fragmentation makes this harder. I can make the decision to use iOS 5 only features and be reasonably sure I can address the majority of iOS devices. Can you make the decision to use ICS only features and be reasonably sure you can address the majority of Android devices?

    If I must test my app on 2.x, 3.x, 4.x across a large variety of hardware and UI skins then this is a burden vs testing my app on 4.x and 5.x on a few hardware devices.

    Increasing developer burden make any program lower quality/capability than it would otherwise have been given the same resources and schedule. Good developer/bad developer doesn’t matter…although it probably penalizes the good developer more. Bad devs don’t mind pushing out crap.

    Even on iOS app testing is somewhat spotty (given the many 1 star reviews about app x always crashing) and that’s WITH Apple’s testing before approving the app.

    I wish I could remember where I read one app developer commenting that the iOS app store approval process with Apple’s review made their app a lot more bulletproof and better because of that review process. Whereas they just let 1.0 go out the door and onto Google Market as it was. This was a major app too, not a 1 person no name app.

    As a dev on both platforms, I find while I like java better (familiarity) I can live with ObjC. The quality of the SDK makes it livable (unlike, say C++). I also find that the bar is higher on iOS in terms of user expectations and quality of competition. A lot of effort is expected on user experience even on internal enterprise apps which is fairly well known for crappy, programmer designed, UX.

    As far as whether fragmentation is an issue for android we made the strategic decision in our group to dump android development despite the Army selecting Android as their mobile OS. Given that the Air Force is thinking of buying 18,000 iPads that probably wasn’t a bad choice for us.

  170. Lots of news of Nokia layoffs this week. They’re letting go of up to 1000 people in Salo, Finland (coincidentally the home town of Finland’s new president-elect) and up to 3000 in Hungary and Mexico. These are manufacturing jobs that are going to Asia, if not just disappearing. I’m not sure if any Nokia phones at all will be “made in Finland” after this point.

    Nokia Siemens Networks is letting go 1200 people in Finland. Accenture wants to get rid of the 1200 of Nokia’s Symbian developers that they hired last year after Elop’s big announcement. Of course, this is the normal operating procedure when a large corporation lays off a bunch of people via Accenture. The unions are moaning about Accenture pressuring people to accept bad terms in their contracts and about Nokia dodging its responsibilities by offloading people to Accenture.

    I just happened to see a news clip about the Foxconn employee suicides in China. That was the first time I saw the pictures of the nets they installed a few years ago in between and around their buildings in order to catch the workers that jump off the roofs. I suppose that’s one way to manage the problem. No surprise that the factory in Salo can’t compete.

  171. >It might open up a nice opportunity though for a business that bought large numbers of unlocked computers and then resold them to anybody who wanted them.

    Exactly. There is no way to prevent the huge volumes of computer hardware built for business use from flooding onto the consumer market, short of legislation to forbid certain classes of person and entities from owning unlocked machines and a huge intrusive enforcement apparatus. Legislation like that would have been politically impossible even before the thumping defeat of SOPA and PIPA – the blocking coalition against it (civil liberties groups, librarians, SOHO users, larger businesses fearing the cost premium from a two-tier morket) is just too broad.

  172. @PapayaSF

    No, 95% of Android users are using a version at least two years old, and most can’t upgrade without buying a new device, so developers can’t take advantage of new OS capabilities without leaving out most users. If that’s not fragmentation, what is?

    Compare that with iOS. Marco Arment put out a tweet today saying:

    89% of iOS devices running Instapaper in the last 60 days have at least iOS 5.0.

    92% if I only go 2 weeks back.

    Instapaper is Marco’s extremely popular iOS reading list app.

  173. @Tom
    And if Linux on the desktop turns out to have an Android UI it is still Linux o .the desktop.

  174. And, by the way, I’d say it is even a stretch to say that ‘Android is Linux’.

    Android is *based* on the Linux kernel. But even the kernel has been substantially altered by Google, and a lot of those changes have been rejected by the Linux kernel devs. Android doesn’t have an X Window system, and it doesn’t have the full set of GNU libraries. Try porting Linux programs to Android and you will find out how much the two are different.

    For the moment, at least, desktop Linux and Android share a common kernel heritage, but they do not share a common kernel.

  175. Yes. Android is barely Linux, and it certainly bears almost no resemblance to GNU/Linux or approximations as we (reading this blog) know them, and this is important. The trouble with the claim “But Android is Linux!” is that it carries a bag of connotations that simply do not hold.

  176. Corporate IT departments (the same that Eric is counting on to scream bloody murder at the prospect of not being able to boot custom Windows distributions) are currently bailing water as fast as they can while simultaneously developing strategies for bring your own device, employee-liable mobile devices, borderless networks, and VDI/cloud services.

    What we’re seeing with the UEFI signed boot is a classic head-fake. Microsoft’s corporate bastion is playing a different game than Eric and the others in the FOSS communities are reading. The current (Windows 8) rules for x86 machines are that x86 Windows 8 systems must allow users to add and remove certificates from the firmware’s certificate store. For example, a Linux vendor could provide a signed operating system loader and corresponding certificate: all x86 Windows 8 systems must permit users to install such certificates. Microsoft calls this “custom mode”, in contrast to “standard mode”, that includes only the Microsoft certificate.

    The story for ARM is rather different, however. On ARM Windows 8 systems, Microsoft’s certification rules prohibit entering “custom mode”—users must not be able to add certificates of their own—and prohibit disabling secure boot completely. The ARM systems will all require the use of a signed operating system loader, and that operating system loader must be signed by Microsoft. Microsoft’s rules also specify that a secure boot failure must be fatal; there must be no option to override the failure and choose to boot the untrusted operating system. Further, manufacturers will be prohibited from shipping updates to their firmware that relax these restrictions; all firmware updates must be protected with digital signatures and must preserve the secure boot settings. They should also prevent the installation of older firmware versions—for example, downgrading the firmware to switch to a version with a known security flaw—though the manufacturer can allow this particular constraint to be manually overridden.

    Meanwhile, Eric conveniently forgets that we are seeing a wholesale industry shift towards ARM as something of a “terminal” and x86 as a something more “workstation”. Combine this with the current IT drive to BYO-PC, and the recently announced plans for Microsoft to ship all Windows 8 tablets with Office included free-of-charge with Windows 8 on ARM. Now your non-legacy IT worker can bring their own inexpensive device to work, based on a rigidly-enforced boot loader.

    So the situation that develops: if your work require some legacy code then you will be using a x86 system. But if all you do is fill out forms and type up reports then you can be issued a ARM system. This will be even more true

    (It should be obvious that this reduces the demand for PCs well-beyond the contraction the market is already witnessing. Potentially this demand becomes low enough that ‘geek’ machines are no longer built in the volumes that enable the razor-thin margins of today. Yes, the geeks can still have PCs, but they’ll be priced like Sun (or higher), not at the commodity levels so-often praised by Eric and his pals.)

    If this is executed well, ARM basically allows Microsoft to put the legacy cat back in the bag, as it becomes a clean break. And they can point to .net and similar managed bytecode as a way forward for those that want to run code on both platform. Code provided to the customer via the Microsoft store, naturally. Microsoft can certainly afford to bundle a copy of Office on ARM in exchange for 30% of all additional software sales, and 30% of any media sales, too. (This also makes Win8 on ARM a very nice sell for big content to promo, as it provides a way for them to “know” that any device playing their content is “secure”.)

    UEFI on ARM for Open systems? Google Chrome notebooks already have a secure boot loader. Yes, there is currently a ‘debug’ switch, but that is easy to remove when it becomes convenient or even required. (Also, unless Torvalds has an issue with UEFI (and i suspect he has, given his dismal view of ACPI), it may provide his wished for unified boot system for ARM based systems.)

    Anyways, this seems all seems like a rerun of the Palladium/NGSCB/TCM shit that there was a bruhaha about a decade ago. Hell, it looks like the same old shit with a new coat of paint to me. But the iPad shows that consumers world-wide are already used to eating sugar-coated shit lozenges and smiling about it.

  177. ” This will be even more true” … as employees are required to supply their own computing interface.

    (sorry)

  178. news clip about the Foxconn employee suicides in China.

    Did that news clip mention that the suicide rate for Foxconn employees is much lower than for China as a whole, and is also lower than every state in the USA?

    Every suicide is a tragedy, but this story has been flogged for page hits for far too long.

  179. ARM basically allows Microsoft to put the legacy cat back in the bag

    Nobody ever wanted Windows on anything but x86. Even when Microsoft was at the height of their monopoly, Windows for DEC Alpha, MIPS, SPARC, and PPC were all dead on arrival.

  180. the coolest native-code platform in the world (Cocoa),

    I’ve been working on Cocoa and NeXTSTEP for decades, and I’m ready for something better. Java wasn’t it. Python wasn’t it. Ruby wasn’t it. If someone offers up a development system for mobile devices with a nice, CLEAN set of UI, media, database, and networking frameworks, based on Smalltalk or Lua or another tolerable language, I’m there.

    In the meantime, I’m sticking with the Mac and IOS because there isn’t anything better. I’ll have a better development platform someday, and chances are it too will come from Apple.

  181. No, 95% of Android users are using a version at least two years old, and most can’t upgrade without buying a new device, so developers can’t take advantage of new OS capabilities without leaving out most users. If that’s not fragmentation, what is?

    Hang on a sec… lets pull apart the hyperbole here.

    60% of android users are on the stable version of android for phones that isn’t < 3 months old (Gingerbread). By the way, Wikipedia lists 2.3.3 as being released on 9-Feb 2011. I don’t know how time travels in your reality but in my there isn’t two years between 2011-02-12 and 2011-02-09.

    So that brings us the next 30% of users are on Froyo which is pretty much Gingerbread with out some of the doodads like NFC. The platform changes for API Level 9 and API Level 10 are neither earth shaking nor are they that critical for the vast majority of apps out there.

    Oh and just to put your 95% of android users on a greater than 2yr old OS fully to bed… Froyo was first released in May 2010 which is closer to one and a half years. If you had’ve said “30% of android users are using a version almost two years old” i wouldn’t have argued too hard but “at least two years” is purely fantasy land bullshit.

    So 90% of android users can use basically everything in the android market other than seriously niche stuff. Yeah… that’s a really compelling case of software fragmentation you’ve got there.

    P.S. My phone is limited to Froyo. So I’m intimately aware of how much bullshit you’re spouting here.

  182. There also isn’t two years between 2012-02-12 and 2011-02-09 either which is more what I was trying to say.

  183. I was counting from the release of 2.3 in 12/2010, more than two years ago. OK, if you want to count fractional releases, use 2.3.3 and knock a year off. So then it’s 95% of Android users who are using an OS at least a year old, most of whom can’t upgrade on the hardware they own. That’s slightly less fragmentation, but still a massive amount.

  184. @Tom
    “Android is not a desktop OS”

    You simply are blind to what is happening under your nose.

    @Tom&David
    “Android is barely Linux”

    linux is not like OSX, you can actually exchange the UI. You simply do not understand how *nix works. That is your problem, not Androids.

  185. @winter

    Android may be Linux. But it’s not gnu/Linux and that’s a win right there in my book.

    And given that android doesnt have a unix userland its not a unix.

    Busybox on top of Vxworks isn’t a unix
    Toolbox on top of Linux isn’t a unix

    As rms is keen to point out, Linux is nothing but a kernel. You CAN build a unix on top of Mach, Linux, QNX, VxWorks etc. but you need someone’s unix userland. BSD for OSX. GNU for GNU/Linux.

    The ability to change UI is not a requirement. Neither is any sort of desktop.

    Perhaps it is you that doesn’t understand how unicies work.

    And you certainly don’t understand OSX given it can run X and with some shoehorning gnome. Or at least I did it with Gnome 1.x and fink back on panther.

    I moved to OSX so I could do any sort of unix development without needing GNU/Linux. And I ain’t got nothing against the Linux kernel. I simply moved to the best desktop unix environment that happened to be based on the bsd userland.

  186. Finally figured out the problem with the original analysis. They didn’t bother to note T-Mobile’s numbers during the same time periods. Perfectly good control factor right there, but the analysts blew it off.

  187. “Android is not a desktop OS”

    You simply are blind to what is happening under your nose.

    Okaaaay….

    Maybe you could explain? I don’t think I made a very controversial statement there. Android is an OS for phones and tablets. It’s a mobile OS, not a desktop OS.

    linux is not like OSX, you can actually exchange the UI.

    I understand that, but you seem to be implying that Android is just some sort of skin on Linux. You are equating it with KDE or GNOME. That is not right. ‘Android’ isn’t just a replacement UI.

    I tried to explain this earlier, so I refer you to my previous comments.

  188. So, Android is not linux, Gnu/Gnome is not linux, Busybox linux is not linux. So there is no linux?

    Android differs from the stuff running in my TV, settop box, router, and PC. All use a linux kernel. Moreover, I can run the Android UI+API in my linux desktop. And there will be Android netbooks and laptops based on ARM.

    While Android takes off in tablets it will creep up to other computers.

  189. @winter

    You specifically stated that some folks here don’t understand *nix. Not that folks don’t understand LINUX.

    Your attempted rephrasing to cover up your own ignorant accusation is transparently bad.

    Android uses the Linux kernel.
    It is not unix. Nor is it like gnu/linux. Saying Android is Linux is as many have stated, misleading. Android could have easily been built on a bsd kernel and still be distinctly android. In fact describing android as a Linux based OS is as relevant as describing OSX as a Mach based OS. True but few will care.

    It is very different in approach and goals vs Gnu/Linux. It’s not a desktop or UI skin on top of Gnu/Linux.

    Gnu/Linux is unix…or close enough that only pendants say “unix like”. Android ain’t.

    If android takes off on the desktop Ubuntu is dead and google will have done what ms never could: embrace, extend and extinguish (desktop) Linux.

    Google has already done so with Java ME. Embedded java is all but dead except as an android variant.

  190. I have possibly used gnu/linux in this thread than all other times combined.

    I object to rms’ whining about using gnu/Linux because in my mind the gnu part is nearly irrelevant. If it did not exist there was a perfectly good bsd userland available Linus could have used. BSD emerged from it’s legal limbo around the same time.

    I am glad that Linus is not the type to insist on calling android Android/Linux. The Linux part is to me also nearly irrelevant. If it did not exist there are perfectly good kernels available to google just as for Bada.

  191. “No, 95% of Android users are using a version at least two years old, and most can’t upgrade without buying a new device, so developers can’t take advantage of new OS capabilities without leaving out most users. If that’s not fragmentation, what is?”

    What fraction of Windows users are on Windows 7? Vista? XP? And then there’s the XP Home/Professional split, and the fact that Microsoft keeps gratuitiously changing the user interface (especially of Office).

    With the business world split between Office 2003 and Office 2007, I live this every day…and there is far more difference between them than between Froyo and Gingerbread.

  192. @cathy some folks might argue that the ms model isn’t necessarily the one folks should emulate :)

    Fragmentation is generally not a good thing and has certain costs on both users and developers. That some products like office can get away with it doesn’t imply it’s not a serious issue AND ignores the great effort on MS’ part to provide converters and minimize the impact.

    No matter how folks want to spin things the fact that device makers are poor at updating ti tge lastest os is a bad thing and costs the ecosystem.

  193. Cathy, I think there are several differences here. Because Windows has been so dominant for so long, sheer numbers and inertia overcome many of the problems of fragmentation. But if most people (and businesses) were buying their first PC today, the fragmentation of Windows would make it much less attractive.

    In contrast, smartphones don’t really have any legacy issues yet, and I suspect that most smartphone users do not want to futz around with them the way they often have to with a PC. Beyond nerds, who wants a phone that has a new and different set of PC-like troubleshooting issues? Thus Android fragmentation is a real issue. (And again, that’s setting aside hardware fragmentation.)

  194. @Nigel
    Android can be ported to whatever kernel you like, but until then it is not Android against linux

    And you might be able to get OSX booting to Gnome, you have little company. It is silly to say Android will kill/displace desktop linux when it has a good chance of becoming desktop linux.

  195. Beyond nerds, who wants a phone that has a new and different set of PC-like troubleshooting issues?

    Sure and if that was actually happening i’d agree with you. But the only people bitching about Android fragmentation are developers. Or to be more precise, the only thing people are bitching about is the potential for fragmentation. Not the reality.

    In reality you can be on Froyo and be having a fantastic experience and not be missing “the latest and greatest features”. More importantly than missing out on the latest and greatest features is making sure the user experience doesn’t take any negative hits. Which brings me back to my original point. Be careful about decrying fragmentation, because all software with more than one version is fragmenting, the only sane question is “is the user noticing?”.

  196. @winter

    Android on the desktop vs Linux on the desktop has zero to do with the kernel and more about the role it fills (desktop os). In the discussion regarding Android vs iOS it sure as hell isn’t a iOS vs Linux battle in anyone’s mind.

    If android gained more desktop mindshare it would be Android vs Windows, OSX and Linux. Of the three the only one it could kill is Linux.

    You would move to a far more appcentric ecosystem incompatible with what most folks think of Linux today.

    That may not be a bad thing really.

    In fact android has essentially killed embedded Linux on smartphones along with JavaME. We’ll see if Samsung and Nokia can resurrect it. Personally, I think Samsung is better off not merging Bada with Tizen but that’s probably because I favor bsd.

    And yes, few want to run gnome on OSX although I assume it’s probably as common on Darwin as other BSDs. On the other hand it still highlights that you don’t know what you are talking about regarding unix and loath to admit you’re wrong about the most trivial things.

    If you’re going to make snide comments it’s best to be right.

  197. @joncb

    Froyo users aren’t just missing out on NFC but improved cut/paste, power management, skins, keyboards and new features that might limit access to some apps like gyro support an sip (games and VoIP apps that want to use the gyro and SIP)

    There are probably a good number of bug fixes too.

    Are the users noticing? Some are

  198. Which will come first: Google Motorola taking 1/3 of Samsung’s market share or Samsung forking Android with its own Market?

  199. @Nigel
    So, linux gets redefined to exclude android. How many legs does a dog have if you call a tail a leg?

    And I was perfectly aware you can run X and Gnome on OSX. Just as it can be run on Windows. And I know that OSX is “unix”. Both of which have nothing at all to do with the original point.

  200. @winter

    So, linux gets redefined to exclude android. How many legs does a dog have if you call a tail a leg?

    It should not be very controversial to assert that when most people refer to Linux they mean the GNU/Linux OS.

    It should not be very controversial to assert that the Linux OS family is a unix OS with many distros.

    It should also not be very controversial to assert that Android/Linux is not part of the GNU/Linux family despite having a kernel based on Linux.

    If you accept those statements it should also follow that the assertion that calling “Android” “Linux” will confuse most people into assuming things they shouldn’t about Android (like that Android is a Unix or that Android belongs to the Linux OS family) isn’t all that controversial.

    This is the point that David, Tom and myself have been making.

    And I was perfectly aware you can run X and Gnome on OSX. Just as it can be run on Windows. And I know that OSX is “unix”. Both of which have nothing at all to do with the original point.

    It has a lot to do with this statement you made to Tom and David who made those non-controversial points above:

    @Tom&David
    “Android is barely Linux”

    linux is not like OSX, you can actually exchange the UI. You simply do not understand how *nix works. That is your problem, not Androids.

    The first sentence is incorrect, you can exchange the UI. If you were perfectly aware of this then why did you make this statement?

    The second sentence states that Tom and David don’t understand what a Unix is. They seem to while you don’t. OSX is a Unix. GNU/Linux is a Unix (I have every belief that RH could get linux to meet the SUS if there was a compelling business case). Android is not Unix nor even *nix. That you don’t seem to understand that is your problem, not Androids.

  201. @Nigel

    Most people consider a hacker a criminal. Most people will not know what GNU/Linux is.

    So, you all know perfectly well what Unix is? Then you should know that Linux refers to the kernel, and Android also uses that kernel. So you know perfectly well that refering to Android and “not linux” is a malicious ploy with words.

    “Do not attribute to malice what can be explained by ignorance”. As you just explained you are all not ignorant, I must conclude the misattribution was malice.

    I admit I was wrong, it was not ignorance about Unix that caused your misattribution, it was malice.

    And I would love see (or hear about) a person who is able to replace the standard UI in OSX at boot time, e.g., Gnome or KDE. I have no idea how that can be done.

  202. FWIW, I’ve come to the conclusion that Honeycomb isn’t actually a meaningful release and it’s unfair to think of it as a major version. Up until ICS was released, 2.3.7 was the most recent relevant Android version. This gives us a clearer picture of the degree of Android fragmentation. Mind you, that 30-odd percent on 2.2 is pretty bad compared to iOS 4 adoption right before iOS 5 was released…

    It also means that Android hasn’t had a significant update from the user perspective for over a year. Remember those thrilling days of yesteryear when Android was going to win because the Android OODA loop was tighter? And yet both Apple and Google are, for the most part, on a nice steady one year cycle for major OS releases. (iOS 5 took more than a year to hit consumer devices as well; no advantage to Apple there.)

    There’s a potential perception issue here for Android due to the perceived lag time between OS release and handset availability. Apple avoids this not because they don’t have lag time, but because they don’t have to announce before the OS is available on handsets. It’s a marketing advantage for tightly integrated hardware/OS. I don’t know if it’s a huge one or not.

  203. @Winter

    You remind me of these guys who are trying to assert that the iPad is a PC. They insist that iPads have a lot of the features ‘traditional PCs’ have (they have to invent a new term for PCs so they don’t contradict themselves) so therefore they are PCs. They says things like ‘oh, just because it doesn’t have a keyboard, that means it’s not a PC? That’s arbitrary!”

    This misses the point that the ‘PC’ is a largely arbitrary set of attributes. ‘PC’ refers to a form factor that includes a vertical, usually non-touch, screen, a keyboard, a mouse, and a desktop-like OS.

    The iPad is a fundamentally different thing, and to throw it in the same bucket with ‘PCs’ means we are missing a big change that is happening with ‘personal computing’ (as a broader subject).

    Likewise you are trying to throw Android and ‘Linux’ into the same category, and in doing so you are losing a valuable distinction between the two.

    To say that Android *is* Linux, and that therefore to speak of Android potentially taking away share from Linux is meaningless, has the consequence of rendering us less able to speak about an important shift that is occurring in computing. That shift is away from PCs and towards a new generation of touch based mobile computers.

    (note, I am not actually saying that Android is going to take share away from Linux, as I explained earlier)

    To lump Linux in with Android, and the iPad in with PCs, on the basis of a few technical common elements, means that we are less able to distinguish between things that need to be distinguished.

    Imagine a zoning administrator who was reporting on the use of land in his area. He is being told that commercial units are on the increase, and they are taking land away that was previously used for residential homes. Instead of duly reporting this, and using the information to better plan for the change, he just replies ‘but what does it really mean that commercial is taking over residential? They’re all buildings aren’t they? They’re all made of bricks and mortar aren’t they? They all shelter people from the elements. Just write down that there are more buildings, and let’s move on.’

    This guy is being too reductionist, and in doing so is missing a real change that is happening. That is exactly what you are doing.

  204. Dell has both consumer and enterprise products. Of course, consumers are allowed to purchase enterprise products, and many recommend that they do because (at least in the past) the build quality and the support was better. Of course, the consumer paid a bit more, but it is hard to make a like-for-like comparison. I think the fear is that large manufacturers with good economies of scale will try to keep consumer sales separate from enterprise sales managing their razor-thin profit margins on consumer sales by (1) maximizing Microsoft marketing incentives and (2) minimizing their support costs. If they think UEFI signed boot gets them there, they will go there. If you want Linux, you will pay more, either for an enterprise-grade product, or to a white-box builder, or an outfit like System76. This will further marginalize general-purpose computing.

  205. Most people consider a hacker a criminal. Most people will not know what GNU/Linux is.

    Exactly. Most people won’t know that Linux is the kernel part of GNU/Linux. To everyone Linux is Linux the OS. Man I hate agreeing with RMS but your persistence in conflating the two has convinced me he has a point.

    So, you all know perfectly well what Unix is? Then you should know that Linux refers to the kernel, and Android also uses that kernel. So you know perfectly well that refering to Android and “not linux” is a malicious ploy with words.

    Actually I would have said that conflating Android WITH Linux was a ploy on words. If all you are concerned about is the use of the Linux kernel why is Unix part of the discussion at all? The Linux Kernel is no more Unix than the Mach kernel. And again, Android isn’t UNIX.

    “Do not attribute to malice what can be explained by ignorance”. As you just explained you are all not ignorant, I must conclude the misattribution was malice.

    I admit I was wrong, it was not ignorance about Unix that caused your misattribution, it was malice.

    Actually, it’s the fact that you keep thinking that Linux the kernel is Unix that is causing you heartache. So the root cause is ignorance…on your part.

    And I would love see (or hear about) a person who is able to replace the standard UI in OSX at boot time, e.g., Gnome or KDE. I have no idea how that can be done.

    Well you do it the same way you do on Linux…start with the bare unix OS and park a new desktop on it. In other words, start with Darwin…running OSX without it’s proprietary core libraries and you’re essentially down to Darwin. None of the Cocoa based apps would work in Gnome or KDE anyway. You have to do it all yourself as I think PureDarwin died sometime in 2010.

    That there isn’t any interest in doing so doesn’t mean it can’t be done.

  206. @larry weirdly in this bastion of the free market profit is unimportant. I assume the same for share price…

  207. weirdly in this bastion of the free market profit is unimportant. I assume the same for share price…

    Oh profit is always important, but in this case (what’s at stake here in the so-called Smartphone Wars) it’s a means not an end. But you knew that, right?

    The people you’re trying to troll with this remark aren’t bluenose scolds, and they aren’t eaten up with envy. That’s generally projection talking.

  208. @nigel:
    “cathy some folks might argue that the ms model isn’t necessarily the one folks should emulate :)
    Fragmentation is generally not a good thing and has certain costs on both users and developers. That some products like office can get away with it doesn’t imply it’s not a serious issue AND ignores the great effort on MS’ part to provide converters and minimize the impact.”

    I’m not disputing that fragmentation can sometimes be a bad thing. What annoys me is the use of language; I always hear people say that Linux is fragmented because there are so many distributions, but those same voices never acknowledge that Windows is equally fragmented, if not more so.

    However, the differences between Froyo and Gingerbread are small enough that I don’t consider them important to users.

  209. Cathy, wouldn’t you say that fragmentation matters somewhat less when the scale is large? I think Windows is so huge that it can afford to be more fragmented than Linux.

    However, the differences between Froyo and Gingerbread are small enough that I don’t consider them important to users.

    OK, but those are both at least a year old. What will matter to users: is their phone upgradable to Ice Cream Sandwich or even Honeycomb? Will it run the new app they want?

  210. “Oh profit is always important, but in this case (what’s at stake here in the so-called Smartphone Wars) it’s a means not an end.”

    Yes, a means to generate network effects and a defense against disruption.

  211. Cathy,

    Linuux is fragmented in space as well as time. Unless you STATIC LINK ALL THE THINGS! there is no way of guaranteeing that any binary you ship will be compatible with any given Linux system, not just because of the variety of libraries a given Linux distro may or may not have, but how those libraries change over time. A lot of binaries fail because they were linked against a specific version of libstdc++ for instance. The only solution, it seems, is to make your code open source and hope the distros will pick it up. Not an option for most vendors; they’ll just stick with shipping for Windows and, increasingly, Mac.

    Until the Linux community makes easy shipment of proprietary software a PRIORITY Linux will suck hind tit on the desktop.

  212. @Jeff – Linux is GREAT with proprietary software! You compile it against Win32 and run it in Wine.

    (I have observed what Jeff describes. Proprietary binaries for old versions of Linux are all but impossible to get working except on the original target system in a VM. Old Win32 binaries work like a charm in Wine. Wine is really very good these days.)

  213. > Yes, a means to generate network effects and a defense against disruption.

    I don’t anticipate many here understanding just what $100B-in-the-bank means for their hopes and dreams.

  214. I don’t anticipate many here understanding just what $100B-in-the-bank means for their hopes and dreams.

    Of ever receiving a dividend? :D

  215. Yelnick,

    Yes we do. A lot of us lived through the Microsoft dominance of the nineties.

    Difference is Apple has the finesse to make their empire a pleasant place for users and vertical developers. Only the openness-über-alles geeks lose out.

    Not all authoritarian regimes are evil. Singapore is business-friendly, virtually crime-free, and smart people are elected to public office.

  216. > Singapore is business-friendly, virtually crime-free, and smart people are elected to public office.

    “Disneyland with the death penalty”.

  217. Unless you STATIC LINK ALL THE THINGS! there is no way of guaranteeing that any binary you ship will be compatible with any given Linux system, not just because of the variety of libraries a given Linux distro may or may not have, but how those libraries change over time.

    Woah, are you serious? Linux still doesn’t have a comprehensive dynamic library solution like the Mac and even Windows do?

  218. AAPL closed @ $502.60

    It’s been a great week for trading AAPL March $ 450 calls.

  219. And I would love see (or hear about) a person who is able to replace the standard UI in OSX at boot time, e.g., Gnome or KDE. I have no idea how that can be done.

    It’s been done. Back when Darwin was released, some people got X-windows running on it. Of course, that effort died from lack of interest because it was pointless thing to do.

  220. @Some Guy

    It’s been done. Back when Darwin was released, some people got X-windows running on it. Of course, that effort died from lack of interest because it was pointless thing to do.

    Indeed. I can’t understand why this was raised as a question. ‘Aqua’ is just like any other desktop environment on *nix. It could be replaced. But nobody is interested because there is nothing better out there (for most people, anyway). Half the point of using a Mac is the desktop environment.

  221. Nigel and Winter,

    Stop insulting each other. You are both incapable of communicating your valid points to each other in a way that compels agreement with your repsective final conclusions. This is the human condition. Welcome to the monkey house.

    Nigel,

    Android has an app which gives it a *nix command line. Since I use *nix with SSH terminals and vi and never with any GUI, Android running that app installed is *nix to me. YMMV.

    Winter,

    And when it’s not running that app, Android is not *nix to me. YMMV.

    Yours,
    Tom

  222. @David Gerard: “yeah, you can do that with Cygwin too … Windows still isn’t Unix.”

    I don’t see anyone saying that OS X is Unix because another UI can be added on top. I only see Winter saying:

    “linux is not like OSX, you can actually exchange the UI. You simply do not understand how *nix works. That is your problem, not Androids.”

    With the inference (although I wouldn’t make it) that it is NOT Unix because the UI canNOT be replaced.

  223. @tom What valid point? The “points” I was discussing was snarky statements were both rude and wrong. Rude is tolerable. Wrong is tolerable. Both together annoys me. Yes, it’s not polite of me to rub his nose in it. I was disinclined to be polite about it the last couple days. Other days I would have let it slide but it still doesn’t make his point valid.

    I see David has pointed out that Cygwin doesn’t make windows a unix. He is correct. You are wrong. This isn’t a very subjective issue and not a question of mileage.

    And humans aren’t monkeys nor have I insulted him much yet. Actually, where have I insulted him? By saying he’s the one ignorant about Unix and not others? Please.

  224. Nigel,

    > Both together annoys me.

    Are you claiming not to have been both rude and wrong? Since both of those are in the eye of the beholder, maybe winter is annoyed at you.

    > I see David has pointed out that Cygwin doesn’t make windows a unix.

    Sure thing. For my co-worker who uses Cygwin, Windows is not technically a *nix. However, for his purposes, it is. Installing Linux in a virtual machine was too difficult, but Cygwin fit the bill. He does not care whether it has a *nix kernel. He wants his shell scripts to run.

    > And humans aren’t monkeys

    Welcome to the Monkey House is a short story by Kurt Vonnegut.

    > Actually, where have I insulted him?

