One of my regular commenters points out an interesting article in the Wall Street Journal,
Sprint Could Yet Strike Out With iPhone. “SPQR” interprets it as follows:
The article states that Sprint’s cost to run the iPhone is $15.5 billion over four years. Unclear to me what that “cost” means from the article. The iPhone won’t have a positive impact on Sprint operating income before depreciation and amortization until 2015. The article then implies that outside of that cost are the costs of upgrades to network to support iPhone data useage on its unlimited plans, resulting in a “cash shortfall of up to $5 billion” through 2013. Again, vague what that exactly consists of. Sprint says that the estimated wholesale cost of the iPhone is 40% or $200 more than other smartphones.
If Apple is running margins that essentially suck the profit out of the wireless phone telcos and into its own pockets, then there is another way that dropping market share can rapidly attack Apple’s margins – and that is by removing their leverage against the wireless phone companies
The article is oriented around a reference made by Hesse, CEO of Sprint Nextel, in a earnings call where he made a reference to “Moneyball” about how smart the iPhone is to him. But the article points out, that the Moneyball theory is low wage value players not high wage players, and claims that Hesse got his metaphor backwards.
SPQR is quite right, but to understand the degree of wishful thinking Hesse is exhibiting here you have to bear in mind the huge Damned Fact that drives the behavior of Sprint and other telcos: the real rates of return on carrier cell networks are negative! The carriers are burning capital, all day, every day.
When ROI is negative, you become desperate to drive down costs or pull up margins. Desperation makes CEOs stupid; Hesse is exhibiting that kind of stupidity by placing a bet that even if he shovels most of Sprint’s present profits down down Apple’s throat, the iPhone will push Sprint’s margins up soon enough for the deal to be a net positive in four years.
Meanwhile, in the real world, Android’s U.S. market share is probably passing 50% right about now. I wonder how long it will take for Sprint’s board to realize they’ve been had and fire Hesse’s ass?
Ah, you found a link. I had a paper copy I stole from the coffee shop at lunch break.
And for the Apple fanboi who want to troll me, be aware that I’m not certain about good Mr. Raymond’s disruption from below hypothesis but I’m pretty certain that Apple’s current business model can’t last in its current form for more than a couple more years.
That’s a very good problem. I better put on my patented Stuponitron Helmet.
The answer’s simple really: Sprint is going to do what the other telcos have done, and precisely what they said they wouldn’t do: start capping and metering data usage.
It’s the only way out.
Jeff Read, brilliant. You’ve solved some small fraction of the $5 billion cost. A mystery how much of it.
Now the fact that Sprint will lose money on iPhones for at least four years? The fact that Sprint will be reporting earnings to analysts before accounting for iPhone subsidies alongside interest, depreciation, taxes etc.?
Accounting games are all well and good, but they can’t be sustained once the checks start to bounce. Is Sprint actually in any danger of a liquidity problem in the medium-term or sooner?
@SPQR Why not more than a couple more years? It’s only lasted a decade or so…well made high margin hardware with lots of value added from excellent software and ecosystem. Or is the implication that without Jobs they wont make it through the next transition to new disruptive product line?
@esr now that Android is 50% I’m sure you’ll tell us again that Apple will be doomed?
Basing your conclusions on a ZDNet article is humorous. Basing it on Craig Moffet who is in the business to push stock is even more humorous. This is the same guy who predicted that the wireless business would collapse back in 2009.
“This industry is collapsing,” Mr. Moffett said in an interview. “The whole wireless business is grinding to a halt.”
http://bits.blogs.nytimes.com/2009/03/05/analyst-wireless-industry-seriously-hobbled/
Reality:
Wireless subscriber connections: 302.9 million, compared to year-end 2009: 285 million for an increase of 6 percent.
…
Total annual capital investment in 2010 rose 22 percent to $24.9 billion. Wireless annual service revenue increased 4.8 percent from a year earlier to $159.9 billion. Wireless data revenue grew to $50.1 billion, representing 31.4 percent of total service revenue, according to the wireless association report.
http://www.dailywireless.org/2011/03/24/u-s-wireless-growth/
Yeah, that’s analysis I’d bank on.
Sprint has to invest in infrastructure to survive, iPhone or no iPhone. The fact that a company has come out and said that they can’t compete without the iPhone must really stick in your craw. Lack of the iPhone was the top reason that customers choose other carriers according to Sprint.
Yes, Sprint might strike out. Then again then need to keep post paid subs high and the iPhone will help greatly there. Their unlimited data is attractive although Sprints 3G speeds much lower than my AT&T speeds. That they too had a record day and sold out their $199 iPhone 4S allotment means that at least they aren’t likely to be sitting on unmovable phones.
> Meanwhile, in the real world, Android’s U.S. market share is probably passing 50% right about now.
Meanwhile, in the real world, Android phones suck diseased donkey phallus.
http://theunderstatement.com/post/11982112928/android-orphans-visualizing-a-sad-history-of-support
Seriously, Eric. If CyanogenMod, et al can manage to update phones, why can’t Google?
SPQR, note that part of that ‘cost’ is to get to LTE, (in accelerated fashion), rather than WiMax. Everyone can now safely assume that the iPhone 5 will be LTE. Why else would Sprint change its long-held WiMax position?
@Mike E> Is Sprint actually in any danger of a liquidity problem in the medium-term or sooner?
The pros are saying that Sprint needs to raise $5B of debt during 2012-2013 to fund Network Vision (the LTE transition), iPhone subsidies and debt maturities. Sprint has existing credit lines that can supply $4B of that.
Sprint’s problem is that, without iPhone, they were fscked. They may be fscked with it, but they were certainly fscked without it.
Nigel, I’m counting Mac computers –> iPod /iTunes –> iPhone / App stores as each being a transition to a new business model. “well made high margin hardware” is not a business model by my definition. At best its a mission statement. Too many people confuse the two.
Nigel, well Sprint says that it can’t compete without the iPhone, but don’t its numbers say that it can’t compete with the iPhone? That’s the point of the WSJ piece, independent of my or ESR’s interpretation.
Wow, Motorola SHIPPED a whopping 100,000 Xooms. No, I didn’t forget to add a zero there. This is in the same period that Apple SOLD 11.1 million iPads.
One reason that Hesse is so bullish on the iPhone is because the ARPU numbers are high while using less bandwidth than the Android phones they have.
http://www.androidpolice.com/2011/10/27/dan-hesse-the-sprint-iphone-is-actually-helping-keep-unlimited-data-around-longer-for-smartphones/
Nigel, there is an amusing contrast between how you describe the story and the tone that the author of the linked piece uses with respect to Hesse’s claim.
@SPQR Apple’s business model is selling hardware at high margins. Is there any debate on that? That the hardware being sold has changed it doesn’t change Apple’s business model.
I suppose you can argue that Apple’s business model is “End-to-End” vertical integration but the bottom line is that they have a hardware centric business model based on lower volume and higher margins.
As opposed to Dell’s which is also hardware centric by based on high volumes and low margins.
@SPQR I quoted it from Android police deliberately among multiple sources. That they still state
“While Hesse didn’t provide any statistics to back this assertion up, the fact that Apple has such network usage regulations in the first place, and that Android does not, makes it seem like a pretty reasonable proposition to us.”
and that it was stated during a conference call means it’s probably not outright lies. The subsidy is higher but coupled with higher ARPU and lower costs it means that their future projections are not fabricated from thin air. Even if increased subs aren’t as high as they expect, reducing churn and retaining high value subs are very important to them to stay alive.
Android sales collapsed heading into the iPhone 4S announcement:
http://www.bgr.com/2011/10/24/ominous-signs-for-android-vendors-after-q3-smartphone-sales-decline-analyst-says/
In fact, if you remember, Google announced they were activating 550,000 per day in July, with a growth rate of 4.4% per week. That would mean that they would have hit a million activations per day last week, something they would have shouted from the heavens during their earnings call a couple of weeks ago. Why the silence? My guess is that the growth rate collapsed between July and September, and remains flat or even lower after the 4S announcement.
http://theunderstatement.com/post/10977941562/october-20-2011-androids-1-million-device-d-day
>My guess is that the growth rate collapsed between July and September
I doubt it; we already have the August ComScore figures and they show no sign of falling off the long-term Android trendline. We’ll know for sure in a week or so when the September figures become available. We hear these “Android sales are collapsing!” stories around every Apple product launch; they’ve never been true yet.
If Sprint has to burn money to upgrade their network to cope with iPhones, doesn’t that tell us that their network needs an upgrade to be not-shit for serious Android users, too?
(Blah blah over 50% Android… yeah, and how many of those are used as glorified feature phones that just happen to be Android? Looking at search counts by platform, and app sales, and the asides in this thread about bandwidth use, the number has to be substantial.
If Android’s “feature” for a carrier is that nobody really uses much data on it that’s just… mock-worthy.)
Fakeaccount asked: Seriously, Eric. If CyanogenMod, et al can manage to update phones, why can’t Google?
Google does; look at your own link. The Nexus phone was kept quite current throughout its lifetime, and is only going to become un-current-from-Google when ICS ships.
Google is only – reasonably – responsible for “updating” the phones it sells – and it’s done a good job there. It can’t force carriers and manufacturers of any arbitrary Android phone to update them.
(Well, it could try, but they’d laugh in Google’s face and just fork Android, I reckon.)
LOL. 3 months ago, Motorola themselves claimed they could sell over a million Xooms this year.
http://www.androidguys.com/2011/07/29/motorola-expects-15-million-xoom-sales-2011/
They’ll be lucky to reach half of that forecast.
Laughingly, Motorola was projected to sell 4 million Xooms in 2011:
http://www.forbes.com/sites/greatspeculations/2011/02/09/xoom-tablet-crucial-for-motorola-mobility/
Something seems amiss in Android-land.
>LOL. 3 months ago, Motorola themselves claimed they could sell over a million Xooms this year.
I said at the time that forecast was full of shit. The Xoom is way, way overpriced.
Louis: But it runs Flash!
More seriously, that sub-site of Forbes is plainly run by idiots – and DigiTimes, their source for “7-800k in Q1” is notoriously terrible at predicting anything.
What confuses me is the way people actually listen to such forecasts as if they’re not completely bullshit. Because they’re almost always bullshit.
(In Motorola’s defense, a projection of 1.3-1.5M sales total after .44M in the first quarter was not implausible; it only required flat sales, no increase. Given the observed curve, they’re not going to make it, but at the end of July that can’t have been as obvious.
If BGR is right (and since it’s a Moto report, they’re probably not inventing it), they’ve already sold 750K this year. So when you said “they’ll be lucky to reach half that forecast”, well… they already have.)
Eric,
Y U NO BLOG ABOUT PROTECT IP ACT?
The Feds are trying to break the internet and you’re strangely silent on it.
@ Sigivald Remember, shipped vs. sold. My bet is they’ve sold at most 200,000 in total since launch.
@Louis4:
1) Android fans have been awaiting new Samsung handsets just as Apple fans have been awaiting new iPhones.
2) When that chart came out, it got me thinking, and I pulled out Apple’s quarterlies. As I’ve already reported in a comment on an earlier article, I believe that Apple was shipping everything they could in Q2 to make it look like the iPhone had momentum. They had firesales (e.g. $50 off at WalMart for almost the whole month of June), to cannibalize their Q3 sales and make Q2 numbers look better, because they knew they’d be shipping a new iPhone by the time the Q3 numbers came out.
The market punished them a bit for a down Q3, but I’m not sure the market fully realizes that Apple deliberately pulled those units back into Q2.
Nigel writes: “@SPQR Apple’s business model is selling hardware at high margins. Is there any debate on that? That the hardware being sold has changed it doesn’t change Apple’s business model.”
Respectfully, Nigel, that’s not a business model. Apple could start selling the contents of Lowe’s and fit your definition of their business model. I think that claiming that from Mac’s to iPod’s to iPhones is all a single business model is simply wrong.
But I can’t figure out what the significance of arguing with me over the definition of business model is, unless you agree that selling iPhones at high margin can’t last more than a couple of years and you want to claim that when they invent a new consumer widget – the iTV or the iCar – with new customer demographics and new sales channels, that they are still carrying out the “same” business model. I don’t see how that really refutes my opinions.
Oh, and I just bought another Apple product. My second. An Apple iBook G4 laptop. Unfortunately it does not help Apple’s quarter since they originally booked the revenue for selling in Q2 2005. Runs Linux Mint faster than the one I had bought previously for $45.
http://modmyi.com/content/5573-sprint-reports-best-sales-day-ever-wishes-iphone-came-sooner-other-carriers-happy-too.html
“Sprint today reported its best ever day of sales in retail, web and telesales for a device family in Sprint history with the launch of iPhone 4S and iPhone 4. We reached this milestone at approximately noon CT/1pm ET. The response to this device by current and new customers has surpassed our expectations and validates our customers’ desire for a truly unlimited data pricing plan.” — Sprint
I’m sure Sprint did the math, and decided that becoming Apple’s third US carrier was worth it. I’m also sure that they had access to far more detailed market info than anyone commenting here.
Meanwhile, in the real world, Android phones suck diseased donkey phallus.
I knew there was an issue with Android phones not getting updated, but I had no idea just how bad it is.
The difference of course is that Apple provides customer support for the iPhone, and their costs are lower if everyone’s on the latest OS possible. Android phones are “supported” by the carriers and some dude in a chat room if you’re lucky, and you have the time and the mad hacker skillz to build your own copy of Android once someone tells you where to find all the bits and pieces you’ll need.