    Calling him ignorant because he doesn’t agree with you.

    > Please.

    Insults are in the eye of the beholder too. If I think you writing in an insulting fashion, you are. Sounds like you were a little insulted by my monkey house comment, in spite of my intentions.

    Yours,
    Tom

  225. P.S. It’s important to note, Nigel, that you and Winter actually seem to try pretty hard to make valid points and not troll, and that is why I’m playing peacemaker. You are both worth it.

  226. Yes, I got the reference. I never cared for Vonnegut…possibly because of the insistence that he and Mary Shelly were the only SF writers worthy of study by my english lit professor who went on to dis Heinlein. Irrational, yes. Either way, I don’t agree with the sentiment.

    I called him ignorant not because he disagreed with me. He’s done that lots and that’s not an issue. I called him ignorant of unix because it’s true if he can’t determine that Android isn’t a unix and is going to be belligerent about it.

    I have been wrong many times. And rude quite a few times. Occasionally both. Not this time.

    It is not a technicality. Stating that Windows is not technically a Unix is like stating that a cat is not technically a dog but for your co-worker they are for his purposes the same because they are both furry when he pets them. And no, I don’t care if both are posix compliant either.

    But yes, I’ve been a bit grumpy lately and I’ve browbeat him enough on this subject. In fact, as near as I could tell it was over until you decided to play peacemaker.

  227. Nigel,

    > Stating that Windows is not technically a Unix is like stating that a cat is not technically a dog but for your co-worker they are for his purposes the same because they are both furry when he pets them.

    No. Cygwin is much more functionally equivalent to *nix than your fur example.

    > And no, I don’t care if both are posix compliant either.

    POSIX compliance is several orders of magnitude more similar than your fur example.

    > In fact, as near as I could tell it was over until you decided to play peacemaker.

    Yeah. I wasn’t just playing peacemaker. I am also exploring who defines what *nix is and what is the definition.

    Yours,
    Tom

  228. @Tom I meant that I don’t care if both dogs and cats are posix compliant.

    As far as the definition of Unix, the owners of the trademark have a specific definition. Folks will stretch that to be anything you can reasonably expect to pass the SUS requirements. Meaning Linux, the BSD, etc.

  229. @cathy One potential impact of android fragmentation for consumers may be fewer apps.

    GameHouse also has to spend an extra two months working on an Android game, vs. what’s needed for an iOS title. That lengthens the time it takes to get it to market by about 30 percent, he says.

    “It’s nowhere near as simple as iOS,” Murphy says.

    The sheer variety of Android devices is one complication. GameHouse has to tweak its games to account for variations in accelerometers and responsiveness in more than 550 different Android gadgets, Murphy says. Apple, by contrast, has just a few models.

    Bill O’Donnell, general manager of mobile products at the travel site Kayak.com, says his company isn’t able to test its software on every variety of Android phone. There are just too many of them.

    “It puts developers in a tough spot,” he says.

    Sometimes individual models require many extra changes. With Amazon.com’s (AMZN) Kindle Fire tablet, two Kayak engineers had to spend a month and a half changing all the company’s apps to work with Bing Maps rather than Google Maps, O’Donnell said. Despite being an Android device, Kindle Fire didn’t come with Google apps preloaded.

    “That was a huge pain,” he says.

    http://www.businessweek.com/technology/android-may-be-losing-ground-in-the-app-war-02142012.html

    Mmm, this is the first time I’ve read about any noticeable impact regarding the Fire fork. Makes sense though.

  230. > As far as the definition of Unix, the owners of the trademark have a specific definition.

    Shhh!

    The volume UNIX platform: MacOS X.

    Not that anyone here would notice that.

  231. “With the inference (although I wouldn’t make it) that it is NOT Unix because the UI canNOT be replaced.”

    I see where this discussion went wrong. As usual, the cause seems to have been me being not perfectly clear (in words and in what the others were writing).

    I simply assumed that those commenters insisting Android is not Linux made the incorrect inference that, because OSX does not allow different UIs to coexist, different UIs must indicate different OSs.

    I now know that I was wrong. The original commenters simply did not want to admit that Linux advanced. Simple spite.

    And I am perfectly aware that OSX is a UNIX derivative/clone/descendant, albeit a locked down one (I have no idea about iOS in this respect). Windows is not “UNIX”, although you can simulate a UNIX system. Just as you can simulate Windows with Wine. Windows Posix compliance is also just a simulation, not a hook into the kernel. Any Turing complete computer can simulate a UNIX system, so that is not a criteria. As usual, a simple side remark about OSX was made the focus of the discussion.

    Whether or not Android can be called a “UNIX” system is a whole different discussion.

  232. @david Interix is Unix sitting atop the NT kernel sorta parallel to the host Windows OS if I remember right.. The kernel doesn’t matter much given Unix has been built atop monolithic and microkernels. Okay XNU is more of an in between than a true microkernel since it’s based on an earlier mach.

    You could get Android to pass the SUS if you park a whole userland on top of it. More likely it, like Interix, would run in parallel to each other with both Dalvik and the unix userland directly atop the kernel.

    But Android today isn’t unix.

  233. I see where this discussion went wrong. As usual, the cause seems to have been me being not perfectly clear (in words and in what the others were writing).

    I simply assumed that those commenters insisting Android is not Linux made the incorrect inference that, because OSX does not allow different UIs to coexist, different UIs must indicate different OSs.

    How many times does it need to be stated that Android isn’t Unix and that Linux the OS is different from Linux the kernel?

    Everyone agrees that Android is Linux kernel based.

    People disagree that Android is part of the Linux OS family. OSX has nothing to do with it.

    I now know that I was wrong. The original commenters simply did not want to admit that Linux advanced. Simple spite.

    ROFL. Right, Linux is more advanced than OSX because it can have multiple desktops. Whatever floats your boat. Gee, can I claim that FreeBSD is more advanced than both because it has both multiple desktops and ZFS?

    From my perspective as a long time unix weenie we always had multiple desktops that all sucked and that was a negative of the ecosystem. One desktop that doesn’t suck as much is much preferable and we finally got that in NextSTEP. NextSTEP was a huge advancement but unfortunately locked into Apple’s ecosystem. The Unix world standardized in CDE instead. Which was major suck. I blame HP for a crappy UI and Sun for caving.

    Gnome/KDE/etc? Not so much an advancement. Wayland? Yes, the next big advancement for desktop unix UI and should benefit everyone except Apple.

    And I am perfectly aware that OSX is a UNIX derivative/clone/descendant, albeit a locked down one (I have no idea about iOS in this respect).

    Nope. OSX is a UNIX period as it passed UNIX 03 certification. Not derivative/clone/descendent. It’s among a handful of unixes that can actually make that claim. At least for Leopard. I don’t know if they bothered to keep certification going.

    As usual, a simple side remark about OSX was made the focus of the discussion.

    If you don’t want to derail your own point refrain from snide comments eh?

    Whether or not Android can be called a “UNIX” system is a whole different discussion.

    Actually it IS the discussion when someone states that Android would replacing Linux on the desktop and you disagree. It is the replacement of a unix OS on the desktop with something not unix. That they have the same kernel is immaterial. Android and Linux are not functionally equivalent nor do they belong to the same family of OS so if Ubuntu’s market share goes down as Android goes up then the market share of the Linux OS drops.

  234. @Nigel
    “Right, Linux is more advanced than OSX because it can have multiple desktops.”

    “Linux advanced” (verb), as in, “Linux progressed [in the world]“. The rest is you projecting your feelings onto me.

    @Nigel
    “How many times does it need to be stated that Android isn’t Unix and that Linux the OS is different from Linux the kernel?”

    So what? “Android is Linux” does not imply “Android is Unix”. This is again you projecting your feelings onto me (mirroring).

    Linux the OS does not exist. There is Linux the kernel. Colloquial language use is to call all OS’ that use Linux as a kernel, “Linux OS'”. You excluding some distributions based on the Linux kernel from “Linux the OS” is simply a rhetorical trick to fool people.

    I have to work on my reading skills, I know. But you could apply some more effort to separate your wishes from your thinking.

  235. Chaperoning 8th graders from Austin to San Antonio.  Three museums in 6 hours.  iPhone’s (esp iPhone 4/4s) everywhere. 

    I’ve spotted two candy bar phones, and a lone Android phone (in the hands of a teacher).  

    These aren’t the “rich kids”, they’re the smart kids, attending a public “magnet” school, and all trying to get into LASA, the #3 public high school in the nation.  

    But Android is winning. 

  236. “Linux advanced” (verb), as in, “Linux progressed [in the world]“. The rest is you projecting your feelings onto me.

    Given that Linux had multiple desktops very very early on this statement isn’t very true either. Like most folks around here I have a CD ROM from the mid-80s with Linux on it and I’m pretty sure I could run a couple different window managers/desktops even then.

    You feel I’m projecting. I feel you’re spinning and trying to dig yourself out of a hole without admitting folks that disagree with you fully understand Android, Linux and Unix.

    Linux the OS does not exist. There is Linux the kernel. Colloquial language use is to call all OS’ that use Linux as a kernel, “Linux OS’”. You excluding some distributions based on the Linux kernel from “Linux the OS” is simply a rhetorical trick to fool people.

    Linux the OS sure as hell exists. I’ve paid a good amount of money to buy the enterprise versions so my RedHat sales rep would disagree with you. :)

    From the colloquial use perspective it’s a very iffy assertion. For example, Android isn’t tracked on Distrowatch as a linux distro. Neither mainline Android nor Android-X86. Folks call all operating systems based on Linux plain “Linux” because until Android essentially all operating systems based on Linux were GNU/Linux (aka Linux the OS or plain ol’ Linux).

    If you really think Android is what most folks consider a linux distro or that Linux the OS doesn’t exist then I suggest running over to Wikipedia and make the relevant edits. Good luck with that.

    You should also mention that to Google:

    “Although Android is built on top of the Linux kernel, the platform has very little in common with the conventional desktop Linux stack. In fact, during a presentation at the Google IO conference, Google engineer Patrick Brady stated unambiguously that Android is not Linux.”

    http://arstechnica.com/open-source/reviews/2009/02/an-introduction-to-google-android-for-developers.ars

    Have they ever publicly retracted that statement? I remember it was a big ballyhoo for some folks.

  237. It’s worth noting for all ya’ll arguing as to whether Android is or is not Unix — there’s an Android app called “Linux Installer” that works on rooted phones. It basically snarfs down a base Debian distro and installs it in a chroot. When it’s finished doing its work you can log in with ConnectBot and run conventional ARM Linux software in this environment.

    So Android isn’t Linux in the sense that Debian is. But Debian is not hard to fit on top of Android.

  238. “Disneyland with the death penalty”.

    Indeed.

    OS X Mountain Lion is shipping with a feature called Gatekeeper. It’s basically a default-allow app approval process:

    1) All apps must be signed in order to run.

    2) Apple can remotely revoke — and prevent the execution of — any app signed with a particular signature if it is found (in Apple’s sole discretion) to be malicious.

    So even Mac OS X is turning into Disneyland with the death penalty.

    Not even Android is really about freedom. Not yours, anyway. Android was designed, from the get-go, to put manufacturers and carriers in the drivers’ seat. To allow them to “customize” (read: load with crapware) the user experience. All the crap that we complain about with Android: the unremovable NASCAR apps, the crufty HTC Sense themes, the updates dribbling out once every two years if that, all of these are what Google deliberately set out to enable with Android.

    And, contrary to Eric’s suggestions above that I’m imagining some vast conspiracy by hardware vendors to force this unbidden down everybody’s throats, I believe just the opposite: this is what people want. When game consoles were totally open, the result was millions of copies of E.T. buried in a landfill in the desert. Along came Nintendo, with an expensive developer program and game-vetting process in order to be allowed to release for their console. The home game market fucking exploded.

    So no, the coming closed-platform revolution won’t be because of evil vendors. Good ol’ market forces will help it along substantially.

    Nobody gives a shit about software freedom.

  239. “Nobody gives a shit about software freedom.”

    This is, sadly, true. No-one cares about software freedom, even freedom 0, unless and until it bites them personally and individually in the arse. It’s like backups in that respect.

  240. @Jeff

    …or you could toggle it to run any app even if unsigned.

    In effect, it offers all the security benefits of the App Store, except for the process of approving apps by Apple. Users have three choices which type of apps can run on Mountain Lion:

    Only those from the App Store
    Only those from the App Store or which are signed by a developer ID
    Any app, whether signed or unsigned

    The default for this setting is, I say, exactly right: the one in the middle, disallowing only unsigned apps. This default setting benefits users by increasing practical security, and also benefits developers, preserving the freedom to ship whatever software they want for the Mac, with no approval process.

    Call me nuts, but that’s one feature I hope will someday go in the other direction — from OS X to iOS.

    From Daring Fireball.

  241. App security doesn’t have anything to do with software freedom…Apple’s solution here strikes me as a good balance.

    And sure, you can run Debian on top of Android. Just like you can run X on OSX. But if Android takes over on the desktop the big loser is still GNU/Linux.

  242. > 1) All apps must be signed in order to run.

    By default, but you can turn it off.

  243. @larry My bad…mid 90s :)

    I still have my first linux box in my basement somewhere. A Dell Dimension P90 running slackware. I wonder if it still boots. Something amusing to try tonight I guess if I can find a VGA monitor cable. Good lord I paid a lot of money for that machine.

  244. True, Nigel, thanks. And that is a sensible solution.

    But I get a feeling this is a camel’s nose in the tent, especially given Apple’s attitude towards legacy apps. Sure, they’ll be supported in the next release, but they may not be supported in the release after that, because fuck you, that’s why. If you don’t believe that, try running a PowerPC-compiled app on a Lion Mac.

  245. @Larry Yelnick:

    Chaperoning 8th graders from Austin to San Antonio. Three museums in 6 hours. iPhone’s (esp iPhone 4/4s) everywhere.

    I’ve spotted two candy bar phones, and a lone Android phone (in the hands of a teacher).

    Sure.

    These aren’t the “rich kids”,

    But they’re well-off enough, and mostly from stable family situations where the credit is good enough that their parents don’t have prepaid, and as we’ve been discussing, the incentives the carriers provide for postpaid actually heavily subsidize Apple. The question is whether or not that will keep occurring. I think it will, at least for awhile, and the major disruption will be facilitated by defections to prepaid.

    they’re the smart kids, attending a public “magnet” school, and all trying to get into LASA, the #3 public high school in the nation.

    Yeah. Both my daughters went to LASA. One’s a senior at UT now, and one’s in medical school (in San Antonio, coincidentally). And they both carry iPhones. (And my wife and I carry candy bars.) What you’re saying here is nothing new. If you want to look at the demographics of who uses Android, you could do worse than to look at some of the Pew reports.

    But Android is winning.

    Yes, it is, so far anyway. Those may not be the “rich” kids, but they’re rich enough — shoot, you live in a rich town. And if (as your anecdote suggests) those kids are magnet students from Kealing or maybe Fulmore, they’re mostly white, to boot. Do you really believe that Nielsen, ComScore, Pew, and all the companies that work really hard to try to count phone shipments are all completely incompetent and that your anecdote trumps that?

  246. @jeff

    Maybe. It strikes me that if you believe Jobs had a vision where Macs are pickup trucks and iPads are cars that if they keep making pickup trucks at all (a debatable assertion but probable) then the pickup trucks will always be able to haul unsigned loads of bricks…or it wouldn’t be a very good pickup truck.

    That I can’t run PPC apps on Lion isn’t a surprise. That they managed that transition as well as they did is. Could they have maintained Rosetta longer? Perhaps. But the transition was six years ago. I’m still running Snow Leopard on some machines (oh no, fragmentation!).

    Apple isn’t MS in fanatically supporting legacy. That’s probably a good thing.

  247. @Jeff Read:

    But I get a feeling this is a camel’s nose in the tent, especially given Apple’s attitude towards legacy apps.

    Could be, but even I am willing to give Apple the benefit of the doubt on this one. And if they keep giving out free developer keys, and if the signing mechanism is easy enough that anybody could sign any app, then there are no real issues here. Apple’s problem with legacy is probably more of a support burden/cost issue than anything else. I’ve actually been amazed at how well Apple has managed the significant hardware transitions on the Mac.

  248. @Patrick:
    “Do you really believe that Nielsen, ComScore, Pew, and all the companies that work really hard to try to count phone shipments are all completely incompetent and that your anecdote trumps that?”

    Which of course is why lots of $$$ are spent doing well-designed, carefully-structured research, rather than relying on anecdotes.

    Larry’s story reminds me of the famous Jules Feiffer cartoon from 1972. It showed two academics, one of whom said to the other, “How could Nixon have won? I don’t know anyone who voted for him!”

  249. @Cathy:
    “Larry’s story reminds me of the famous Jules Feiffer cartoon from 1972. It showed two academics, one of whom said to the other, “How could Nixon have won? I don’t know anyone who voted for him!”

    Which, itself, seems to be an anecdotal story / urban legend, attributed to many different speakers…

  250. Tim Cook talked about pre-paid markets the other day. Not much real meat there, other than confirmation that he’s aware of the issue. He did say that China Unicom tried post-paid at Apple’s suggestion and it worked very well, but without numbers I’m not sure what that means. I guess we’ll see if anyone else in China gives the model a try.

  251. The Nixon statement is commonly attributed to Pauline Kael, but may well be apocryphal.

  252. Man, if MacWorld is correct, Gatekeeper is really weak security. Once you open an app for the first time, you clear the quarantine key and you can always open that app — even if the developer is added to the developer blacklist. That’s highly unoffensive for me.

    I am alarmed that some of the new APIs (iCloud, Notifications) may be restricted to App Store apps. That is a trend I dislike a lot.

  253. @Bryant:

    Interesting quotes from Tim Cook. Thanks for the link.

    Last year, as you know, we covered price points in the subsidized markets from zero on up. And of course, that doesn’t translate to 0 to prepaid markets. But it does translate to lower in the prepaid market, and so we’re covering more price points there.

    That implies some segmentation, either geographically or by functionality. If “we’re covering” means currently, it’s geographically and he’s not really discussing the US. This is the most likely meaning, as the following quotes seem to make clearer.

    But I would come back to the paramount thing is the product. It is the focus. And of course, distribution, we’ve recognized the differences there. We’ve recognized the difference in purchasing power.

    It will also be interesting to see how, in the prepaid markets where he charges less, he manages to keep the arbitrageurs at bay. There are plenty of folks who would love to bring in a few boatloads of cheap real iPhones and make $10 a phone, and probably at least one carrier who wouldn’t mind if you attach an iPhone that wasn’t bought inside the US. I’m sure there will be technical measures, but that runs counter to the idea of reducing the number of SKUs, and winds up being exactly the same mindset that leads to DVD region coding.

    And, by the way, unlike probably many people, I don’t subscribe to the premise that a prepaid market is a prepaid market is a prepaid market.

    He doesn’t seem to be concerned about prepaid getting more popular in traditional postpaid territory. He’s more interested in converting prepaid territories into postpaid territories. This may be the right thing for him to focus on, but for all the guys here talking about how Apple is all about taking power from the carrier and giving it to the consumer, I gotta say this emphasis on “look — here’s how you can lock the customer in for an extended period of time! (and btw, we can sell our phone for better profit)” runs the wrong way.

  254. > And if (as your anecdote suggests) those kids are magnet students from Kealing or maybe Fulmore, they’re mostly white, to boot.

    There were brown kids on the bus. Not many, but more than a few. Still, mostly white, as you surmised.

  255. @patrick I think that the statement “we’re covering more price points” means exactly that. The lower price points are covered by the iPhone 3GS at pretty much the same price across the world. Most price oddities geographically are due to currency exchange rates and VAT.

    The iPhone 3GS in India appears to be Rs.20599 or around $420 USD.

    http://www.mobilepriceindia.co.in/price-list/apple

    In the UK it is £319.00 or $519 but that includes VAT. The cost of the phone itself is £271 or around $430.

    http://store.apple.com/uk/browse/home/shop_iphone/family/iphone/iphone3gs

    You aren’t going to have the scenario where it is profitable to ship in iPhones from a different region unless there’s some large disparity in the current exchange rate and when Apple set it’s prices. Sometimes it is a significant amount but it’s not Apple really setting a different price for a different market as much as Apple not wanting to reset prices very often.

    This seems to be true historically as well:

    http://www.iphonehail.com/list-of-countries-selling-factory-unlocked-iphone-4-333.html

    The lower price of the 3GS after the release of the 4S allows Apple to be more price competitive in pre-paid markets and folks seem to prefer the 3GS to many other options at that price. It will be interesting to see if the 3GS survives the launch of the iPhone 5 or whatever it’s called. The lower resolution panel and now much lower end ARM should make it very price competitive by then and still support iOS 5.

  256. @Nigel:

    You could be right — I may have been reading more into his price points statement than was really there. Still, it was pretty obvious that he thinks that postpaid is a better deal for Apple than prepaid…

  257. @Larry Yelnick:

    There were brown kids on the bus. Not many, but more than a few. Still, mostly white, as you surmised.

    I think LASA might have more minorities than the middle school (and there are a heck of a lot of blacks downstairs at LBJ).

    But after thinking about it a bit more, I think the situation is slightly more complicated than I first mentioned. In addition to the well-off people who can afford postpaid data (where iPhones are practically free), there will certainly be some others who have less money and/or different priorities. The thing is, though, that a kid who can make it to Kealing or Fulmore or LASA is going to be disciplined (compared to others in his peer group) and come from a disciplined family. So, for example, when my youngest was at LASA she had a cellphone. But it was a candy bar and you never would have seen it, because it was, in fact, prepaid, and was really just for making emergency calls. And, of course, a lot of kids with postpaid cells might not have data plans, either.

    I would guess that most of the kids with smartphones had iPhones, but that a lot of the kids didn’t have smartphones. I would also guess almost all the rest of the kids had cellphones (but no reason to take them out of their pockets).

    Finally, I would guess that practically all the kids had internet at home. (According to what I remember from a Pew report last spring, for a lot of minorities, the prepaid Android smartphone is the way they access the internet.)

  258. @nigel
    “Given that Linux had multiple desktops very very early on this statement isn’t very true either. Like most folks around here I have a CD ROM from the mid-80s with Linux on it and I’m pretty sure I could run a couple different window managers/desktops even then.”

    Who said something about desktops? I never did. And you are still projecting. “Advancing in the world” can be said of anything in any respect (read Dickens or Austen). I certainly did not have multiple-desktops etc. in mind.

    The assertion “A is (technologically) more advanced than B” always implies a direction in which the advances have carried A. You might have strong feelings about which directions matter (eg, those OSX or iOS are following). But I do not see one direction for technical advancement that clearly outweighs all others. Why you think multiple desktops should be the differentiator is beyond me. But if you feel that way, more power to you.

    So, to stop this silly discussion.

    If you claim Android is not UNIX, be my guest. There are several meanings of UNIX (eg, Trademark, Standard, Original AT&T codebase), and it seems Android is not UNIX in any of them.

    If you claim that Android will displace (kill) the current crop of Linux based distributions aimed at desktops, I won’t argue with that. That is a real probability and the future will show whether they become real or not.

    If you want to use a definition of “Linux” that fits your (and Google marketing’s) rhetorical needs to be able to proclaim “Linux DEAD”, you will end up with saying:
    “Linux is DEAD and replaced by, uh, linux”.

    Which I still consider a very silly thing to say. And if you go back, that was the whole meat of my comment.

    And do not hide behind word-use of people who consider a hacker a criminal, MS Office an OS, and microwaves a kind of nuclear fission fall-out. This is supposed to be a platform for people who think.

  259. Winter, nice evasion and use of strawmen. Kudos. Bonus points for being pretty much wrong on every one of your assertions and still getting in a snide comment that this supposed to be a platform for people who think.

    Evidently it’s also a platform for people who get by via using ad homs and pounding on the table.

  260. @nigel
    Lets say I was thinking the exact same about your comments, word by word. So we finally agree.

  261. Heh, Google being Google:

    “Google Inc. and other advertising companies have been bypassing the privacy settings of millions of people using Apple Inc.’s Web browser on their iPhones and computers—tracking the Web-browsing habits of people who intended for that kind of monitoring to be blocked…”

    http://online.wsj.com/article_email/SB10001424052970204880404577225380456599176-lMyQjAxMTAyMDEwNjExNDYyWj.html

    New Motto: Don’t be evil when we’re caught

  262. “New Motto: Don’t be evil when we’re caught”
    Yawn, see:

    https://www.eviscerati.org/comics/comic/hd/2012/01/cheerful-admission

    and the follow up story line. There are many more at “Help Desk”.

    I have always had the impression that Eric as well as every other person posting here believed Google acted from well-understood self-interest, and self-interest only. Obviously with the exception of those posters who proclaimed Google acted out of stupidity.

    Case in point, but still not news:
    Why Android will never be closed source
    http://www.itworld.com/mobile-wireless/250628/why-android-will-never-be-closed

    Businesses like Google (and Facebook and Yahoo and any other content-driven venture) have built empires on one really simple idea: let others do as much work as they can. For Google, every search result correction, every contribution to Google Plus, and every tweak on Google services is a way to improve their product line with minimal impact on their bottom line. Opening and giving away Android to any device vendor who wants to use the mobile operating system is just an extension of that strategy.

    …..

    Closing Android to go exclusively with one vendor would run completely counter to that success story, and suddenly put the much more of the burden of developing, marketing, and controlling Android right back on Google. Remember, at the end of the day, Google is about revenue through advertising and data–why would it suddenly want to become a mobile operating system vendor? Because, whether you think so or not, that is not what Google wants to be.

  263. Closing Android is simply another business decision made on the basis of whether allowing forks like the Fire or Yi is in Google’s best interest.

    Why would Google want to become a mobile operating system vendor? I dunno. Apple at $500/share? Apple makes a boatload more money than Google and if Google is going to enter tha hardware business as some folks think (home entertainment system) then closing Android isn’t off the table. With MMI they will be making smartphones and tablets too. Why let Samsung have the lions share of Android profits?

    As poorly managed as iAd has been it’s not really that far behind AdMob. 18% vs 24.8%.

    http://www.slashgear.com/apple-cuts-iad-pricing-amid-declining-ad-share-15213590/

    Can Google afford to let Samsung fork Android like Amazon did? It’s not like Samsung couldn’t find a willing partner in Microsoft willing to bend backwards to suit their needs. Or a three way with NHN covering Korea, Japan and China and Bing/MS covering US/Europe.

  264. @Nigel
    “Why let Samsung have the lions share of Android profits?”

    Just pure speculation, but maybe for the same reason Volkswagen does not mind letting others have the lions share of profits from gaz/petrol sales to Volkswagen drivers? Because it is not their Business?

  265. @ Winter

    “Just pure speculation, but maybe for the same reason Volkswagen does not mind letting others have the lions share of profits from gaz/petrol sales to Volkswagen drivers? Because it is not their Business?”

    Horrible analogy considering the profit margin on gasoline (at least in the US) is extremely small. Volkswagen DOES want to sell you VW brand floormats, bras, car covers and other profitable addons.

  266. @pinhead
    Distributing gaz is an entirely different operation from distributing parts. Just as selling ads has nothing in common with building handsets.

    Combining them raises costs without increasing revenue

  267. > I would guess that most of the kids with smartphones had iPhones, but that a lot of the kids didn’t have smartphones.

    Possible, and even quite probable.

    There were 97 8th grade students on the tríp, we took 4 busses. On my bus I counted 27 iPhones (3 iPhone 3s, the rest 4/4s models) in the student’s hands, and two more in the possession of the four adults (teachers or chaperones) on the bus. In my group of 10 students, there were 5 iPhones and one candybar. The twin boys didn’t seem to have a phone.

    I’m not sure I saw them all, of course.

    My point was that nearly all who had a choice of smartphone seem to have picked iPhones.

    And if that’s true, and it’s also true that in post-paid, lifetime carrier and acquisition costs for an Android phone are nearly identical to iPhone, how is it that Android is “winning”?

  268. Just pure speculation, but maybe for the same reason Volkswagen does not mind letting others have the lions share of profits from gaz/petrol sales to Volkswagen drivers? Because it is not their Business?

    A proper analogy has Volkswagen owning a faltering oil company as well. Google just made handset manufacturing part of their business.

    Why let Samsung make all the profits off Android at the expense of their own Motorola division?

    There are a several reasons but they become less and less compelling if Motorola handset and tablet P&L center records more and more losses. They could, of course, simply dump the handset and tablet portion. That should make shareholders happy.

    Given that the Fire is locking them out of the 7″ tablet market except for basic search and Silk is devaluing that by anonymizing the search somewhat AND they can switch over to Bing at any time I can see a time when Google stops funding open Android development in favor of something else.

    Or making it GPL and only selling a commercial license to close partners to lock out Amazon and Baidu from taking new features.

  269. @nigel
    Your advice to Googe is to take a failing handset maker and to use it to kill the industry that recruits Google’s customers?

    Sounds like the Advice Apple and MS would give to Google.

  270. Is Android always recruiting Google customers?

    Amazon Fire. No. Will dominate Android use on 7″ tablets.
    Baidu Yi. No. Will dominate Android use in China.
    B&N Nook. No. Will take large chunk of remaining 7″ tablet market left over from Fire.

    My advice isn’t to close Android (you need to read better).

    My assertion is that the business case to keep Android open isn’t a foregone conclusion.

    The primary scenario I can see to make Google change its development stance in some way (closed, GPL or abandon) is if Samsung forks Android and parters with NHN/Microsoft and locks Google out of the rest of Asia (given Baidu’s dominance of China) and makes a run for the money in Europe/US (using Bing/MS infrastructure).

    NHN has a strong search capability and gaming capabilities. It’s another one of those big Korean companies with reach and is the #5 search engine worldwide. GIven Yahoo’s troubles it’ll probably become 4th sooner or later.

    Would Samsung do such a fork? They are well positioned for such a thing and it makes them more competitive against Apple if they can control all aspects of the software, hardware and backend integration like Amazon does with the Fire. I’d say the odds are 1 in 5 they do so. Not likely but a significant chance they could.

    Samsung makes the lion share of Android handsets. That’s something Google has to respond to or its whole strategy goes out the window if Android is driving mobile ad and search revenue to some other company. AdMob is not the dominant force that Google is.

    Millennial Media is a close third. Want to bet that it has crossed Steve Ballmer’s mind to buy it?

    Google’s Android permissive license strategy using Apache might go down behind Sun’s disastrous open source strategy as an example of what not to do in order to successfully monetize open source. GPL would probably have been the way to go along with a free commercial license for Google Android Market partners. Leave our ecosystem then pay $x in royalties per handset.

    I can see this happening (moving to GPL) if Google does an Android II that fixes some of the Android architecture issues. Say merging Chromium with Android into some “new” thing with a “new” licensing scheme while keeping some backward compatibility with legacy Android apps.

  271. @nigel
    If Google’s partners try to betray them, Google can retaliate. If your spouse betrays you, you can get a divorce.

    Read the Decamerone to see what happens to people who distrust their spouses on a precautionary basis.

  272. Its rather comical to read advice from people who claim Apple iOS should rule that implies that Google should stop doing what made Android successful.

    Instead Google should repeat the mistakes that tanked WinPhone and Nokia.

  273. Business is generally more mercenary than marriages. Except perhaps those of the very rich…and there pre-nups seem to be the rule rather than the exception.

    Android’s success for Google isn’t market share but profit. Free market and all that. From that perspective Apple and iOS does rule. Depending on how much of the Motorola purchase to Android Google has spent billions on Android. How long before they break even?

    WinPhone, Nokia and RIM all tanked by not moving fast enough when Apple iOS massively disrupted them. Not by being closed source.