Verizon sold 2 million iPhones in Q3. They sold 5.6 million smartphones total. “Over half” of those were Androids. All of the iPhones were year old models.
Some Guy, the article above has some numbers from the Sprint reports. What math did they do? You say that you are “sure” that they did “the math”. What math did they do that led them to conclude that spending $15 or 20 billion with a revenue neutral point (I’m interpreting “positive impact on Sprint operating income” to not mean actually paying back the up front costs) four years out made sense?
What math is that?
The telcoms need to add more knowledge into their service offerings. No one can make a profit selling debt (i.e. dumb infrastructure).
I wonder if technologically they could offer a higher throughput or call quality, in exchange for premium fee, on a per use basis. Perhaps they could offer a smartphone software that toggles such a service level on and off. The advantage would be that you would know you have the ability to supercede all other non-premium users when you need it urgently. Business people would probably love such a feature. When you are in hurry, price is secondary.
Imagine it could be like an auction feature. Where you up your bid, to increase your throughput, as you compete against the bids of other users on your same tower and backhaul.
This sounds like Google and GOTO.com’s PPC advertising bid model. I was one of the first GOTO.com (Overture) customers.
Any other ideas for the telcoms to escape their monopolistic mode and become competitive knowledge business models?
Some Guy, my Motorola Android phone has been well supported to date. And I got updated to Android 2.3 a couple of months ago. I didn’t actually build my own copy or root the phone – but only because the one year warranty hasn’t expired yet.
> my Motorola Android phone has been well supported to date.
MMI continues to lose money.
Thread jack:
@Jeff Read & esr, I also wish influential freedom proponents would comment publicly about the US legislative push to censor the internet. (the link I provided is not a reliable source of facts, but the linked page contains mainstream links and I am in hurry)
Life as we know it Says, well great you learned “Non sequitur”. Now for a good “No True Scotsman” you should have included something like … oh, I don’t know, maybe … “Gingerbread is not True Android!”
@SPQR:
Much as I hate to admit it, Life as we know it’s comment may possibly be relevant, if he can somehow make a case that it’s impossible for a phone vendor that properly supports Android to stay solvent.
@Patrick Maupin, (apologies to esr fix 3x post but I must respond to this), the free market is not functioning if there aren’t entities that fail. If every Android hardware company succeeded, then I would be trying to identify some subsidy preventing the free market from optimizing via competition.
@Shelby:
I understand that perfectly. What did I write that indicated that I didn’t?
My mistake, I see now that is what you were saying. I guess it was the opening part “Much as I hate to admit it” that threw me off. I see at the end you are challenging him to prove that isn’t just the free market optimizing.
@Jeff Read: In case you hadn’t heard, just after starting to sell the iPhone, Sprint has announced that it’s dropping unlimited data plans and instituting caps — but only for tablets, USB connection cards and hotspots (including tethering). So, they still have the advantage that non-tethered phones will be unlimited. But, if I have this correct, unlike Verizon, existing plans will not be grandfathered.
What math did they do?
Read and learn:
http://en.wikipedia.org/wiki/Cost–benefit_analysis
I didn’t actually build my own copy or root the phone – but only because the one year warranty hasn’t expired yet.
What does the warranty have to do with it? Surely, no Android vendor would make the warranty contingent on whether you fucked up the software on the phone? Isn’t Android all OPEN and shit?
>My guess is that the growth rate collapsed between July and September
http://thenextweb.com/google/2011/07/14/there-are-now-550000-android-phones-activated-every-day/ , July 14th, Page announced 550’000 phone activations per day and a total of 130 millions devices in the wild. October 13th, the same Page talks about 190 millions devices. 3 months, 60 millions additional devices, do the math: 666’000 activations per day. Is that “collapsing” for you? With ICS and holiday sales approaching, and with Nokia Windows Phones which do not stand out, I guess there’s no collapse in sight.
@SPQR Why would you assert that given that Macs are still high margin hardware that’s been around decades and still making Apple a lot of money ($5B in revenue)? Hardware that increased sales 26% YoY in Q4.
Likewise iPods are high margin hardware that’s been around 10 years that still makes Apple a lot of money ($1.1B)? Declining because the iPhone (touted as the best iPod ever when launched) has largely cannibalized it but still a healthy amount.
While content and software generates a good amount of income the bulk comes from hardware sales…because their business model is to create high value for their hardware through excellent ecosystems to justify the high margins.
iPhones will be high margin hardware that will be generating a lot of money for Apple for a long time. Even if the iPad revenues eventually exceeds it just like iPad revenues are higher than the Mac.
Name a core hardware product type that stopped generating at least $1B revenue for Apple with high margins and maybe you have a case that the iPhones ability to generate high margins might be short lived.
@Nigel
“Apple’s business model is selling hardware at high margins.”
“Why would you assert that given that Macs are still high margin hardware that’s been around decades and still making Apple a lot of money ($5B in revenue)?”
So you would describe Warren Bufffet’s business model as “buying low and selling high”? Not exactly an informative use of “business model”.
@Some Guy, openness of the OS to multiple hardware vendors and to run any s/w on the devices, is orthogonal to the issue of efficient upgrading of the OS version on an older device.
There is a cost/benefit analysis. A chaotic market of many hardware vendors makes upgradability much more costly to coordinate (evidence is the Windows bloat to maintain backward compatibility), yet on the benefit side is cell phones are approaching “throw away” or “hand me down” devices below $100.
And device specifications are changing so rapidly, many people want to upgrade the hardware.
If upgradability becomes a paramount priority, then the market will demand it and Android vendors will compete to fulfill. But it has a cost, and markets apparently currently prioritize $150 Android phones and diversity of price points, features, and styles.
I have written before that I think Apple has played an important role in the development of the smartphone market and are appreciated. That doesn’t stop me from agreeing with the projection that they will be disrupted by the more open and chaotic Android model, which will eventually stabilize and tertiary priorities will be met eventually by Android (and possibly forks).
@Nigel, I don’t think Apple’s high-margin model scales into the non-indebted world. That is why I am putting a 2 – 4 year timeframe on their peak marketcap. We even see the telcoms are subsidizing Apple via debt, in this blog. If the telcoms were smart and found a way to escape the debt model early (before it is forced on them), by perhaps selling bandwidth in real-time auctions, Apple’s subsidy might disappear sooner.
“Meanwhile, in the real world, Android phones suck diseased donkey phallus.”
So true, so true.
@Winter:
“So you would describe Warren Bufffet’s business model as “buying low and selling high”? Not exactly an informative use of “business model”.”
Buffet doesn’t generally sell.
Buffet’s business model is using regulatory and tax loopholes of calling oneself an insurance company to act as if he is a capital management firm and bank holding company.
@SPQR
A ‘business model’ is just an answer to the question ‘how do you make money from your business?’
Apple makes money by creating products (a combination of hardware and software) and selling them to customers in exchange for money (usually at a high margin).
That has not changed, even though they have introduced a lot of new products.
@Tom
“Apple makes money by creating products (a combination of hardware and software) and selling them to customers in exchange for money (usually at a high margin).”
That business model is called “Manufacturing”. The part about the High Margins is a subset called “Luxury Products”.
I am not arguing that Apple does not do this. But I consider calling it a “business model” rather comical.
@Winter
Manufacturing is not quite the ‘model’. As you know, Apple outsources manufacturing to its partners in Asia. Manufacturing is one activity goes into making Apple’s products.
I don’t know why you find what I said ‘comical’. It’s their business model, plain and simple.
Google’s business model is making things that people use and then selling advertising on those things.
McDonald’s model is to collect franchise fees and royalties from people it allows to run its branded fast food restaurants.
Apple’s model is simply to sell high-margin products to people for money.
“SPQR is quite right, but to understand the degree of wishful thinking Hesse is exhibiting here you have to bear in mind the huge Damned Fact that drives the behavior of Sprint and other telcos: the real rates of return on carrier cell networks are negative! The carriers are burning capital, all day, every day.”
Okay, being a carrier is a crappy business: networks are utilities. And the iPhone demands greater subsidies than other smart smartphones. However, does the iPhone equate to a higher ARPU for the carriers than other phones from the competition? That seems to be the question of primary importance.
Here’s a question — if Microsoft can insist that its partners update phones on a schedule, why can’t Google do the same with OHSA members? The players on the carrier and manufacturer side are essentially the same, just swap Nokia with Motorola.
Yes, Microsoft screwed up but good with the first over-the-air updates but the Mango release, to their credit was smooth. I talked to someone on the Windows Phone team about the update issues and he conceded that it was a tougher problem than they anticipated because of having to test with not only each phone but each carrier combination. Still, they saw it as a problem that must be solved.
It wouldn’t be difficult for Google to say, if you want to get advanced access to our shit, you must agree to comply with our upgrade terms. If HTC and Samsung and LG can do it for one platform, why can’t they do it for the one that makes them more money?
@tom it must be made more complex (or denigrated) in order to make some claim that it is doomed based on some fancy term that makes the claimant appear to have expertise.
The concepts are simple, it is the execution that is hard.
Warren Buffet model is to buy low (and rarely sell). It’s the buying low that’s hard to do right. More accurately Buffet studies the fundamentals of a business (not just the metrics) to determine if a company is undervalued and if so buys it. The devil is in the details of determining valuation.
It’s funny that folks here are so adamant about stuff that’s just plain weird to claim. The iPhone isn’t a business model…it’s a product line. It’s also funny that folks ignore when you point out their basic premise is flawed so any subsequent logic is meaningless.
Like that whole Moffet CAPEX ROI is negative thing*. Or McDonalds and Burger King is a duopoly and therefore WP7 is doomed because there’s NEVER a strong 3rd contender ever.
—
* Currently wireless CAPEX ROI is low because its been driven by a past technical decision on the part of VZW. By going CDMA and now needing to go LTE to stay competitive their CAPEX expenditures are high. But they’ve offset this somewhat by making LTE/4G such a huge selling point that AT&T has to push forward their LTE deployment as well when HSPA+ for the next few years makes a lot more sense for them. The estimate in CAPEX reduction in staying HSPA+ longer for AT&T is estimated to be around $1B. Real world data rates wont exceed what HSPA+ could provide anyway even at only half the theoretical peak (84 Mbps down). But getting pounded by the LTE/4G stick is more painful than upgrading early.
steg Says:
October 28th, 2011 at 12:36 am
>My guess is that the growth rate collapsed between July and September
http://thenextweb.com/google/2011/07/14/there-are-now-550000-android-phones-activated-every-day/ , July 14th, Page announced 550’000 phone activations per day and a total of 130 millions devices in the wild. October 13th, the same Page talks about 190 millions devices. 3 months, 60 millions additional devices, do the math: 666’000 activations per day. Is that “collapsing” for you? With ICS and holiday sales approaching, and with Nokia Windows Phones which do not stand out, I guess there’s no collapse in sight.
Uh, this actually proves that the GROWTH RATE has collapsed. If they had continued to GROW at 4.4% per week, as they claimed in July, they’d be at 1 million activation per day 2 weeks ago. In 3 months, the rate has gone from 550K to 666K per day, and increase of only 116K, or about a 20% growth rate over 3 months. If their growth rate had kept going on as they claimed, it should be about almost 100% growth since July. The GROWTH RATE has definitely collapsed.
Not sure where to put this, but I’ll post it here because I think it’s relevant:
http://www.bbc.co.uk/news/business-15489523
During the last quarter, the iPhone was outsold not just by Android in general, but by a *single* Android vendor (Samsung). (And rightly so — I just picked up a Galaxy S II and I’m convinced it is *the* finest smartphone available today, bar none.)
The Apple apologists are, predictably, touting the “Apple’s customers were waiting for the iPhone 4GS” rhetoric, but that line of talk has always been proven incorrect in the past.
I really remain baffled why Tom and Nigel think that arguing their version of “Business model” refutes any of my opinions. That said, I’m quite serious and I suspect so is Winter, when I say that “sell high margin hardware” is really not a business model. To properly execute a business model requires quite a bit more understanding of the kind of product you are selling (more detail than “high margin”), the customer profile and the sales channels. Changing any of those are large undertakings in business and do result in a different business model.
@SPQR
I’m not talking about how to execute the business model. That is something entirely different. I am simply telling you what the business model is.
You claimed that their business model has changed over the years as they have expanded from the Mac to the iPod and to iPhone and iPad. This is incorrect because their business model has always been simply to create products that can command high margins and sell them to customers for money.
Yes, they have introduced new products, of course, but the model has remained the same.
Maybe “sell high margin hardware” is their ‘mantra’ or ‘Mission statement’?
@David Scott Williams
Actually their mission statement is:
To me that is not really a mission statement, or not a very good one anyway.
But if Apple says it is their mission statement then I guess it is their mission statement.
I suspect that they might have a different one internally. Or maybe they just don’t care about mission statements.
Those are smartphone SHIPMENTS, not SALES. There is a significant difference — often a vast gulf of one. Only a tiny fraction of Galaxy Tabs SHIPPED were actually SOLD. I suspect the percentage is larger for Galaxy S phones, but not 100%.
You haven’t tried an iPhone 4S. Its CPU is about the same, its GPU is faster, and it runs 100% native code (no VM layer). It objectively outperforms the Galaxy S II by a considerable margin. And it’s easier and more pleasant to use.