    If Open Source licensing was what made Android automatically successful then Sun would have bought Oracle and not the other way around. Instead it was a combination of many things but primarily very good execution by Google. The question is whether these early gains will end up profiting Google or someone else in the long run and how much some of the early short cuts will cost Google in the end.

  274. @nigel
    All obvious. But there is also an old advice not to wreck a winning strategy.

    Google will not spend more than they intend, I expect. And it is not they are losing money or marketshare.

  275. To date, Samsung is the only (non Apple) smartphone company *in the world* to earn sustained profits (of any size) off smartphones, and they’re a giant South Korean conglomerate.

    Android is winning? Really? No one making money doesn’t seem like “winning” to me.

    What if we are all wrong about Google’s rationale for acquiring Motorola? Some say it was for the patents. I say it was for the patents and to build feature phones with limited Google services — and only Google services, and also to help Google storm back into China. If they call these feature phones “Android”, will they count in the comScore marketshare numbers?

    And what if instead, the reason Google spent all that money on Motorola was simply to keep the platform afloat? To keep the OEMs from suing each other into the ground?

    Mobile ad revenues are comparatively a pittance for Google. There’s almost zero chance Google makes a *net* profit off Android during the entirety of this decade. Read that again.

    “Show me the money!” is a line from a movie. Motorola, HTC, Sony, Nokia and LG are extremely capable, advanced, global, high-tech companies… and they haven’t made DOLLAR ONE off smartphones since the launch of iPhone.

    Apple launched the iPhone and fundamentally changed the market. Everyone else was caught off guard, and are still trying to catch up. What if they never catch up? What if Android is absolutely the wrong strategy? How many years will these companies watch Apple generate billions in profits while they make nothing with Android? One? Three? Eight?

    In a few years, will the only company that offers Android devices be only Google, via their Motorola subsidiary?

    Perhaps the assembled here need to alter their thinking. Stop thinking that Android will eventually win because it’s a market share numbers game and pay attention to who is making money at this very expensive game.

  276. Oh, I think Android will be offered by many companies for quite a while, because it’s the cheapest option and the only thing they can offer that is anything like iOS. (Ignoring Windows Phone, as so many will.) But as you say, the profits are important. Apple plays a long game, and huge profits + R&D + design focus + hardware/software integration + economies of scale means that Android is doomed to play “me, too!” for the foreseeable future, regardless of their market share.

    And how long will that marketshare grow? With lower user satisfaction, fewer app purchases, and handsets that quickly go obsolete and often can’t be upgraded, how loyal will the average Android user be to the OS? I suspect that many will switch to iPhones as soon as they can.

  277. >I suspect that many will switch to iPhones as soon as they can.

    Comparative userbase growth figures suggest that most of the switching is going in the other direction.

  278. From what I’ve read, userbase growth is coming from people buying their first smartphones. When it comes to their next one, though, the figures I’ve seen are that around 30% of Android users are thinking of switching to iPhones, and that around 36% of iPhone buyers are coming from Android/Blackberry/Windows.

  279. There’s a difference between obsolescent and obsolete, btw. The rapid release cycles of Android handset makers means the average person’s phone may rapidly become obsolescent (meaning, no longer the latest but still useful) but certainly not obsolete (meaning, no longer effective or useful). It’s pure psychology, and I notice this ‘obsolete’ thing only being pushed in one direction, with selective blindness in the other. My G2 is one whole hell of a lot less obsolete than a 3GS, despite one being eol and the other still selling at retail.

    Oh and do those figures also list percentage of iPhone users thinking of switching to Android? Or percentage of Android buyers coming from iPhone/BB/Windows?

  280. Well, OK: obsolescent, then. But while my 3GS may be two generations back in hardware, it’s running the latest version of iOS. Are you running the latest version of Android?

    And sure, it’s “psychology,” but so what? That’s how people choose what they like. My point is that user satisfaction matters, and Android doesn’t come close to iOS in many ways.

    I wasn’t able to find any figures regarding your last two questions. I suspect we’d have heard if iPhone defection rates were anywhere near that high.

  281. The basic question remains whether Apple is willing and able to replace all 5b mobile phones in the world? And is Google cs?

  282. Why is that “the basic question”? I think the answer is clearly no. They only want to make the best smartphones and make money doing so. There will always be a market for other, low end phones.

  283. @Papayasf
    People want the Internet like they want mobile phones and TV or Radio.

    There is a market for 6 billion Internet access points. That makes 5 billion low cost smartphones. 4.5 billion of those will not be iPhones. So, who “wins”?

    ps, I expect Internet acvess to be done by cheap or free WiFi.

  284. Apple has sold a total of 316 million iOS devices to date, including iPads (55 million) and the iPod Touch (what looks like about 77 million), and they sold 37 million iPhones last quarter, so they’re on their way towards what you see as their upper limit. And given Apple’s engineering expertise and economies of scale, don’t write them off as unable to sell ever more affordable iPhones. (The first iPod was $399. Now the 2G iPod Shuffle is $49.)

    As for who “wins”: I’d say the side with the best devices, the happiest customers, and the most profit wins over the side with the greater numbers of crappier devices, less happy customers, and less profit to use toward R&D. But that’s just me.

  285. @PapayaSF
    “And given Apple’s engineering expertise and economies of scale, don’t write them off as unable to sell ever more affordable iPhones.”

    That is not the point. Apple does not make their own phones. There are Chinese factories that do that. Apple makes huge profits, not these factories. If the price goes down, at some point it will be more profitable for the Chinese to cut out the middle man (Apple). Actually, it most likely already is.

    I would even dare to predict that within three years, the Chinese will have their own branded mobile platform (phone/tablet) that competes with the iPhone on “cool”. The Japanese did exactly the same with products in the 1970s. At the end of the 1980s, Japanese bought Japanese products. At the end of the 1990s, the Korean bought Korean products. I expect this shift in China within the next 5 years.

    @PapayaSF
    “I’d say the side with the best devices, the happiest customers, and the most profit wins over the side with the greater numbers of crappier devices, less happy customers, and less profit to use toward R&D. ”

    That sounds a lot like saying that the athlete who gets the most sponsor money, the most cheers, and the most interviews wins the Olympics.

  286. Apple does not make their own phones. There are Chinese factories that do that. Apple makes huge profits, not these factories.

    Chinese factories assemble the phones. Core components do come from elsewhere. Even then Apple invests billions into the factories to be able to manufacture the phones. Some of this is moving elsewhere.

    Maybe if you count Taiwan as part of China.

    I would even dare to predict that within three years, the Chinese will have their own branded mobile platform (phone/tablet) that competes with the iPhone on “cool”.

    Right, in three years they will have a worldwide brand that can challenge both Samsung and Apple.

    Not even Lenovo has such branding power.

    The Japanese did exactly the same with products in the 1970s. At the end of the 1980s, Japanese bought Japanese products. At the end of the 1990s, the Korean bought Korean products. I expect this shift in China within the next 5 years.

    Not likely. I know a lot of Chinese folks and pretty much the sentiment is the same. You often have no idea what you’re buying. Chinese buy chinese brand either based on price or availability. Given the (economic) option they buy Taiwanese, Korean, Japanese or American products because it’s safer. You know, like milk, candy and other luxury items like that.

    Funny you should mention the olympics. Some folks bought heavily into condos in the Olympic village because they knew that the government wouldn’t be as likely to tolerate the standard crappy construction endemic to chinese building construction due to corruption. Those folks are making out like bandits from the resale value.

  287. @esr> “Gartner reports Android widened its lead over Apple in 4Q2011, says Android has 50.9% global market share”

    LOL. Nice try, Eric, but not even close.

    If you read the actual report, rather than the fanboy interpretation, you’ll see that Garter had Android in a decline sequentially over the previous quarter, where it had captured 53 percent of the global OS share. Gartner attributed this to the surge in demand for the iPhone 4S, which was reflected in the Apple’s 24 percent share of the smartphone handset market (up from 15.8 percent in Q4, 2010).

    Gartner expects Android to regain any lost ground over the January to March period.

  288. > Given the (economic) option they buy Taiwanese, Korean, Japanese or American products because it’s safer. You know, like milk, candy and other luxury items like that.

    One need look no further than the recent milk contamination events.
    http://en.wikipedia.org/wiki/2008_Chinese_milk_scandal
    or the pet food scandal from the year prior.
    http://en.wikipedia.org/wiki/2007_pet_food_recalls

    It’s obvious that ‘winter’ doesn’t have a clue about the situation on the ground in China.

  289. >LOL. Nice try, Eric, but not even close.

    The title of the article is “Android widens lead over Apple”. The 50.9% figure is direct from a table in the article. This is not “fanboy interpretation”, these are the facts being reported.

    I understand you are disappointed that Apple’s record quarter failed to change the strategic picture, but the facts are what they are.

  290. @nigel & Larry Yelnick
    The extend to which a prediction will come true can only be known after the fact. So I have no grounds to insist that it will come true.

    It is much more interesting to look under which circumstances it might come true. And whether these circumstances are likely today.

    Your well known stories of fraud and deception in China are paralleled by stories from Japan in the 1980s (See Van Wolferen, The Enigma of Japanese Power). This split between good export products and bad products for the home market was common in the Asian Tigers. It did not stop them.

    With a growth rate of 8-10% a year, in three years, Chinese GDP will have risen by well over 25%. That will impact domestic consumption. The Chinese government is very keen on steering that consumption to domestic products. They invest heavily in domestic production (and branding!). QA will be part of that stimulus package.

    Is see the quality problem solvable (needed anyway for exports) and the investment in Chinese brands real. That was the basis for my prediction.

    You forgot the real problem: Will the Chinese government be able to let the middle class share in the growth. Or will the elite succeed to keep it all to themselves. Another looming disaster is the real-estate bubble. If the bubble bursts or the middle class does not share in the growth, massive unrest will follow.

    But obviously, I am just an ignorant fool who has no clue about what happens in the world. So my predictions will falter catastrophically.

  291. > But obviously, I am just an ignorant fool who has no clue about what happens in the world.

    This remains to be seen.

    For now, you appear to be a rube who has read a few too many xenophobic screeds while also having never negotiated with the Chinese.

  292. > This is not “fanboy interpretation”, these are the facts being reported.

    To be clear, I was labeling the author of the article, not you. It was “fanboy reporting” (aka yellow journalism).

    If you go back to the original source (Gartner), you’ll find that the situation is as I stated. Gartner shows Android as having contracted market share during the quarter, and states that this is directly attributable to Apple’s blowout quarter.

    The headline could state that pigs fly, and it still wouldn’t be based in fact.

    I can’t give you access to the Gartner report, but you might want to google Gartner, iPhone and Q4 to read the rest of the press.

    As Yoda said, hate will blind you.

  293. That sounds a lot like saying that the athlete who gets the most sponsor money, the most cheers, and the most interviews wins the Olympics.

    Not really. In business, winning is usually judged by profit. From a purely technological point of view, it might mean the best/coolest/most influential device. Perhaps from an ecological point of view, marketshare might be the decisive metric… but in that sense, bacteria have “won” on Earth, far surpassing human beings or even the entire animal kingdom.

  294. China Telecom joins China Unicom! Of course, China Mobile is the really big fish.

    Kantar Worldpanel also had good news for Apple: “It’s a slightly different story in the US with Apple continuing to make gains on Android. Apple now has 48.4% of the US market compared with Android’s 42.6%.” Of course, we can only assume that Kantar Worldpanel has become a lousy consumer research agency since last April.

  295. So the conclusion of all the new iPhone offerings seems to be that the iPhone sells well if other subscribers subsidise it?

  296. How about: “The iPhone sells so well that carriers bend over backward to offer it”?

  297. “The iPhone 4S launch helped Apple nearly reclaim balance in the US, at 43 percent of the fall’s market share to Android’s 48 percent. Google’s core appeal was to newcomers who had never owned a smartphone…”

    All this talk about Apple selling “more” smartphones than Android combined leaves me with a puzzle.

    There are 700k Android Google activations per day worldwide. That is over 60M per quarter. In their best quarter EVER, Q4 2011, Apple sold 37 million iPhones and 15 million iPads.

    That is still only 62% of Google activations, and even phones and tablets combined are much less than Android phones alone. And, as many are fond to say, there are much more Android phones around than are activated with Google.

    And while this is stellar growth for Apple and the iPhone, it follows a quarter with only 20 million iPhone sales. So in Q3+Q4 2011, less iPhones were sold than Google activated Android appliances in Q4 alone (as everybody says, almost exclusively phones).

    With the iPads added, Apple will come somewhat closer to Android in units. But not in users, as an iPad user is likely to own an iPhone too.

  298. The obvious conclusion would be that 700k figure reflects the holiday season, given that Rubin announced it on December 20th. It wasn’t the sustained figure for the entire quarter, and they may very well not be activating 700k a day right now. Since Google announced 550k per day a month earlier, in November, I think that’s a pretty solid conclusion.

  299. >It sells well only when it is free? That is, consumers want it when they do not know
    >how much it will costs them?

    Aside from the fact that this completely discounts the runaway success that the iPhone was when it was offered unsubsidised, you’re also pretending you’re making some profound statement when in reality you’re simply stating a fact of life: “as the price goes down, more people are willing buy your good.” See also the Touchpad firesale.

  300. @Bryant,
    “Since Google announced 550k per day a month earlier, in November, I think that’s a pretty solid conclusion.”

    Nope, the 550k was already announced in the summer of 2011. It is more likely that in November, only the old numbers from the previous announcement were used. Especially as the number of Android activations has grown by an average of 30k/month since early 2010.

    There is no indication that the 700k was a holiday spike. The fact that Google updates the daily activations stats only every few months is not evidence that these daily numbers actually stay flat for months and then jump by 150k in a week, never to come down again. (activation stats are commercially and market sensitive data, Google will be careful with what they say)

    This was already discussed here:
    http://esr.ibiblio.org/?p=4015&cpage=1#comment-355092

    There was a holiday spike, 3.7M activations on 12/24-25:
    http://esr.ibiblio.org/?p=4024

  301. > Since Google announced 550k per day a month earlier, in November, I think that’s a pretty solid conclusion

    The obvious conclusion is that the pros don’t believe even the 550k/day activation rate.

    550k/day would be 49.5 million activations per quarter. Apple reported 37 million phones sold..

    Those total to 86.5 million. Comscore reported 97.8 million smartphone subscribers in December.

    49.5 is 50.61% of 97.8. This would be Android’s share of Q4 if the 550k/day rate were accurate. The highest quotes from NPD and Gartner have Android at just over 50% for the quarter.

    Given that Apple reported 37.8% of comscore’s assumed market, the analysts are questioning the Android ‘activation’ rate.

    QED.

  302. @Larry Yelnick
    “550k/day would be 49.5 million activations per quarter. Apple reported 37 million phones sold..”

    Google activations are worldwide, as are Apple sales. Comscore is USA. You are mixing global and USA only data.

    More on the continuous increase in activations. 550k/day this summer to 700k/day in December is in line with the number of smartphones sold worldwide.
    http://esr.ibiblio.org/?p=4015&cpage=1#comment-355351

  303. you’re also pretending you’re making some profound statement when in reality you’re simply stating a fact of life: “as the price goes down, more people are willing buy your good.” See also the Touchpad firesale.

    You’ve almost got it, soooo close. Hint: we’re not the ones claiming that the firesale proved the Touchpad was awesome.

  304. @Bryant:

    Of course, we can only assume that Kantar Worldpanel has become a lousy consumer research agency since last April.

    A bit heavy on the snark, don’t you think? And it would be useful if you included a link next time…

    In any case, here’s more data from Kantar than I’ve usually seen collected in one place:

    http://www.guardian.co.uk/technology/2012/feb/21/android-smartphones-os-uk-apple?newsfeed=true

    I particularly like this chart:

    http://static.guim.co.uk/sys-images/Technology/Pix/pictures/2012/2/21/1329818696665/Screen_Shot_2012-02-20_at_19.25.52.png

    That chart shows Apple really focusing on the US. I don’t want to start yet another interminable argument about carrier subsidies, but whether this is a new front (4S not yet rolled out properly around the world) or a rear-guard action (unreplicatable firesale on heavily subsidized older kit) remains to be seen.

    I will note that in the table below that chart in the article, it shows that Apple is bravely almost holding their own. In the countries Kantar tracks, Android increased its share of installed base 83%, from 20.1 to 36.9%, while Apple’s share of installed base decreased 2.4%, from 29.2% to 28.5%. Obviously, Apple can do that for a long time and still make beaucoup bucks, because the installed base is growing so fast.

    OTOH, because the installed base is growing so fast, that means that Android’s installed base is growing really fast. Now, what were the rest of you saying about Android’s fudged activation numbers?

    @greg:

    soooo close

    Too subtle by half. Whether it’s a profound statement or a simple fact of life is immaterial. It’s what I’ve been arguing for a couple of years now…

  305. Winter, could not the discrepancy be accounted for by a higher rate of Android deactivations (returns and disposal of obsolete phones)?

  306. BTW, the countries that Kantar was tracking in that chart are all relatively wealthy. If Apple can’t maintain market share there, it will be interesting to see how they fare in the wider world once they saturate the pent-up demand for luxury smartphones.

  307. Patrick,

    > If Apple can’t maintain market share there, it will be interesting to see how they fare in the wider world once they saturate the pent-up demand for luxury smartphones.

    By the looks of things (chart based on the Kantar data), Android’s gains are based on Symbian’s losses, not those of iOS.

    http://www.asymco.com/wp-content/uploads/2012/02/Screen-Shot-2012-02-21-at-2-21-3.04.58-PM.png

    Thus Microsoft’s strategy of increasing the cost structure for Android through IP licensing, and my above-mentioned potential strategy of Google purchasing Motorola Mobility both for the IP, as well as their well-proven ability to build feature phones. Look for feature phones with Google .. er, ‘features’ in the next year or so.

    No doubt these will also be called ‘Android’ phones, just to keep up the appearance of growth of the platform.

  308. Patrick: I expect Apple’s “luxury smartphones” to progress the same way that Apple’s “luxury” MP3 players did: start expensive and work their way downmarket over time. That’s already happening in the US, where the 3GS is free with contract at AT&T.

  309. @PapayaFS
    The estimates of total number of Androids sold are very close to the integrated worldwide activation rates.

    So there is no discrepancy.

    Anyhow, if Google lied about activation numbers they would get a visit from the SEC.

  310. @Patrick — hm? I did include a link. Here it is again, tho: http://esr.ibiblio.org/?p=3137

    I think the 3GS pricing is going to be an ongoing thing — if Apple has a way to consistently sell 2 year old models at $0 upfront cost, why wouldn’t they? I know that over the summer the 3GS sales were mostly refurbs, but I don’t think that’s the case any longer. I always felt Apple had a problem in that they didn’t have a low cost entry model; I was not smart enough to see old models as a potential solution. I don’t see any reason why, next winter, we won’t be seeing the iPhone 4 take over that price point.

    The iPad 3 release should be pretty informative there, actually. If Apple’s strategy is to use older models as the lower price points, the iPad 2 will stick around. I’d bet that it will but I could be wrong.

    Hm. And now I wonder if we aren’t seeing Tim Cook’s hand. I am totally hypothesizing here, but I can easily imagine Jobs deciding not to sell anything but the best possible product. Cook, who was always the logistics master, might well believe that it’s OK to sell old product in order to attack the lower price point.

    Android will, of course, continue to own the really low price points. An unlocked new 3GS still runs… $375. Pricy!

    Oh, FWIW, I think Larry’s on drugs if he thinks the Android activation numbers were fudged. I also don’t think we’ll see feature phones labelled Android. It triggers certain pleasure points in the human brain to think of this as a war, but I bet Google would be fairly happy with a constant 65% of the market and Apple would be happy with 35% of the market. Steve Jobs was personally offended by Android. I’m not sure Tim Cook is.

  311. Winter, I am suggesting that the discrepancy in the various numbers might be because some are counting 1) sold/activated and others are counting 2) phones currently in use. I suspect Android phones go out of use faster than iPhones, but don’t have anything to prove that.

  312. So I’m on the T and I bump into an acquaintance of mine — a developer, as it happens, on the new BioShock game. He was telling me about how some of his friends recently joined up with Google on the Android development team.

    “But the thing is,” he snickers, “three out of the four of them own iPhones!”

    Yelnick is right. The smart kids are on iOS. And where the smart kids are now, everyone else will soon follow.

    Android is a booby prize you settle for when you can’t get an iPhone.

  313. Anecdotes and data.

    Take the case of a friend of mine and his wife, he’s a technophile software guy, she’s a technophobe medical doctor (you all know the type). They’re both smart as you could want, and have the ‘small technical school on the Charles’ sheepskin. A few years ago when the G1 was current, he advised me to stay away from Android because it was basically beta software, that it was interesting but not ready yet, not sufficiently usable. He and his wife wound up with iPhones. The last conversation I had with him about phones, they’ve both got Androids- Motorola for him, Samsung for her. She switched when she saw how much better the Android phone was.

    iPhones are fashion accessories. As technology, they’re nothing but training wheels that even technophobes eventually outgrow. *shrug*

  314. On a different note…The nook color got a price drop and B&N introduced the lower end Nook Tablet at Fire priced. With CM7 on the NT now I’d buy one if I wasn’t waiting for the iPad 3 come March 3.

    With MS Office coming to the iPad it becomes an even more compelling platform for the student. I love Keynote but I can’t send it to anyone except as a PDF. Assuming the iPad Powerpoint doesn’t suck and can Airplay I can see kids using it over a laptop to give their presentations.

    I also think that Enterprise adoption is going to move even faster for the iPad. I have no clue what MS is thinking.

  315. @Bryant:

    > Here it is again, tho: http://esr.ibiblio.org/?p=3137

    I meant a link for your quote for Kantar’s current research: “It’s a slightly different story in the US with Apple continuing to make gains on Android. Apple now has 48.4% of the US market compared with Android’s 42.6%.”

    I think the 3GS pricing is going to be an ongoing thing — if Apple has a way to consistently sell 2 year old models at $0 upfront cost, why wouldn’t they?

    Sure, but (a) looking at Kantar’s data, that seems pretty US-centric so far. Especially since Apple/AT&T/whoever started really pushing 3GS last spring (when Verizon got the iPhone 4) it seems that if Apple wanted to (or could) go worldwide with that strategy, we might have seen some different results in other countries for Apple’s growth rates if it were applied consistently.

    The interview with Cook you found indicates that he’s very interested in other price points. He seems to have found a painless way to get there in the US, and he’s interested in replicating that method (switching prepaid to postpaid) in other countries. Will be interesting to see (a) how well that works; (b) what he does if it doesn’t work; and (c) whether the strategy is just a flash-in-the-pan even in the US, e.g. if state-of-the-art Androids become more interesting to consumers than equivalently-priced older iPhones.

    As you say, the iPad rollout will be interesting, but I am also very interested to see what happens internationally (for both iPhone and iPad).

  316. Oh, doh, sorry. Kantar makes it fricking impossible to link into their site. This link should get you to where you wanna go; the UI is atrocious and I feel bad just linking to something so non-Web-like.

    Agreed on the relatively US-centric approach. Vodafone does a free 3GS with contract in the UK. Not sure about other countries. The developing countries are the really key markets. Mind you, if Apple’s succeeding in getting China carriers to do a free 4S with contract, they don’t need to worry about the free 3GS as much. Maybe a less egregious contract would be a selling point for the 3GS? Or maybe it would confuse the marketplace, I’m not sure.

    But yes. It’s way interesting times, and I too am fascinated by the international strategies.

  317. BTW, this http://www.androidcentral.com/ubuntu-android-becomes-real-looks-take-motos-webtop-experience seems to be realization of something ESR talked about a while back.

    Sounds sweet.

    Though, not everybody can get it. It’s only available as an OEM feature. Which means it’s subject to the same vendor-dribble that Android releases themselves are. Meaning I probably won’t be able to get it on my Transformer.

    Hell, ICS on that thing (I didn’t get the Prime) is turning out to be like sex in the champagne room: you’ll never get it, you know you won’t, but oh how Asus love to tease!

    Meanwhile, users of all but the oldest and weakest iOS devices get the latest iOS release on launch day, no exceptions.

    The smart kids use iOS.

  318. @PapayaSF
    I have no idea. I read an average of 18 months use for Android phones first owner. No idea where or how they got these numbers. Could be completely fictional. I have no link anymore.

    I have my Nexus One for almost two years now and have no reason to trade it in. So 18 months does not seem unreal.

    When the contract ends I will have paid 720 euros for phone, speech, SMS, and unlimited data over the two years combined. Was a good deal then.

  319. Ubuntu on Android is a seriously cool development. OEM only… will they work with CyanogenMod? Will the speed be there? Lots of important questions, but it’s worth watching, to say the least.

  320. New Nielsen study results on smartphone penetration vs. age and income:

    http://blog.nielsen.com/nielsenwire/online_mobile/survey-new-u-s-smartphone-growth-by-age-and-income

    It confirms what we all knew — younger people are more likely to own smartphones, and richer people are more likely to own smartphones. According to Nielsen, 48% of those polled owned smartphones, but 69% of recent acquirers bought smartphones. I have seen this sort of installed base/new acquirer split a lot over the last year, but have a hard time figuring out its relevance. It seems to conflate two distinct numbers: people moving to smartphones, and rich people with smartphones replacing them more often than poor people with dumbphones. (I say this because of AT&T’s and Verizon’s quarterly reports, which consistently show smartphone sales percentages that don’t seem to be reflected in their customer base.)

  321. Ubuntu on Android is a seriously cool development. OEM only… will they work with CyanogenMod? Will the speed be there? Lots of important questions, but it’s worth watching, to say the least.

    Unless something has seriously changed, Ubuntu doesn’t do anything under a proprietary license. Whatever magic they did to get this to work should be usable by anyone.

  322. > That’s already happening in the US, where the 3GS is free with contract at AT&T.

    The iPhone 4S is free with contract in China.

  323. @Larry Yelnick:

    > The iPhone 4S is free with contract in China.

    Yes (where, of course, as usual, “free” means you don’t pay for it right away).

    But…

    Unlike AT&T, where you just need any contract with data, and that data contract would be the same price whether or not you got the iPhone, and the contract is always two years long, in China the fact that you’re paying extra to have the phone over time seems pretty obvious from the different contract options.

    http://www6.pcmag.com/media/images/335380-china-telecom-iphone-4s-plans.jpg?thumb=y

  324. > Anyhow, if Google lied about activation numbers they would get a visit from the SEC.

    Given the repeat offenses of Google found by the FTC, DoJ, EU, etc, I’m not sure that Google would go out of its way to not get caught.

    Besides, what is an Android activation, anyway?

  325. There’s a hands-on with the Ubuntu/Android implementation on the Verge. Canonical says it’s in discussions with manufacturers — I would have thought it’d be really open too, but the signals being sent are not quite that. More data wanted.

    Some of the commenters point out the obvious: this would be a really good enterprise feature for Microsoft Windows 8.

  326. I’m about halfway there already with the Linux Installer app. I’ve got Debian squeeze running on my Android tablet, and so far it appears to run really well. In one machine I now have two things I’ve always wanted:

    1) a small, always-on portable Linux box with extra long battery life

    2) a machine on which I can do some ARM assembly hacking with little fuss

    Man, I’m stoked. The Linux-desktop-cohabiting-with-Android future probably holds appeal to a vanishingly small minority of geeks (most geeks already have iPads and MacBook Airs and are content with those), but I’m one of that minority, and I say it’s about time.

  327. Jeff, may I just say that I deeply admire your ability to combine honest Linux appreciation with eloquently snotty pro-Apple commentary? Also, which tablet are you using?

  328. @Larry Yelnick
    “Besides, what is an Android activation, anyway?”

    Google and Andy Rubin are your friend:
    https://plus.google.com/112599748506977857728/posts/Kkjf8oESTZs

    For those wondering, we count each device only once (i.e., we don’t count re-sold devices), and “activations” means you go into a store, buy a device [and] put it on the network by subscribing to a wireless service.

    Here is an animation of worldwide activations:
    https://www.youtube.com/watch?v=fqFpq9WXbJo

    Just give up and accept reality:
    There are more than 250 million Android users and that fits in very well with the sum of all activations since the start:
    http://phandroid.com/2012/01/19/over-250-million-people-using-android-phones-and-11-billion-market-downloads-says-google/

  329. @esr
    “It would be even more interesting if Ubuntu Unity weren’t unusable shit, but as long as I can bring up XFCE in this environment I won’t care.”

    If it can run Ubuntu, it will probably run any of the derivatives soon. Personally, I prefer Mint (or even better, #!). I would really like it when I could leave my laptop at home and work on my phone instead.

    And it has been done, sort of:
    http://androidcommunity.com/this-is-what-a-galaxy-nexus-desktop-computer-looks-like-20120210/

    Obviously the Android interface isn’t meant for a 20-inch screen, but thanks to the Galaxy Nexus’s native 720p display (which is mirrored on both the phone and the monitor) it works surprisingly well. Replacing a desktop browser or document editor technically works, but the mobile interface holds you back when blown up. That said, the increasingly cloud-based nature of Google’s Android apps makes for pretty good integration.

  330. @winter

    Andy Rubin is not my ‘friend’, nor do I ‘like’ him. Get thee back to Facebook with that language.

    Otherwise, better people than I have done rudimentary analysis on Google’s announced ‘activations’, and found them… lacking.

    http://www.asymco.com/2011/12/21/how-many-android-phones-have-been-activated/

    On the other hand, if you believe than an Android handset is worth $14/year to Google
    (http://www.zdnetasia.com/study-ios-android-users-average-us14-per-in-app-buy-62301373.htm)

    then 700K activations/day are $9.8M per year for each day’s worth of activations, or approximately
    $3.6 billion a year. Two points to make here:

    First, Oracle sure would love a slice of that pie.
    Second, Apple makes this much in less than a week.

    Remember, the smart kids are using iOS.

  331. @Larry Yelnick
    “Otherwise, better people than I have done rudimentary analysis on Google’s announced ‘activations’, and found them… lacking.”

    So, what is your problem? The guy finds a 10% difference between his calculation of the total number of activated phones and the cumulative number of shipped phones with Android.

    There is so much wrong with his analysis. Just the two glaring ones:
    1) A 20 million units difference is somewhat more than 1 month worth of Android activations. A reporting lag between activations (immediate) and sales (have to be communicated) of just 1 month explains the difference

    2) The daily activations are given in round numbers (50k and higher) and always refer to unspecified dates. These then have to be integrated (squared dependency). That alone is worth a 10% uncertainty margin.

    @Larry Yelnick
    “Remember, the smart kids are using iOS.”

    If you are meant to be an example, they also flunk statistics.

  332. Andy Rubin is not my ‘friend’, nor do I ‘like’ him. Get thee back to Facebook with that language.

    Pedantry fail.

    Interesting that a Dutchman has a better grasp of English colloqualism than you do. The expression ” is your friend” has nothing to do with Facebook friending, or liking anything. It means ” has the information you’re looking for, you should look there already”.

  333. I must correct myself. Above, I stated that most iPad owners will also own an iPhone.

    I might have been utterly wrong:
    http://www.typhone.nl/aanbiedingen/samsung-galaxy-y-apple-ipad-2-wifi–vodafone-bel-sms-web-smart-150-24?utm_source=3674&utm_campaign=3674&utm_medium=AFC&utm_content=6bsc_apresski

    Translated: An offer for a free iPad2 WiFi & Samsung Galaxy Y on a 2 year contract with 150 minutes call, unlimited SMS, and 1200MB/month.

    This comes down for 1000 Euro for the two years combined.

  334. @nigel
    “Well that a kind of a bizarre offer but I guess you can sell nearly anything if you offer a free iPad with it.”

    That was my interpretation too.

    But what does it tell us about Apple’s offerings?