@SPQR Then ignore the semantics simply address the challenge:
Name a core hardware product type (what you mistakenly call “business model”) that stopped generating at least $1B revenue for Apple with high margins and maybe you have a case that the iPhones ability to generate high margins might be short lived.
Answer: You can’t and hence arguing what is or isn’t a business model even when you’re using some arcane definition you made up. Your opinion isn’t based on anything.
Re Samsung outselling Apple:
Nokia outsold Apple until recently. Apple being #1 was always going to be a short lived affair given their business model.
Nigel, well I guess I see no reason to continue treating you with respect.
>Louis4 says:
>Uh, this actually proves that the GROWTH RATE has collapsed. If they had continued to GROW at 4.4% per week, as they >claimed in July, they’d be at 1 million activation per day 2 weeks ago. In 3 months, the rate has gone from 550K to 666K per day, >and increase of only 116K, or about a 20% growth rate over 3 months. If their growth rate had kept going on as they claimed, it >should be about almost 100% growth since July. The GROWTH RATE has definitely collapsed.
I’d say the growth rate has indeed slowed down, but certainly not collapsed. Besides, note that 666K is the average of 3 months, it’s very likely that in July that number was close to 550-600K and right now it’s certainly above 700-750K. Now, indeed, the growth rate has notably slowed down. Why? Is there some seasonality, i.e. people waiting for holiday sales and new models? Does the looming economic crisis play a role in this? I don’t know, but it looks like Apple’s sales followed a similar trend during that period (less sales than anticipated).
SPQR, so in your opinion, what exactly is Apple’s “business model”?
@Nigel
Selling stuff for profit is not a business model, but simply business. And Apple is not just selling high margin beverages or hand bags. You argumentation is silly.
Also, Android is outselling iPhone 2:1. Someone has to sell them so there is room for a producer to outsell Apple. Given that sales are likely distribute with Zipfs distribution, that is even likely.
Zipfs distribution also explain why being third on the market is so hard.
Actually if you look at it ZTE now sells more handsets than Apple. The list for 2011 is Nokia, Samsung, LG, ZTE and then Apple.
That Apple took #1 in smartphones the last qtr was really an aberration based on the brief moment where Nokia was falling too fast for Samsung to overtake it. Apple’s growth has steady since the initial burst but never atmospheric since because production constraints always limit growth.
It’s fairly surprising that Apple is #5 really. In Q3 ’09 the top 5 were: Nokia, Samsung, LG, Sony Ericsson, Motorola.
http://www.businesswire.com/portal/site/home/permalink/?ndmViewId=news_view&newsId=20091029006638&newsLang=en
A year later it got to #4. Nokia, Samsung, LG, Apple, RIM
http://www.gartner.com/it/page.jsp?id=1466313
Numbers for Q3 2011
http://www.idc.com/getdoc.jsp?containerId=prUS23112511
Note that Nokia shipped 107M units and Samsung shipped 88M units. Even if Apple meets UBS’ wildly exuberant expectations of 42M handsets in CY4Q (IMHO not a chance – Apple likely can’t MAKE that many) that’s a far cry from the number of units Nokia and Samsung sells.
I’m a little surprised that LG has slipped so much though. Another year like this one and they’ll be #4 behind ZTE and possibly Apple. 21M vs 17M isn’t as big a gap as I would expect.
@SPQR My apologies if I offended you. Lets forget the whole, what the hell is a business model thing for the moment. We can debate that again later.
However, there’s no answer to the challenge then?
On what basis do you suggest that the iPhone will only be able to produce high margins for Apple for a limited time when every other major Apple product line has produced high margins for a decade or more?
@steg “I’d say the growth rate has indeed slowed down, but certainly not collapsed. Besides, note that 666K is the average of 3 months, it’s very likely that in July that number was close to 550-600K and right now it’s certainly above 700-750K.”
Larry Page said it was 600,000 just a few days ago during earnings.
@Nigel “Actually if you look at it ZTE now sells more handsets than Apple. The list for 2011 is Nokia, Samsung, LG, ZTE and then Apple.”
Apple was never #1 for handsets. Nokia, Samsung, and LG always led them for total handsets. They may well remain #1 for smartphones for the full year. We won’t know until after the holiday.
“Apple’s growth has steady since the initial burst but never atmospheric since because production constraints always limit growth.”
Apple’s growth has been “atmospheric”, approaching 100% year over year consistently. What isn’t “atmospheric” is its market share growth. But their own unit growth has been astounding.
ESR: http://esr.ibiblio.org/?p=3548#more-3548
“Apple is now relying on smartphones for 68% of revenue, so they’d be very vulnerable to an actual drop in marketshare. I’ve taken a lot of flak for saying the company looks like a late-stage sustainer with a principal product line about to experience disruptive collapse, but this is yet another straw in the wind.”
http://www.osnews.com/story/25269/Samsung_Surpasses_Apple_as_World_s_Top_Smartphone_Seller
“The strong results in Samsung’s telecommunications divisions also means this division now accounts for 60% of Samsung Electronics’ operating profit, surpassing the chip division for the first time” … Uh oh!
@winter zipf’s is yet another statistical approximation, not reality (and something of a stretch to apply here anyway). Every example you have produced I have shown a 3rd major contender. Many quite close in size to the #2 contender.
You might argue that handset market is a duopoly with Nokia and Samsung controlling 49% of the market but in ’09 it was split 37%-21%-11% between Nokia, Samsung and LG. And it didn’t keep a new entrant from coming in from 0% in 2007 and taking the #4 spot for a year.
@Tim F. I agree that unit sales growth has been “atmospheric” but the portrayal has been the market was dominated by Apple (never really true even for smartphones) and that it’s “losing” its dominance to Android. In the larger scheme of things, however, 20M handsets isn’t all that much.
Certainly they have more than exceeded their initial goal of 1% market share which many thought was quite ambitious.
The only place that Apple has dominance in has been MP3 players and now perhaps tablets. Given the iPad is still so new its hard to say but Apple has a shot at tablet dominance.
steg:
“I’d say the growth rate has indeed slowed down, but certainly not collapsed… Now, indeed, the growth rate has notably slowed down. Why? Is there some seasonality, i.e. people waiting for holiday sales and new models? Does the looming economic crisis play a role in this?”
Why is anyone surprised that growth rates follow an S-curve, and slow down as you pull more of the available market? It’s perfectly understandable that the early adopters are already in, the late adopters are the current market, and at some point you’ll have to pull in the laggards from the dumbphone world. Each group is harder to pull than the previous one.
On the somewhat related TV front the strategy of Apps vs Airplay should be interesting. I hold that Apple should have an aTV app store but it nice to see that Google TV will have that.
Still, Airplay solves the issue of expensive game controllers. Have to have at least one to do AirPlay at all. Interesting enough, one of my son’s games now has iPhone remote to play on an iPad main screen as well as an AirPlay to aTV mode. I guess that means I ought to buy an aTV for christmas.
@Tom, agreed Apple’s business model has always been to build niche products that can command premiums, based on their vertical market ideas about priorities in computing. Generally speaking computing is about hardware openness, because a computer should be a general purpose programmable device with a wide diversity of physical configurations and software customizations. However, when they applied their “end-to-end” integration philosophy to consumer electronics market with iPod and iPhone, their ideas about hassle-free use, hit a mainstream chord. Afaics, they have been helped by the massive debt subsidy in the developed markets, that lets westerners afford to pay $1 per song (instead of hassling with “illegal” file-sharing), and $700 per phone (hidden in the debt subsidy of the telcoms).
@Nigel, has Apple ever been able to sustain a significant market share with any product? I don’t think that is their goal. They want to build tightly integrated products. No company dominates in consumer electronics with a quality, high-margin product line. Even the Sony Walkman and Handycam succumbed to low-priced competition. Even Sony was disrupted from below and had to give up their proprietary media formats and adopt those that became ubiquitous. This is the fate I expect for iOS, as Android becomes ubiquitous.
@Tim F, “Okay, being a carrier is a crappy business: networks are utilities.”
I don’t agree that networks have to be utilities, because afaik networks have enough information to charge market pricing, where users could compete on price to get the throughput they need in real-time. Seems that the telcoms need think out-of-the-box, and realize they too can be software (knowledge valued added) companies.
To allocate finite resources, one must sell portions each to the highest bidder, i.e. a free market. It is collectivism to give the same price to all bidders (flat fee to all customers for network bandwidth), and can only be sustained by debt, because it doesn’t maximize the knowledge in the system.
If the electric grid could be so dynamically allocated with smart meters and multiple suppliers, then perhaps another utility could be converted into a competitive free market.
@Christina Warren, it is a reasonable conjecture that requiring upgradability would have a cost in terms of slower advancement of other areas of the ecosystem, such as expending maximum effort to get the lowest priced smartphones and/or the most diversity of phones. And by not requiring it, it frees the priorities to be determined by market demand. If the market demands upgradability and is willing to pay more for it, then probably some vendors will fulfill the market. Whereas, if it were forced by Google on all vendors, then it would be an unnecessary cost on all users, who may not have that as their priority need (not willing to incur the tradeoff of it, i.e. less rapid advancement of cheaper and/or more diverse phones choices).
@Louis4, if the entire smartphone market growth has slowed (seasonal?), then even if activation growth rate has slowed, Android share growth rate may not have slowed.
@Shelby “I don’t agree that networks have to be utilities, because afaik networks have enough information to charge market pricing, where users could compete on price to get the throughput they need in real-time. Seems that the telcoms need think out-of-the-box, and realize they too can be software (knowledge valued added) companies.”
I’m trying to address the core of ESR’s post here, and I’m unsure if anyone has or if anyone could find an argument.
To me, it appears to be: any investment or attempt to improve carrier margins is useless because they ultimately have negative return. Sprint is making a particular aggregious investment because it’s Apple. I would like to understand the basis and logic to ESR’s post. I haven’t seen it yet.
>To me, it appears to be: any investment or attempt to improve carrier margins is useless because they ultimately have negative return.
Er, does your iPhone emit stupid-rays or something? I wrote nothing even remotely like this.
What’s idiotic about Sprint’s move isn’t the attempt to raise ARPU, it’s accepting such a long period before the expected net goes positive that Sprint’s gains basically disappear down the net-present-value event horizon while Apple makes all the prompt money. Brutal cost-cutting now would serve Sprint better; this deal reeks of the kind of “market development” traditionally lubricated by bribery, hookers and cocaine.
Tim F. wrote:
>Larry Page said it was 600,000 just a few days ago during earnings.
You sure about that? Not only I never heard that number but the other numbers I discussed in my previous comment are inconsistent with that 600K.
Cathy wrote:
>Why is anyone surprised that growth rates follow an S-curve, and slow down as you pull more of the available market? It’s >perfectly understandable that the early adopters are already in, the late adopters are the current market, and at some point >you’ll have to pull in the laggards from the dumbphone world. Each group is harder to pull than the previous one.
Heh, probably because I just got my hands on my first, dead cheap Android phone running Eclair only recently (a 19€/month 1 year contract for a budget plan and the phone was free with it). No idea if I’m a late adopter or a laggard but to me the market is still booming and in the early stages. Then again, it’s quite possible that Europe is lagging a bit behind the US.
@Winter/SPQR
I think you guys might have a mistaken idea of what a business model is. It’s not a business plan or business strategy. I have pasted below the definition from Investopedia, which you might find helpful:
Business model is a buzzword that everybody used (or overused) during the dotcom boom. In fact, poorly thought out business models were the downfall of many dotcoms.
However, the business model dates back to the earliest days of business; it merely describes the way in which a company makes money. A business model can be simple or very complex. A restaurant’s business model is to make money by cooking and serving food to hungry customers. A website’s business model might not be so clear, as there are many ways in which these types of companies can generate revenue. For example, some make money (or try to) by providing a free service and then selling advertising to other companies, while others might sell a product or service directly to online customers.
Mentioned before in this thread (I think), off topic (in specific) .. but ..
Eric — any linkable thoughts on this:
http://news.cnet.com/8301-31921_3-20126165-281/copyright-bill-revives-internet-death-penalty/
..(the nearest at hand article at time of writing this comment)
@ESR “I wrote nothing even remotely like this.” It’s entirely unclear what you are trying to insinuate.
“What’s idiotic about Sprint’s move isn’t the attempt to raise ARPU, it’s accepting such a long period before the expected net goes positive that Sprint’s gains basically disappear down the net-present-value event horizon while Apple makes all the prompt money.”
‘What’s idiotic about Sprint’s move isn’t the attempt to raise ARPU, it’s accepting such a long period before the expected net goes positive that Sprint’s gains basically disappear down the net-present-value event horizon while Apple makes all the prompt money.”
This is much clearer; however, I disagree that 3 years is a long period. In fact, in terms of carrier investments, I would rate it as “near” or “short” term. Nor do I see how Apple’s profits factor into the value of the investment for Sprint at all.
“Brutal cost-cutting now would serve Sprint better; this deal reeks of the kind of “market development” traditionally lubricated by bribery, hookers and cocaine.”
Cost cutting would do nothing to retain old or attract new customers; Sprint’s largest problem is they are losing their customer base.
Possible relevance to the discussion (at least of profitable cell phone manufacturers, if not of carriers)? (Someone with an actual finger on the pulse of the smartphone market can discuss the numbers better than I)
http://techcrunch.com/2011/10/28/the-king-has-fallen-samsung-dethrones-apple-as-smartphone-leader/
‘McDonald’s model is to collect franchise fees and royalties from people it allows to run its branded fast food restaurants.’
Sort of. They are also one of the world’s largest real estate companies.
It’s either that, or become the only U.S. carrier without the world’s premier smartphone — the iPhone.
The latter option could have been a business killer for a company in Sprint’s position.
David Scott Williams,
Again — Samsung is reporting shipments. Apple is reporting sales. It’s hard to get a hard lock on Samsung’s sales numbers because they can’t track it the way Apple can — but I’d be willing to guess that Apple still has Samsung beat in the sales department, and definitely in the “number of satisfied customers” department.
Remember, Samsung is the company that bragged about shipping a million Galaxy Tabs, but only actually sold 20,000 of those.
@Jeff Read Samsung hasn’t even said one word as to how many they shipped — all of the 27-28 million shipment numbers are based on analyst’s estimations.
However, I totally believe that Samsung may have surpassed Apple — but I don’t see how it’s relevant to this conversation. It seems like a rather unremarkable fact if it is, in fact, true.
Indeed if your competitors are selling resources below cost by using debt, Sprint has to do the same to not bleed share. But one could look for a smarter pricing model as I suggested to potentially find a way to compete against the debt bubble in advance of its implosion. With more finely grained market based pricing (where customers are bidding in real-time geographically for the throughput cost vs. benefit they desire), perhaps a telcom could gain more information about where to allocate capital spending in order to maximize customer needs and thus profit. Such a higher knowledge free market system could in theory outperform over the long run, after some shrinkage due to the transition costs. Perhaps there are details that make the impractical or hassle for customers which outweigh any benefits. I imagine a meter which shows the month rate extrapolation of current bandwidth spend rate and recent avg throughput, then the user dials it up or down when ever they wish. They only need to hassle with it, when they have need to up their throughput or reduce it.
“I enjoy cocaine because… it’s a fun thing to do.
I enjoy the company of prostitutes for the following reasons: …oh, because it’s a fun thing to do. Much like cocaine. If you combine the two together, it’s probably even more fun.”. — Robert Wexler
@Jeff Read:
> I’d be willing to guess that Apple still has Samsung beat in the sales department, and definitely in the “number of satisfied customers” department.
I’d take that bet on both counts. Samsung’s been selling cellphones for a long time, and even in Apple’s much-touted Q2 sold 3.5 times as many cellphones as Apple. You don’t keep doing that for that long without having satisfied customers. You also don’t keep doing that for that long while keeping lots of inventory locked up in the channel.
Even in the rich (and as Shelby notes, debt-subsidized) US, Samsung’s installed base is over 2.5 times as many phones as Apple’s. Even if 60% of Samsung’s US cellphone customers described themselves as not satisfied, and 100% of Apple’s customers described themselves as satisfied, Samsung would win the “number of satisfied cellphone customers” contest. But that’s not the case. Most Samsung cellphone customers are very happy with their phones, although Samsung still has some work to do to make people as happy with its smartphones as they are with Samsung dumbphones or with iPhones. See, for example:
http://gadgetbox.msnbc.msn.com/_news/2011/09/08/7673715-iphone-again-no-1-in-customer-satisfaction-jd-power
This might not mean much to those cognitively-dissonant folks who simultaneously think that (a) cellphones and smartphones aren’t in the same market, and (b) smartphones and tablets are in the same market. But people like me, who think that smartphones are just better cellphones, think Samsung has a pretty good shot of converting a lot of its installed base of dumbphone users into smartphone users without Apple’s “help.”
And I am very impressed with how Samsung has turned things around. Although Eric isn’t tracking the comscore manufacturer numbers, IIRC for awhile, right when the iPhone became wildly popular, Samsung was bleeding marketshare just like Nokia. But lately they’ve been ramping steadily, just like Apple, but from a much bigger installed base.
@Patrick Sure samsung has brand loyalty. Their products don’t typically suck…in marked comparison to some of the dismal Moto and Sony phones I’ve used. Although to be honest the first Samsung phone I used was the blackjack.
I never quite understood how Sony could consistently screw up their walkman phone models somehow. If there was ever brand that should have had the phone + mp3 player part lock down long before the iPhone it should have been Sony.
The launch of this iconic device resulted in Sprint’s best ever day of sales in retail, web and telesales for a device family in Sprint history. The response to this device by current and new customers has surpassed initial expectations. The iPhone is expected to be accretive for Sprint, and iPhone users are expected to be among Sprint’s most profitable customers.
Wait, what was that, “iPhone uses are expected to be among Sprint’s most profitable customers.”
Oh Really?
To my surprise, it appears Blackberry style keyboard phones (not Blackberry brand) are very popular in the SMS capital of the world. I would opt for a slide-out, because I want to SMS less and want to use two thumbs. Here the females want to SMS as often as possible (see them typing while walking in the mall and glued to their cps). The cp + keyboard can then be operated with one hand, leaving the other hand free for carrying a coffee.
Dunno why Sprint is so excited about this in the first place. The iPhone isn’t all that great a device in the first place. It’s really just a fashion statement — overpriced and overrated.
The iPhone 4S is the fastest, most advanced smartphone on the planet. The benchmarks place it well ahead of any Android phone. From a UX standpoint, even its predecessor was smoother and snappier than recent-model Androids. There is a lot to be said for using native code and fully leveraging the GPU. Oh, and as fake account said, Apple is the only company that can deliver this kind of UX, because they have a lead of several years over everyone else in resolving subtle compositing bugs.
There are areas were Android has a clear lead, like mapping. But Google can soon kiss its lead in mapping goodbye.
@Jeff Read:
Interesting article. But, because of some things that some posters (not you, I don’t think) keep harping on, I wanted to stress this text from the article.
This. sort. of. thing. is. why. google. needs. Android.
Anybody who says “but Apple gives them lots of ad hits. What’s the problem?” is missing the bigger picture.
@Patrick
Is that not a perfect summary of Apple appologists?
Missing the big picture
> This. sort. of. thing. is. why. google. needs. Android.
You’re right, this is exactly why Google can’t stop developing and defending Android, no matter the cost.
($12.5B for MMI patents!) The only product that Google has that generates appreciable revenue is web advertising.
btw, Jobs offered Google a permanent place on the home screen of iOS if they would stop Android development.
Now if only Eric would write things like this: http://bytebaker.com/2011/10/19/ubuntu-should-zig-to-apples-zag/
instead of blathering on about the cellular industry, where he has shown exactly zero expertise.
@the adrenalist
Such an offer by Apple would be highly illegal in Europe. Worthy of billions in fines.
@patrick while its true that Google would prefer to control as many pieces as possible, originally there had been no real overlap between Apple and Google. Without Android there’d be no need for Apple to develop iAds or their own mapping services. Apple was never a threat to Google revenue until they made them an enemy.
Big picture wise open sourcing Android was a mistake. Not open sourcing Android would have prevented Kindle Fire and Baidu Yi.
@Nigel
Apple’s fundamental philosophy is vertical integration. They’d have gotten around to doing their own maps with or without Android. And eventually search.
The Kindle Fire is good for Android. If it succeeds, developers will be strongly motivated to create high-quality Android tablet apps. Google might prefer that you buy a “pure” Android tablet than a Fire, but they’d much rather you buy a Fire than an iPad.
“This. sort. of. thing. is. why. google. needs. Android.
Anybody who says “but Apple gives them lots of ad hits. What’s the problem?” is missing the bigger picture.”
This is a chicken-and-egg problem. Not only in the sense of: would Apple be competing with Google if Google wasn’t competing with Apple? But you also have the scenario of: Google is withholding key features of Maps/Navigation from other platforms in order to make Android more attractive, which it has to do to make sure Google Services are attractive, but if it does so on all platforms, it makes Android less attractive… Lather and Repeat.
It’s true that it’s great to find a defensible market, create your own platform that guarantees a market for your core profitable services; however, more often than not this can not be created and/or there are still plenty of avenues to compete. Suggesting that any threat of competition requires controlling a platform to secure market presence rather than any other means of competition seems very weak-kneed to me.
If any threat of any competition was sufficient justification for owning your own platform, then no one besides Google should be using Android. No one other than Microsoft should be using Windows. The web is everyone’s enemy. Etc. However, we more often than not advise traditional means of competing: reduce costs, pricing strategies, marketing, build brand and product loyalty, differentiation, etc while actually advising some form of dependency on a platform (Windows, Android, Java, the Web) which prevents them from guaranteeing against “winning” and actually places them on equal footing with a number of other competitors.
“Apple’s fundamental philosophy is vertical integration.” Which never achieves market dominance so it was never a threat. Whether or not it was inevitable that Apple would compete with Google is irrelevant if it would only would occur within a minority of the market place (25% at the more optimist end of the fanboy perspective or far less than 10% at the “inevitable, open always win” extreme).
Tom Says: ” A restaurant’s business model is to make money by cooking and serving food to hungry customers.”
I have never seen a restaurant that uses this fictional business model. There may be a few that are the exceptions to prove the rule. If you as McDonalds, they are in the “retail and real estate” business. They make money on the locations they buy, and on toys. The food business just pays the basics until they make money on the toys and real estate.
Most non-fast-food restaurants lose money on food, and make all of their profit from the booze. The food is the loss leader to get you in the door. Even without a liquor license, most restaurants make money off drinks. How much did that fountain Coke cost them?
@Rim F.
“Which never achieves market dominance so it was never a threat.”
Except that Apple was sweeping the market for high end phones. It was obvious from the start that that would be the future of personal computing.
So, for everyone who thinks from the standpoint of Google and other competitors of Apple, creating a competing platform was the most rational action.
@Winter
Doesn’t seem to be doing them much good though, does it?
Nobody answered my question: how would Apple have gained by denying Google the default search engine slot on the iPhone?
The tacit argument here is that it is only by creating Android that Google somehow put pressure on Apple to keep them as the default. But this has never satisfactorily been demonstrated.
The simple fact is that Google’s services get 2/3 of their mobile traffic from iOS (despite Android’s marketshare), and there is no reason to think this would have been any less had Android not have occurred.
The best thing Google could do right now is to shut down Android and write it off as a failure. Then they can start concentrating on improving their core services and making sure that iOS has first class access to them.
@Rim F.
Okay, let’s assume that. So, who would have gained market dominance, and would they have been friendly to Google?
Well, I look at who became the primary competitor to Apple in the computer OS space, and I see Microsoft. And I look at who became the primary competitor to Palm in the PDA OS space, and I see Microsoft. And I see Microsoft with an OS for the phone space . . .
Maybe Maemo/Meego manages to get somewhere. Maybe WebOS gets licensed to everybody (and, now, how much influence can Microsoft apply to PC/printer maker HP to favor MS services over Google?). Or maybe Windows Phone winds up the dominant platform.
@Tom
How much is Microsoft offering to pay Apple to switch to Bing in a world without Android? How much would it hurt an Apple that refused it for, for example, MS Office Mac development to suddenly suffer delays and the product cease being available at academic pricing? If one company has market dominance, that one company can be bribed into excluding Google . . . or alternately use its gatekeeper position to demand tolls from Google to let people use Google.
(Apple wouldn’t let anything detract from user experience in order to make money? Yeah, explain the 30% toll demanded of Amazon for sales through the Kindle app for iOS.)
At which point either Android survives anyway as the product of a coalition of smartphone makers based on Ice Cream Sandwich, who are all pissed off at Google for putting them to the trouble . . . or everybody who isn’t Apple ships Windows Phone 7.5 devices that default to Bing, and MS takes the license fees so gathered and buys the search default on iOS.
@Steven Ehrbar
So why isn’t it happening now? Given that Apple commands apparently 2/3 of the mobile traffic Google receives why is MS not pressuring them?
“Okay, let’s assume that. So, who would have gained market dominance, and would they have been friendly to Google?”
The point is that identifying Apple as the primary threat and primarily guiding Android strategy as an answer to the Apple threat is incorrect. It is exactly my point that Microsoft or other platforms remain the primary direct threat, and Google chose poorly in creating an enemy of Apple.
It’s also striking to me how little faith Google seems to have in its product. Does it really have to rely on enforcing its engine as the default to get users? If so (and I would argue not) what does that say about how they have failed to differentiate?
I think people still regard Google as far and away the best engine, and would probably switch the default even if it were set to Bing or another provider.
Apparently Google doesn’t have such confidence.
“Okay, let’s assume that. So, who would have gained market dominance, and would they have been friendly to Google?”
Moreover, this statement implies that Google does not want or need market dominance, that it is only seeking to guarantee that no others achieve market dominance and blocks them out. I fundamentally disagree with this argument and think it’s plain to see that without market dominance from Android (official, trademarked, complying with the CDD Android), the strategy will never deliver on the profits that Google desires and/or will be no more successful than competing simply by improving their core products.
Has anyone noticed this yet?
http://www.guardian.co.uk/technology/2011/oct/28/iphone-4s-battery-apple-engineers
>Has anyone noticed this yet? http://www.guardian.co.uk/technology/2011/oct/28/iphone-4s-battery-apple-engineers
Oh, yeah. I was figuring I’d write about it my next Smartphone Wars post. “Superior user experience”, my ass!
@winter> Such an offer by Apple would be highly illegal in Europe. Worthy of billions in fines.
you don’t actually know that, of course.
You suppose it, but don’t know it.