    My impression is that the iPhone has lost a lot of its “Cool” factor with respect to Android phones. At least enough to make it Cool to own an iPad & Galaxy Y(oung, really) io of an iPhone.

  335. @Nigel:

    Somewhat less snarky, techbargins released survey results that suggest that half of Kindle Fire
    owners intend to buy a iPad 3.

    Well, they might… eventually… or not.

    My all-time favorite survey about what consumers are going to do was the uSamp one from a year ago:

    http://macdailynews.com/2011/02/03/bloodbath_54_of_verizon_android_blackberry_users_to_upgrade_to_iphone_day_1/

    “more than a quarter of current AT&T customers (26 percent) intend to upgrade to Verizon’s iPhone on the day it becomes available.”

    “54 percent [of Verizon's Android and Blackberry customers] are very likely (25 percent) or somewhat likely (29 percent) to upgrade to iPhone on February 10.”

    Even with a well-constructed survey (which I’m sure this wasn’t) people don’t follow through on their intentions. The only thing less reliable than surveys about what people are going to do is web-based usage metrics.

  336. >But what does it tell us about Apple’s offerings?

    I don’t think it tells us anything more than rooms to go will offering you a free samsung TV with purchase of a complete living room set tells us about samsung products. I mean, I guess it means some companies view the iPad as something they can use as a loss leader, but that’s hardly new. Car dealerships have been giving away iPods with car purchases for quite a while now.

  337. @tmoney
    I think it is Erics 50%. Android has become the default smartphone platform.

  338. > Interesting that a Dutchman has a better grasp of English colloqualism than you do.

    Interesting that you didn’t understand that the entire thing was tongue-in-cheek. (You–>#fail)

  339. > Android has become the default smartphone platform.

    Android may be about to fall.

    Samsung could quite easily make Tizen look and feel like Android, and with some licensing deals (Hello, Oracle) it might be able to run Android (Dalvik/Java) apps, *legally*. That’s the thing with Android: in essence, it is an open source project, and beyond the developer support provided by Google, there is no intrinsic reason for Samsung to stick with it. All of its functionality, look, and feel can be replicated — and in the process, it can waive the royalty it must currently pay Microsoft in order to use Android, and gain oodles of maneuverability.

    Tizen will be a fully open OS hosted by the Linux Foundation, and steered by Intel and Samsung; very similar to how Mozilla runs Firefox. There won’t be any “hidden” branches, and OEMs (including Samsung) will be free to modify the codebase without having to appease a commercially-conflicted third party. (Hello Moto!)

    Meanwhile, of course, we must not forget that Intel is Tizen’s co-founder — and Intel would absolutely love Samsung to build phones around its upcoming Atom SoCs. Samsung might love this too, given that Intel will be at 14nm in a couple years. Can you say, “power envelope”?

    My prediction: You’re seeing a near crest of the Android wave. The high water mark.
    Android will soon splinter into 1,000 shards. In 5 years it will be a mere note in the history of the Smartphone Wars.

  340. Interesting that you didn’t understand that the entire thing was tongue-in-cheek.

    You’re really that oblivious to the 60 cycle hum?

    Android may be about to fall.

    You can say that about anything. The goal is freedom, choice, good software running on good enough hardware. The Android alliance just happens to be the best bet to provide that right now, but if a better option came along then great! It’s the Apple folks who have their identity invested in their platform.

  341. >Android may be about to fall.

    You just totally don’t get it, do you? If the scenario you describe were to happen, it wouldn’t be failure at all – it would be success, in both Google’s terms and mine.

    Remember: Google’s end goal isn’t to control mobile users, it’s to block anyone else from controlling a chokepoint in Internet delivery and being able to charge tolls on Google’s ad business. If the effect of Android is to have opened the way for Tizen or something even less amenable to single-point control, Google wins. And so, not at all coincidentally, do I.

  342. > You just totally don’t get it, do you?

    I believe I do, actually. You’re the one shouting in fury.

    > Google wins.

    Google loses (a great deal of) its ability to collect information and display ads in the mobile space. Not a win.

    > And so, not at all coincidentally, do I.

    Yes, open source (and openness) would win. I didn’t say it wouldn’t. You could use the collapse of Android as (yet another) example of calling something “open source” without a real OSS process being implemented.

  343. @Larry Yelnick
    “You could use the collapse of Android as (yet another) example of calling something “open source” without a real OSS process being implemented.”

    So, Android too is “No true Scotsman“?

  344. @Larry Yelnick
    “and with some licensing deals (Hello, Oracle) it might be able to run Android (Dalvik/Java) apps, *legally*.”

    About Android being somehow an “illegal” program somehow infringing Oracle “precioussss IP”, you did not follow the developments did you?

    http://www.groklaw.net/article.php?story=20120223073144912

    It makes one wonder whether Oracle wouldn’t be better off withdrawing its patent infringement claims and proceeding only on the copyright claims. The court would likely insist that Oracle withdraw the patent claims with prejudice, meaning Oracle could not reassert the seven patents originally asserted in this case, but that would not preclude Oracle from bringing another patent infringement suit in the future against Google … provided it can find any of the Sun patents that are valid.

    Their damages claims have been shot down twice and the third attempt is again extremely weak.

    Almost nothing is left of Oracle’s grandiose claims. They really are following SCO here, with “Billions and Billions” of damages.

  345. TMobile has a lousy Q4:
    http://www.bgr.com/2012/02/23/t-mobile-reports-big-subscriber-losses-in-q4-says-it-will-launch-lte-in-2013/

    “However, not carrying the iPhone led to a significant increase in contract deactivations in the fourth quarter of 2011. In 2012 and 2013, T-Mobile USA will invest to get the business back to growth, including an incremental $1.4 billion investment in its network modernization initiative, which will total a $4 billion investment over time.”

  346. T-Mobile is dying because they don’t have the iPhone. They mentioned the word iPhone 7 times in their earnings press release.

    The other 3 major carriers are going to survive because of the iPhone. Sure, they take an upfront hit, but the overall revenue per iPhone is spread out over 2 years, and is early over $2000 per user.

  347. Wow, T-Mobile took a hit. And yeah, the answer is pretty clear at this point. Apple’s successfully forced the iPhone down US carrier throats. It does have an effect on their profits, because Apple is changing the balance of power. (Neither a good or a bad thing; either way it’s not consumers who hold the power.) But not taking the iPhone hurts even more.

    @Winter — you might want to read Florian Mueller’s take on the damages question. Yeah, I know, we all hate him because he’s biased. But he’s not afraid to report bad news for Oracle, either. I think his opinions are at least worth taking into account. It’s not like Groklaw is any less biased.

  348. @Bryant
    “you might want to read Florian Mueller’s take on the damages question. ”

    If you can point out a case where FM predicted something right, or actually showed he understood the legal and FLOSS aspects of the case, I will try to read his accounts.

    As far as I know, FM has been able to get every single prediction about the legal troubles of FLOSS *Wrong* with a capital W.

    Remember that FM is not a lawyer, has no formal legal training, and has no experience as a developer of any type of software. Neither is he a journalist. And it shows, he got everything legal, CS, and journalistic *Wrong*. He was involved in the anti-Patent front in the EU a few years back. But no one seems to know where he came from and why.

    For all we know, FM is a lobbyist or free-lance analyst.

  349. @Louis4:

    > T-Mobile is dying because they don’t have the iPhone.

    I don’t think it’s quite that bad. In fact, with the new spectrum and cash they got from AT&T, T-Mobile is better positioned than they have been in awhile. Sure, they lost a lot of customers when Verizon and Sprint got the iPhone, but it remains to be seen whether that is a one-time pop (as T-Mobile seems to be gambling) or whether that is a continued drain (as Sprint seems to be gambling with its multi-year multi-billion dollar contract with Apple).

    The bright spot in T-Mobile’s picture seems to be prepaid data — their “monthly 4G” plan. A few tweaks to that, and they could be sucking in customers fleeing from high prices on other networks.

    BTW, T-Mobile is not ignoring the iPhone. Not at all. They negotiated with Apple, and decided that the costs were too high. They brag about how they have over 1 million iPhones on their network right now. They are busy repurposing (“refarming”) their 1.9 GHz spectrum, which will allow smartphones (including iPhones) to connect via 3G (instead of 2.5G “edge”) and they are handing out (over the internet) free miniature SIM cards suitable for direct insertion into an iPhone — no scissors required.

    Another thing working in T-Mobile’s favor is that the bulk of used iPhones, which won’t work on Verizon or Sprint, will work quite happily on T-Mobile’s network — a bit slowly now, perhaps faster when they get done with their 1.9 GHz changeover. Of course, you have to unlock the things, but there are people who will show you how to do that, do that for you, or sell them, new or used, already unlocked.

    When it comes to spending real bucks, T-Mobile apparently decided to spend them on LTE rollout, rather than simply handing them to Apple. This is a strategic decision they may come to regret in a few years, but I doubt it. The current hold Apple has over the carriers can’t last forever if Apple continues to push into international markets — the true global price of an iPhone that can work with TMobile will become apparent, and the true cost of the subsidy for that phone will also become apparent. T-Mobile are being criticized for a “late” LTE rollout, but I’m not sure that was a bad decision — the way electronics work, they’re probably getting much better gear for much less money than if they had rolled it out earlier.

    It was pretty obvious T-Mobile was going to lose some customers when Sprint and Verizon got the “latest” iPhone. It’s not nearly so obvious that significant losses will be ongoing. I think a lot of that is in T-Mobile’s hands. Even if iPhone is a must-have for a third of smartphone customers, that leaves two-thirds of them for which it’s not, not to mention that half of all subscribers aren’t even on smartphones yet.

  350. BTW, as I mentioned in comments on earlier posts, if I was T-Mobile and losing money, I’d blame it on iPhone, too. It’s very convenient, especially as it’s true (and obvious to everybody) that they have some subscriber losses due to the iPhone. But it’s not a given that there’s nothing they can do to counter to improve their bottom line, although the way they’ve been operating over the last few years it might be a given that they can’t figure it not. They did pretty well on the AT&T deal, though — maybe they’re making better decisions these days. I certainly think focusing on LTE is a good one.

  351. @Bryant
    I forgot. If someone argues “bias” I consider it an admission that that person is unable to understand the topic.

    You can *read* the documents online. You can form your own opinion.

  352. > For all we know, FM is a lobbyist or free-lance analyst.

    FUD much?

  353. @Patrick Maupin “BTW, T-Mobile is not ignoring the iPhone. Not at all. They negotiated with Apple, and decided that the costs were too high.”

    No, the iPhone hardware doesn’t support TMobile’s weird 3G frequencies.

    My guess is Apple couldn’t get a chipset that had CDMA/GSM and also had TMobile’s weird 3g freq. And they didn’t want to do a TMobile specific SKU. Hence, they ignored TMobile. But it’s a bit unclear…..

    This is probably the next chipset for the iPhone. I don’t know if it works with Tmobile: http://www.qualcomm.com/media/releases/2012/02/21/qualcomm-announces-fifth-generation-embedded-data-connectivity-reference-p

  354. @Patrick Maupin

    Oh, the other big factor with TMobile/iPhone is all parties probably thought the AT&T deal was going to go through so the iPhone would just use AT&T’s network until TMobile got upgraded.

  355. @phil:

    No, the iPhone hardware doesn’t support TMobile’s weird 3G frequencies.

    And it didn’t support Verizon’s weird modulation until last year. It’s entirely likely that T-Mobile, like Verizon, could have had an iPhone by now if they had enough cash. And the news is that T-Mobile is working harder to support people who use iPhones on their networks (at edge speeds) right now.

    all parties probably thought the AT&T deal was going to go through so the iPhone would just use AT&T’s network until TMobile got upgraded.

    There’s no doubt that if AT&T had purchased T-Mobile, any money that T-Mobile might have paid Apple would have been wasted. But I’m sure there were some contingency plans. One of those contingency plans may have even involved Sprint. A Sprint acquisition of T-Mobile might not be quite as good a fit for technical reasons, but would probably be approved by the regulators. If that’s the backup plan, then (a) T-Mobile not having an iPhone is still no big deal, and (b) the large number of handsets Sprint is contracted to buy from Apple over the next few years doesn’t look quite so daunting, with another 32 million customers to amortize them over.

  356. @Larry Yelnick:

    There is no question that FM is a paid consultant, and that Microsoft has been one of his clients. What FUD are you referring to?

  357. >BTW, as I mentioned in comments on earlier posts, if I was T-Mobile and losing money, I’d blame it on iPhone, too.

    You speak my very thought, Patrick. It’s easier to say “waah, no iPhone” than for T-Mobile’s management to admit that larger strategic decisions they made and are still deeply committed to might have been stupid. They’re ass-covering in a way that looks plausible to investors, nothing more.

  358. If a fully-loaded Linux OS with a real GNU userland becomes a serious contender in the mobile space, I say GOOD! It’s the wet dream I’ve been hoping would come true since 1995. I don’t think our host will have many complaints.

    I don’t think that will happen though. Android architecture sucks donkey balls, but Android the company is formed from ex-Applers, which probably has a lot to do with why Android’s UX isn’t complete shit. (It’s still shit, just not completely.) That also probably has a lot to do with why destroying Android was such a fervent deathbed wish of Jobs’s: his own people betrayed him.

    If Android is destroyed and cellphone builders look to KDE or GNOME for the basis of the next big smartphone platform, all it will take is an incremental iPhone update (“Now Siri laughs when you tell it a joke. Funny. Reinvented. Only from Apple.”) and they will be swept into the sidelines. And that’s leaving aside the fact that the open-source community can’t get its shit in one sock when it comes to things like battery life and power consumption.

    Betting against Apple at this point is a fool’s game. They’re the largest and most profitable company on earth, they’ve been at the vanguard of every microcomputing revolution, and if open source ever does dominate the smartphone landscape, they can just change the game again and make everybody else catch up. It’s their field and their ball, after all. Has been since 1976.

  359. I think Jeff is correct. I don’t think either Apple or Android (or forks/OSS equivalents) will be going away anytime soon, but Apple is going to be dominant here for a long time, and Android will be playing catch-up on everything except some features and maybe market share.

  360. @esr
    “Gartner reports Android widened its lead over Apple in 4Q2011, says Android has 50.9% global market share”

    Another take at the Gartner data:

    Reality check: Windows Phone less popular than Samsung’s Bada
    http://thenextweb.com/microsoft/2012/02/17/reality-check-windows-phone-less-popular-than-samsungs-bada/

    Well, here’s a slap in the face for fans of Microsoft’s nascent mobile line: it sells less units per quarter than Samsung’s Bada platform, according to Gartner.

    …..

    By several hundred thousand units, Bada beat Windows Phone in global shipments, despite Nokia’s new handsets basking in a pool of warm media love. Ask yourself this question: what was the last time you read a story on a Bada handset?

    @Larry Yelnick
    “Otherwise, better people than I have done rudimentary analysis on Google’s announced ‘activations’, and found them… lacking.”

    Notice the 76 million Android phones sold in 4Q2011? 76 million in 3 months equals 840000 units per day.

    So, if activation rates are really only 700k/day, then around 20% of Android phones sold are not activated at Google.

    According to Gardner’s sales numbers, Google’s Android numbers are too conservative.

  361. “They negotiated with Apple, and decided that the costs were too high.”

    You keep saying this while presenting no proof. In fact, you’ve presented evidence to the contrary.

  362. > “They negotiated with Apple, and decided that the costs were too high.”

    Apple in T-Mobile’s boardroom:

    “You want us to WHAT????”. LOL. “Seriously? Oh, wait, you *are* serious? OK. Wait, you know your freqs are freaks, right? RIGHT?!? ”

    “OK. Special GSM iphone, just for you. Yours for the low price of: “.

    Apple: “Hey, we have costs, too.”

    “Tell ya what, we’re building a next-gen iPhone for everyone else built on LTE. Take some of that cash and spectrum that AT&T just gave up, and build a LTE network rather than your joke HPDA+ network, and you can have the LTE iPhone on your network.”

    “Sure! Spin it anyway you wish. See you next year.”

    (LOL)

  363. > Notice the 76 million Android phones sold in 4Q2011?

    Who generated this number? How? Using what facts?

  364. @Larry Yelnick
    “Who generated this number? How? Using what facts?”

    Link was given. And it was stated that these were Gartner’s data. I think I agree with Bryant that you are on some kind of drugs, or maybe just mad.

  365. @Tim F.:

    How does “[The iPhone is] a good device and we have always said we’d like to have it on the right terrms” contradict “the price was too high?”

    @Larry Yelnick:

    Obviously Apple hates multiple SKUs and adding a feature just for a single carrier. Again, this doesn’t contradict that the price was too high — after all, I think we all can agree that Apple could have had a T-Mobile iPhone a few years ago if it was important, right? You may be right that T-Mobile will carry it in a year or so. Certainly the cost to Apple will be better after new chipsets come out and/or T-Mobile “refarms” some of their spectrum. The real question is are there additional terms that Apple wants to impose that T-Mobile would still find too onerous? One of the second tier cell companies got in bed with apple; the other (US Cellular) declined…

  366. @Jeff Read
    A company sitting on heaps of cash does everyone a disservice. It also shows massive market failure. If Apple has no productive use for the cash they should pay it to their shareholders.

    Two obvious targets for criticism are onerous patent laws that limit competition and “hidden” carrier subsidies that hide real prices and even make non users pay for iPhone sales.

  367. @winter all large tech companies are sitting on mountains of cash.

    Well, other than the ones going under.

    Google, MS, etc all have large warchests. Apple spends billions on supply chain. Google scares the hell out of cable and phone companies with bids on spectrum or giving away fiber connectivity.

    As much as I might bitch about Google being evil I’d use them over Comcast in a heartbeat.

  368. I find it funny that random blog posts or questionable articles from flaky sources can be treated as gospel when it aligns with your argument but when T-Mobile comes right out and says “we’re losing lots of customers because of not having the iPhone” it’s spin to cover up other inadequacies and Sprint highlight that the iPhone is so important to their ability to compete that are willing to spend billions that they are simply being taken.

    Gee, maybe these companies aren’t getting conned but Apple provides significant value to both them and their customers?

    Nah, can’t be. Android is winning.

  369. @larry Tizen doesn’t strike me as much more likely to be viable than Meego, Moblin, Maemo, etc.

    Nokia going with Meltemi is ignoring that all prior Linux phone variants that started with M have not really ended well. Perhaps Intel figured a name starting with T might work out better.

    What strikes me as more likely and important to Google is that Samsung forks Android like Amazon and Baidu. And I guess QQ now has done the same by removing Google services (that don’t work anyway in China) in favor of it’s own.

    When it’ll really sting is if Facebook does the same. Take away enough eyeballs and eyeball revenue from Google and the whole Android strategy is moot.

    And don’t expect the Android ecosystem to stay all that open. The Android tablet market will be driven by the Fire and Amazon’s more closed market. That the Amazon Apptore appears to bring in more money than the Android Market means good things for Amazon and less good things for Google.

    An Amazon branded phone might not be far off in the future. I think a good number of folks will buy a Fire phone.

    http://techcrunch.com/2012/02/21/for-some-developers-amazon-appstore-now-brings-in-more-money/

  370. > Obviously Apple hates multiple SKUs and adding a feature just for a single carrier. Again, this doesn’t contradict that the price was too high — after all, I think we all can agree that Apple could have had a T-Mobile iPhone a few years ago if it was important, right?

    Patrick, I want to agree with you, but T-Mobile wasn’t going to get the iPhone before the AT&T exclusivity ended.

    And T-Mobile is a second-tier carrier. I’m sure Apple had Verizon in the first slot.

  371. IIRC Apple offered Verizon the first iPhone at the same time, or even before, they offered it to AT&T, but Verizon demanded then-standard carrier preferences (e.g. crapware, their own app store, etc.) that Apple wouldn’t agree to.

  372. Another twist on “Android is stolen”. I like the banksy artwork in the article.

    If Android is a “stolen product,” then so was the iPhone
    http://arstechnica.com/tech-policy/news/2012/02/if-android-is-a-stolen-product-then-so-was-the-iphone.ars

    High-tech innovations are often developed by laboratory researchers long before they’re introduced into the commercial market. Multitouch computing was no exception. According to Bill Buxton, a multitouch pioneer now at Microsoft Research, the first multitouch screen was developed at Bell Labs in 1984. Buxton reports that the screen, created by Bob Boie, “used a transparent capacitive array of touch sensors overlaid on a CRT.” It allowed the user to “manipulate graphical objects with fingers with excellent response time.”

    In the two decades that followed, researchers experimented with a variety of techniques for building multitouch displays. A 1991 Xerox PARC project called the “Digital Desk” used a projector and camera situated above an ordinary desk to track touches. A multitouch table called the DiamondTouch also used an overhead projector, but its touch sensor ran a small amount of current through the user’s body into a receiver in the user’s chair. NYU researcher Jeff Han developed a rear-projection display that achieved multitouch capabilities through a technique called “frustrated total internal reflection.”

    This is also a good example of the lethality of the “Piracy” madness addressed in the “Letter to Dodd” post.
    http://esr.ibiblio.org/?p=4155

  373. So, with some 20% of Android phones sold according to Gartner unaccounted for in Google activations, there remains a question. How easy is it to run an Android phone without getting Google involved?

    According to the article, not that difficult.

    Can an Android phone run without Google?
    Does Android really allow for a smartphone experience where you don’t share with Google?
    http://www.itworld.com/mobile-wireless/251508/can-android-phone-run-without-google

    When you first turn on your phone, or after you perform a “factory reset” on it, you’ll be asked to sign in with a Google account. Don’t do it, and look for the “Skip” option. Your phone can run without a Google account, and you can add other accounts to fill out your contacts and calendar and the like–Microsoft Exchange, Facebook, Twitter, and more. Also skip the options to send feedback about your usage, back up your settings to Google, and so on. Skip just about everything.

  374. Apple saw its iPhone sales more than double in the last quarter when it took a 24 percent share of the smartphone market, helped by a 15-month period before the 4S model was launched, which created a lot of pent-up demand.

    Gartner said Apple’s strong performance will continue into the first quarter of 2012 as availability of the iPhone 4S widens, but without the benefit from delayed purchases sales would decline from the previous quarter.

    In contrast to Apple’s blow-out quarter, former smartphone market leader Nokia saw its share of that market fall to 12 percent from 30 percent just a year ago.

    The guys from Smartphone Test

  375. “How does “[The iPhone is] a good device and we have always said we’d like to have it on the right terrms” contradict “the price was too high?””

    Why do you cherrypick two random bits of information to claim they aren’t contradictory, while still have not presenting a single shred of evidence that T-Mobile turned down the iPhone, while quoting the contradictory information (T-Mobile has said over and over they want the iPhone)?

  376. @Tim F.:

    Why do you cherrypick two random bits of information

    They’re not really random. Certainly a comment from the CEO isn’t. And the rest of your comment doesn’t really parse very well, but I agree that T-Mobile has the hots for the iPhone. Never meant to say anything otherwise. But, there are a lot of things I have the hots for I just know I can’t afford, too.

  377. I see “Intel” and “smartphone” together and envision a return to the days of holding hot bricks to our heads…

  378. > I see “Intel” and “smartphone” together and envision a return to the days of holding hot bricks to our heads…

    Jeff, if Intel can perform on-plan, releasing a 22 nm Silvermont Atom SoC in 2013 and the 14 nm Airmont Atom SoC in 2014/15, it’s going to be the ARM parts that melt in your hand.

    AMD is expected to unveil Hondo, a APU with power consumption of less than 5W specifically for tablet PCs, at Computex and plans to launch a 28nm-based 2W APU, Temash, in 2013.

    If Intel and AMD can lower prices, their processors are going to give the ARM parts a serious run in the Android space, especially given that Motorola/Google seems to be moving away from ARM.

  379. No surprise here:

    Google: 850,000 Android devices activated daily
    http://www.h-online.com/open/news/item/Google-850-000-Android-devices-activated-daily-1443475.html

    Google+
    https://plus.google.com/u/0/112599748506977857728/posts/Btey7rJBaLF

    Walking around Mobile World Congress and seeing the Android ecosystem at work is quite impressive. There are over 450,000 applications in Android Market. Some of the developers who built these apps are showing their latest and greatest in the Android stand. But the ecosystem doesn’t sleep. Our partners have activated a total of 300 million Android phones! Currently they are activating over 850,000 phones and tablets per day.

    These are the reported daily activation rates of Android I commented earlier. This was as official as I could get.
    http://esr.ibiblio.org/?p=4015&cpage=1#comment-355351

    60,000 01-04-2010
    160,000 01-07-2010
    200,000 01-09-2010
    350,000 01-02-2011
    400,000 01-05-2011
    500,000 01-06-2011
    550,000 01-07-2011
    600,000 01-10-2011
    700,000 15-12-2011
    850,000 15-02-2012 (actually, Andy tweeted it today 27-02-12, but I stick with the half month)

    Sustained 150k/day more activations in just two months comes down to 75k more activations/day each month. That is would be a definite jump from the 30k/month we have seen until December last year.

    The 50M increase in phones sold from 250M to 300M corresponds to 2 months worth of activations at 850k.

  380. @Jeff Read:

    As I’ve said many times before here, Intel and AMD didn’t need to worry about power consumption until recently. Now it’s important, so watch out!

    @Larry Yelnick:

    If Intel and AMD can lower prices

    I think this is a given for Intel. Their only “problem” would be not wanting to cannibalize their PC parts. But if they do an MCM with a controlled amount of RAM internal, that becomes (mostly) a non-issue. Enables some sweet little mini-PCs, too. Fancy WiFi router with Intel inside and 512MB of RAM, anybody?

    their processors are going to give the ARM parts a serious run in the Android space

    They need to do this or die trying. I won’t put any money on them dying.

  381. Err, well, actually, I won’t put any money on Intel dying…

  382. Nonetheless, I think Apple will have an easier time moving towards the low-end market than Android will have capturing much of the high-end. It’ll be the iPod all over again.

  383. @PapayaSF:

    (1) Android already has a lot of the high end of the market.

    (2) The price delta between an Apple and a no-name brand for a cheap MP3 player is much less noticeable to the consumer than the price delta between an Apple and a no-name cellphone.

    (3) For it to be iPod all over again, Apple needs the cellphone equivalent of iTunes. Siri isn’t it, at least not yet.

  384. Yeah, I don’t think it’s necessarily comparable. Apple invented the market as a result of creating an ecosystem with the iPod. The iPhone could have done the same thing but Google recognized the opportunity and got the Android Market in there before Apple had the chance to establish the same kind of lead. Which is a good thing.

    I would love to see smartphone sales split up by the unlocked cost of the smartphone. I more and more think there are really at least two markets — the market the iPhone plays in, and the market at around $200 or less. Note that if there are, practically speaking, multiple markets… a lot of the arguments about market share domination become irrelevant.

  385. The App Store doesn’t have a lead on the Android Market? And the Android presence at the high end hasn’t yet translated into profits and (AFAIK) long-term customer loyalty (on a large scale, at least).

    I don’t think the success of the iPod was so much iTunes as the fact that the whole experience was simple and reliable and pleasurable. I think the iPhone does that as well, with or without Siri.

  386. @Bryant:

    Interesting article. It appears that Android outsells Apple 12:1 in Greece, but only around 6:1 in Portugal. I wonder what the difference between the markets is, e.g. perhaps something like Portugese carriers having some level of subsidies (possibly with more transparency, like paying extra every month).

    Or maybe Portugal is one of those countries, like France, that proportionately simply loves Apple.

  387. Greece is broker than Portugal: massive unemployment, soup lines, etc.

  388. @PapayaSF:

    The App Store doesn’t have a lead on the Android Market?

    Bad example. Whatever lead there was is rapidly shrinking.

    And the Android presence at the high end hasn’t yet translated into profits

    Samsung might beg to differ:

    http://www.pcworld.com/article/248831/samsung_smartphones_tvs_help_drive_strong_4q_profits_even_as_rivals_falter.html

    and (AFAIK) long-term customer loyalty (on a large scale, at least).

    Changing the goalposts? In any case, a few years ago, most phone customer would have sworn loyalty to Nokia or Motorola…

  389. The App Store has a lead, but it’s not the same kind of lead the iPod/iTunes ecosystem had.

  390. @Bryant:

    Note that if there are, practically speaking, multiple markets… a lot of the arguments about market share domination become irrelevant.

    I don’t agree with this. The whole reason that market share matters is the network effects causing positive feedback loops. Assume for a moment that there are two different smartphone markets, and that Apple (via 3GS in the US for the low end) and Android both play in both markets right now. An Android version of Angry Birds makes all the Android customers in both markets available to the Angry Birds developer. So, to the developer, that’s one big market (perhaps segmented a bit by willingness to pay, but there are more potential customers there in any case), so he develops a game. Now, to the consumer, whether he’s in the high-end or the low-end market, the availability of a game like Angry Birds might make a difference to him. So, if developers are looking at consumers in both high-end and low-end markets, and making titles available based on their projections, then game title availability will cause more consumers in both markets to consider that OS.

    There’s no question that Android and iOS both have momentum and will probably both have the next Angry Birds available. But if Android only played in a single market, that might be less true for it. That’s why market share matters.

    In some ways, though, overall market share actually matters somewhat less for Apple (for any company with a lot of sales to rich people, for that matter) than for Android. Since the developers know that Apple has a lot of price-insensitive customers, they will develop for iOS even if it has a low market share. For instance, even though Androids outsell iPhones 12:1 in Greece, software developers in Greece would probably be foolish to ignore the iPhone for the simple reason that it’s a high-end device with wealthy customers.

    If, OTOH, Apple only had 1% of the overall Greece market rather than 5%, it might be a different story.

  391. Patrick Maupin: “(3) For it to be iPod all over again, Apple needs the cellphone equivalent of iTunes. Siri isn’t it, at least not yet.”

    Siri? What are you talking about? How is that in any way comparable to a content distribution systems like iTunes?

    Anyway, I’m trying to figure out how you’ve overlooked this little thing called the App Store, which is functionally almost identical in nature to iTunes and provides an analogous content download system that iTunes provided to the iPod.

    And bear in mind that when Steve Jobs unveiled the iPhone, he didn’t say it was just a phone. He said it was also the best iPod they’ve ever made (thus, the existence of the iPod Touch.) So, objectively viewed, iOS has not just one, but two content download services supporting it, whereas the iPod had only one.

    Still scratching my head over why you chose to focus on Siri though. I’d love to hear your rationale behind comparing a voice-based search feature to iTunes.

  392. Here’s why your argument doesn’t make any sense. The differentiator should NEVER be hardware in the first place. The notion that a carrier can compete by having some unique device is nonsensical.

    Superficial differences aside, all Android phones do the same thing, and it’s basically the same as the iPhone (except of lesser quality, according to consumer surveys). No carrier will ever win by carrying the best new device because their customers are locked into long term contracts anyway – only a fraction of their customer base can actually upgrade to the how new phone of the month.

    This whole argument is nonsense. Carriers should be competing on price and reliability. Deliver a better experience and the customers will come, because, as you point out, you can get an iPhone anywhere these days.

    Dropping the iPhone, however, would signal to customers that carriers are more concerned with short term profit scraping than long term customer satisfaction – so in other words its a terrible idea.

  393. wow. one of the dumbest blog posts ever.

    sure, nothing lasts forever. sure, someday the iPhone won’t be the tops in demand and Apple won’t be able to command a premium price from telcos for it. what insight!

    but that ain’t 2012. and there is no evidence to think it will happen in 2013 either (wishful thinking does not count as evidence), which is about as far as anyone can foresee given how fast things change. no one knows when.

    meanwhile, T-Mobile is bleeding subscribers becuse it doesn’t have the iPhone. that is real evidece. what’s going to happen to them in the next two years?