@Brian 2> Apple’s fundamental philosophy is vertical integration. They’d have gotten around to doing their own maps with or without Android. And eventually search.
They’re unlikely to build their own search engine, unless you count Siri. I think Apple’s strategy is to move people onto mobile (native) apps, rather than web apps. It’s a “better experience”, and it’s starting to look like the media / journalism industries are paying attention.
That this move “remove(s) the oxygen from the room” for Google’s primary source of revenue is only an unfortunate side-effect, of course.
iOS already lets you select Mobile Safari’s search engine (your choices are currently: Google, Yahoo! or Bing). While the default is ‘Google’, very little stops a motivated iOS device owner from changing to Bing.
> The best thing Google could do right now is to shut down Android and write it off as a failure.
GOOGLE HAS NEVER DONE THIS WITH A PRODUCT!
and even if they did, cyanogenmod would take over development.
@esr
Superior user experience does’t mean the product is perfect, or that it is 100% free of bugs. Bugs happen to all computing products, and this is obviously an example.
The fact that every Android fan jumps on every tiny problem Apple has is evidence of how few problems Apple users experience.
If you want some facts about user experience you only have to consider that Apple consistently tops user satisfaction polls year after year after year, and by a comfortable margin.
@The Adrenalist
Yeah, they’ve got about as much focus as an ADHD kid who missed his morning ritalin.
@Adrenalist
Never heard of “Google Wave” then?
@Tom
Actually i’d point out that the problems iOS has had (and i’m thinking the Antenna, Battery and the Alarm/Appointments issues here) have all been fundamental doozies that (for the most part) a) should have been picked up in testing and b) are fundamental to “the smartphone experience”.
And as I said with the antenna issue, the part that should concern every Apple fan is part a. This SHOULD have been picked up in testing and fixed before an actual release. Either they knew and released anyway, or they didn’t know and their testing regime is incomplete.
That first sentence in reply to Tom is poorly worded.
I’m sure that those three aren’t the only problems iOS has had but the others are either truly insignificant and not worth jumping on, philosophical in nature or I haven’t heard about them. The all references all three of the major ones I know about.
@JonCB
This antenna thing has been so thoroughly debunked that you should be embarrassed about bringing it up. The whole affair was little more than press hysteria.
If it were really true that the antenna had fundamental problems then where are the ongoing complaints? There are millions of iPhone 4 users out there happily using their phones every day with exactly the same antenna that was the subject all the hoopla last year. Have the antennas matured with age perhaps?
Instead of continuing complaints all we hear from iPhone 4 users is a chorus of ‘very satisfied’ to any satisfaction survey that comes along.
And, what’s more, Apple just shipped their new phone with basically the same antenna design!
Eric,
Come on, man. The 4S gets better battery life than ANY top-end Android phone, while still beimg more powerful by a goodly margin, and you think Apple’s claims to superiority are undone by it falling short of Apple’s previously established stringently high standards for battery life?
I know my Samsung Epic would be out of juice if I left it in standby even for 72 hours.
@Tom:
What do you mean “enforce?” If a vendor wants a cut of revenues from google, they need to let google make some money. If the vendor doesn’t care, they don’t even need a search engine. See GB&N Nook for an example.
What do you mean “little faith?” Google has enough faith in their search engine that they open-sourced Android, figuring that they would win…
@Patrick
Ok, ‘enforce’ might be a bit misleading. What I mean is, why does Google feel it is necessary for them to have their search engine as the default? Can’t they compete on the merits of their engine and have confidence that people will choose to use it?
Their whole strategy with Android has been motivated by the fear that their engine will no longer be used if smartphone vendors give pride of place to other ways of getting information.
> Never heard of “Google Wave” then?
BETA-ONLY! NOT A PRODUCT! :-)
> Yeah, they’ve got about as much focus as an ADHD kid who missed his morning ritalin.
SQUIRREL! (http://www.youtube.com/watch?v=SSUXXzN26zg)
> And, what’s more, Apple just shipped their new phone with basically the same antenna design!
To be fair, the iPhone 4S supports antenna switching, so when you short one side with your thumb, the other side still gets signal.
> hey open-sourced Android, figuring that they would win…
Android isn’t really open, unless your a fanboi.
Leaving aside the whole Honeycomb debacle, Google Search, Google Maps, YouTube, Gmail, Talk, Messaging, Google Voice and the Android Market are all proprietary, and you don’t get to brand the phone ‘Android’ unless you put them on the phone.
See also:
http://www.businessinsider.com/google-skyhook-emails-2011-5
They’re unlikely to build their own search engine, unless you count Siri.
Siri’s not a search engine, it’s more like an AI agent that talks to search engines, among other things. Apple doesn’t care if the back end for searching is from Google, Microsoft, Yahoo, Wolfram, Bloomberg, or all of the above. They just write the appropriate adaptor on the server side.
Either they knew and released anyway, or they didn’t know and their testing regime is incomplete.
..or it’s actually a trivial issue that was blown out of proportion by people trolling for page hits. The fact is, the antenna on the iPhone 4 outperforms all previous iPhones. If you’re in a marginal signal area where the iPhone 3 wouldn’t have even been able to connect at all, and you detune the antenna by bridging the gap, you can sometimes make it drop the call.
Where the rubber meets the road is customer satisfaction, and the return rate for the iPhone 4 was about half that of its immediate predecessor, the iPhone 3Gs.
>See also:
http://www.businessinsider.com/google-skyhook-emails-2011-5
I hadn’t seen this before.
“It’s not like it isn’t obvious to the OEMs that we are using compatibility as a club to make them do things we want” – Dan Morrill (Google)
Well, that should not be a surprise to anybody, but it’s nice to have it there in black and white, just in case anybody was still deluded about Google’s intentions.
@Some Guy> Siri’s not a search engine
We are shouting in agreement.
Another juicy titbit:
For background, they are formulating a plan to respond to Moto’s move to use Skyhook Wireless instead of Google’s own wifi positioning system (referred to here as ‘NLP’). GMM is the google maps app on Android.
There is also an email thread between Motorola and Google in which Moto complains that Samsung has shipped phones with a build of the software that Moto was ‘not permitted to ship’ and requests a ‘waiver’ to allow them to do so in order to allow fair competition.
At the end of the thread Motorola is assured by Google that ‘Samsung confirmed they stopped shipping the devices with Skyhook’.
Tim F. et al, Android is not a defense against iOS dominance, rather it is to:
1. Defense against iOS incompetence on maximizing accelerated rollout to 7 billion people.
2. Defense against MS buying Nokia, etc..
etc..
The point is to get smartphones out to the most people and pre-empt any attempts to slow down the process with approaches that are less open.
Is Android perfectly open? No, and that is desirable, because some amount of cathedral development is necessary to accelerate R&D. But it is pen for multiple vendors to build phones on, and iOS is not. If MS was allowed to fill that void with WP7, they either would have continued to fumble and fail as they did for years prior on mobile, or they might succeed to stifle with a closed and predatory business model.
This is interesting.
Extract from an email from Motorola to Goolgle:
“As you know, Motorola has certain contractual obligations that generally require Motorola’s Android devices to be Android Compatible Devices as defined by Google.”
I didn’t realise that the OEMs actually have contracts with Google that *require* them to remain Android Compatible Devices.
I always thought that Google was using the carrot of ACD status to get OEMs to fall into line, but it seems that the OEMs are actually obliged to be compatible. It’s not just a lure, but a requirement.
Hmm. So if Android fans weren’t criticizing, that would be evidence that Apple products are bad?
I can run software of my choice without the vendor telling me I’m a criminal. I can load my own code on my own hardware without writing a check and begging for permission. I can modify the source code and distribute my own OS build. What more do you want?
The specific claim is incorrect (http://source.android.com/faqs.html#how-can-i-get-access-to-the-google-apps-for-android-such-as-maps), and it’s irrelevant regardless. Firefox isn’t non-open because Debian had to rename their version to Iceweasel.
@Brian 2
>Hmm. So if Android fans weren’t criticizing, that would be evidence that Apple products are bad?
No.
Which would fall under the “they knew and released anyway” category.
It’s funny how spin control worked on the Antenna issue. Suddenly a signal strength that would give excellent reception suddenly becomes “marginal”.
Considering that i see a lot of rubber cases around iPhones which is known to “workaround”/”solve” the issue… i hardly think that is a reliable metric.
Suddenly a signal strength that would give excellent reception suddenly becomes “marginal”.
Reading comprehension isn’t your long suit. I’m talking about the antenna’s performance in low-signal areas, which is superior to its predecessors.
“You might think, like Tim F., that lockout-avoidance isn’t a sufficient reason for Android, but it’s clearly a valid contributing reason. After all, the argument that Google is using its power with Android to force a superior Skyhook service off Android phones in itself demonstrates that there’s a real threat of a superior product being locked out of devices in favor of an inferior service if the owner of that service can apply other types of pressure.”
You’re conflating two separate arguments. I argue that “Android Devices” that do not deliver Google Apps and Services is not sufficient motivation for their strategy (i.e. Google is perfectly happy as long as no platform is dominant) because Google needs Android dominance to drive Google Apps and Services and Ads to receive a sufficient return on their investment. I don’t see how this relates to them locking out competition on “Pure Android,” which I know very well they can and will do, despite the rhetoric, because they will most certainly lose out if they don’t close out competition and make their “partners” dependent on their Apps and Services, even when they are, in fact, inferior.
> The specific claim is incorrect (http://source.android.com/faqs.html#how-can-i-get-access-to-the-google-apps-for-android-such-as-maps)
OK, fanboi. From that same link, 4 lines down:
, if a manufacturer wishes to use the Android name with their product, or wants access to Android Market, they must first demonstrate that the device is compatible.
While what I claimed was: “you don’t get to brand the phone ‘Android’ unless you put them on the phone.”
Motorola is part of the inner circle of “most favoured vendors”. They could have signed that contract in order to get early access to things like honeycomb. I doubt that every OEM is contractually obliged to be compatible… if that was the case then Amazon would need to step very carefully or risk getting reamed for contract violation.
I was talking about the Antenna issue.
Superior performance in low-signal areas except when you touch it the wrong way doesn’t strike me as a good compromise but hey it takes all kinds.
Throwing some chum in the water here:
http://www.forbes.com/sites/ericjackson/2011/10/28/why-siri-is-a-google-killer/
@the Adrenalist
“You suppose it, but don’t know it.”
You do not live in the EU, do you? Apple offering Gooogle an incentive to kill of a product:
$COMPANYA offers $COMPANYB $INCENTIVE to divide the market =>
“These fines are not fixed and can extend into millions of Euros, up to a maximum of 10% of the total worldwide turnover of each of the undertakings participating in the infringement, although there may be a decrease in case of cooperation and increase in case of recidivism. Fines of up to 5% of the average daily turnover may also be levied for every day an undertaking fails to comply with Commission requirements.”
http://en.wikipedia.org/wiki/European_Union_competition_law
@All arguing “Google Android is a mistake”
Just silly arguments. Android is the fasted growing platform with the biggest market share. Google makes loads of money.
If this is a failure, how do you define success?
@esr> I doubt it; we already have the August ComScore figures and they show no sign of falling off the long-term Android trendline. We’ll know for sure in a week or so when the September figures become available. We hear these “Android sales are collapsing!” stories around every Apple product launch; they’ve never been true yet.
http://mobile.bloomberg.com/news/2011-10-31/htc-sales-forecast-misses-estimates-on-iphone-competition-1-.html
(esr: “nothing to see here!” htc: “oh sweet mother of god, we’re going down!” chortle)
> You do not live in the EU, do you?
No, but I understand that the intent and basis for those laws is to prevent the formation of cartels, and that fundamentally, a cartel must maximize the profits for its memberS. (emphasis mine)
http://en.wikipedia.org/wiki/Cartel#European_Union
@The Adrenalist
“intent and basis for those laws is to prevent the formation of cartels, and that fundamentally, a cartel must maximize the profits for its member”
Nope. Profit is irrelevant for the law. Any communication between competitors about pricing, market share, or any other aspect of business relevant to competition is strictly forbidden. An offer for consideration if a competing product is pulled from the market would cause a Red Allert all over the offices of the competition comissioner. Even if the offer would reduce profits and prices for all involved.
@Winter Google isn’t making loads of money from Android. They are losing at least 10 billion this year alone on Android.
@Winter (Or to be more kind) Google is “investing” at least 12.5 billion in Android this year alone by purchasing MMI.
@winter
>If this is a failure, how do you define success?
I have an old fashioned view of business, but in my mind it might be nice if they turned even a little profit on the project, rather than being billions in the hole.
@JonCB
Of course, vendors like Amazon who aren’t really building “Android” devices will not have the contractual relationship.
I suppose it is obvious in retrospect that those vendors who are using the Android name and want access to the Android marketplace will have a contract with Google that requires them to remain compatible. I was just surprised to see it written in such stark terms.
@Tim F.
“@Winter Google isn’t making loads of money from Android. They are losing at least 10 billion this year alone on Android.”
Which is a kind of reasoning that is called “Penny wise and pound foolish”. Both Android and Google are flourishing. Without any deep insight in Google’s internal structure and accounting, you know why they would be better off ditching a successful project because you think they lose money on it.
If Apple had used this kind of reasoning, they would have gone bankrupt a long time ago. Actually, it was the kind of bean counting that was used by Sculley in the 1990s.
“Both Android and Google are flourishing. Without any deep insight in Google’s internal structure and accounting, you know why they would be better off ditching a successful project because you think they lose money on it.”
I have no problem saying the platform is flourishing. Just as I have no problem saying it has no sign of making a profit for the company 6 years after purchase and a full 3 years into “flourishing” and that there is no indication that it is better for Google’s mobile revenue than the smaller share of Apple iDevices. However, I have not said they should shut it down. I have suggested that they’d be better off running it as a closed, proprietary project.
>Superior performance in low-signal areas except when you touch it the wrong way doesn’t
>strike me as a good compromise but hey it takes all kinds.
It’s a good thing you don’t own an android then. My LG Optimus loses two bars of signal strength just going from holding it in my hand to dial to holding to me head to talk.
@The Adrenalist
And that claim is wrong. Passing the compatibility test is necessary in order to distribute the proprietary Google apps, but you don’t have to distribute the apps in order to pass the compatibility test and use the Android trademark.
When one Android manufacturer of many has poor sales, the platform is doomed. When the only iOS manufacturer has poor sales, that just proves how eager everyone is for their upcoming products. Amazing how blatantly you guys will write the conclusion first and then fill in the arguments.
@Tim F: a trenchant analysis, provided the fair market value of MMI is $0.
Worked great for Honeycomb.
Tom & Jerry, loss leader definition:
Android in intended to increase or sustain growth of Google’s ad profits by maximizing web growth.
The benefits of a FLOSS platform:
Android smartwatch smackdown!
http://deviceguru.com/android-smartwatch-smackdown/
I think here is one reason so many pro-apple commenters argue Google should either close Android or abandon it all-together: Too many people and consumers benefit from an open Android platform. In the Apple world, any positive externalities would “rightfully” flow back to Apple. The fact that Google does not insist on this is unfair competition.
@ Winter,
Having fun with those straw men you’re constructing? Perfect season for it I know.
All the arguments here about vendor contracts and what the vendors are allowed to do miss one very important piece of the puzzle: google shares ad revenue with handset vendors and carriers.
I’m sorry, but big bad google threatening to withhold a carrot if a vendor doesn’t do what they want is not at all the same as google threatening them with a stick…
Winter,
I can’t help but giggle and think “Brain? Come in, Brain! Follow Uncle Gadget!” We’re now officially in THE FUTURE.
Although… Android 1.6?
The difference is, Apple pours R&D funds into the development of products that actually make them money.
Android needs Google’s money in order to stay afloat. It is a cost center for Google, not a profit center. If Google needs to tighten its belt, how will Android-the-platform stay alive? CyanogenMod? It’ll become something for krelboynes to flash onto their Windows Phone devices, and not much else.
By contrast, iOS is not only making money for Apple, but iOS devices are how Apple makes most of its money.
“@Tim F: a trenchant analysis, provided the fair market value of MMI is $0.”
If and when they sell it off in chunks, I’ll modify my thinking. But for now, I will accept ESR’s thinking (even if I disagree) that it is solely for patents. In which case, yes, it’s value does approach $0. What is clear, is that Google most assuredly has had to make a huge investment in the strategy that has not yet, nor will Google discuss when it will, make a profit.
“Worked great for Honeycomb.”
The colossal failure that is Honeycomb wasn’t due to it being closed.
“I’m sorry, but big bad google threatening to withhold a carrot if a vendor doesn’t do what they want is not at all the same as google threatening them with a stick…”
Yeah, it’s the difference between a beating and being starved.
@Patrick
Google’s rev share is just one more way for them to ensure that they get what they want on the platform. They are aligning their interests with that of their OEMs.
What is now absolutely crystal clear is that Google created Android and the OHA with the intention of having a dominant mobile platform that it could control sufficiently for their advertising to be prominently featured.
Everybody who has previously claimed that Google created and open-sourced Android in order to ‘give up control’ must surely now recant.
“I think here is one reason so many pro-apple commenters argue Google should either close Android or abandon it all-together: Too many people and consumers benefit from an open Android platform. In the Apple world, any positive externalities would “rightfully” flow back to Apple. The fact that Google does not insist on this is unfair competition.”
I could care less who reaps the rewards, Apple or otherwise. I’ve lived in a world where the majority of positive externalities haven’t flowed back to Apple “rightfully” or not for more than 30 years. I think Google’s strategy is a bad one for themselves because it doesn’t appear to yield a benefit proportional to their investment or greater than they can achieve by other means while also empowering the means for competing with them directly with their own strategy.
@Patrick
>big bad google
And, again, you don’t hear any pro-Apple people here saying that Google is ‘bad’ or evil. We are happy to leave the overblown sense of moral significance to you guys.
My extracts from the Google emails aren’t intended to show that Google is being mean towards the OEMs, they are merely intended to demonstrate what I have said all along: that Google has a degree of control over the Android platform that allows it to ensure the prominence of its services. This was the strategy all along.
@Jeff Read
“I can’t help but giggle and think “Brain? Come in, Brain! Follow Uncle Gadget!” We’re now officially in THE FUTURE.
Although… Android 1.6?”
I wrote earlier that Apple apologists are characterized by their inability to see “the big picture”, that is, how the future is being shaped now.
I appreciate it when you want to help me by supplying examples of this short-sightedness. However, you should not exaggerate. People might think you are not serious.
http://www.lifein3d.net/2011/10/30/why-iphone-is-better/
@Winter
As William Gibson is fond of saying: ‘The future is here, it’s just not evenly distributed yet.”
It’s always the case that technological advances start in the lab, spread to industry as high-cost items and gradually (or rapidly) decrease in cost so as to allow more people access.
So what?
I don’t know about you, but I’d rather have the best technology as fast as possible, which is why I like to buy Apple products (among other reasons).
I don’t know why this constitutes an ‘inability to see the big picture’.
@Tom, “Everybody who has previously claimed that Google created and open-sourced Android in order to ‘give up control’ must surely now recant.”
Who claimed that?
The only thing that I have seen claimed by Esr, is that Google gave up control in the sense that Android is open source. Only a communist would expect them to give up control of their ability to drive profits. Afaics, the Magic Cauldron is about how open source can be profitable using derivative business models, and Android is a shining example, as Google’s ad profits continue to grow.
I read between the lines that Apple apologists want to paint Android as equivalent to closed source. Tim F is even wishing they would be forced to make it closed source. Hell will freeze over first. Google is very profitable and profits are growing with no end of growth in sight. Whereas, Apple is approaching the apex of their logistic S curve growth on iPhone.
“Tim F is even wishing they would be forced to make it closed source.”
I’ve said no such thing. I think it would be a more successful strategy, but I could care less what they do. In fact, since I’m not a fan, they might as well proceed along on this present path.
@Tom
“I don’t know why this constitutes an ‘inability to see the big picture’.”
The big picture that is failed to be seen is how the humanity will “compute” in the next decades. What will be relevant in 5-10 years for those billions of people who will use computers and mobile phones? iPhone? Android? Windows Phone?
That is the question Eric tries to address in his blog posts.
But when this question is posed we see comments like yours (I want it now) or things like iPhone can do $INCREDIBLETHING and Apple makes $HUGENUMBER of dollars profit and Android does not. In the mean time, Android is cleaning out the market for the rest of us.
@Shelby
>Who claimed that?
Let me help you:
esr: “Remember that what Google is aiming it is not control but preventing anyone else from controlling.”
esr: “these “people [allegedly] close to the deal” seem to have no comprehension at all of the grand strategy behind Android. Google isn’t chasing control, it’s chasing the prevention of control.”
patrick: “google isn’t hoarding the control. They’re giving it to the free market, to give to the consumers.”
patrick: “By giving away code and releasing source code, google is relinquishing control.”
esr: “Er, and they “control” it how? Android phones are general-purpose devices, customizable by their users. Android is open source. Google is deliberately giving up the possibility of control”
MMI bleeds 800 jobs
http://www.businessweek.com/technology/job-cuts-to-cost-motorola-mobility-31-million-10302011.html
@Winter
In 10 years I doubt we will be using smartphones as we know them today. Likely we will have moved away from the touch interface that we have today towards voice and perhaps other modalities (I think perhaps even neural interfaces will start to come onto the market by this time). Screens will likely have been replaced by displays in our eyeglasses and contact lenses. In short, I think the current iPhone/Android/WP7 debate will look like ancient history.
As to what brand will dominate, Apple, Google or Microsoft, I find this completely impossible to predict. Far too much can happen in a decade to make such specific predictions.
I enjoy thinking about the future and broad scale trends as much as anybody, but I also like to take pleasure in the little things. I don’t see any harm in this.
As for why Apple’s profits keep coming up, it’s only because Eric and others keep saying that Apple is doomed/fallen/going-the-way-of-the-dodo.
@Tom
“brand”
Another example (unneeded, but thanks anyway).
This blog is not about brands, but about platforms and OS development. Who cares about “brands”?
Eric does not seem to care about brands.
@Winter
Jesus Christ. OK, sorry, would you be happier if I had said:
“As to what specific company/platform will dominate, Apple, Google or Microsoft, I find this completely impossible to predict. Far too much can happen in a decade to make such specific predictions.”
@Tom
You still dont get it, dont you?
@Winter
>You still dont get it, dont you?
Get what?
@Tom
Get the big pictue Eric is writing about.
How 4 dexades of Unix, and 3 decades of personal computing and Internet will develop in this decade. What will be important and what not. Who will control what 7 billion people can do with their mobile computer/phone and what they can not do.
@Winter
Yes, I get it, but I just disagree.
I do not think, unlike Eric, that closed-source and closed-ecosystem platforms pose any real-world risk to the future freedom and prosperity of mankind. I’ve never seen any evidence for it.
I also think that open-source platforms mean nothing to most of the 7 billion people you are always going on about. To most people it makes absolutely no difference whether a platform is closed- or -open source. All most people care about is how good the experience of using the device is.
@Tom:
It was always crystal clear that google wanted to work deals with handset vendors and carriers for advertising. It was apparently crystal clear to google that Microsoft and/or Apple could get in the way of that, and that controlling their own destiny was a good idea. It is now clear that if you want revenue sharing from google, you have to jump through a few hoops, although anybody could easily have predicted this.
Anybody can put Android on a Kindle or a Nook or whatever. Where’s the control?
There is no doubt that some of the ecosystem stuff that goes with Android is a carrot that Google uses for partners, but Android itself is about as control-free as it gets.
@Patrick
Nobody denies that, for the most part, the system software itself is open-source, and can be taken and forked. That is clear.
But a mobile platform is much more than just the literal OS software. When we say ‘Android’ we mean the whole platform including the 3rd party apps, the marketplace, the internet services provided by Google, the physical handsets, and (yes, sorry Winter, but it is important) the branding and marketing.
Google has allowed *one part of* the platform to be free, but that doesn’t mean that they have not created a platform that they still exercise a great deal of control over.
When you and Eric talk about the Android platform and ecosystem that is constantly gaining marketshare, that is a platform that was created by Google with the intent that they would control it to ensure a continued audience for their advertising. They didn’t do it to ‘relinquish control’ they did it to give themselves control.
@esr do you stand by your iPhone 4S comment “Me-too voice recognition features? Really, Apple? Is this the best you can do?” ??
All convenience function with fairly easy workarounds. Google is using a light hand indeed here.
Ah, you mean the “royal we.” You certainly don’t speak for me in this matter.
OK, now you’re being ridiculous. Google shouldn’t have control over their own branding?
@Patrick
Of course they should! I’m not saying that Google is wrong to do any of this, I’m just saying you guys should face up to reality.
You said it yourself: it’s their branding. It’s their marketplace. To a substantial extent it is their platform.
It’s really almost unbelievable that you continue to contest the point. Google has created a mobile platform over which they exert a degree of control sufficient to ensure that its advertising is protected.
This is blindingly obvious and uncontroversial to almost everybody except, apparently, you and Eric.
We agree why they did it. We agree that it’s working. We agree that they have created an ecosystem they control (which might or might not be replicated by Amazon, B&N, etc.) We apparently disagree on what constitutes “Android.”
We also disagree on your statements “It’s also striking to me how little faith Google seems to have in its product. Does it really have to rely on enforcing its engine as the default to get users?” and “Can’t they compete on the merits of their engine and have confidence that people will choose to use it?”
Those are silly at best. If google didn’t have any faith in its search engine and wanted the ad revenue from Android, they’d choose bing, to beef up the Android experience. But google does have faith that they have the best search engine, and wants Android users to have a good search experience, so they choose it.
Maybe you don’t think those are silly, so let me ask you:
a) Is the out of the box user experience enhanced by having a default engine for search?
If so,
b) Is there any good reason for google to make that default be one of its competitors’ search engines?
@Life as we know it:
On the one hand, HTC does seem to be saying in that article that competition is stiff. On the other hand, you were using that article to imply that Eric is wrong about the trendline. Looks like that theory will have to wait until we have data from another quarter:
http://www.bgr.com/2011/10/31/htc-takes-top-smartphone-spot-in-u-s-according-to-q3-data-from-canalys/
That’s right, ladies and gentlemen — in the US last quarter, HTC shipped more than Samsung, who shipped more than Apple…
@Patrick
>Those are silly at best. If google didn’t have any faith in its search engine and wanted the ad revenue from Android, they’d choose bing, to beef up the Android experience.
The whole *point* of creating Android was that it would use the Google engine! Google doesn’t make any money on Android directly.