  394. @Patrick — I think we’re more or less agreeing. If Android has significant market share in the high end app buying market, the benefits will trickle down to the low end market. Assuming there are in fact two markets, Apple doesn’t need to worry as much about the low end market.

    It’s a big assumption, though, and I haven’t seen enough information to cause me to make up my mind one way or the other.

  395. @Andre Richards:

    Siri? What are you talking about? How is that in any way comparable to a content distribution systems like iTunes?

    It’s not. Obviously. Which is why, despite everybody gushing over Siri as the killer product differentiator, it’s not at all the same sort of product differentiator that iTunes was for Apple in the MP3 market — iTunes is what made the iPod a category killer.

    Anyway, I’m trying to figure out how you’ve overlooked this little thing called the App Store,

    The app store is not a significant product differentiator. Android has one of those.
    which is functionally almost identical in nature to iTunes and provides an analogous content download system that iTunes provided to the iPod.

    And bear in mind that when Steve Jobs unveiled the iPhone, he didn’t say it was just a phone. He said it was also the best iPod they’ve ever made (thus, the existence of the iPod Touch.)

    There is no question that the smartphone is the “eater of gadgets.” Google for esr’s previous post on this. But at this point in time, lots of people have figured out how to use iTunes with non-Apple kit, and also figured out how to get their music elsewhere. In the consumer’s mind, iTunes doesn’t lock in a phone like it does an MP3 player.

    So, objectively viewed, iOS has not just one, but two content download services supporting it, whereas the iPod had only one.

    I assume and hope you get to them the same way (or at least in a compatible way). Apple’s not stupid enough to have two different incompatible storefronts, are they?

    Still scratching my head over why you chose to focus on Siri though. I’d love to hear your rationale behind comparing a voice-based search feature to iTunes.

    Lots of people claim that Siri is some sort of magical product differentiation faerie dust. OTOH, for MP3 players, iTunes actuall was a sort of magical product differentiation faerie dust. I claim that Apple hasn’t yet invented such dust for the iPhone yet, although many here will tell you that Apple is a JIT vendor of exactly the dust that is needed, slightly before it is actually needed and well before any competitors can come up with their substitute vastly inferior dust.

  396. @Bryant:

    Assuming there are in fact two markets, Apple doesn’t need to worry as much about the low end market.

    I think we are in agreement on this. The root of your concern (“I haven’t seen enough information to cause me to make up my mind one way or the other”) is simply this: at the end of the day, if the market bifurcates into two markets, it could be great for Apple, unless the size of the higher-end market shrinks to the point where it’s not profitable to develop for. I don’t forsee that happening, but I must admit my crystal ball is quite cloudy.

    Others who claim they have much better crystal balls are quick to point out that nobody has stayed on top of the smartphone heap for very long, but that may ignore network effect realities. No previous smartphones had very serious network effects, and the reality is that the sheer absolute size of the Android and Apple ecosystems are daunting. A lot of “apps” are moving online, and other apps are coded with platform-agnostic tools, but the killer product differentiator apps are likely to remain those that are lovingly ported to each supported platform.

  397. OTOH, for MP3 players, iTunes actuall was a sort of magical product differentiation faerie dust.

    *sigh* I’ve been hearing this for years and I still don’t understand it. Many years ago I received a first gen iPod shuffle as a gift. iTunes wasn’t “magical product differentiation faerie dust”, it was a horrible punishing obstacle I had to surmount in order to get the damned iPod to do anything. The obstacle never became noticeably less punishing- even just changing the music I had on the iPod hurt.

    As for Android-related faerie dust, IMO HTC enormously failed in 2011 with Beats, paying $300 million some eq settings and a logo, purchased from a huckster (Dre) fronting for more hucksters (Monster, the cable company). They might have something in 2012 with the new camera design, but it remains to be seen. Having the best smartphone cameras didn’t do Nokia much good.

  398. “And the rest of your comment doesn’t really parse very well, but I agree that T-Mobile has the hots for the iPhone. Never meant to say anything otherwise.”

    You keep saying that T-Mobile turned down the iPhone. You have not presented ANY EVIDENCE to support it. Please stop making up stuff.

  399. you said
    “And though T-Mobile (the one carrier without iPhone) ain’t doing so well, nobody in the industry thinks lack of iPhone – as opposed to, say, weak execution and lack of the capital mass to pursue its buildout – is its problem.”

    T-Mobile said (in their earnings report)
    http://www.vision2mobile.com/news/2012/02/customers-flee-t-mobile-for-iphone-4s.aspx

    “Philipp Humm, president and CEO of T-Mobile USA, said in a statement that “not carrying the iPhone led to a significant increase in contract deactivations in the fourth quarter of 2011.”

    “Sequentially, the decline in branded net contract customers was driven primarily by higher branded contract deactivations as a result of the launch of the iPhone 4S by three nationwide competitors in mid-October,” T-Mobile USA said.

    so T-Mobile doesn’t get it? the 700k subscribers they lost isn’t because of something else?

  400. Daring Fireball peeps: you’re right, of course. But before you reiterate the obvious, skip back to this comment where Eric rejects the idea that T-Mobile really meant that they were losing money on lack of iPhone. Eric knows that T-Mobile says that the lack of iPhone is a problem. He just doesn’t believe them.

  401. >Eric knows that T-Mobile says that the lack of iPhone is a problem. He just doesn’t believe them.

    That’s right. T-Mobile’s problems with customer retention, churn, lack of the financial mass required for competitive buildout, and elsewhere are continuous with problems they were having well before “lack of iPhone” was a plausible explanation.

    That’s why I think that they’re now using the iPhone as an excuse. To believe otherwise I’d have to think all those other problem magically stopped being problems at exactly the right time for lack of iPhone to continue the same trend curves. I don’t believe in coincidences that elaborate.

  402. Don’t over-think this.

    Carriers: the iPhone is a better, easier-to-use product for the vast majority of consumers. Better, easier-to-use products are what you should be selling to your customers. Stop selling iPhones, and—given the current state of Android phones—you’re not offering your customers what they want.

    Forget about the numbers; sell better products. If not the iPhone, then figure out how to make your Android phones as good. And don’t kid yourselves; none of today’s Android phones, loaded with crapware that can’t be removed, are as good as the iPhone. The numbers will follow.

  403. For the record, in the two years previous to the quarter we’re discussing, T-Mobile’s problems with customer retention looked like this:

    2010 Q1 -77,000
    2010 Q2 -93,000
    2010 Q3 137,000
    2010 Q4 -23,000
    2011 Q1 -99,000
    2011 Q2 -50,000
    2011 Q3 126,000

    And then in Q4 2011, it’s -706,000. I’m not entirely sure I’d characterize that as “continuous.” Going back to 2009, the biggest delta between quarters is +371,000 in Q4 2009 followed by -77,000 in Q1 2010. And that’s half the size of the difference between Q3 and Q4 2011.

  404. In general, Apple is selling tons of apps vs. Android users, who typically go for the freebies. It might be a reasonable assumption that if iPhone users are more willing to part with their money, they are more likely to purchase their providers over priced add-ons. $20 for this and $40 for that each month times X amount of iPhone users is money in the bank. A loose parallel is Gillette, they don’t make money from the razors, but from the blades.

  405. i don’t understand. if the carriers wanted out, why haven’t they bailed yet? we’re on the eve of the 6th gen hardware, and it appears the only carriers who are unhappy are the one(s) who don’t have the iphone.

    FTA “In an interview with CNNMoney in October, just after Sprint announced it would begin selling the iPhone, Sprint CEO Hesse said the No. 1 reason why customers had left Sprint prior to October was because it had no iPhone.”

    CEO Dan Hesse said he expects that, eventually, the iPhone will be “our most-profitable device.”

    I don’t understand the economics of it, but as an iphone user.. ATT may have realized a loss because I bought a new phone this year.. but how do they account for next year. when I don’t buy hardware, but I continue to pay my monthly.. or the phones I’ve resold? carrier paid ZERO subsidy, but is getting the monthly. I figure at least 2 (my old 4 and 3S) are still running somewhere out there in the world. I gave them 2 data plan paying, non-subsidy sucking customers and they won’t let me upgrade at the best price for 2 years).

  406. Tim F.:

    You keep saying that T-Mobile turned down the iPhone.

    Yeah. They don’t have an iPhone. So either they don’t want one at any price, or they asked about it. If they asked about it, either Apple turned them down flat, or Apple offered to make one that would work on their network at some price and/or schedule.

    On the evidence, T-Mobile management has some issues, but even I don’t believe they’re incompetent enough not to ask Apple about the iPhone. On the evidence, Apple would have at least talked to them — they are dealing with one smaller carrier in the US, and another small carrier claimed explicitly that they turned apple down. So the the question is, did Apple just say “no” or did they say “we could do that but here’s what it would cost and how long it would take”? From T-Mobile’s CEO’s statement “[The iPhone is] a good device and we have always said we’d like to have it on the right terms” I have to believe it’s the latter, but you’re of course free to believe in the tooth fairy or whatever else floats your boat.

    So, if you don’t think that any kind of serious negotiations ever took place, do you feel it’s because T-Mobile really is that stupid, or because Apple is too incompetent to support T-Mobile’s frequency plan?

    You have not presented ANY EVIDENCE to support it. Please stop making up stuff.

    I’ve given my opinion of what happened and the evidence of publicly available information that I think supports it. YMMV and obviously does, but I don’t know why you’re so insecure about your position you have to ask me to shut up.

  407. @Bryant:

    There is no question that T-Mobile defections spiked. But for defections to be ongoing, they had to be doing something else wrong, unless you think that every one of their lost customers was going to AT&T to get an iPhone all along.

    There is no question that the iPhone is a popular device, and its sudden availability on Sprint and its sudden cheapness on Verizon would attract a lot of people. But were those people otherwise happy, or were they at the trigger point where the proverbial one more straw (T-Mobile not getting the iPhone) was enough to push them out?

  408. I think the 4S triggered it; my guess is that they don’t lose as many next quarter.

  409. @mark:

    I figure at least 2 (my old 4 and 3S) are still running somewhere out there in the world. I gave them 2 data plan paying, non-subsidy sucking customers and they won’t let me upgrade at the best price for 2 years).

    I agree that the increased subsidy the carrier pays is due to perceived value to the carrier, and have even discussed both old phones and the reduced cost of support that the iPhone probably offers for the carrier.

    The problem comes when Sprint claims they won’t see profit from the iPhone until 2014, e.g until some time after all last quarter’s contracts run out. But in two years, people get to buy a new phone (and an opportunity to switch carriers). As part of the phone oligarchy, they’re probably used to planning this far out, but Apple’s going to have different plans. Maybe a must-have iPhone that doubles as your desktop. Who knows? But if history is any guide, the price of an iPhone, and what it will cost the carrier to hand one of those out to the customer, is only going up.

    So, if AT&T’s experience matches Sprint’s, it’s only after you give away the phone (or are too lazy to upgrade or to demand a lower monthly price from the carrier) that they make any money on the iPhone.

    This is why AT&T won’t unlock an iPhone, even off contract. But third parties will do it for you, and a lot of those are already running on T-Mobile. If T-Mobile “refarms” their spectrum as planned, it will make it a lot more attractive to unlock an off-contract AT&T phone and use it on T-Mobile.

    Sprint may have it a bit better (might be harder to move one of their phones — not sure), but at the end of the day, if there are any chinks in the wireless oligarchy and contract prices drop (or more people go with prepaid), then that payoff in 2014 won’t actually happen.

    Android phone prices are dropping like crazy. Look at the article Bryant posted about Greece. If, in two years, a customer can buy a really nice Android phone for $150 off contract, or an Apple iPhone for the same price on contract with Sprint ponying up another $400, then the customer is going to demand a lower monthly price for the Android phone if he buys it up front. (T Mobile already has plans that let you pay an adder for a contract phone — essentially adding visibility to the fact you are financing the phone and letting you stop paying for it after two years.)

    So there are a few possibilities here. (1) The situation could stay the same. (2) Carriers might start charging less for off-contract smartphones, like T-Mobile does already. (3) Carriers might be able to keep prices where they are now for people buying the $150 Android phones by simply charging extra per month for iPhone subsidies.

    The question is whether (1) is stable. Given the telecom oligarchy, it might be and Apple will be laughing all the way to the bank. But there are already cracks in the oligarchy — prepaid is rising like crazy, and if T-Mobile and the smaller carriers and MVNOs have any sense they’ll do something disruptive as well. (3) would be great for the carriers, perhaps not so good for Apple. (2) wouldn’t be good for either the carriers or Apple.

  410. @Bryant:

    I think the 4S triggered it; my guess is that they don’t lose as many next quarter

    I agree, to an extent. I think that any T-Mobile subscribers who would have had AT&T as first choice would have been long-gone if iPhone was an issue. That leaves the customers leaving for iPhone on Verizon, who could have left last January but might have hung around until the imminent summer release of the iPhone 5 eventually morphed into the 4S, and the customers leaving for iPhone on Sprint, who couldn’t have even done that until last quarter.

  411. Hmm, wasn’t clear. I definitely agree that it seems unlikely that their spike in lossage continues into this quarter, partly agree that 4S the device had something to do with it, and additionally clarify that availability of any iPhone whatsoever on Sprint also had something to do with it.

  412. Sprint is in immediate trouble, right now. Their Board called off a MetroPCS buyout that was apparently nearly a done deal and the only plausable reason I’ve heard is that they couldn’t afford it. Also they’re looking to float $2billion worth of notes right now. Their network partner is in such grim shape that Google is looking to offload a major chunk of their stock at a ~92% loss.

  413. Two things. First. It’s going to be years before sprint profits off of ANYTHING at this point.

    Second, the iPhone is keeping customers upgrading. http://news.cnet.com/8301-13579_3-20030654-37.html

    Third. Apple is only taking 4.2 percent more than other smartphone vendors.

    Fourth, the iPhone is the only phone that people seem to actually like. Across the board, satisfaction rates for everyone else is abysmal.

  414. @Taiki:

    > Third. Apple is only taking 4.2 percent more than other smartphone vendors.

    What do you mean? And where do you get this number?

  415. @Taiki:

    Fourth, the iPhone is the only phone that people seem to actually like. Across the board, satisfaction rates for everyone else is abysmal.

    Depends on who is asking and how they ask. JD Powers is fairly reputable and gives point information rather than simply “are you happy”. Yes, Apple is on top, but not really by that much. In their survey last September, Apple scored 838 out of 1000, and HTC scored 801.

  416. @Taiki:

    Heh. Just noticed your link was to the bogus usamp survey. That’s over a year ago and is my favorite “this is why asking consumers what they are going to do is a terrible idea” survey. Did everybody really switch on the first day of iPhone availability like they said they were going to?

  417. @Greg:

    [Sprint's] Board called off a MetroPCS buyout that was apparently nearly a done deal and the only plausable reason I’ve heard is that they couldn’t afford it.

    The board probably figured the premium was too high, especially with everything else going on.

    Also they’re looking to float $2billion worth of notes right now. Their network partner is in such grim shape that Google is looking to offload a major chunk of their stock at a ~92% loss.

    I assume you mean clearwire. Right, to Sprint the stock apparently looks cheap. Supposedly the reason for the $2 billion (and money is cheap right now) is to simply outright buy clearwire.

    Which might not make much sense unless you’re Sprint, and just noticed that all the cable operators just entered a deal to sell spectrum to Verizon, and then for both Verizon and the cable companies to resell each others’ services and offer cable/internet and wireless on a single bill.

    Add that to lightsquared’s troubles at the FCC, and suddenly Sprint’s future became very cloudy. I haven’t been following closely enough to know whether adding clearwire instead of MetroPCS makes sense for them, but maybe they are somewhat unrelated.

    Maybe clearwire looks like is a great opportunity (pushed by Verizon and cable perhaps) and happened around the same time that T-Mobile started making overtures to them after AT&T fell apart. Sprint/T-Mobile would stand a much better chance of getting through the FCC unscathed than AT&T/T-Mobile.

  418. Pingback: Links 28/2/2012: More Than 850,000 Androids Activated Daily | Techrights

  419. After my earlier comment on the 850k/day activation rate of Android, I was wondering about the jump from a growth in activations of 30k/day each month to 75k/day in the last two months (from 700k/day to 850k/day). At this pace, there will be 1M Android activations per day in early May.

    Why?

    I guess it is the magical 50% boundary. In people’s mind, the majority of smartphones are Android phones. So that has become the default choice. Users of RIM, Nokia, and other (feature) phones who want to upgrade simply buy an Android phone.

    Android phones have the widest choice of handsets, the widest range of prices, and it is a safe choice. All the alternatives, except for the iPhone, have a sword hanging over them. For each of them there is a fear that they might not be around in a years time.

    So, I think Eric was right to put a mark on the 50% marketshare point. We are now seeing it happening. But I do not think Apple will be bankrupt anytime soon, though.

    However, I still think MS will start to divest this year. Their outlook in the mobile space is becoming more desperate by the day. And with it, their future in Personal computing.

  420. @esr> T-Mobile’s problems with customer retention, churn, lack of the financial mass required for competitive buildout, and elsewhere are continuous with problems they were having well before “lack of iPhone” was a plausible explanation.

    That’s why I think that they’re now using the iPhone as an excuse. To believe otherwise I’d have to think all those other problem magically stopped being problems at exactly the right time for lack of iPhone to continue the same trend curves. I don’t believe in coincidences that elaborate.

    Now compare: http://esr.ibiblio.org/?p=2534
    “I actually like my cell carrier.”

    and http://esr.ibiblio.org/?p=3285
    “HTC’s carrier partner T-Mobile, as the #4 in the U.S. market, was polishing a valuable reputation as the carrier that doesn’t screw you in hopes of gaining market share.”

    (etc)

  421. Everyone saying that T-Mobile doesn’t carry the iPhone is wrong. Of course T-Mobile carriers the iPhone – just not T-Mobile USA, Inc. http://www.t-mobile.de/iphone/0,18389,22260-_,00.html

    Perhaps the parent of T-Mobile USA has realized what’s alluded to in the article and does not deem it profitable enough to bring it to the US market?

  422. >Now compare: http://esr.ibiblio.org/?p=2534 “I actually like my cell carrier.”

    Yes, as a customer I actually like T-Mobile. This doesn’t keep me from noticing their weaknesses and problems. I’m not like an Apple fanboy; I don’t feel any need to worship my service providers or any inconsistency in criticizing them.

    Sadly, having a deserved reputation for not screwing their customers does not so far seem to have offset T-Mobile’s other problems. I’ve actually been thinking I should write something about this, but I’m not sure I understand the causation well enough.

  423. I assume you mean clearwire. Right, to Sprint the stock apparently looks cheap. Supposedly the reason for the $2 billion (and money is cheap right now) is to simply outright buy clearwire.

    Which might not make much sense unless you’re Sprint, and just noticed that all the cable operators just entered a deal to sell spectrum to Verizon, and then for both Verizon and the cable companies to resell each others’ services and offer cable/internet and wireless on a single bill.

    Yes, I meant Clearwire. Sprint’s experiment in “partnering” in order to outsource paying for a network buildout has turned into a massive failure. Clearwire is losing lots of money, thanks to the Verizon deal has no prospects for major new customers, and is saddled with dead-end network technology that needs to be replaced. Google’s willingness to take a 92% loss is a good sign of how bad things are. Sprint might as well buy them out while they’re so cheap- if you have to pour money down a hole to prop up an investment, it might as well be your investment instead of someone else’s. But that still leaves Sprint needing to pour money down the Clearwire hole.

    Wimax is a failure, Clearwire is a failure. Sprint has problems now, and to fix them it needs a great deal of money. But instead what it has is debt, and a truly stupendous new financial commitment to Apple. Sprint’s ambitions seem to be much greater than their resources or capabilities. The MetroPCS situation seems to be an indicator that someone inside Sprint is realizing they’re out of money. (Yes, Hesse is in serious personal trouble, too.)

    As for the $2billion they’re trying to raise, buying out Clearwire shouldn’t take all of that but then they still wouldn’t have the cash for a proper LTE buildout. (Which ironically might be part of the reason they wanted to buy MetroPCS- for their new LTE network.)

    The iPhone may actually have killed Sprint.

  424. @esr It is possible to make assertive statements without insulting folks in the process. Doing so simply shows you don’t believe your own hype.

    Apple fanboys criticize Apple and Apple products lots. All fans do. They are often the most nitpicky of critics but their criticism is in the context of how to make things better. Not to simply be haters. In any case, Apple has lots of fans because they make really good products. That’s pretty much as it should be.

    As far as your analysis goes, you’re very selective in what data you accept. Both Sprint and T-Mobile have stated that the iPhone is critical to success in the US market. If Android phones are that much superior to the iPhone then the iPhone wouldn’t be a factor nor a very good excuse.

    That all three iPhone carriers massively gained subs at T-Mobile’s expense indicates it’s a very valid and very important factor.

    The iPhone hasn’t had it’s high water mark yet in US share. There’s not going to be any imminent collapse in market share that you’ve been predicting/hoping for a long while now.

    Booming demand for the latest iPhone model helped Apple beat all phones using Google’s Android platform in the U.S. smartphone market in the fourth quarter, data showed on Wednesday.

    Research firm Kantar Worldpanel ComTech said Apple’s share of the U.S. market doubled from a year ago to 44.9 percent in the October to December period, just beating Google’s Android smartphones, which slipped to 44.8 percent from 50 percent.

    “Overall, Apple sales are now growing at a faster rate than Android across the nine countries we cover,” Sunnebo added.

    http://www.reuters.com/article/2012/01/25/apple-google-microsoft-idUSL5E8CO4QP20120125

    Android may have slipped a little this past qtr in US share.

    (Reuters) – Google Inc’s accelerated efforts to carve out a position in the fast-growing mobile and social networking markets leapt into the spotlight Friday, a day after the giant Internet company reported a rare earnings miss.

    Google’s heavy investments in mobile and social network initiatives — to stave off competition from rivals Apple and Facebook — and its planned $12.5 billion acquisition of smartphone maker Motorola Mobility Holdings Inc have raised investors’ concerns.

    http://www.reuters.com/article/2012/01/20/us-google-idUSTRE80I26020120120

    Mmmm….

    If the iPhone was a competitive disadvantage none of the other carriers would have jumped on board, or in Sprint’s case bought their way on board. T-Mobile simply couldn’t afford to do what Sprint did and still build out their LTE network. Even a partially built out LTE network is of more value when getting purchased by someone than iPhones built for a defunct 3G network.

    Deutche Telekom is looking for the best exit strategy. Not long term survival strategy. T-Mobile is still hosed though. Their LTE buildout will trail everyone else despite $3B and some spectrum from AT&T and they’ll have no iPhones of their own.

    I hope you T-Mobile customers will like being part of Sprint in a year or two.

  425. Drop the iPhone because it kills margins? The author doesn’t seem to understand the simple economics of competition.

    The choice for the carrier is between lower profits selling iPhones and even lower profits by not selling iPhones. Those pre-iPhone high-margin days are gone, the iPhone’s extreme desirability made sure of that. Even if Apple sold the iPhone for less, the carriers will still drive down iPhone margins as they compete with each other to capture the iPhone customer. The author is saying if no carrier sold the iPhone then it’s as if the iPhone never existed and we can all go back to the salad days. That’s just ridiculous. You might as well ask for a time machine.

    Now every iPhone carrier would certainly encourage their competitors to drop the iPhone thus making them an exclusive vendor. But what carrier will do their competitor that favor?

    Sorry, the whole article just defies simple economic logic.

  426. >The choice for the carrier is between lower profits selling iPhones and even lower profits by not selling iPhones.

    You’re not paying attention. What the analysts are telling us is not that iPhone leads to low profits but to no profits.

  427. @esr

    You’re not paying attention. What the analysts are telling us is not that iPhone leads to low profits but to no profits.

    BS. What analysts are saying this? And pointing to your own post about the subject doesn’t answer the question. Show me a link to where analysts are saying the iPhones leads to no profits.

  428. >Show me a link to where analysts are saying the iPhones leads to no profits.

    The links are in my post. Now stop wasting my time, please.

  429. @esr> I’m not sure I understand the causation well enough.

    Many believe you don’t understand the primary drivers of Android’s “success”, either.

    @esr> What the analysts are telling us is not that iPhone leads to low profits but to no profits.

    Wrong.

    Rapid iPhone adoption by a carrier customer base leads to short-term negative cash-flow, due to acquisition costs. Given that each (US at least) iPhone customer is under contract for 2 years, and has a mandatory data plan, they are (ahem) quite profitable over the term of the contract.

    That’s what the analysts are telling us. Another example of your fury at Apple blinding your thinking.

  430. Problems with these assumptions:

    1. All phones are “subsidized” and the original iPhone was not when it first became available. It wasn’t until the second generation iPhone where subsidies were offered with a contract.
    2. The carriers with iPhones have incredible sales figures
    3. The carriers are evil
    3a. – RIM & Android smartphones cannot release an OS update without carrier approval
    3b. – It is to the carrier’s (and manuf) advantage to get people to buy a new phone to get new OS
    3c. – Carriers charge for SMS when it costs them ZERO to support it
    3d. – Some Carriers (AT&T) now only charge unlimited SMS for $20/mo!
    3e. – Carriers are bandwidth data capping heavy users who still have unlimited accounts
    3f. – Carriers force custom apps onto a new phone that you cannot even remove
    4. Apple free’d customers from the evil carriers by supporting iOS upgrades to even older devices even if all the new features didn’t work. Apple controls the iOS update releases not the carrier.
    5. Apple supports free texting over the Internet between iOS devices (BlackBerry did this first with BBM, but Apple has market share to actually make a difference).

    Android is growing worldwide but it’s split across many different devices. Plus, many previous Android users have been switching to the iPhone in droves as their contracts expire. So this number is really important but we won’t likely see the numbers on how many former Android users switched to an iPhone!

    I support Mobile devices in a huge enterprise environment. I can tell you that Android adoption is nowhere near the iPhone numbers. Many of my companies Android users end up switching to the iPhone when they are upgrade eligible. The end users just want their device to work and be reliable they don’t care about customizing it like most computer savvy folks do. They just want the cool apps and a device that works easily. Android is confusing to many regular users.

  431. >Many believe you don’t understand the primary drivers of Android’s “success”, either.

    True. There’s a technical term for these people. We call them “fools”.

  432. Not to mention, most new Android phones are enormous is size. Won’t fit in a pocket anymore and requires two hands to use effectively. What is this? Is it the 1980’s with Boomboxes getting ever larger? First the phones got smaller and smaller and now they are getting bigger and bigger. Except for Apple, who seems to understand that the current size of an iPhone is ideal and to make it much larger is silly. They just increased the resolution making the screen much more readable.

  433. November 11 2008 “The iPhone should be feeling teeth in its ass right…about…now.”

    November 11, 2008: http://esr.ibiblio.org/?p=578&cpage=1#comment-229089
    “Android has Sprint and the rest of the lockin-mad mobile phone vendors shit-scared and scrambling to adapt. The smart ones will join Android rather than trying to beat it.”

    No, Eric. The smart kids are on iPhone.

    November 11, 2008: http://esr.ibiblio.org/?p=578&cpage=1#comment-229095
    “[T-Mobile] have a justified expectation that the openness of the stack will grow their business, which is exactly the argument for open-source hardware drivers I’ve been making for ten years.”

    History proves otherwise it seems.

  434. >History proves otherwise it seems.

    Oh? So Apple closed the 20% gap in market share while I wasn’t looking?

    You are amazingly blind and fanatical. It’s repulsive but morbidly fascinating to watch, like a bad auto accident in progress.

  435. @esr> There’s a technical term for these people. We call them “fools”.

    You’ve got wrinkles round your eyes and grey in your hair,
    You’re puttin’ on a little bit of weight but you don’t seem to care,
    Fool say, “Hey slick you lookin’ good!”, but it’s a lie.
    That fool in your mirror is singin’ the same old song,
    But it still don’t change the fact that your theory is wrong.

  436. >Not to mention, most new Android phones are enormous is size. Won’t fit in a pocket anymore and requires two hands to use effectively. What is this? Is it the 1980?s with Boomboxes getting ever larger?

    It’s not “most”, but I think I can tell you why it’s happening. I read somewhere that a lot of these super-large phones are being bought by women. Android handset makers have identified a market segment that wants their phones to be, effectively, small browser tablets that can fit in a purse.

  437. Quoting Gruber

    My hypothesis has long been that Android has very little traction in and of itself. What has traction is the traditional pattern where customers go to their existing carrier’s retail store to buy a new phone, listen to the recommendations of the sales staff, and buy one of the recommended phones. Tens — hundreds? — of millions of people have done this and walked out of the store with a new Android handset. (By my theory, this is why Android phones are so under-represented compared to the iPhone in terms of usage — things like mobile web traffic. A lot of people think of them just as phones.)

    There is no such traction for the idea of going into your phone carrier store and buying a computer. That’s why carrier-subsidized netbooks didn’t take off, and that’s why carrier-subsidized Android tablets haven’t either.

    This all in reference to Rubin’s latest, ““there’s no organized way for consumers to recognize [Android] as a viable platform” quote, in reference to Android tablets. Rubin is also quoted as stating that Google will ‘double down’ on Android tablets in 2012.

    But if you look closer, it serves as an explanation of how for most of the market, Android is merely a set of training wheels for moving into the smartphone space. Once folks discover what they want out of a smartphone, they’ll move to a better (more expensive) Android device, or move on to iOS or perhaps WP.

    The issue for Android is if Bada takes over the low-end in China.

    As for this:
    @esr> “So Apple closed the 20% gap in market share while I wasn’t looking?”

    As of the end of 2011 Apple had 316 million iOS devices. Android has 300 million as of the Rubin reported numbers.

    Next question.

  438. >As of the end of 2011 Apple had 316 million iOS devices. Android has 300 million as of the Rubin reported numbers.

    Wow, that’s desperation talking. Now count phones.

  439. Rubin is essentially admitting that Android’s rise in smartphones has less to do with Android, and more to do with the carriers.

    In that light, Google’s decision to turn their backs on their original ideals for Android and instead get in bed with the carriers was a smart one.

    The very same carriers who are sacrificing short-term profits to get the iPhone to their customer base. You might wish to ask yourself, “Why?”

    Across all the various OEMs that make Android tablets, 12 million have been sold in total, ever. In contrast, Apple sold 15 million iPads last quarter.

    I predict that even if Google does “double down”, they’ll be far behind the iPad by the end of 2012, especially given the launch of the iPad 3 in 7 days.

  440. @esr> Now count phones.

    Google counts “activations” across the platform.

  441. Nigel: Show me a link to where analysts are saying the iPhones leads to no profits.

    Eric: The links are in my post. Now stop wasting my time, please.

    They aren’t, actually. The article you linked to claims that iPhones are reducing profit margins, but it does not at any point claim that they are reducing them to zero. Furthermore, the article concludes with an analyst claiming that “The iPhone in the long term will turn out be profitable for carriers once they raise their price points.”

  442. >The article you linked to claims that iPhones are reducing profit margins, but it does not at any point claim that they are reducing them to zero.

    OK, this is unsettling. The article I was thinking of has apparently been edited some time in the last two hours. I could tell at a glance before I reread the prose, because the paragraph weights are different. Then I went looking for the zero-profits quote I reread two hours ago (because I actually did check) – and it was gone.

    I’ve heard about news sites retroactively editing stuff without time-marking the change, but this is the first time I’ve ever had it actually bite me. It’s Orwellianly disturbing. Shit. Might mean I’m going to have to start keeping local copies of stuff I reference. I apologize for snarking at you; what you read justified your remarks.

  443. Furthermore, the article concludes with an analyst claiming that “The iPhone in the long term will turn out be profitable for carriers once they raise their price points.”

    I haven’t read the altered article, but even this quote is enough is parsed properly. It’s very mealy-mouthed, but look at what it’s actually saying:

    1) Right now, in the short term, the iPhone is *not* profitable for carriers and there’s no way to make it so.
    2) Even in the long run, unless something changes, the iPhone still will not be profitable for carriers.
    3) In order to salvage something from the situation and make the iPhone profitable, and even then it will only be effective over the long term, the carriers will be required to raise prices. (Which, oddly enough, they’ve begun to do.)

  444. @Nigel:

    [The analysts] aren’t [saying the iPhone reduces carrier profits to zero], actually.

    Right. Eric was being generous. One quote from the article (that Eric in fact repeated with emphasis in his post) was:

    “A logical conclusion is that the iPhone is not good for wireless carriers,” says Mike McCormack, an analyst at Nomura Securities. “When we look at the direct and indirect economics that Apple has managed to extract from the carriers, the carrier-level value destruction is quite evident.”

    “Carrier-level value destruction” sounds like way past zero to me, and the article makes a point of saying other analysts agree with that viewpoint.

    Furthermore, the article concludes with an analyst claiming that “The iPhone in the long term will turn out be profitable for carriers once they raise their price points.”

    Right, and (1) the article doesn’t say any other analysts agreed with this, and (2) there’s a major assumption there about carriers having service pricing control, and (3) even if the carrier has some service pricing control, that’s for new iPhone customers going forward. Anybody who Sprint sold an iPhone to last quarter is a loss for them until at least 2014. Might even be worse. Some articles say 2015. But what happens when Sprint tries to raise the price at the end of a contract? How many customers will stand for that?

    The domestic carrier strategy relies on two things: (a) the hope and prayer that the average service life expectancy of an iPhone is way past two years; and (b) the fact that when a domestic carrier subsidizes an iPhone, some sort of unholy alliance, probably led by Apple means that even after you have paid the thing off, you still can’t use it on another carrier.

    So the carrier subsidizes the phone for two years, hoping that it remains useful after that, and knowing that if it remains useful after that, it will only be useful on their network.

    Historically, this has probably been true, but a couple of things could change this. If government intervened and forced smartphone unlocking after the contract, that would be bad for carriers, especially for agnostic handsets like the iPhone 4S. Or if new desirable Android phones keep getting cheaper, the lock-in power of an old iPhone might be dramatically reduced.

    It will all be entertaining to watch, but the very idea of “investing” in something you don’t actually own (the customer’s handset) with no guaranteed returns, in a volatile market (for handsets) seems, uh, creative. More power to ‘em if they actually make anything off it. Meanwhile, Apple and the customers seem quite happy.

  445. >“Carrier-level value destruction” sounds like way past zero to me, and the article makes a point of saying other analysts agree with that viewpoint.

    To be fair to Nigel, it sounds like way past zero to me too, but there is a non-crazy interpretation that it’s merely devastating margins rather than eating capital. So I wanted to find the more specific quote about wiping out margins, which seems to be gone.

  446. @esr No problem. I double checked to make sure they didn’t say that before calling you out. I didn’t know how the original article read.

    @patrick

    Anybody who Sprint sold an iPhone to last quarter is a loss for them until at least 2014. Might even be worse. Some articles say 2015. But what happens when Sprint tries to raise the price at the end of a contract? How many customers will stand for that?

    Anybody who Sprint sold a Galaxy S2 to last quarter is also a loss for them. Just a shorter period because the subsidy is lower.

    What happens when Sprint raises the price at the end of the contract? Depends on how and how much. I assume that data plan prices are subject to change although most will grandfather you in to existing rates. Switching from a legacy Sprint contract with a minor price bump to Verizon for an even more expensive plan strikes me as a non-optimal from a purely economic perspective. You might do so because the Sprint network sucks in comparison or other reason anyway.

    As much folks bitch about AT&T I’ve like their rollover minutes and they are price competitive with Verizon. If they had rollover family data plans I’d be even happier. Whomever does that, I’m jumping ship to.

    That’s also why I like the iPad. I can drop the data plan for months I don’t expect to leave wifi. When I’m wrong I pick it up for $20 for the rest of the month. That’s a nice savings for us. One driven by Apple to improve the user experiences of their customers. Amazon gets kudos for the original kindle too from that regard.

  447. To be fair to Nigel, it sounds like way past zero to me too, but there is a non-crazy interpretation that it’s merely devastating margins rather than eating capital. So I wanted to find the more specific quote about wiping out margins, which seems to be gone.

    I do not have a very high opinion of AT&T despite liking rollover minutes (like most we have more than we can use anyway).

    I still can’t imagine very many scenarios where I get to do something where they consistently end up losing money by having me as a customer. I can believe they’ll tolerate making less money on me than me to hold me with a new iPhone every couple years. But an outright loss?

    What? They’re going to make it up on volume?

    Then I have to believe that both Verizon and Sprint management are dumb enough to want customers with a negative ARPU as opposed to laughing at AT&T all the way to the bank.

    I can believe that T-Mobile management has a different view based on different goals. DT want’s out so I can imagine they’re maximizing resale value. Typically the most important aspect is customer base. In this case I believe they’re going for network value if they can’t actually stop near term erosion of the customer base. Having a 3G iPhone compete against 4G iPhones next year isn’t a good value proposition for them.

  448. Anybody who Sprint sold a Galaxy S2 to last quarter is also a loss for them. Just a shorter period because the subsidy is lower.

    That’s actually the whole controversy in a nutshell. The subsidy being lower shortens the time until Sprint is making money to less than the length of the contract. Wanted to emphasize that. Because the way that one article is worded, an iPhone that Sprint sells in 2012 doesn’t become profitable until 2014 or 2015. On a 2 year contract. WHAT?

  449. @Nigel:

    I can believe that T-Mobile management has a different view based on different goals. DT want’s out so I can imagine they’re maximizing resale value. Typically the most important aspect is customer base. In this case I believe they’re going for network value if they can’t actually stop near term erosion of the customer base. Having a 3G iPhone compete against 4G iPhones next year isn’t a good value proposition for them.

    If I were T-Mobile, I would probably (a) beef up the network like they’re doing (including the “spectrum refarming” that makes them compatible with global handsets including the iPhone) and then (b) sue or lobby for all out-of-contract phones to be unlocked.

    A successful campaign could easily catch on at the grassroots level. Sure, AT&T can claim that you read the fine print when you bought the phone, but everybody knows that’s not really true…

    So if the government forced AT&T to admit that a non-contract phone or a contracted paid-up phone was the customers property and then to unlock it, all sorts of interesting things would start to happen.

  450. @Nigel:

    Anybody who Sprint sold a Galaxy S2 to last quarter is also a loss for them. Just a shorter period because the subsidy is lower.

    Right, but this misses the point that the back-end profit is probably theoretically guaranteed in the case of most Android phones, because the carrier will make some of it before the 2 year contract is over. If you look at the dates, Sprint’s relying on not making money on iPhones until they are no longer under contract.

  451. Re: “what happens when Sprint tries to raise the price at the end of a contract?”

    The vast majority of the customer base doesn’t even notice. This isn’t hypothetical; I work in a subscription-based industry and I happened to participate in an involuntary experiment along those lines which is still under NDA. If a bill is being auto-paid, most people won’t even think about it. The carriers are not really encouraging you to go to autopay for the sake of the environment.

  452. >Honest to $DIETY, there’s no change between the 24th and now.

    Weirder and weirder. It’s not a substantive point now, because Patrick Maupin has done a pretty good job of demonstrating that “zero revenues” is deducible without the quote I went looking for. But…what the hell actually happened?

    /me is confused

  453. @Bryant:

    > The vast majority of the customer base doesn’t even notice.

    I absolutely believe this to be true in the general case. But when you crunch a few numbers and realize that the spike on a new iPhone introduction is basically a lot of people who were waiting for that iPhone model to happen, and then you extrapolate that enthusiasm to other quarters when people are going off contract, I think the iPhone may currently be unique in its ability to drive customer upgrades. It’s no surprise that Verizon dropped their “new every two” promotion at the same time as they rolled out the iPhone, but it will be interesting to see if they manage to make that stick. After all, after the contract expires, the user is a free agent.

    And Apple’s pumping out a new extremely desirable handset roughly once a year. As we have discussed, a lot of iPhone upgrades mean that the old iPhone is being sold/gifted to a different user, so those are still wins for the carrier, and are probably one reason why the carriers don’t mind so much. But if there is ever any shift in the market that means that older iPhones don’t have some “must-have” feature that is available on a shiny new sub-$150 Android, the subsidy of that particular iPhone might be a loss. And, as I mentioned, if the legality of post-contract carrier lock-in is successfully challenged, any “investment” the carrier made is moot — now the carrier needs to compete with others for the very phones they financed.

  454. @esr:

    > /me is confused

    There are a couple of links from that article to others that might contain other info you read at the time.

  455. @Bryant:

    One more point: while it’s in the carrier’s interest (and apparently, at least in the case of Sprint, a requirement to generate any profit whatsoever) for a user to keep using his iPhone longer than two years, it’s in Apple’s interest to sell him a shiny new one at least once every two years. I think, no matter our other disagreements, we can all agree that Apple has been executing on its plans much better than any of the carriers…

  456. > But…what the hell actually happened?

    Nothing happened. The article was a professional hit piece, designed to leave an impression, and it (apparently) worked.

    BGR (an Apple fan site) copied the same quote from the CNN article. Check it.
    http://www.bgr.com/2012/02/08/carriers-hate-the-iphone/

    And here is a fundamentally sound analysis that you might wish to read:
    http://untether.tv/2012/its-not-the-iphone-carriers-hate-its-apple/

    You got led astray by the Gartner report, too. You saw the 50.9%, but haven’t ever acknowledged that Gartner previously had Android at 53%. When you crowed about the 50.9%, you were crowing about a down period for Android (based on Gartner’s numbers).

    Primary sources, my friend. Primary sources.

  457. @Larry Yelnick:

    > And here is a fundamentally sound analysis that you might wish to read…

    From my perspective, it really says pretty much the same thing as the other articles, and some of the more flattering things it says about why the carriers are addicted to the iPhone are unsupported, such as “Ultimately, the iPhone user is the most profitable customer on all the telcos’ books. Hefty subsidy or not, these customers that are responsible for lifting ARPU to 1.9x more than a regular cell phone user.”

    Well, kinda, sorta. The AT&T quarterly info she references describes smartphone users as doing that, and doesn’t distinguish between Apple and non-Apple smartphones.

  458. @patrick

    Read it again. It doesnt say that any iPhone sold today doesnt make money until 2015 but they won’t break even on the 15.5B spent until after 2015. They are spending this over 4 years so they don’t stop investing until 2014.

    They don’t suddenly book 15.6B profit in 2015 so iPhone customers are expected to be profitable during that 2 year contract period. The higher iPhone ARPU means that they are generating much more profit every year. Short term margins are depressed until the iPhone subsidy is paid back…probably late year 1 or early year 2 given the size of the subsidy.

  459. > Short term margins are depressed until the iPhone subsidy is paid back…probably late year 1 or early year 2 given the size of the subsidy.

    They also expect iPhones as a percentage of installed base to grow over those 4 years. So the revenue growth will track / follow growth of new sales, too. Classic waterfall revenue stream modeling.

    As I’ve said before, a next-gen iPhone will have LTE, (the chipsets are getting better in terms of power budget with each passing quarter), and all three major carriers are either actively building out LTE plant, or have announced a radical shift in strategy away from WiMax to LTE (is Sprint buying Clear to get out of its contractual obligations?), and this is in (possibly large) part due to iPhone.

    These buildouts cost real money, and Sprint may be allocating some of these costs against future iPhone ‘costs’. (They probably had to agree to move to LTE to get the iPhone. Perhaps this is what blocked T-Mobile.)

  460. > Well, kinda, sorta. The AT&T quarterly info she references describes smartphone users as doing that, and doesn’t distinguish between Apple and non-Apple smartphones.

    Patrick, your bias is showing.

    The iPhone accounted for about 80 percent of the smartphones AT&T activated in the fourth quarter of 2011, up from 70 percent just before Stephenson made his vow to aggressively market other smartphone platforms a year ago.

  461. @Nigel:

    Read it again. It doesnt say that any iPhone sold today doesnt make money until 2015 but they won’t break even on the 15.5B spent until after 2015.

    That’s a distinction without a difference — some iPhones will make money sooner, but in the aggregate they won’t make more than the 15.5 billion until 2015.

    They don’t suddenly book 15.6B profit in 2015

    so iPhone customers are expected to be profitable during that 2 year contract period.

    Doesn’t follow. iPhone customers are expected to essentially repay the 15.6 billion (which is somewhat PAYG — there’s no way Sprint ponied that up up front) over the two year contract period. But the point stands that for each customer, Sprint only really starts to see profit (as in makes more than they spent) after two years, and how well that works out depends heavily on customer inertia at the end of the time frame.

  462. @Larry Yelnick:

    Patrick, your bias is showing.

    In what way?

    The iPhone accounted for about 80 percent of the smartphones AT&T activated in the fourth quarter of 2011, up from 70 percent just before Stephenson made his vow to aggressively market other smartphone platforms a year ago.

    And that has what, exactly to do with whether an iPhone customer generates more ARPU than a non-iPhone smartphone customer?

  463. @Larry Yelnick:

    > They also expect iPhones as a percentage of installed base to grow over those 4 years.

    Well, yeah, they kinda bet the farm on that when they agreed to buy over 30 million of the things…

  464. > Well, yeah, they kinda bet the farm on that when they agreed to buy over 30 million of the things…

    If they don’t convert any of the installed base, Sprint’s iPhone base (over all subscribers, not just smartphone subscribers) will be 36% at the end of four years. There are scenarios where Sprint is 50% iPhone across all subscribers four years from now.

  465. @esr:
    “I think I can tell you why it’s happening. I read somewhere that a lot of these super-large phones are being bought by women. Android handset makers have identified a market segment that wants their phones to be, effectively, small browser tablets that can fit in a purse.”

    I totally buy this. I am a large-purse woman, and in fact my Nook (repurposed to a general-purpose Android tablet) is often carried around in my purse, where it fits nicely.

    On the other hand, trying to browse the web on my Optimus is a real pain, while I can read it comfortably on my Nook.

    Unmarried men — heck, perhaps even some married men — may not realize that most women’s outfits (excluding jeans) do not have pockets, which makes “pocket-sized” irrelevant for products targeted at us.

  466. > And that has what, exactly to do with whether an iPhone customer generates more ARPU than a non-iPhone smartphone customer?

    Not what she said. What she said was, “these customers that are responsible for lifting ARPU to 1.9x more than a regular cell phone user.”

    “regular cell phone user” .vs smartphone user, and 70-80% of those are iPhone customers on AT&T.

  467. Seen in the WSJ today:

    “New Nokia Phones Disappoint”

    At the Mobile World Congress trade show, Nokia showed the Lumia 610, which runs WinPhone 7.5. However, they did not show any devices running the next version of WinPhone (Windows 8), and none of the devices they showed cost less than $100. When you remember that Nokia’s roots are in the feature-phone world at lower price points than smartphones, this could take on some significance; will current Nokia consumers be interested in a substantially more expensive phones, and if they are, won’t they just jump to Android or iOS? The article comments on “cheap Chinese smartphones running Android…[sold around the $100 mark].”

    Nokia also showed a new $600 smartphone based on Symbian. Apparently they haven’t quite walked away from Symbian yet.

    Great quote from an analyst: “We were looking for a cheap Windows phone at $200 on a new version of the software. We got a phone at $260 on the old software.”

  468. “Weirder and weirder. It’s not a substantive point now, because Patrick Maupin has done a pretty good job of demonstrating that ‘zero revenues’ is deducible without the quote I went looking for. But…what the hell actually happened?”

    No snark intended: you read what you wanted to read. I mean… eh, you’re not going to hear this in the spirit I mean it in, but Eric, you have a serious case of confirmation bias. When someone says something you want to believe, you don’t examine their motives and credibility before you believe ‘em. When someone says something that contradicts your desired reality, you tend to decide that they aren’t worth listening to. Sometimes it’s because they’re leftist, sometimes it’s because they’re Apple fanatics. But it’s a distinct tendency.

    Do what you will with the observation.

  469. @Larry Yelnick:

    And that has what, exactly to do with whether an iPhone customer generates more ARPU than a non-iPhone smartphone customer?

    Not what she said. What she said was, “these customers that are responsible for lifting ARPU to 1.9x more than a regular cell phone user.”

    Way to selectively quote. She said that right after saying “Ultimately, the iPhone user is the most profitable customer on all the telcos’ books.” In other words, “these customers” are “iPhone customers.” A bold assertion that she acted like she gave some data to back up, but the data says nothing of the sort. First of all, it says nothing about the other carriers (and we’ve all heard about how Sprint is reeling — certainly not all their current customers who upgraded to the iPhone when it became available would have fled the network — for that particular subset of customers, the iPhone subsidy is a sunk cost with zero net benefit) and second of all, she only discussed ARPU, not subsidies or other support costs, and used an ARPU number that doesn’t distinguish iPhones from other smartphones. So that does nothing to prove her assertion. For all I know, the ARPU of a non-iPhone smartphone customer on AT&T is higher than for iPhone customers, and the support costs and subsidies are lower, and she provided zero evidence to the contrary.

    I thought you were just counseling Eric not to be taken in by cleverly worded articles that don’t prove what they assert, but perhaps that was the other Larry Yelnick.

  470. The WSJ has an interesting recap article on the subsidy issue:

    http://online.wsj.com/article/SB10001424052970204653604577247471036145902.html?mod=WSJ_WSJ_News_BlogsModule

    Besides discussing the IDC data about US and UK vs. Portugal and Greece, they point out that in Denmark, the carriers had been offering subsidies, but stopped:

    Carriers around the world are having second thoughts about subsidizing phones, presenting a risk for all phone makers—not just Apple. That became clear in Denmark last year after several leading wireless carriers stopped subsidizing phones and lowered their monthly rates to keep up with lower-priced competitors. “We saw that the customers valued lower prices on calling plans, and simpler calling plans, higher than the subsidy on the phone,” said Jon Erik Haug, CEO of the Denmark unit of Oslo-based wireless operator Telenor ASA.

    In the second half of last year, after Danish carriers stopped offering subsidies, phone sales declined about 10% from the second half of 2010, according to IDC analyst Francisco Jeronimo. Smartphone sales continued to grow, but at a much lower rate than in 2010.

    Obviously it would be difficult for that scenario to play out here. Sprint has a contract, and if the article Larry linked to is accurate, Apple will play hardball enough to rotate through the carriers and make sure that one of them always has the kind of contract that keeps them subsidizing the phones, and then the others have to follow suit or lose customers. It’s a treadmill that’s very hard to get off of if Apple really threatens cutting off the carriers’ supply to the iPhone with “shortages” if they don’t play ball.

    Maybe if a class action suit (there are a couple now) to force Apple and the carriers to unlock phones gets any traction (might be difficult after recent Supreme Court decision about arbitration) it will actually help the carriers in the long run by reducing their reliance on Apple’s goodwill. I can see it now: “bring your own iPhone from another carrier and sign up for 2 years, and we’ll give you $200 cash on the spot.”

  471. @Patrick Maupin
    This whole discussion about whether or not iPhones are sold on merit or on subsidies, and whether the subsidies are worth make little difference in the world wide onslaught Android is performing right now on all other platforms.

    In Europe, unlocking phones on request after a fixed period of time is mandatory for pre-paid phones. Phones on contract are generally sold unlocked (you have to pay your contract anyway). Unlocking phones yourself (25 euro at a market stall or phone shop) is legal in Europe too.

    In the Netherlands, you pay a price difference in your contract if you want a more expensive (iPhone) model. So there is much less cross subsidy, if any. I suspect this is the same all over Europe. And I think also in other markets. I cannot imagine, e.g., the Chinese doing otherwise.

    In the end, it matters little whether the US providers bankrupt themselves or get incredibly rich from giving away iPhones.

  472. @esr
    “Might mean I’m going to have to start keeping local copies of stuff I reference. I apologize for snarking at you; what you read justified your remarks.”

    That is why I always literally quote the part I refer to. Because the Internet is too fleeting.

    In your case, you might want to keep dated screen shots if you base a post on specific pages. If you get influence, your posts will change the web real-time. Pamela Jones of Groklaw had extensive experience with that and seems to keep screenshot of *everything*.

  473. Android 4.0 for x86 processors approaches with RC
    http://www.h-online.com/open/news/item/Android-4-0-for-x86-processors-approaches-with-RC-1444566.html

    The Android-x86 Project has published a release candidate of its Android 4.0 “Ice Cream Sandwich” port for the x86 platform. Aimed at netbooks and tablets, it includes the 3.0.8 Linux kernel with KMS enabled and, its developers say, should be able to run at most netbooks’ native resolution.

    Android-x86 Project – Run Android on Your PC
    http://www.android-x86.org/

  474. > She said that right after saying “Ultimately, the iPhone user is the most profitable customer on all the telcos’ books.” In other words, “these customers” are “iPhone customers.”

    True. (And the data shows that most smartphone customers are iPhone customers on AT&T.)

    > First of all, it says nothing about the other carriers

    She was talking about AT&T. The “1.9x more” was linked to http://www.att.com/gen/press-room?pid=22304&cdvn=news&newsarticleid=33762, which is a press release about AT&T’s 4q11 results.

    > (and we’ve all heard about how Sprint is reeling — certainly not all their current customers who upgraded to the iPhone when it became available would have fled the network — for that particular subset of customers, the iPhone subsidy is a sunk cost with zero net benefit)

    category error

    As for ‘data’, here is AT&T discussing the 4Q11 results:

    “In the year, when our competitors began selling the iPhone, we outsold them in every single quarter. We also led all competitors in total wireless subs added, 7.7 million for the year. And we’re #1 in just about every key mobile broadband growth metric: smartphones, emerging devices, postpaid data ARPU and customers on tiered pricing plans.”

    and

    “Strong smartphone sales also drove impressive revenue growth. Total wireless revenues were up $1.5 billion or 10% and wireless service revenues increased by $0.5 billion. We grew postpaid ARPU, as we have done for every quarter for the last 3 years. We also grew it from a much higher base than anyone else. Our postpaid ARPU is $6 higher than our closest competitor. And in 2012, we expect postpaid ARPU to continue to grow and to grow 2% for the year, thanks in part to new pricing plans that we announced last week.”

    As previously-discussed, a heavy percentage of AT&T smartphone customers are iPhone customers.

    So if 70-80% of AT&T’s smartphone customers are iPhone customers, and Sprint has to execute a situation where at least 30% of its customer base are iPhone customers, who is buying all the Android phones? Verizon sells a lot of Droids, sure.

    but the rest? MetroPCS? Soon to be Sprint, and Sprint has to move iPhones into customer hands
    T-Mobile? T-Mobile is shedding customers at an unprecedented rate, and blames the lack of iPhone for its troubles.

    Leap? Cricket will move some Android units, for sure!

    CenturyLink? With a subscriber base of 219k, they’re a rounding error at the big guys. How would they make a difference?

    In the US, Apple’s iPhone surpassed the smartphone market share of all Android devices combined during the final three months of 2012. http://www.reuters.com/article/2012/01/25/apple-google-microsoft-idUSL5E8CO4QP20120125

    Nielsen says something similar. According to Nielsen, 44.5 percent of customers who purchased a smartphone during the last three months of 2012 bought an iPhone, and 57 percent of those buyers got the iPhone 4S. At Nielsen, Android still remained the No. 1 smartphone category, with 46.9 percent of buyers going for a Google-powered device. Also

    http://www.digitaltrends.com/mobile/apple-iphone-4s-sales-cut-into-android-market-share-nielsen/

    Quoting that, “Out of all smartphone users, an average of 46.3 percent own an Android-based device, while 30 percent own an iPhone.” Which suggests that Eric’s 20% delta is shrinking, if not collapsing.

  475. Doesn’t follow. iPhone customers are expected to essentially repay the 15.6 billion (which is somewhat PAYG — there’s no way Sprint ponied that up up front) over the two year contract period. But the point stands that for each customer, Sprint only really starts to see profit (as in makes more than they spent) after two years, and how well that works out depends heavily on customer inertia at the end of the time frame.

    No, it doesn’t hold. Each phone customer is expected to be profitable after 2 years because after 2 years they get a new iPhone.

    Here’s a much simplified example. Assume:

    * Sprint spends $5B per year for a bulk iPhone buy on Jan 1
    * All customers sign up for a 2 year contract on Jan 1
    * Sprint makes $3B per year for each of these customers.
    * All customers get new phones after 2 years.

    So:

    2011 Jan 1 Sprint pays for $5B iPhone 4S and signs up customers called group 1
    2011 Dec 31 Sprint has made $3B on iPhone 4S owners and is $2B in the hole
    2012 Jan 1 Sprint pays for $5B iPhone 5 and signs up customers called group 2.
    2012 Dec 31 Sprint has made another $3B on group 1 for a total of $6B on their initial investment of $5B. They have made $3B on group 2. They have spent a total of $10B and made $9B and are $1B in the hole
    2013 Jan 1 Sprint pays for $5B of iPhone 5S for Group 1 who all wants new phones for a new 2 year contract.
    2013 Dec 31 Sprint makes $3B on Group 1 under the new 2 year contract and another $3B on Group 2. They have spent $15B and made $15B and finally have broken even.
    2014 Jan 1 Sprint pays for $5B of iPhone 6 for Group 2 who all wants new phones for a new 2 year contract. Suddenly $5B in the hole again from $0…
    2014 Dec 31 Sprint is finally showing a profit having made another $6B over the year and a total $21B made for $20B invested. $1B in profit.
    2015 Jan 1 Group 1 customers decide to defect and Sprint gets no more money from them. Group 2 is on their final year of contract.
    2015 Dec 31 Sprint has made a total of $24B after 5 years on a $20B iPhone investment made over 4 years for a $4B profit.

    Every customer is profitable after 2 years. Every customer gets a new phone every 2 years. There is no requirement that a customer uses a single phone for 3+ years for Sprint to make money.

    Assuming otherwise requires criminal negligence on the part of Sprint management. You can quibble with how much money Sprint makes per customer or how long in the 2 year contract before Sprint is making money but to assert that Sprint doesn’t make money on a customer until year 3+ is bogus because almost every customer can and will get a new phone at the end of year 2 and restart the clock.

    So that’s not what the analysts have said. Just that the DEAL doesn’t produce profit until year 4 and I just showed you how that works.

    So your assertion here:

    Right, but this misses the point that the back-end profit is probably theoretically guaranteed in the case of most Android phones, because the carrier will make some of it before the 2 year contract is over. If you look at the dates, Sprint’s relying on not making money on iPhones until they are no longer under contract.

    is clearly false. Sprint MUST rely on making money on EVERY iPhone BEFORE they leave contract because there is no other way they can do so. Just like with every subsidized Android phone. It’s just that it takes longer into the 2 year contract because the subsidy is higher, NOT after.

    Because after simply doesn’t happen.

  476. @Winter:

    In the end, it matters little whether the US providers bankrupt themselves or get incredibly rich from giving away iPhones.

    Sure, in the end, we’re all dead, but if the reason the carriers aren’t offering me a reasonably priced data package is because they want to take money from me and hand it to Apple, I care. Obviously your mileage varies.

    @Larry Yelnick:

    Her: “Ultimately, the iPhone user is the most profitable customer on all the telcos’ books.”
    You: She was talking about AT&T.
    Me: What???

    As far as the rest of your comment goes, if we assume that the extra ARPU for an iPhone customer is around $6 (which is a bad assumption because AT&T has more total smartphone customers than anybody else in any case), then that could easily be lost in the excess subsidy if the user upgrades continuously. So that still doesn’t show that iPhone customers are more profitable. Neither does the EBITDA.

    @Nigel:

    No, it doesn’t hold. Each phone customer is expected to be profitable after 2 years because after 2 years they get a new iPhone.

    Where do you see that assumption?

    Assuming otherwise requires criminal negligence on the part of Sprint management.

    Yeah, pretty much the thesis of Eric’s post. But seriously, can you show me where that assumption is made? IIRC even on Sprint last quarter only 80% of upgrade eligible customers upgraded. Not everybody always wants to spend $200 for the latest when they can, and that seems to be what Sprint is banking on.

    Your example is probably not that accurate. There was probably some NRE, and there is probably some cash to be delivered, not at the start of every year, but whenever a phone actually changes hands. Your thesis about profitability probably requires that the NRE be large and the per-handset to be small. Given the numbers I remember from Sprint’s quarterly, I assumed it was the other way around, but I could have missed something.

  477. @Nigel:

    “Honestly, we’re not doing very well in the tablet market,”

    That’s awesome. The first step is recognizing that you have a problem…

  478. “Sure, in the end, we’re all dead, but if the reason the carriers aren’t offering me a reasonably priced data package is because they want to take money from me and hand it to Apple, I care. Obviously your mileage varies.”

    ‘zactly. I don’t care what the relationship is between Apple and the carriers, but I care very much if I can’t buy an Android phone and put it on a carrier’s network without being required to subsidize someone else’s iPhone. If the iPhone is as popular as Larry keeps asserting, why is it necessary to subsidize it?

    (Note that “subsidize it” is not the same as “finance it”, a distinction that seems to have been overlooked in most of these discussions. Yes, some people don’t want to pay several hundred dollars up front, and would prefer to pay that over time. This can be done without any cross-subsidies.)

    The bigger worry I have is that neither the carriers nor Apple seem very enthused about prepaid. It’s pretty clear that Tim Cook wishes it would just go away. I consider prepaid absolutely essential as a way of avoiding unwanted entanglement with long-term contracts, cross-subsidies, non-transparent pricing schemes, and the like. In my ideal world, we’d have more prepaid options on all of the major carriers’ network. Where is Verizon’s equivalent of Virgin Mobile USA, Boost Mobile, or T-Mobile’s prepaid offerings?

  479. @patrick

    To flip your question: Where on earth did you see the assumption that an iPhone contract doesn’t generate profit within 2 years?

    My example is to show where the DEFAULT assumption (phones contracts make money within 2 years despite subsidies) holds true even given statement made by the analyst. It is necessarily simplified but clearly shows how both can be true. So where do YOU get that the analyst is claiming that at the end of a 2 year contract Sprint is STILL upside down on the subsidy?

    Here is the text from the article that clearly states this:

    All smartphones weigh on carriers’ margins, since wireless carriers pay a hefty subsidy up front to buy the phones from the handset manufacturers. They make up the difference over the life of a two-year contract.

    They did NOT state “They make up the difference over the life of a two-year contract except for the iPhone.” They also don’t state that carriers make nothing because of the subsidies. Just that it depresses the margins.

    Here’s the text regarding the Sprint deal:

    Sprint (S, Fortune 500) revealed in October just how onerous those subsidies can be: The company said it has committed to paying Apple roughly $15.5 billion in up front costs over the next four years, and the carrier does not expect to make money on the deal until 2015.

    Where does it say here that iPhone contracts do not make money for Sprint until after the expiration of the contract? It doesn’t. Just that the deal means that Sprint is in the hole until 2015 over the whole deal where they’re paying into over 4 years.

    Which is exactly the example I provided.

    If you’re referring to another article I missed it. If you can repost the link I can look at it.

  480. That’s awesome. The first step is recognizing that you have a problem

    Problem 1: Amazon is selling tablets at no profit making them up in selling content. We have no content to sell and no ecosystem to sell it with. Not in the same scale that Amazon uses to make it worthwhile anyway. Trying to sell tablets at no profit will probably get the whole division fired.

    Problem 2: Apple can make tablets cheaper than we can because they’re moving millions of units and buying in bulk because they can afford to. If we buy in the same bulk to get the same discounts we probably end up with several warehouses full of unsold Galaxy Tabs. That will probably get the whole division fired.