>a) Is the out of the box user experience enhanced by having a default engine for search?
Yes.
>b) Is there any good reason for google to make that default be one of its competitors’ search engines?
No.
But, I’m not really sure what you are getting at here.
Original report is here:
http://www.canalys.com/newsroom/htc-takes-lead-us-smart-phone-market
I note for Tim F.’s edification that Apple is already apparently down to around 14% of the global smartphone market… (Yeah, yeah, I know, new phone, blah, blah, blah…)
> But, I’m not really sure what you are getting at here.
You’re the one who claims that google making their search engine the default shows some sort of insecurity or something.
@Patrick
>That’s right, ladies and gentlemen — in the US last quarter, HTC shipped more than Samsung, who shipped more than Apple…
“shipment estimates”
Doubly useless.
@Patrick
>You’re the one who claims that google making their search engine the default shows some sort of insecurity or something.
It’s not that, so much as the whole Android project. Android is based on Google’s fear that people wouldn’t use their services on smartphones.
Turns out that they should have just had confidence in their services, because people *are* using their services on smartphones, even on those phones whose vendors are not being strong-armed by Google.
Unfortunately for them we are now getting to the point where interesting new ways of getting information (e.g. Siri, Wolfram-Alpha) actually *are* going to be a threat. But during the time that Google could have been pushing this field forward themselves (‘organise the world’s information’) they have been distracted by a project that does not serve their core mission.
@Tom:
No, canalys is usually pretty accurate about what shipped. Sometimes off perhaps an aggregate of 10% or so. But if you want to see really useless, take a look at any market forecast…
@Tom:
> Android is based on Google’s fear that people wouldn’t use their services on smartphones.
No, Android was based on Google’s fear that Microsoft and Apple would lock up the market and not LET people use their services on smartphones. That’s a huge difference, and it’s shameful of you to keep glossing over it.
@Patrick
>if you want to see really useless, take a look at any market forecast…
There we agree!
>No, Android was based on Google’s fear that Microsoft and Apple would lock up the market and not LET people use their services on smartphones. That’s a huge difference, and it’s shameful of you to keep glossing over it.
Your argument makes some sense from for Microsoft, but absolutely no sense for Apple. Before it became clear to Apple that Google was entering the smartphone space Eric Schmidt was on the Apple board. He and Steve Jobs had appeared together at events. Apple was heavily integrating Google services. Google had every reason to expect that Google services would be extremely well represented in iOS. And even after the split, after Apple learned of Google’s plans to compete against them, their services remain extremely well represented and drive 2/3 of Google’s mobile traffic.
But there is a broader argument here. People want to use Google services and any platform that forbids the use of these services will be considerably the poorer for it. The market will respond appropriately and either force the vendor to change their policy or push them out of the market.
Google could have further ensured this by continuing to innovate their information services to make them even more essential and useful. But they didn’t do that. Instead they chose to create a new platform (something outside their core mission) in order that they could control it.
To me that speaks of a lack of confidence (misplaced in my view) that the market would not demand access to their services.
@Tom
>The whole *point* of creating Android was that it would use the Google engine! Google doesn’t make any money on Android directly.
>It’s not that, so much as the whole Android project. Android is based on Google’s fear that people wouldn’t use their services on smartphones.
Then why did I have to void the warranty on my Fascinate to remove Bing as the default search engine so I could use Google?
@Kevin Snyder
I am not really sure what a ‘Fascinate’ is, but I assume it is some kind of WP7 device?
Let’s just see how well they do in the market.
@Tom:
Last I heard Google still employed a few engineers to work on search and mail.
Your proposal is that Google should be content to be Apple’s vassal while they and (presumably) Microsoft split the market. That’s a very fragile position with almost no growth opportunity and many ways to fall.
Well, that’s a fundamental disagreement. I can think of several plausible threats enabled when almost everyone is on a locked-down platform:
– Direct abuse by government, e.g. banning anonymity, mandating tracking software or the camera kill switch.
– Abuse by special interests via purchased legislation. A key reason that the CBDTPA failed was that it would have been utterly impossible to enforce. But if users can only run software approved by a central authority, it becomes feasible.
– Abuse by the platform vendor(s). If Microsoft in the 90s had the kind of power over the Windows ecosystem that Apple has over iOS, the web as we know it wouldn’t exist because they would have prohibited Mosaic and Netscape.
@Tom
Sorry, Samsung Fascinate, the Verizon version of the original Galaxy S line. Stock Verizon ROM has Bing as the default search engine. Possible to use Google via the browser but the hardware search button is locked to Bing. PITA. Only way around it is to root the device and do some other stuff, thus technically voiding the warranty. Got mine in October of 2010, still have it. Good hardware. I broke my warranty within ten hours of owning the phone to get rid of a search engine I didn’t want.
Obviously some money changed hands between MS and Verizon to make that happen, but it is an example of Google not requiring their services on the device.
And all it would take is the belief within Apple that they could make a better vertical integration product with their own search engine(say for example they bought yahoo) and suddenly Apple can foreclose all of Google’s options. They’re locked out with no chance of reply. Even if 100% of Google’s iOS revenue disappears tomorrow (and it won’t because some of that is based on developer choice to incorporate google ads into their app), Android at least gives them a point of reply.
Saying that no-one would expect Apple to do this is massively naive in the extreme. It is PRECISELY the sort of move that Apple is famous for. Siri is about half a step away from it.
I’m with Tom on Apple vs. Google: in the years prior to the iPhone, there was a strong sense that Google and Apple were allies against Microsoft. But then Google decided to get have a smartphone OS and one or two other things that Apple (not unreasonably) saw as aimed at them. Apple didn’t get into the search engine biz, and only got into online ads after the split.
@Tom, I agree with you on a couple of points, which I interpreted as:
– One could fathom that Google’s vested interest could in the future potentially shift naturally from spreading open-source, to hindering it.
– Technological innovation is accelerating, and Android’s dominance of smartphones could possibly be supplanted by some new technologies in the future. (you may even be hoping Apple will pull another proprietary rabbit out their hat, e.g. Siri).
However, I disagree with you on the more fundamental points, which also eliminate the above two points:
– the push of open source is not ideological (i.e. religiously motivated), nor it is slowing down or under any threat of a resurgence of closed-source, because it is an economic phenomenon, i.e. it is economically more efficient and it is just recently that the ubiquitous networked computing has enabled it to come to fruition. It will completely wipe-out closed source within 10 years, including any vested interests (inertia) of Google and any others that might try to hinder it. Btw, I used to be a closed-source proponent, before I learned more about economics, even though I aced “Economics” in college (which is correct, but lacks depth of conceptual understanding).
– closed-source is an economic and civil liberties threat as Brian 2 noted, and the more fundamental economic reason and threat is that centralized decisions have lower degrees-of-freedom, i.e their dynamic fitness is less adaptable. This is essentially a test of the economic superiority between statism (collectivism) and anarchism (freedom) politics, which imho might explain why many Apple proponents appear to be either liberals or conservatives (both are statist).
– you continue to miss the point that I made in a prior comment, that Google didn’t just do it to prevent disruption of their services, but also to accelerate the spread of smartphones and the web, by increasing the degrees-of-freedom and fitness to the market, in a way that Apple, MS, and Nokia (on smartphones) were failing to do. Now we see the Taiwanese and the Koreans competing head-to-head (HTC vs. Samsung) in smartphones thanks to Android. Without Android, they might still be mired in hanging on to proprietary markets they have in dumb phones and fumbling their way around as MS and Nokia had been doing. Btw, I see that Acer, LG, etc. are now shipping smartphones here. Massive onslaught this Xmas of Android in the developing world. Not an iPhone any where in retail here.
Shelby, if you look at Apple and all you see is “closed source,” you are missing a lot. Excellent design is not an “economic and civil liberties threat,” for heaven’s sake, but it is something difficult or impossible for Android to achieve at anywhere near the level Apple can achieve it. Apple also has profit, economies of scale, hardware/software integration, and lack of fragmentation on their side.
I’ve never been one to say that Android will “fail,” but there are many reasons that Apple will continue to succeed with iOS in ways that it never did with Mac OS or OS X.
Looks like Sprint doubled down on a winner:
http://gigaom.com/apple/thanks-to-iphone-4s-ios-market-share-rockets-in-october/?utm_source=social&utm_medium=twitter&utm_campaign=gigaom
>Looks like Sprint doubled down on a winner:
I’ll believe it when I see it in the comScore report, and not sooner. Breathless stories like this are often built on seriously cooked numbers.
Louis4,
I love how that article points out that the illusion of Android dominance is due to the snipped-vs-sold distinction in volume reporting that some of us have harped about.
Apple is evil, and the iOS ecosystem per iciously so, but the open source crowd is going to have to wise up and learn that evil will always triumph because good is dumb.
>Looks like Sprint doubled down on a winner:
Was the point not that Sprint actually lost money on every iPhone they “sold”?
But if Apple can keep up this rising market share for the comming 12 months, then they will obviously outgrow Android. Just like they did last year. And the year before that. ;-)
@esr:
The numbers aren’t cooked. They’re just… wrong.
For a start, there is already empirical evidence available that the average iOS user uses more bandwidth than the average Android user. And if you look at how the statistics are gathered, there are a few things that might raise your eyebrows. In particular, this stands out:
Oooh! Shiny!
@Jeff Read:
The most relevant passages from the article are these:
Given that the source data doesn’t break out handsets separately than tablets, and given that the target audience for the data is advertisers, this all makes sense.
Let’s do a little thought experiment here. You (I think) and certainly others have been harping about sold vs shipped here for at least three quarters. What happened to the excess that was shipped in the two preceding quarters? Do you think the vendors just decided to landfill it and start over? Or did it eventually make it into the hands of consumers? Would the companies accepting shipments from, e.g. Samsung, say “well last quarter’s stuff didn’t sell, but sure, send me some more on over?” or would they slow down their orders?
Now, since Apple actually sold fewer handsets last quarter than the quarter before, while Samsung apparently shipped 35% more smartphone handsets than they did the quarter before, even if there is some sort of shipped vs. sold major discrepancy, can you really still not see that the trend so far has been that Android is kicking some serious Apple butt in shipments?
@Patrick
Can you estimate the cummulative losses of dumping these “shipped only” phones and tablts in a landfill or shredding them?
Because that is an excercise never done by these “oh they are only shipped”. It is not as if they are copying digital Windows licenses.
> I am not really sure what a ‘Fascinate’ is, but I assume it is some kind of WP7 device?
No, it’s an Android phone, built by Samsung’s Galaxy S, available on Verizon in the US. Microsoft paid Verizon between $550M and $650M over 5 years for locking Bing into some of their phones.
No True Scotsman^WAndroid fanboy would buy one, but the public sees a cheap pretty phone and goes all, “gimme”
Revealing your performance is a powerful signal. Not only of your own reliability but if you consider your products as platforms it signals to the whole ecosystem that your platforms are sound.
Case in point, Amazon not revealing any data about Kindle sales does not signal to me anything about whether I should consider their platform for publishing an eBook. How many readers am I supposed to imagine exist?
Same question to Samsung about shipped .vs sold? Why should I develop for their platform if I don’t know how many users they have.
This is most peculiar with respect to Google. Why don’t they have a web site showing real time stats on how many Android users, where they are and so on?
Do these companies really think they have an *advantage* in not letting anybody know what their performance is? I’ve never met anyone who admires a company that keeps its performance under wraps.
Google seems to like the PR potential of releasing activation numbers intermittently. They’re releasing activation numbers roughly every quarter, which is after all how often we get numbers from Apple. Google certainly could give us more data, but then Apple could also give us specific model sales numbers broken down by market – they could give us the number of active devices (which they’d know due to the push notification service which all ios devices connect to).
There is always more data to give, and obviously that data has value – to partners, to competitors and to investors. For Samsung neither partners nor investors are relevant, so their main consideration is competitors.
@The Adrenalist
“Microsoft paid Verizon between $550M and $650M over 5 years for locking Bing into some of their phones.”
One example of how MS squandered $6B on Bing. It is telling that MS did not use it to make Bing a better search engine.
@The Adrenalist
“Why don’t they have a web site showing real time stats on how many Android users, where they are and so on? ”
Brilliant!
Real time stats on sales are what you competitor needs to kill your marketing. It is the same reason Apple fights tooth and nail to keep upcoming models secret.
@The Adrenalist:
I also think that sort of data is good for consumers. But obviously most consumers don’t care. That’s what allows the companies to be more secretive.
@patrick
“All convenience function with fairly easy workarounds. Google is using a light hand indeed here.”
Skyhook would disagree.
@winter
“Can you estimate the cummulative losses of dumping these “shipped only” phones and tablts in a landfill or shredding them?”
Ask Motorola given they only sold 100K Xooms in the last qtr. 800K total…which pre-iPad was probably decent enough numbers.
http://www.latimes.com/business/la-fi-motorola-jobs-20111101,0,7185936.story
I’m not a huge fan of Enderle but we agree that the numbers game that folks put so much credence to is suspect.
http://www.itbusinessedge.com/cm/blogs/enderle/apple-vs-samsung-why-you-can-t-trust-market-share-numbers/?cs=48960
Eric favors the comscore numbers because they support his assertions. All other numbers are therefore suspect. The fact is all these numbers have to be taken with a large grain of salt. Less so for Apple since they provide their numbers in their financials where companies are required to lie less or at least more artfully. Even there Apple never breaks stuff down in sufficient detail to make very specific statements.