    Hmm…lets just sell very large pseudo-tablet cellphones to women and hope more men start carrying man bags.

    Not all problems are very tractable.

    If it were me I’d build $150 7″ tablets with very meager specs that shifted the CPU and GPU requirements to a BT 4.0 paired Galaxy S3 phone. Phone in pocket while using tablet in lap for games, movies, etc. Without pairing, it’s essentially a high quality 7″ photo frame with a couple GB worth of storage for music and pictures. The biggest problem will be lag.

  481. @cathy

    The bigger worry I have is that neither the carriers nor Apple seem very enthused about prepaid. It’s pretty clear that Tim Cook wishes it would just go away. I consider prepaid absolutely essential as a way of avoiding unwanted entanglement with long-term contracts, cross-subsidies, non-transparent pricing schemes, and the like. In my ideal world, we’d have more prepaid options on all of the major carriers’ network.

    AFAIK the iPad is sold unlocked and I have the freedom of simply ordering 3G connectivity when I need it. No contracts. Usable on T-Mobile (albeit on Edge). Replace the SIM. How to set up payment I dunno but if it’s a live SIM it should just work.

    http://www.redmondpie.com/3g-ipad-2-unlock-in-united-states/

    Given Apple’s desire for electronic SIM that the carriers vetoed and other moves I have no idea where you get the idea that Apple is hostile to avoiding unwanted entanglements with long-term wireless contracts. Carriers don’t like this but have to put up with it to a certain degree for Apple products. Sometimes Apple gets its way. Sometimes (like with the electronic SIM) not so much.

    As to why subsidies for the iPhone? That’s the way the game is played in the US.

    As to whether the iPhone is competitive in non-subsidized markets not today. If the market shifts to largely pre-paid then Tim would be happy to oblige with lower end models. The iPod touch is only $199 retail.

    What carriers would really hate is a no contract 3G iPod Touch for $299 with Messages, Skype and a $30/month 3GB no contract data plan.

  482. > What carriers would really hate is a no contract 3G iPod Touch for $299 with Messages, Skype and a $30/month 3GB no contract data plan.

    Skype was purchased by Microsoft last year, so that’s not going to happen.

    FaceTime, however, is Apple-controlled. And Apple could certainly figure out something like SkypeOut or GoogleVoice and run it out of its datacenter(s).

  483. @Nigel:
    “Problem 1: Amazon is selling tablets at no profit making them up in selling content. We have no content to sell and no ecosystem to sell it with. Not in the same scale that Amazon uses to make it worthwhile anyway.”

    So partner with someone who has content to sell, but doesn’t want to get into the hardware business.

    Problem 2: “Apple can make tablets cheaper than we can because they’re moving millions of units and buying in bulk because they can afford to. If we buy in the same bulk to get the same discounts we probably end up with several warehouses full of unsold Galaxy Tabs.”

    You have to create the demand before you can scale up like that. Apple didn’t get there overnight.

    Nobody said this stuff was easy!

  484. “AFAIK the iPad is sold unlocked and I have the freedom of simply ordering 3G connectivity when I need it. No contracts. Usable on T-Mobile (albeit on Edge). Replace the SIM. How to set up payment I dunno but if it’s a live SIM it should just work.”

    We’re talking about phones. Tablets are a different market. iPhones are NOT sold unlocked. I have a phone and a tablet and I do not use them in the same way. My tablet can’t make phone calls, and my phone is not a practical device for reading books or complex documents on the Net.

    “Given Apple’s desire for electronic SIM that the carriers vetoed and other moves I have no idea where you get the idea that Apple is hostile to avoiding unwanted entanglements with long-term wireless contracts.”

    Partly from Tim Cook’s comments about seeing opportunities to convert prepaid consumers to postpaid, which means he sees postpaid as more potentially lucrative. Mostly because I am not aware of any way to purchase a current-production iPhone in the U.S. and put it on a carrier prepaid. Since I *can* do that with an Android (and have), I assume — perhaps incorrectly — that the difference is Apple.

    “As to why subsidies for the iPhone? That’s the way the game is played in the US.”

    Really? Who is subsidizing Android phones? If only Apple products are subsidized by carriers, then that is not a level playing field and is simply stealing my payment dollars to subsidize someone else’s iPhone, resulting in higher-than-market prices for my service.

    Again, you need to distinguish between subsidization and financing. Yes, U.S. consumers don’t like to pay up front, so financing phones is the way it’s usually played in the U.S.

  485. @Cathy:

    > Who is subsidizing Android phones?

    To be fair, the carriers do this. But according to comments from Sprint’s CFO and others, apparently Sprint’s iPhone subsidy is $450, and Android phone subsidy is $250.

  486. Again, you need to distinguish between subsidization and financing. Yes, U.S. consumers don’t like to pay up front, so financing phones is the way it’s usually played in the U.S.

    With financing, the price would come down at the end of the contract period. Normally this doesn’t happen, but as Bryant points out, that’s mainly due to consumer inertia. Consumer could ask for a new phone and/or a lower price, but most of them don’t.

    Note that T-Mobile has started having more transparency here — you can actually finance a phone for extra dollars per month for a couple of years, and then the price goes back down.

  487. If iPhones get a bigger subsidy, doesn’t this simply prove that iPhones are more desirable to the carriers than the average Android phone? I’ve been following this discussion but it seems odd to me that this state of affairs seems to be considered unfair and unsustainable or whatever. It’s supply and demand: Apple has something the carriers want more, at least for now. Plus, in retail, different products often have different margins. It’s not really a matter of “fairness” to one product or another.

  488. “If iPhones get a bigger subsidy, doesn’t this simply prove that iPhones are more desirable to the carriers than the average Android phone? I’ve been following this discussion but it seems odd to me that this state of affairs seems to be considered unfair and unsustainable or whatever. It’s supply and demand.”

    I don’t see the consumer in there anywhere. That’s precisely the problem: the carrier has become the customer, and the consumer is ignored.

    I don’t care what’s more desirable to the carriers. I see myself as the customer, and look at this from the standpoint of what’s desirable to me. Because ultimately I am the one paying for it.

  489. > Apple wouldn’t push skype but it doesn’t mean a bunch of folks wouldn’t use a 3G iPod touch instead of a phone.

    Doesn’t matter. Eric still wouldn’t count them against Android.

  490. To flip your question: Where on earth did you see the assumption that an iPhone contract doesn’t generate profit within 2 years?

    Well, according to Sprint, they are hopeful that the investment will be profitable sometime:

    Moreover, service providers frequently offer wireless equipment, such as devices, below acquisition cost as a method to retain and attract subscribers that enter into wireless service agreements for periods usually extending 12 to 24 months. Equipment cost in excess of the revenue generated from equipment sales is referred to in the industry as equipment net subsidy and is generally recognized when title of the device passes to the dealer or end-user subscriber. The cost of multi-functional devices, such as smartphones, including the iPhone, has increased significantly in recent years as a result of enhanced capabilities and functionality. At the same time, wireless service providers continue to compete on the basis of price, including the price of devices offered to subscribers, which has resulted in increased equipment net subsidy. We have entered into a purchase commitment with Apple, Inc. that increases the average equipment net subsidy for postpaid devices resulting in a reduction to consolidated results from operations and reduced cash flow from operations associated with initiation of service for these devices until such time that retail service revenues associated with customers acquiring these devices exceeds such costs.

    But, of course, maybe it won’t be:

    We expect to incur expenses to attract new subscribers, improve subscriber retention and reduce churn, but there can be no assurance that our efforts will result in new subscribers or a lower rate of subscriber churn.

    To take what they say here and put it in a concrete example: last quarter, Sprint sold 1.8 million iPhones. Over a million of those were to existing subscribers. Some portion of those million handsets (over $450M in subsidies) were presumably for subscribers who were already using smartphones and not inclined to jump ship. To the extent they are merely selling iPhones to existing smartphone customers who were otherwise happy, they aren’t attracting new subscribers or reducing churn. Even if they’re upselling a current customer with a featurephone that already had data, my reading of the 10K suggests that’s only an additional $10/month ARPU, or $240 over the life of the contract.

    So, it’s not entirely clear to me or to them when or if they will break even and start making a profit on this investment. I understand what you’re saying, that the recurring profit somewhat lags the recurrent investment, but occurs on each phone before the end of the contract. OTOH, I would assume that most Sprint customers who really want an iPhone will get one ASAP, which would have the tendency to front-load the costs and profit. I would also assume that Sprint management (unless they’re completely incompetent) would also make this assumption, and factor this into their 2015 projection, which would mean that the average iPhone is perhaps barely profitable at the end of the contract.

  491. > according to comments from Sprint’s CFO and others, apparently Sprint’s iPhone subsidy is $450, and Android phone subsidy is $250.

    Apple gets a piece of the recurring revenue stream.

    But the WSJ reported “40% higher”:

    “In the U.S., carriers pay Apple an estimated $400 every time a customer buys an iPhone with a two-year contract, analysts say. Sprint has acknowledged that iPhone subsidies run 40% higher than they do for other smartphones on average.”

    and $250 * 1.4 is only $350. The Android subsidy would need to be around $285 for the 40% higher iPhone subsidy to be $400.

  492. @Patrick:
    “With financing, the price would come down at the end of the contract period. Normally this doesn’t happen, but as Bryant points out, that’s mainly due to consumer inertia. Consumer could ask for a new phone and/or a lower price, but most of them don’t.”

    Sorry, I should have said “cross-subsidization” rather than simply “subsidization”. While I don’t like the fact that some carriers don’t want to drop your service fees after the phone is paid off, the real issue I’m trying to focus on here is the use of profits from Android subscribers to make iPhones cheaper relative to Android than they would be at a free market price.

    And yes, I get annoyed when I cannot buy items at a free market price.

  493. > Because ultimately I am the one paying for it.

    So buy a phone off-contract, and stop complaining.

  494. @Cathy:
    “Who is subsidizing Android phones?”

    @Patrick:
    “But according to comments from Sprint’s CFO and others, apparently Sprint’s iPhone subsidy is $450, and Android phone subsidy is $250.”

    So unless Sprint charges more per month for iPhone users, which I doubt, Android users are subsidizing iPhone users, not the reverse.

  495. @Larry Yelnick, Nigel:

    > Doesn’t matter. Eric still wouldn’t count them against Android.

    Snark aside, what would be the distinguishing characteristic between a 3G iPod touch and an iPhone? I always thought an iPhone (well, at least the original one) was a 3G iPod Touch…

  496. Cathy, the consumer is there because that’s who the carriers are selling to. When your grocery sells (e.g.) milk at a lower markup than wine, it’s muddling things to say that wine is “subsidizing” milk. Retailers pay what they have to in order to stock what consumers want, and sell each item for as much as they can. Different items have different markups. This isn’t unfair to Android or you, it’s just the way the market works.

  497. @Larry Yelnick:

    Sprint’s CFO Joe Euteneuer said that iPhone subsidies are $200 more than other smartphones. Other sources say that the subsidy Sprint pays for iPhones is roughly $450. Perhaps the WSJ wasn’t paying attention — I have seen claims from Sprint that “The cost of adding an iPhone customer is about 40% higher than the cost for the average non-iPhone customer.” A lot more than the subsidy is counted in the acquisition cost.

    > So buy a phone off-contract, and stop complaining.

    But even off-contract, it’s hard to get a better monthly deal, unless you go second tier prepaid.

    @Cathy:

    So unless Sprint charges more per month for iPhone users, which I doubt, Android users are subsidizing iPhone users, not the reverse.

    Exactly. With, of course, the caveat that perhaps the average Android customer is a higher support burden, or causes more churn or whatever.

  498. @PapayaSF:

    Your analogy is somewhat flawed. If you want a grocery store analogy, it would be that some brand of wine is popular enough that grocery stores notice that it drives traffic. So all the grocery stores have to sell this brand of wine. Now the winery notices this and demands that the grocery stores sell the wine below cost (because, obviously, the store can make money in other ways from the customers coming in to buy the wine) or they won’t get any more deliveries. If all the grocery stores in an area succumb to this demand, it is unfair both to the consumers who prefer a different brand of wine, and to the other brands of wine.

    You would think all the grocery stores could simply stop doing this, but the winery cleverly signed up one store for an exclusive, and then renegotiated the deal, and then signed the others up. They are all in long-term contracts, and whenever one of the contracts expires for one store, the other stores are still under contract, so that one store can’t stop selling the wine because it can’t afford to see its business drop that much until the other stores can also get out from under contract.

    At least that’s the way it’s explained to me.

  499. OK, the analogy is incomplete because a phone comes with a contract and thus gets paid for over a long time, unlike wine. But Apple doesn’t force anyone to sell anything “below cost.” They are selling iPhones for what they can get, and carriers are pricing their products and services the way they think is best.

  500. > So unless Sprint charges more per month for iPhone users, which I doubt, Android users are subsidizing iPhone users, not the reverse.

    If true, it’s only because Sprint sees (a lot) more demand for iPhone than Android. (You walked straight into that one.)

  501. > Snark aside, what would be the distinguishing characteristic between a 3G iPod touch and an iPhone?

    Technically? The iPod Touch doesn’t have (A)GPS, though the 3G data iPads do, so perhaps that would go away. “Not much” (the ability to run cellular telephony) is really the answer.

    With a huge installed base of iCloud users, Apple *could* move to a 3G/LTE data-only architecture. Jobs was certainly interested in that eventuality, and the 700MHz Block C open access rules could make such a device possible without carrier (Verizon, mostly) approval.

  502. > They are selling iPhones for what they can get, and carriers are pricing their products and services the way they think is best.

    Actually, there probably are terms & conditions that restrict the price that carriers can charge for the iPhone.

  503. @Larry Yelnick:

    If true, it’s only because Sprint sees (a lot) more demand for iPhone than Android.

    Or, more likely, they segment the market into satisfied customers and “dissatisfied customers + non-customers.” They obviously don’t need to do much for the satisfied customers. The dissatisfied customers who just want a lower price will probably self-identify over the phone, and for all the others, well there’s always the iPhone.

    With a huge installed base of iCloud users, Apple *could* move to a 3G/LTE data-only architecture. Jobs was certainly interested in that eventuality, and the 700MHz Block C open access rules could make such a device possible without carrier (Verizon, mostly) approval.

    It’s an interesting question as to whether the carriers see it in their interest to insure that the average TCP connection has low enough latency for a satisfactory voice call. The answer to this question might change (positively or negatively) depending on circumstances. (Customer paying $2.00/MB? Booyah! Customer paying $0.04/MB? What’s a few hundreds of ms latency here and there?). One would expect that (eventually) many or most data-only voice calls would be satisfactory, but at some point it wouldn’t be too surprising to see some of the same sorts of shenanigans that various carriers pulled on VOIP when it first came out.

    The open access requirment’s allowance for the spectrum holder to engage in “reasonable spectrum management” is probably big enough to drive a whole fleet of trucks through.

  504. @PapayaSF:

    > But Apple doesn’t force anyone to sell anything “below cost.”

    How do you know this?

  505. @PapayaSF:

    And even if it is, in fact, literally true that Apple doesn’t force anybody to sell things below cost, what are the practical aspects of negotiating a deal where a company with 49 million postpaid subscribers has to buy 30.5 million iPhones over the course of 4 years? Do you really think they could sell all those at the $650/unit they have to give Apple?

  506. Ummm… common sense? How does a wholesaler force a retailer to lose money on their product? Are there proven examples of this? And no, the fact that the product is demand, and that there is price competition with other products and retailers, still doesn’t “force” anyone to sell at a loss.

  507. They obviously think that can sell them at a cost which, combined with the ongoing services they sell, will make them money.

  508. Sorry, got some bad data. Apparently, Sprint has 33 million postpaid subscribers…

    Man, I tell you what. If I were an Android vendor, I would steer way clear of sprint. They have no incentive whatsoever to push my product for 4 years.

  509. @PapayaSF:

    They obviously think that can sell them at a cost which, combined with the ongoing services they sell, will make them money.

    More likely they think they can lose less money selling them than not selling them. Mind you, they don’t believe this strongly enough to let the Apple expenses eat into the executive bonus plan…

  510. @PapayaSF:

    How does a wholesaler force a retailer to lose money on their product?
    Are there proven examples of this?

    There have been “tying” cases in the past where sellers forced buyers to take goods they didn’t want in order to get goods they did want or need, but I can’t think of anything exactly like the current Apple behavior.

    From an outside perspective with limited information, the Sprint deal seems to be one of a seller refusing to do business with a buyer unless the buyer buys a large minimum quantity.

    In most markets, this wouldn’t be a problem (or would be solved by a middleman). But somehow, even though the iPhone 4s is theoretically carrier agnostic, Apple and the carriers collude (Apple because it gives them this control; the carriers because, if they finance a handset, they want to continue to reap the benefit of it) to insure that iPhones are not fungible — in the US they are always locked to a carrier. This started out as a technical accident on AT&T, but Apple apparently realized that, while they wanted to be able to build a single device for cost reasons, it would be awesome to be able to continue to ship it as different segmented devices.

    So there can be no middleman. Apple apparently insures that the only way that a carrier can receive its product is to deal directly with it, on whatever unique terms it decides to demand from that carrier.

    Even that might not be a problem, except that the minimum quantities negotiated apparently put the deal on par with the old Microsoft MS-DOS deal that severely punished dealers for selling other operating systems. The Apple deal is cleverer by half — the carrier is free to sell other devices, but runs the risk of not meeting its minimum purchase commitment to Apple if it does so in too serious a fashion.

    DVD region coding is another example of sellers artificially segmenting the market for what some would consider nefarious purposes.

  511. We’re talking about phones. Tablets are a different market. iPhones are NOT sold unlocked.

    No, you were talking about Apple which makes both iPhones and iPads. You called out the company and Tim Cook the CEO as against pre-paid. The iPad shows that this is not true and that Apple has pushed the no-contract concept where allowed by carriers.

    The iPhone IS sold unlocked. You can buy an Apple iPhone 4S unlocked and contract free from Apple’s website in the US as well as some other countries. Apple itself has no problems with selling unlocked iPhones. You can’t in places where the carriers block Apple from doing so.

    The unlocked 16GB iPhone 4S is $649. You can even finance the purchase 6 months no interest through Apple.

  512. Sorry, I should have said “cross-subsidization” rather than simply “subsidization”. While I don’t like the fact that some carriers don’t want to drop your service fees after the phone is paid off, the real issue I’m trying to focus on here is the use of profits from Android subscribers to make iPhones cheaper relative to Android than they would be at a free market price.

    And yes, I get annoyed when I cannot buy items at a free market price.

    @cathy

    That’s hilarious. This IS the free market price. What some folks here are advocating is that the GOVERNMENT force carriers to both unlock and not subsidize phones. Evidently because it favors Apple or carriers or something.

    The “free market” is not free as in liberty or beer but free as in unconstrained by regulation. Corporations being what they are will be predatory.

    Unless you’re of the impression that someone is forcing carriers to subsidize the iPhone more than Android. iPhone subsidies are driven by consumer behavior and the huge value of the iPhone to the carriers. Not regulation or coercion.

  513. @patrick

    I read your post and I’m no longer sure what you are arguing. I was under the impression your assertion was that customers with iPhones would not make Sprint any profit until AFTER the end of the two year contract. Meaning an iPhone contract sold today wouldn’t make Sprint any money until months after Feb 29, 2014.

    Nothing in what Sprint wrote or you say in that comment has anything to do with this assertion. If that’s not what you meant then okay but it was unclear.

    If you ONLY mean that Sprint’s $15.5B might not pay out in the end that’s true.

    But ONLY if they don’t manage to sell all the iPhones they contracted for. Whether that iPhone is sold to an existing customer who is happy, an existing customer for might have jumped ship (reducing churn) or a new customer is immaterial. They will have had sold 30M of the most profitable post paid contracts in the US market over the next four year period via the iPhone.

    That’s insanely great given that they ONLY have 33 million post paid subs now. Not all on smart phones and which was undergoing huge churn. Now they’re probably stealing customers from T-Mobile that don’t want to be on AT&T or Verizon and stabilized their own losses.

    IF they manage to sell all 30M iPhones folks believe they have contracted for they’ll be doing great and made great money on the iPhone deal.

    IF they manage to only sell 15M iPhones they’re probably dead because they’ll have both a warehouse full of unsold iPhones AND have little post-paid customer base left.

    That’s the risk. Not that they lose money on every iPhone sold until long after the 2 year mark.

  514. @Nigel:

    I was under the impression your assertion was that customers with iPhones would not make Sprint any profit until AFTER the end of the two year contract. Meaning an iPhone contract sold today wouldn’t make Sprint any money until months after Feb 29, 2014.

    I believe that to be absolutely true for at least some of the contracts. As I have said, if Sprint sells an iPhone to an existing smartphone customer who wasn’t planning on leaving, they just left $450 on the table for zero return. I can’t believe that all their customers were unhappy and planning to leave — here’s a good chance that Sprint would have hit equilibrium without the iPhone, withing a few quarters of the introduction of the 4S.

    Whether that iPhone is sold to an existing customer who is happy, an existing customer for might have jumped ship (reducing churn) or a new customer is immaterial.

    I completely disagree. How does giving $450 to an existing happy customer for no incremental ARPU make sense (other than the simple fact that they have no way of distinguishing those)?

    If you ONLY mean that Sprint’s $15.5B might not pay out in the end that’s true.

    That was my original premise, then we got sidetracked breaking down how the failure occurs.

    IF they manage to only sell 15M iPhones they’re probably dead because they’ll have both a warehouse full of unsold iPhones AND have little post-paid customer base left.

    Sure, but that won’t happen. In for a penny, in for a pound. If the contract with Apple is airtight, they’ll shift those phones, even if they have to sell them for $50 rather than $200 and take a $10/month hit on ARPU. Mind you, that deal will probably be for new customers only unless they get really desperate.

    That’s the risk. Not that they lose money on every iPhone sold until long after the 2 year mark.

    Not sure that’s true. There’s a lot of ways things could go bad for them. Their 10K also discusses that they’ll probably be selling contracts with subsidized phones to people who are not terrific credit risks, because all the good customers are already taken.

  515. @Nigel:

    That’s hilarious. This IS the free market price. What some folks here are advocating is that the GOVERNMENT force carriers to both unlock and not subsidize phones. Evidently because it favors Apple or carriers or something.

    I, for one, don’t think the carriers need to be forced to not subsidize phones. I think that forcing them to unlock off-contract phones (now that most smartphones are capable of running on most carriers) would probably be a suitable remedy for past bad behavior.

    What behavior? Simply the lack of full disclosure (and I don’t mean in the fine print) that, when you “buy” an iPhone, even though it’s technically capable of running on more than one carrier, you absolutely will not be able to move it to a different carrier, even after your contract expires.

    I’ve posted a link to this article before, but note the “Last month I purchased an iPhone 4S from Best Buy Mobile for use on my Verizon account. I paid full retail so there would be no restrictions from Verizon with regard to unlocking the handset. ” and how that didn’t work out so well…

    http://www.forbes.com/sites/marcwebertobias/2011/12/22/how-u-s-carriers-fool-you-into-thinking-your-iphone-4s-is-unlocked/

  516. One more note on the unlocking. Page Plus, a Verizon MVNO, has apparently been coerced into not allowing iPhones to be activated. Verizon would hate for somebody to use an iPhone with cheap prepaid.

    Not only that, but Verizon apparently doesn’t allow its MVNOs to advertise that they are using the Verizon network.

    So if unsuspecting consumer bought an iPhone 4S because it should work on any network (even T-Mobile, albeit slowly) thinking they could use it on some BYOD MVNO, they’re in for a rude surprise.

    Caveat emptor and all that, but these are not technical limitations…

  517. @patrick What does any of that have to do with Apple being against pre-paid/no-contract/easier for the user?

    Fact: You can buy an iPhone 4S unlocked in the US.
    Fact: Apple pushed AT&T into a no-contract data plan for the iPad.
    Fact: Every example you presented was CARRIERS preventing the users from unlocking or using these devices on other networks.

    Cathy was wrong and the current state of affairs is largely a result of the free market decisions made by carriers to try to optimize their profits.

    I am not a free market solves all proponent. I’m for rational regulation of markets because the truly unfettered free market is rapacious. I just find it either hilarious or maddening that so-called free market proponents favor forcing companies to do stuff for the “consumer’s benefit” when it is THEIR consumer benefit in question. But when it’s MY benefit, oh no, the free market should decide.

    This is the same problem I have with pacifists.

  518. I believe that to be absolutely true for at least some of the contracts. As I have said, if Sprint sells an iPhone to an existing smartphone customer who wasn’t planning on leaving, they just left $450 on the table for zero return. I can’t believe that all their customers were unhappy and planning to leave — here’s a good chance that Sprint would have hit equilibrium without the iPhone, withing a few quarters of the introduction of the 4S.

    Leaving $450 on the table is not the same as not making any profit on the customer. It doesn’t matter if the customer is happy or planning on leaving. Either way the contract WITH subsidy is profitable at the end of 2 years.

    They did not believe that would have liked where the equilibrium would be without the iPhone.

    I have no idea why folks want to continue to ignore statements and evidence that the iPhone is a competitive requirement in the US market. Sprint sold 1.8M iphones in the 4th qtr and of these 40% were to NEW customers. 720,000 new contracts with the highest ARPU.

    “Wireless service revenues for the fourth quarter increased more than 7 percent year-over-year, driven by Sprint platform postpaid ARPU growth of $3.69 – the largest year-over-year increase on record across the U.S. wireless industry.

    The company reported total net subscriber additions of 1.6 million during the fourth quarter of 2011 – the best quarterly result in six years – bringing total ending subscribers to the highest level in the company’s history.

    “Our strong fourth quarter performance illustrates the power of matching iconic devices like the iPhone with our simple, unlimited plans and industry-leading customer experience,” said Dan Hesse, Sprint CEO. ”

    http://newsroom.sprint.com/article_display.cfm?article_id=2179

    How on earth you spin that negatively is beyond me. The iPhone deal is paid off for Sprint in Q4. EVEN on the 1.2M other iPhone sales. They just need to sell a lot more iPhones.

    But wait, it’s easier, somehow, to believe that Hesse is spinning about what made Q4 strong (having the iPhone) for Sprint and Humm is just covering his ass when he states what made Q4 suck (not having the iPhone) and OMG Sprint doesn’t make ANY money on iPhone contracts until year 3 and it’s all because of evil subsidies and if only it were a free market without Android users being forced to support iPhone prices it would all be different.

    :rolleyes:

  519. I am not a free market solves all proponent. I’m for rational regulation of markets because the truly unfettered free market is rapacious. I just find it either hilarious or maddening that so-called free market proponents favor forcing companies to do stuff for the “consumer’s benefit” when it is THEIR consumer benefit in question. But when it’s MY benefit, oh no, the free market should decide.

    Libertarians have a name for Gramscian false consciousness. They call it “libertarianism”.

  520. > I have no idea why folks want to continue to ignore statements and evidence that the iPhone is a competitive requirement in the US market.

    Nigel, please.

    The crowd here can’t stand the implications of iPhone being a competitive requirement in the US market. If that is true, wither Android?

    The smart kids have iPhones. Everyone else gets (sloppy) seconds.

  521. I am not a free market solves all proponent. I’m for rational regulation of markets because the truly unfettered free market is rapacious. I just find it either hilarious or maddening that so-called free market proponents favor forcing companies to do stuff for the “consumer’s benefit” when it is THEIR consumer benefit in question. But when it’s MY benefit, oh no, the free market should decide.

    Project much?

    Where exactly is anyone saying that the gov’t should force carriers to not subsidize the iPhone? This discussion is about how the carriers paying users extra to choose iPhone instead of Android affects Apple (almost unbelievable profits but unsustainable, beware the letdown) and especially the carriers themselves. We’ve observed how it hurts them, and posted links to analysts observing how it hurts them, as support. We’ve suggested that the carriers placing a higher value on iPhone customers over Android customers is wrong (as in mistaken). We haven’t said what the carriers are doing is immoral and should be make illegal, we’ve said what they’re doing is *stupid* (and in the case of Sprint, possible fatal).

    Some Android customers have expressed unhappiness that the prices they pay may be inflated, in part to help offset the downsides of the (we say “bad”) bet carriers have made on iPhone and iPhone carrying customers.

  522. Grr. Damn quoting. Let’s have another go.

    The crowd here can’t stand the implications of iPhone being a competitive requirement in the US market. If that is true, wither Android?

    That’s not quite it. I don’t think anyone has any problem with the concept of iPhone being a desirable product, we all know enough people who carry them. But it’s amusingly dissonant to argue that iPhone is in such demand, that carriers have to pay people to use them.

  523. @Nigel:

    Leaving $450 on the table is not the same as not making any profit on the customer. It doesn’t matter if the customer is happy or planning on leaving. Either way the contract WITH subsidy is profitable at the end of 2 years.

    Assuming this is true, it says that the price of wireless data is way too high. So how long will it stay true?

    But wait, it’s easier, somehow, to believe that Hesse is spinning about what made Q4 strong (having the iPhone) for Sprint and Humm is just covering his ass when he states what made Q4…

    Parser fail. Not sure exactly what you’re saying. BTW, in Q3, SPrint’s postpaids on the Sprint platform (excluding the dying Nextel platform) was positive. Before the iPhone. Also, Q4 was really strong in expenses. Will be interesting to see if this investment really pans out.

    Cathy was wrong and the current state of affairs is largely a result of the free market decisions made by carriers to try to optimize their profits.

    You can’t completely exonerate Apple. They are happy to sell the unlocked handset, but they don’t sell that many of those, so they focus on the carrier. They are quite happy to lock the handset for the carrier, but apparently extract their pound of flesh in the volume agreements. Apparently, Sprint doesn’t get any sort of real discount on the phones. They just have to buy lots of them to get them at all.

    @Larry Yelnick:

    The crowd here can’t stand the implications of iPhone being a competitive requirement in the US market. If that is true, wither Android?

    If the iPhone really is a requirement, then economic theory as practiced in this country suggests that Apple will need to be heavily regulated. I don’t think that’s the right answer, but I do believe that full disclosure on the fact that iPhone 4S handsets aren’t fungible from carrier to carrier is necessary, and that forcing carriers to unlock phones after the contract period is over might be desirable, and that forcing carriers such as Verizon to not discriminate against iPhones on their MVNO might be supportable, as might forcing carriers such as AT&T to not require a data plan when a customer attaches a smartphone to the network.

    Especially when the smartphone becomes the computer, the user needs to have his choice of smartphone. When he has an AT&T smartphone and then moves to an area that AT&T doesn’t service, he should be able to use his device on another network after paying the AT&T contract termination fee. Somehow, we need to disentangle devices and carriers, and I think a few simple rules about how when a customer buys a device, it’s his, and he ought to be able to use it as he sees fit, could go a long way to removing market distortions.

  524. Hmm, but if you run the math on their May survey, it looks like 23.7% of smartphones were iPhones back then.

    That was lower than comscore then, and the current number is likely to be higher than comscore (though that remains to be seen, that’s the trend). Of course, comscore measures consumers 13 and older, and I think Pew is measuring 18+, but still the data points might be in conflict.

  525. Actually, the numbers might mesh with comscore fairly well, considering the large number of iPhones sold last quarter. If Pew is right, one out of every 6 American adults (39 million people) carry iPhones. That doesn’t quite match my observed reality, but then I live a sheltered life.

  526. @nigel:
    “Fact: You can buy an iPhone 4S unlocked in the US.
    Fact: Apple pushed AT&T into a no-contract data plan for the iPad.
    Fact: Every example you presented was CARRIERS preventing the users from unlocking or using these devices on other networks.

    Cathy was wrong and the current state of affairs is largely a result of the free market decisions made by carriers to try to optimize their profits.”

    I acknowledge that I was wrong about the ability to purchase an unlocked iPhone in the U.S. It does appear that this is possible at Apple stores, and that the device can then be used on various networks.

    However, nowhere in my laments did I ask for regulation. Reread them and you’ll see this is true. I am frustrated at what looks to me like a market failure (demand for device-agnostic prepaid service), but I expect it to fix itself over time as consumers sort out what they really want and what they will (and won’t) pay for it.

  527. So, yeah, as Greg pointed out yesterday, iPhone may be hurting Sprint really badly — keeping it from necessary spectrum growth. So why would they do it? Well, apparently it makes their bonuses better…

    I’ve been thinking about this, and I’m not sure that’s it. Changing the rule so that initial iPhone expenses don’t affect bonus calculations is a little thing, a small easy win for management, not the base of a strategy. :) (HOW do you meet 106% of your bonus targets in a year you lost a fuckton of money, including $1.3billion in one quarter alone? *shakes head*)

    In the short term, Sprint is in crisis. So why would they make a major financial commitment to Apple that kills their short-term maneuvering options, and prevents them from effectively dealing with the immediate crises? (The MetroPCS deal falling through is a symptom.)

    Hesse may have given up, and decided to roll the dice. I read somewhere that Sprint is already $20billion in debt. At least part of the $2billion they raised recently is for debt service. And there’s that loss of $1.3billion dollars last quarter. They’ve been losing money every year for a while now, and look like they’re going to keep losing money every year for the forseeable future. Sprint is already a dead man walking. The iPhone deal is essentially a double-or-nothing bet. If it pays off, Sprint gets to live. If it doesn’t, well no surprises there.

  528. @nigel etc.
    I am not sure whether I understand you correctly. It is claimed that iPhone users are/will be profitable for the networks. Moreover, they are/will be generating more profit than Android users. The networks subsidize iPhones (much) more than Android phones. Therefore, the upfront costs to the networks of an iPhone user are (much?) bigger than the costs of an Android user.

    I conclude from this that at the end of a contract, the average iPhone user has paid (much) more to his network than the average Android user: First to pay for the extra “costs” of the iPhone to the network and second to generate the extra network profit iPhone users are supposed to generate over Android users.

    The networks are very good at hiding this supposed extra revenue they extract from iPhone users. None of the knowledgable people here was able to dissect where this extra revenue was hiding. So we can be very confident that the majority of the iPhone users too is unable to see how much they are going to pay for their “free” iPhone contract. Especially, these customers will be unable to decide how much choosing that iPhone is costing them extra than choosing an alternative.

    My conclusion of all this is that customers chose for an iPhone mostly because they are unable to see what the real costs of an iPhone to them is. My corollary is that the iPhone’s market share would get a serious hit if customers find out what the cost is. Unless, of course, the networks make a heavy loss on iPhone users. But that would confirm our host’s analysis.

    Both cases predict dire times for the iPhone. And all of that is according to your own analysis.

    A few extra headlines that say it all:
    (not all real news, but just for the record)

    Android is most-used smartphone OS in UK
    http://hi-android.info/2012/02/24/android-is-most-used-smartphone-os-in-uk/

    Android sales form half the global smartphone market
    http://www.knowyourmobile.com/blog/1246285/android_sales_form_half_the_global_smartphone_market.html

  529. The only way the iPhone subsidy situation makes any sense is that the carriers see there being a relatively limited pool of highly desirable iPhone-demanding customers. (This seems to have some basis in reality, there is a limited population of non-price sensitive buyers and those people seem to prefer Apple.)

    Where everything goes off the rails, is that the carriers have engaged in a price war with each other to attract those customers. Yes, they’re giving away the family china in a desperate price war to attract the non-price-sensitive customers. Well done! And what they’re actually going to GET, is opportunists sniffing a bargain.

    Simply handing a customer, any customer, an iPhone is not going to magically turn him into a wealthy, big spending, non-price sensitive hipster trustafarian. Just like buying underachieving neer-do-wells a house and a car isn’t going to turn them into middle class producers, but I digress.

  530. @patrick “It’s an interesting question as to whether the carriers see it in their interest to insure that the average TCP connection has low enough latency for a satisfactory voice call.”

    Normally VoIP is implemented over UDP.

  531. @Larry Yelnick:

    > Normally VoIP is implemented over UDP.

    Yeah, slip of the keyboard — obviously care about latency much more than guaranteed delivery.

    My point was that the carriers could fuck with the latency. Whether they feel it in their interest to do that or not, or whether others could pressure them to stop doing it, is, in my mind, an open question.

    Certainly, at the end of the day, you would expect things to be fine. It’s just the unsettling period before we get there. And even at the end of the day, you might expect (and might even wish for) network traffic management that (e.g. during a disaster or huge sporting event) prioritizes ‘n’ good voice calls over ‘n+1′ terrible calls.

  532. @Winter:

    > The networks are very good at hiding this supposed extra revenue they extract from iPhone users.

    I have seen it asserted here and in other places that the carriers give additional monthly revenue to Apple. But that’s obviously not the whole story. If the user pays $200 and the carrier pays $450 for a phone that retails at $649, obviously the incentives go the other way (except, of course, for phones bought and activated in an Apple store. Just like any other third party business that activates phones for the carriers, they will certainly get a cut of that contract.)

    We know google shares revenue with the carriers, and we know that Apple collects more revenue per user in its app store than google does, so Occam’s Razor says that the carriers are getting a cut whenever a user buys something from Apple.

    This is the most plausible explanation as to why iPhone customers are perceived to be so valuable to the carriers. It remains to be seen whether the newer iPhone customers (especially the ones who buy the older one for $0) are as profitable as the early adopters. This is quite possibly a major component of the risk alluded to in the Sprint 10-K.

  533. This is the most plausible explanation as to why iPhone customers are perceived to be so valuable to the carriers. It remains to be seen whether the newer iPhone customers (especially the ones who buy the older one for $0) are as profitable as the early adopters. This is quite possibly a major component of the risk alluded to in the Sprint 10-K.

    Got a post stuck in moderation that talks about this. IMO it looks like a cargo cult. Every carrier wants more of those profitable early iPhone adopters, and is now engaged in handing out iPhones in an effort to create them.

  534. @winter the additional cost to the customer for an iphone vs android is the difference in out of pocket costs for “buying” the handset. For the 3GS that out of pocket cost is $0.

    Voice and data plan costs are identical for all smartphones, android, ios, whatever.

    Top end Android handsets are probably more profitable than iPhones but both top end Android and iPhones both live in the desired high ARPU post-paid market (low end Androids in the pre-paid market not as much).

    Twisting the analysis to end up with “dire times” predictions for the iPhone is silly and dishonest. The iPhone, like the iPod, will get cannibalized eventually. Perhaps even soon. But Apple prefers to do the cannibalization and has a good track record of managing to do so.

    A iPod Touch with a no-contract 3G/4G data plan could be such an object IF the carriers allow it.

  535. @Greg:

    Every carrier wants more of those profitable early iPhone adopters, and is now engaged in handing out iPhones in an effort to create them.

    Well put. Obviously this is excellent for Apple, both from a direct revenue standpoint and a network effects standpoint. How well the carriers do depends on how compelling their new less-well-heeled customers find the apps. Here’s an interesting study:

    http://www.travel-impact-newswire.com/2012/02/the-rise-of-the-planet-of-the-apps-7-key-trends/

    “In a global comparison of average revenue per app downloaded, Norway is #1 at $0.37 per download, nine times more than China.* App Annie’s research finds there is only a loose correlation between GDP per capita and $ per download by country: factors such as smartphone penetration and local currency billing are more important.”

    So, just to pull some numbers out of my ass: if Sprint gives a $450 phone subsidy to a preexisting happy customer, and said customer buys 20 apps a month at double the Norway (highest in the world) average, and Apple shares half its 30% cut with the carrier, then over the life of the contract, that phone makes Sprint 24 months * 20 apps/month * 0.75$/app * 15% = $54.

    So they only lost $396, not $450. But it’s OK, because when that customer ponies up $200 for his next iPhone and gives the old one to someone else, that other person will spend an extra $20/month to go on contract and get data, so Sprint can make back their money in under 4 years. Yay, Sprint.

    Of course, if the initial customer follows the poorer Chinese model and downloads fewer apps, Sprint stands to make less than $2 on him through app purchases. But that’s not really any big deal — Sprint’s “investment” in the phone will now pay for itself in 47 months instead of 44 months.

    Unless, of course, the phone gets broken in the meantime. Which segues into the other way carriers get extra money from the iPhone: “Nice phone you just purchased there — it’d be a shame if anything happened to it, wouldn’t it? You’d be out another $650, and you can’t afford that!” So that’s another $10/month of almost pure profit for the carrier, which means that the carrier’s phone investment really does pay for itself slightly within 2 years. Now they just have to hope that the phone stays in use after that.

  536. Oops, bad last example — would be 3 years instead of 2 years.

    Of course, that’s for a phone that went to a preexisting happy customer. As Nigel points out, if Sprint can actually upgrade a customer who didn’t have data, or can pull a customer from a different carrier, the investment pays back more quickly. But as Greg points out, once the carriers spend all their time and money poaching each others’ iPhone customers, it becomes a zero-sum game. And as the Sprint 10K points out, an attempt to escape this zero-sum game by extending credit to less credit-worthy customers comes with its own set of risks.

  537. “Voice and data plan costs are identical for all smartphones, android, ios, whatever.”

    I think you meant, “Voice and data plan costs are identical for all postpaid smartphones, android, ios, whatever.” This is not a small difference.

    “Top end Android handsets are probably more profitable than iPhones but both top end Android and iPhones both live in the desired high ARPU post-paid market (low end Androids in the pre-paid market not as much).”

    But it’s hard to know what the demand would be for top-end Android or iPhones would be in the pre-paid market, since as far as I am aware they aren’t offered. I would certainly consider a high-end pre-paid no-contract Android.

  538. @greg every post-paid smartphone user is in the higher ARPU bracket. Every iPhone is in this bracket. Subsidies on the iPhone are higher than for high end android but there are advantages to the carrier to pay more to have the iPhone.

    With the iPhone on AT&T only you saw very little churn among iPhone users because it was worth putting up with even AT&T to have the iPhone vs using a high end Android phone on Verizon.

    Here’s some supporting data. The iPhone apparently costs carriers less for tech support.

    “Using the iPhone as a baseline, ClickFox found BlackBerry users required additional support (e.g., call transfers, followups) 37 percent of the time. Android users required more help a staggering 77 percent of the time.

    ClickFox estimates the total cost of additional tech support is US$46 million higher for BlackBerry users and $97 million higher for Android users — per year.”

    http://www.tuaw.com/2011/08/11/tech-support-for-iphone-is-cheaper-than-blackberry-android/

    The $97M is probably in the noise level…however the fewer tech support calls supports surveys that show a higher user satisfaction for the iPhone (JD Power, Change Wave, etc) which I believe leads to a halo effect for the carrier. Churn with the iPhone is probably going to be much lower than with Android users although not to the degree seen when it was AT&T exclusive. We’ll see in the next couple years how this plays out but if your phone is making you very happy overall the impulse to switch carriers is much lower unless you live in an area of bad coverage or the carrier dicked you over somehow.

    This customer satisfaction with the iPhone creates the most halo effect for Apple itself as can be seen by the massive increase in Mac sales in comparison to the rest of the PC market. But given how crappy AT&T can be some of that halo clearly transferred to the carrier as well.

    You guys probably don’t want to believe this but Sprint is likely to see much smaller churn among their iPhone customers and that 75% overall “very satisfied” user response with the iPhone is a significant contributing factor. The iPhone 4S ranks even higher at 77% “very satisfied” with 19% “somewhat satisfied”.

    Android, at 47%, is not as compelling in generating good vibes for the carrier.

    http://www.changewaveresearch.com/articles/2012/smart_phones_20120109.html

    http://www.changewaveresearch.com/articles/2011/iphone4s_20111201.html

    Opportunists seeking bargains already buy android. Paying $99 for an iPhone 4 on Verizon vs a Droid 3 for $9.99 doesn’t strike me as something someone very price sensitive will do. Or a 4S for $199 vs $99 for a Droid Bionic.

  539. But it’s hard to know what the demand would be for top-end Android or iPhones would be in the pre-paid market, since as far as I am aware they aren’t offered. I would certainly consider a high-end pre-paid no-contract Android.

    This is an exploitable market niche for T-Mobile. They’d have to accept being a “value” carrier, closer to a dumb pipe than the big 2 (and the #3 wannabe) are willing to accept but they’ve already made some efforts in that direction. They have an excellent prepaid offering we’ve already talked about, the $30 Monthly 4G plan. You can BYOD, they’ll activate any clean IMEI. The main problem is finding compatible devices (T-Mobile branded devices on the used market are the easiest to come by.) Their spectrum refarming should help with that.

  540. Opportunists seeking bargains already buy android. Paying $99 for an iPhone 4 on Verizon vs a Droid 3 for $9.99 doesn’t strike me as something someone very price sensitive will do. Or a 4S for $199 vs $99 for a Droid Bionic.

    Or $0 for a 3GS. We’ll have to see.

  541. @patrick

    Why guess? Sprint has stated that thus far 40% of their iPhone customers are new to Sprint.

    You’re presuming worse case and then making the seem like Sprint is losing money on an iPhone sold to a satisfied customer. They aren’t. They are making less money on them over the two year contract by providing a $450 subsidy over a $250 subsidy for another smartphone. Sprint aways makes back their money during the 2 year contract. They never have to hope for anything after the 2 year mark. Their iPhone investment pays off in 24 months. Not 44 months or 47 months or whatever other bullshit number you want to postulate.

    EVEN assuming your worst case scenario of “leaving money on the table” and ignoring the fact that iPhone user satisfaction is 77% vs 45% for Android and support costs are lower, we’re talking $200 not $450 so your math is still wrong regardless.

  542. @nigel
    Whatever you say, iPhone users have to pay more for the networks to earn more. The question is, are they aware they are going to pay more when they make the choice? If not, how will they react when they find out?

  543. @cathy

    I think you meant, “Voice and data plan costs are identical for all postpaid smartphones, android, ios, whatever.” This is not a small difference.

    Of course I meant that. It should be obvious from the next sentence I wrote AND that you quoted where I clearly differentiate between the high ARPU post paid Android devices and lower end Android devices sold in the pre-paid market.

    Was that supposed to be a “Gotcha!”?

  544. @winter iphone users don’t pay more than non-iphone users. That’s the point. Cathy doesn’t like that because she feels that Android users are cross-subsidizing iPhone users.

    It’s not the iPhone users that would get pissed about that so your whole contention falls flat on it’s face. No “dire outcome” sorry.

    If you expect there to be some Android user uprising then maybe US carriers will switch over to the non-subsidized model.

    Don’t hold your breath.

  545. @Nigel:

    Why guess? Sprint has stated that thus far 40% of their iPhone customers are new to Sprint.

    What did I say that said I was guessing about this? But in any case, how do you know how many customers they would have gained (or not lost) if they had simply dropped their rates? Or how many of the 60% (over 1 million) customers who were already on their highest plan? They don’t discuss that.

    You’re presuming worse case and then making the seem like Sprint is losing money on an iPhone sold to a satisfied customer. They aren’t. They are making less money on them over the two year contract by providing a $450 subsidy over a $250 subsidy for another smartphone.

    Sprint’s not making money. They’re losing money. You guys keep talking about how awesome the iPhone is. To the extent that’s true, it’s partly an attractive nuisance. If an existing Sprint customer buys an iPhone from Sprint instead of not buying a new phone, yes Sprint gets the benefit of locking them in for another two years, but that benefit is worthless if they weren’t going anywhere anyway. So basically, for that customer (who I agree is not the entire customer base, but still has to be averaged in) Sprint loses an extra $450 as soon as they hand him the iPhone. As discussed, maybe they get $50 back through app purchases, and $240 over the life of the contract for device insurance. And maybe they save $10 in support costs. But if Apple manages to make a shiny new device that attracts that customer in 2 years, he’ll want it, and Sprint can drop another $450 on him. If Sprint is lucky, he’ll give his old one to someone who will actually bump up a couple of plan grades to be able to use it, and they’ll get $20/month for awhile.

  546. EVEN assuming your worst case scenario of “leaving money on the table” and ignoring the fact that iPhone user satisfaction is 77% vs 45% for Android and support costs are lower, we’re talking $200 not $450 so your math is still wrong regardless.

    No. I’m talking about a perfectly satisfied customer. You’re one of the ones who’s talking up Apple’s awesomeness and helps convince him he’s missing out and should switch. The choice for this customer isn’t between a new Android (which still might cost him $200 out of pocket if it’s a primo one) or an Apple — it’s whatever he already has in his pocket or Apple. Still $450 to Sprint.

    (And BTW, I have mentioned the possibility of lower support costs as a reason that Apple is attractive to the carrier in the past, so I am certainly not ignoring that possibility. But in the case of my hypothetical perfectly satisfied customer, those would have been minimal or nonexistent anyway.)

    We don’t have enough visibility to know how many of Sprint’s 1 million preexisting customers who switched to iPhone match this hypothetical profile. We don’t have enough visibility to know how many of Sprint’s new customers would have switched if Sprint had a lower priced data offering rather than the iPhone, or how many of their exiting churned customers would have stayed around. And, of course, we don’t know details on support costs, app store kickbacks from Apple, or anything like that. So it’s all speculation, other than that we know that Sprint has a lot of expense commitments related to the iPhone, and they fervently hope and believe that the extra income will outweigh the expenses three or 4 years from now.

    Oh, and we also know that, if you’re right that the iPhone pays for itself within 2 years, the extra revenue (over having no new phone, and ignoring reduced support costs) must be at least $18.75, and extra revenue over an Android phone for the same customer must be at least $8.33. The latter number comes close to being believable due to the extra insurance and minor app store kickbacks, but it still seems iffy.

  547. @nigel:

    @winter iphone users don’t pay more than non-iphone users. That’s the point.

    Agreed. If the wheels ever fall off the bus, this isn’t how it will happen. It won’t be the iPhone customers revolting, and it won’t even be (directly) the Android customers revolting. If an Android customer demands a lower price or he’ll walk (especially after expiration of a contract) he’ll probably get it. That’s what the “retention” departments are all about.

    It has to be carrier-driven, and it will probably be subtle when it comes, so as to avoid the wrath of Apple. Unless, of course, the government orders the carriers to unlock off-contract phones. That would probably cause a huge realignment.

  548. If an existing Sprint customer buys an iPhone from Sprint instead of not buying a new phone, yes Sprint gets the benefit of locking them in for another two years, but that benefit is worthless if they weren’t going anywhere anyway.

    How many smartphone owners don’t get a new phone every 2 years? Show me that’s the norm and your hypothetical has more meaning. Otherwise that “phone in his pocket” is more likely a dumb phone without a data contract.

    It appears that the US smartphone replacement interval is 19 months. 16 months for Samsung. 22 months for Apple and RIM.

    https://www.npd.com/lps/SmartphoneTrackPR/

    So your scenario is an edge case.

    EITHER way, Sprint WILL make money on that customer. It’s built into the cost of the contract. If the customer doesn’t take advantage of a new phone every 2 years then of course Sprint makes more money. But evidently iPhones have a longer replacement cycle than the industry average and particularly in comparison to Motorola and Samsung. Meaning there are probably more 3+ year old 3GS users out there than 3+ year old Android phone users out there.

    16 month average replacement cycle. Well that’s one contributor to Android activations.

  549. @ Nigel:

    In terms of carrier profits, “average” smartphone lifespan is, by itself, a meaningless statistic. Was it replaced under warranty? Did the user upgrade? Was the contract extended? What’s the standard deviation?

    > 16 month average replacement cycle. Well that’s one contributor to Android activations.

    And you act like you know all about this stuff.

  550. @Cathy:
    “I think you meant, “Voice and data plan costs are identical for all postpaid smartphones, android, ios, whatever.” This is not a small difference.”

    @nigel:
    “Of course I meant that. It should be obvious from the next sentence I wrote AND that you quoted where I clearly differentiate between the high ARPU post paid Android devices and lower end Android devices sold in the pre-paid market. Was that supposed to be a ‘Gotcha!’?”

    I’m not into “gotcha”, and I knew you got this point from your context. You just seemed determined to ignore it.

    I was trying to make the point that the prepaid market really matters, and that attempts to hand-wave it away distort the discussion. There is no fundamental reason why there couldn’t be prepaid iPhones or high-end Android phones; that issue is political (in the sense of corporate politics and goals, not national politics), not technical.

  551. @Patrick:
    “Several prepaid vendors now offer BYOD. For example:
    [links deleted]”

    Thanks Patrick. These are very interesting options. Separating device sales from sales of service seems like a logical next step for the industry.

  552. In terms of carrier profits, “average” smartphone lifespan is, by itself, a meaningless statistic. Was it replaced under warranty? Did the user upgrade? Was the contract extended? What’s the standard deviation?

    That’s a whole lot of “I ain’t got no supporting data so I’m gonna ask a bunch of unimportant questions that can’t be answered without buying the $5000 report”.

    Seriously, do you believe most smartphone users DON’T get a new phone every 2 years? It doesn’t matter what the reason was for Samsung phones to last only 16 months before replacement. That data shows that the majority of smartphones have a less than 2 year replacement cycle, even for your “perfectly happy customer”.

    Given how much Samsung dominates the Android market that 16 month replacement cycle is the norm for Android. Even perfectly happy iPhone customers appear to upgrade the second the carrier allows them to in order to have a 22 month average.

    So the average lifetime data supports the contention that those 1M Sprint iPhone owners are largely EITHER smartphone owners that would have upgraded anyway OR dumb phone owners that upgraded to a smartphone.

    In terms of carrier profits, your “perfectly happy customer that wouldn’t have upgraded his phone” scenario is a meaningless edge case.

  553. That’s a whole lot of “I ain’t got no supporting data so I’m gonna ask a bunch of unimportant questions that can’t be answered without buying the $5000 report”.

    More like “I’m not going to let Nigel blow smoke up my dress with one data point taken out of context.”

    Seriously, do you believe most smartphone users DON’T get a new phone every 2 years?

    I’m sure a lot of them don’t, and the percentage probably increases as the years go on and the phones get more capable. (The number you reported was about people replacing their phones — what was their last phone? So the number’s always going to be out of date (because it doesn’t include people who aren’t replacing their phones), plus I didn’t see anything about the sample period…)

    Given how much Samsung dominates the Android market that 16 month replacement cycle is the norm for Android.

    Umm, no. At least not in the US. At least not for phones that are over several months old. Pay attention, man! Samsung’s only now starting to get serious smartphone traction here.

    http://blog.nielsen.com/nielsenwire/online_mobile/in-u-s-smartphone-market-android-is-top-operating-system-apple-is-top-manufacturer/

    That data shows that the majority of smartphones have a less than 2 year replacement cycle, even for your “perfectly happy customer”.

    That data is for people getting a new phone. So it doesn’t include new smartphone customers, for one thing. It will have a lot of early adopters (e.g. rich people), and won’t have people that got their smartphone 3 years ago and are quite happy. Without more detailed information that’s worse than useless, because it’s designed to make people like you think it’s meaningful, and it’s obviously not.

    So the average lifetime data supports the contention that those 1M Sprint iPhone owners are largely EITHER smartphone owners that would have upgraded anyway OR dumb phone owners that upgraded to a smartphone.

    No, it doesn’t.

    In terms of carrier profits, your “perfectly happy customer that wouldn’t have upgraded his phone” scenario is a meaningless edge case.

    You’re welcome to believe that but you certainly haven’t proved it.

  554. @nigel
    You are saying that networks earn more from iPhone users and pay more to Apple. Still, the iPhone users do not pay more than Android users.

    Why do I not understand these simple economics?

  555. @Winter:

    He seems to think that the carriers give out free Android phones once every 16 months or something.

  556. @winter because I didn’t say that.

    I said that iPhone AND post-paid high-end Android users are in the top ARPU tier. These make more money for carriers than pre-paid Android smartphones and dumb phones.

    I said that high-end Android users are more profitable than iPhone users because the subsidy is around $200 lower.

    I said that the iPhone is still a good value proposition for carriers over high end Android phones because user satisfaction is much higher, support costs lower and most likely a far lower churn rate despite the $200 subsidy difference. Circumstantial evidence is Sprint betting the company on the iPhone. The iPhone may not SAVE Sprint but evidently they felt it would doom them to not have the iPhone.

  557. More like “I’m not going to let Nigel blow smoke up my dress with one data point taken out of context.”

    One data point that you don’t like because it completely destroys your argument. For which you have thus far provided ZERO data points.

    Oh wait, perhaps you googled for data to disprove mine (something I didn’t do). Let’s quickly try that now.

    Gee, this indicates that the worldwide replacement cycle is 11.5 months:

    “And then we can calculate the average replacement cycle.. The full smartphone population is less than 23 months old (complete replacement of total smartphone installed base). Thus the average replacement cycle of smartphones today is.. 11.5 months!”

    http://communities-dominate.blogs.com/brands/2011/12/smartphone-penetration-rates-by-country-we-have-good-data-finally.html

    That seems a bit short and dubious. Lets see what else we might find. Next relevant link.

    “The data set in Table 1 shows that the United States has consistently had the shortest handset replacement cycle, while India and Brazil have the longest. In 2010, Americans replaced their mobile device after one year and nine months,

    This proves conclusively that Americans use the newest handsets in the world. As we all know, new affordable handsets are a key decision factor for consumer when chosing their mobile carrier. Hence the level of handset subsidization is an indicator of how competitive a market is. In no other country are consumers upgrading faster than in the U.S., so no one else has comparable access to the latest handsets, technology and services than the American consumer. The rapid handset replacement cycle has put new smartphones in more people’s hands faster than anywhere else in the world.

    Addendum

    Handset replacement cycle: The handset replacement cycle describes the length of time in months that a device owner keeps his handset before purchasing a new one. It is calculated by first subtracting the number of new subscribers from the subscribers at the end of a year to get the number of long-term device owners. Then we subtract from the total number of device sales the number of net subscriber additions to determine the replacement device sales. Then the number of long-term device owners is divided by the number of replacement device sales and multiplied by 12. This results in the handset replacement cycle in months.

    ((subscribers – net subscriber additions) / (device sales – net subscriber additions)) * 12 = handset replacement cycle in months”

    This evidently this is the replacement cycle for ALL handsets. Not just smartphones.

    http://mobilefuture.org/page/handset-replacement-cycle.pdf

    I’m sure a lot of them don’t,

    How are you sure? Gee, I guess it’s not just “one” data point anymore. Perhaps repeated assertion IS better than data. Who knew? (other than Fox News)

    Yes, that’s snark. I think in this case it’s deserved. You didn’t bother to google to see if you could still baldly assert that the replacement cycle is longer than 24 months? Instead you simply decided that going on the offense was better than 5 minute research? Really?

  558. @nigel
    Ok. That are quite a lot of ifs and maybes.

    Not a very convincing refutation of the original post. But without evidence saying otherwise it could be true. In the USA.

  559. @nigel
    These replacement times in a growing market are complete nonsense. You cannot use averages in such skewed distributions.

    The relevant measure is the half time, the time after which half the owners have replaced their handset.

  560. @Nigel:

    That’s much better data. However, I like how you throw out the data you yourself don’t like, and then get on my case for discounting unsupported data. Seriously, that’s the pot calling the highly polished flatware black, especially when it’s obvious from the chart in that PDF that people, real people, around the world keep their smartphones for many years. Finland is not a poor country (and it’s the home of Nokia!), but an average 6 year replacement cycle is huge. Don’t for a minute imagine that no people in the US keep their handsets that long.

    It’s a better big picture, but what I still don’t see in that data is information about things like lost/stolen/broken phones. Stuff I have seen earlier seemed to indicate that even iPhones died or were damaged at a rate approaching 30% within 2 years. Cheaper Android models, maybe more. Carriers don’t usually pay for the new phone (except for insurance), so that might skew the data a bit more. And finally, all phones (smartphones, dumbphones, well-designed phones, crapphones) are all aggregated, so the conclusions we can draw are a bit limited.

    And Winter’s right. The data is still incomplete, and somewhat meaningless when people are rapidly transitioning from dumbphones to smartphones. This is why I mentioned that the average, itself, is somewhat meaningless. Let me give you a more concrete example.

    Let’s ignore dumbphone conversions for now, but let’s say that 10% of all consumers find their smartphone unbearable and get a new one within the first month (I have seen this multiple times, so believe the number might actually be low). Let’s further say that 30% of all smartphones die within the first year, so we’ll give those a lifespan of 1 year. To simplify, we’ll just sum those numbers and not try to worry about replacement phones dying.

    So 10% * 1 month + 30% * 12 months + 60% * x = 21.7 months.

    x = 30 month average for phones that didn’t die/aren’t stolen/weren’t replaced immediately.

    Now, let’s further say that my hypothetical completely happy customer makes up 30% of T-Mobile’s subscriber base, and 10% of the customers are going to buy a new phone every year no matter what, and the other 60% are going to buy one after two years when they are eligible (which doesn’t happen on Verizon anymore btw unless you pay extra).

    .30 * x + .10 * 12 + .60 * 24 = 30

    x = 48, or in this scenario, my (non-existent according to you) happy customer would hold onto his phone for 4 years on average.

    Now, do I know that these numbers are right? No, but the point is: you haven’t showed that they are wrong. I can come up with a lot of plausible scenarios that have both a 22 month average replacement rate and a hell of a lot of people that hold onto their handsets for over 24 months. In fact, it’s really hard to come up with a plausible scenario where the average is 22 months that doesn’t have a lot of people holding on to their handsets longer than 24 months.

    But the real question from the carrier’s perspective is not whether the original owner replaces the handset within 2 years. It’s whether the handset itself is used for more than 2 years or not. Interestingly, the methodology given in that PDF, based on handset sales, really addresses this second question, anyway.

    This is why it is so critical for AT&T and Verizon (and, I’m sure, now Sprint) to refuse to unlock iPhones. They really count on the money after the contract period is up. There was even a lawsuit, with the outcome that AT&T will unlock every phone except an iPhone.

    Now, the next question is: is the iPhone such an attractive nuisance that an otherwise happy, non-handset-changing customer would spring $200 for one when it’s available? Hey, you’re the one arguing it’s a must-have, and I’ll state up front that even I, if I were happy with my wireless service and wasn’t planning on going anywhere for 2 years, might consider an iPhone, depending on what else was offered and what the subsidy was. It has a lot going for it, but it has a lot more going for it if the carrier is going to give me $450 to buy it.

    And the final question is: how many replacement phones are purchased outright by customers? If 25-30% of handsets die within 2 years, and the customer (or insurance) has to buy a replacement, that doesn’t eat into the carrier’s margins like a customer demanding a new subsidized phone direct from the carrier every two years (but it does figure into those averages).

    Believe it or not, a lot of people aren’t true gadget freaks. They learn how to use something, fall in love with it, and absolutely hate when the next one doesn’t work as well. No doubt the carriers count on this and have the historical statistics to back it up.

    What they don’t have (Sprint, anyway) is statistics for devices from a company like Apple that makes things that get better without a learning curve. If a customer knows that his new iPhone will work just like his old one, except that it does ‘x’ better, he’s more likely to buy a new one. And Apple’s really, really good about making sure that his experience doesn’t change, or if it does change, they train him by giving him the new software on the old phone.

    This is to Apple’s credit, but it’s pretty bloody obvious from the numbers that it drives upgrades, which makes it a double-edged sword for the carriers. And when all of them have the same sword (except poor T-Mobile) then it’s just the arms merchant who profits.

    So my thesis remains: if carriers were forced to unlock phones that aren’t under contract, the landscape would change considerably.