Comscore’s methodology is as problematic as NetApplications:
http://www.comscore.com/About_comScore/Methodology
“Central to most comScore services is the comScore panel, the largest continuously measured consumer panel of its kind. With approximately 2 million worldwide consumers under continuous measurement, the comScore panel utilizes a sophisticated methodology that is designed to accurately measure people and their behavior in the digital environment.”
So the core data is based on 2M users and then adjusted by demographics and traffic monitoring. Traffic monitoring with it’s own share of controversy:
“Comscore CMO Linda Abraham published a detailed response to the outrage about Comscore’s decision to force sites to become Comscore customers if they want Comscore to report their traffic properly.
Since for many sites Comscore’s new methodology results in a boost in traffic, this amounts to blackmail: Pay Comscore or Comscore will under-report your traffic.”
http://articles.businessinsider.com/2010-01-24/tech/30046614_1_methodology-sites-traffic
Their smartphone methodology is somewhat more opaque.
“MobiLens data is derived from an intelligent online survey of a nationally representative sample of mobile subscribers age 13 and above. Data on mobile phone usage refers to a respondent’s primary mobile phone and does not include data related to a respondent’s secondary device.”
Yah, Eric’s numbers are based on online surveys.
Found on-line, thought it was worth sharing here.
(http://www.reddit.com/r/Android/comments/lweir/this_subreddit_largely_misunderstands_what/)
Google is iterating as quickly as possible in attempt to build the software that is the foundation of computing for decades to come. They don’t care about any single user (hence your lack of software updates and your lack of customer support) as those users do not matter in the long term. Their view is: if they can build the best software and be the first to market with it, repeatedly, then they will gain market share (done) and be in a position to set the trends for the future rather than follow them.
Each time there’s an update that you don’t get, it’s because you (a massive nerd, one in a million, literally) don’t matter. Each time they have a product launch that doesn’t attract half the world, it’s because they’re not trying to. Each time your gmail gets deleted, it’s not because Google accidentally sucks at customer service but rather that a mistake they made probably isn’t worth correcting.
ALL OF THIS IS INTENTIONAL.
@nigel
Enderly meets with MS to discuss marketing. He has never been right. If he doubts Samsung and supports Apple, I am concinced Apple is in dire straights.
@esr “I’ll believe it when I see it in the comScore report, and not sooner. Breathless stories like this are often built on seriously cooked numbers.”
How would US market share data reflect the success or failure of a Sprint business strategy? Wouldn’t it make sense to look at Sprint’s financial results and comments for the next couple of years rather than OS share which is completely irrelevant to them?
@The Adrenalist
That strategy is copied from MS. Being first to market beats being the best every time. MS are the dying example: As long as they were first to market, they were king. When they trail, they lose, big.
> As long as they were first to market, they were king.
Not saying I agree with this, but… wasn’t Google second, or even third?
@nigel:
>> Google is using a light hand indeed here.
> Skyhook would disagree.
Yes, I’m sure they would. As will anybody whose business model involves simultaneously (a) selling services to companies that are getting money from google; and (b) siphoning off data that google would otherwise see.
@The Adrenalist
Huh, Google did not exist when MS were king.
@winter enderle isn’t important to the discussion. That all of these numbers are suspect is.
The fact that the eric market share numbers are derived from user surveys is interesting given the reliance he places on them as the true gospel.
That said, they probably are reasonable approximations of the truth. The problem with eric’s “analysis” is that he insists that Apple is about to crash and burn…and only relies on this as the primary data point for his analysis.
So lets all agree that Android has reached 50% US smartphone marketshare. What’s next?
esr: Apple will suffer from massive disruption from below and crash and burn because 65% of revenue is from the iPhone.
Reality: Apple will ship more iPhones than last year. And in 2012 the expectation is pretty much the same as before. Apple will ship far more iPhones than in 2011.
Samsung reported profits of $2.2 billion from smartphone sales last quarter, 60% of the electronic giant’s total.
iPhone sales last quarter, while less than expected, generated revenues of nearly $11 billion (39% of Apple’s total, not 65%).
Assuming, conservatively, 60% profit margins on each iPhone sold, that’s an estimated profit of $6.5 billion, leaving No. 1 Samsung in the dust.
Only in the distorted reality of a fanboy does selling more of a thing while making less profit equate to success.
My apologies it’s not 65%…this is what esr wrote in early August:
“Apple is now relying on smartphones for 68% of revenue, so they’d be very vulnerable to an actual drop in marketshare.”
Which was factually wrong when he wrote it. The article he quoted stated that iPhones were 47% and iPads 21%. Which he combined to make 68% iPhones and 21% iPads. But we’ve all made math errors. What comes next is the doozy.
“I’ve taken a lot of flak for saying the company looks like a late-stage sustainer with a principal product line about to experience disruptive collapse, but this is yet another straw in the wind. If next month’s figures show an actual share drop, expect it to be self-reinforcing and get the hell out of Apple stock.”
Yah, not so much.
He repeats this theme later in the month:
“Foremost among Apple’s problems is Android. 68% of the company’s profits come from its smartphone business, and another 21% from the iPad, leaving only 11% from other sources…
“I’ve said before that I think Apple looks just like sustaining incumbents often do just before they undergo catastrophic disruption from below and their market share falls off a cliff…
The near-term threat of an Apple market-share collapse to the 10% range or even lower is, in my judgment, quite significant – and comScore’s latest figures whisper that we may have reached a tipping point this month.”
Again, not so much.
Of course this started back in April and a possibility of this collapse within 90 days:
“I think we’re looking at the end stage of a successful technology disruption on the classic pattern. The question is no longer whether Android can be stopped, but when Apple’s market share will fall off a cliff. I think that could easily happen as soon as the next 90 days; one of the patterns in technology disruptions is that collapse often follows the victim’s best quarter ever.”
But here we are in November. And again, not so much.
If he only wrote that Android is doing awesome I doubt anyone would disagree (or notice for that matter). It is doing awesome. But so is iOS. Given that the two don’t really compete except at the upper end of the smartphone market both can do extremely well at the same time. But he as to go on about Apple being disrupted from below and folks need to be selling their Apple stock. With a P/E ratio of 14. Who’s gaining worldwide phone market share. Who’s selling more and more iPhones every year.
Nope nope…Apple is doomed. Disruption from below is imminent.
Yah, okay. I suppose it drives hits when he gets fireballed. But now the metric that he’s been waiting for (50% US smartphone marketshare) is upon us…so I’d presume that Apple’s share will start falling off a cliff and will have 10% or less marketshare by New Years.
Given I was standing in line for a new iPhone 4S this past Sunday I’m thinking maybe…not so much.
Why MS is doomed (falling apart in the coming year, I expect): They simply cannot handle change in business.
Whatever bad things I can say about Steve Jobs and Apple, they can handle changes in course.
The inside story of how Microsoft killed its Courier tablet
http://news.cnet.com/8301-10805_3-20128013-75/the-inside-story-of-how-microsoft-killed-its-courier-tablet/
This is a story much like the demise of Bing:
Birth and Death of Microsoft Bing
http://myprasanna.posterous.com/birth-and-death-of-microsoft-bing
I give Ballmer (much) less than a year. MS will start to shed failing division the coming year. Bing and Phone are high on the list.
> Apple’s share will start falling off a cliff and will have 10% or less marketshare by New Years.
Patrick Maupin’s prior comment estimated 14% global share for shipments (not sales) of iPhone, which I deduce is based on the 17.1 million Q3 iPhone shipments.
steg’s prior comment provided a link that said there were 55 million activations for Android in Q3, and 25 million for iPhone. If activations can be roughly correlated with shipments for proportional effects, then Android global share must be only 31%.
But how can that make sense? Per the linked article, Nokia has 14%, so who has the other 41% global share?
One of the linked articles said about RIM, “US market share slipped from 24% in Q3 2010 to just 9% in Q3 2011”, yet Esr’s ComScore data showed RIM at 35% at end of Q3 2010 and 20% in August 2011. Even extrapolating the highest US share for RIM to the world, doesn’t reach 41%.
Some searching reveals that it may be possible to skip activation on Android, and a Google spokesperson confirms that they don’t count devices that don’t use Google services:
Is it possible that Android sales are thus double the number of activations being reported by Google? Is there any other explanation for global share discrepancy?
Are Google services even available in China? One of the linked articles mentions “explosive” share increase for low-end Android phones in China, apparently taking Nokia’s dumbphone upgrades away. My thought is that these users might not bother to activate Google services that require a login. I don’t know anybody in Philippines that uses Google for anything that requires a login (i.e. only web search, web maps, etc), they use Facebook, YM, and the web in general. People here hate to remember passwords. They lose access to their Facebook account after a couple of weeks of non-use.
replicated typo, “other 45% share?”
Rats! I was correct the first time, there was no typo. 14 + 14 + 31 = 59%. Who has the other 41% global share?
@shelby you’re adding numbers which vary widely from different sources for different things (global vs US smartphone share, smartphone vs ecosystem, etc).
@Nigel, I used only global (not USA) shipments share for iPhone’s 14% and Nokia’s 14%. See the canalys and bbc links I provided.
Then I used the ratio of global (not USA) activations for Android and iPhone to project the 31%. So the 31% is dubious, which I noted in my prior comment. Samsung alone globally shipped 27 of 120 million, i.e. 23%. So it seems unlikely that the 31% is too low.
My point is that the activations ratio is inadequate (31% is too low) to explain where the rest of the market share is going.
I understand that activations is ecosystem and shipments is smartphone, but my point is that what is the ratio of global shipments for Android compared to iPhone? Afaics, it must be much higher than the activations ratio, because I can’t see where the other 120 million global smartphone shipments would be coming from on a share percentage basis. Seems the Google activations are about half of Android shipments.
Can anyone offer another possibility for the discrepancy?
P.S. The link I provided to steg’s comment, did not mention the 25 million iOS activations for Q3, here the link that does.
@Shelby:
I think the issue is that your “iPhone activation numbers” are probably not right. Probably are iOS activation numbers, including tablets and iPods.
The Canalys report says that 120 million phones were sold in the quarter. If there were 60 million Android activations, that’s roughly 50% of the market, compared to Apple’s 17.1 million units at roughly 14% of the market.
Shelby, the key is to try to use numbers from one source because other than the numbers supplied in financials everything else is guesswork. If you do what you did you can end up with pretty bizarre numbers.
For Q32011 from financials you see that globally:
Apple shipped 17.1M smart phones
Nokia shipped 16.8M smart phones
HTC shipped 13.2M smart phones
Unfortunately Samsung no longer provides smartphone shipments in their financials.
Analysts have said that:
RIM shipped 11.8M smart phones (mentioned as part of how HTC overtook RIM by 1.4M units)
The problem being that most of these places will tell you number for the top 5, but anything more they want $$$ for.
So for BB, iOS and Symbian you get a combined total of 45.7M out of 117M…about 39%.
Everything else is Android, WP, Linux (Maemo, Meego, etc), Bada, etc. I’d assume that when considering Android vs Other, the Other makes up less than 10%. Call it 6%.
So you can reasonably guess that Android’s global smartphone market share is on the order of 55%. Given that everyone’s estimate for 2Q 2011 is around 43% (Gartner) to 48% (Canalys) that sounds about right given the share that Symbian and BB have been losing every qtr. This is enhanced a bit from Apple not shipping the iPhone 4S in 2Q. 4Q will probably see Apple around 18-20% and Android not grow as much as in 3Q if Apple does ship the 28M smartphones some analysts thinks it will.
Google activation numbers don’t equate to smartphone shipments. But 55M activations vs 64.4M shipped is pretty close. There’s around 10M random Android handsets shipping out there not hitting Google servers…guessing mostly Chinese ones.
We’ll call tablet activations a wash in all the approximations. A few million here and there wrong on both sides of the equation probably cancel out. :)
Note: the reason that it’s unfortunate that Samsung doesn’t report is that we don’t have solid numbers for Bada anymore.
I’m going to guess that 3Q2011 Android global smartphone market share is between 53% and 58% probably toward the lower end.
I’m going to guess that 4Q2011 Android global smartphone market share will be between 55% and 60% toward the high end.
The reason for the conservative estimates on Android 4Q growth is more WP7 sales in 4Q than in 3Q (that’s not hard I wouldn’t think) and more iOS sales in 4Q and 3Q (which pretty much everyone is predicting). Meaning Apple goes up a bit and Nokia share losses slows down in the EU where they’re actually shipping WP7 phones and are traditionally strong anyway.
>I’m going to guess that 3Q2011 Android global smartphone market share is between 53% and 58% probably toward the lower end.
I’m going to guess that 4Q2011 Android global smartphone market share will be between 55% and 60% toward the high end.
I think these are reasonable, if slightly conservative, extrapolations. If they go wrong, they’ll go wrong by being low – but the plausible top of the range is no more than 10-12% above these.
@winter that courier was cancelled is indicative of poor vision by Gates and Ballmer. To extrapolate this to Microsoft is going to fall apart next year comes under the heading of “wishful thinking”.
Get it right, keep it straight. At all times since the mid-70s, Apple was changing the course. Before the iPhone, smartphones were horrid, Windows Phone- or Palm-based devices nobody wanted to actually use. People booked face from their PC. That’s what it was there for.
Now, between the iPhone and the iPad, PC’s are all but obsolete.
And Microsoft is killing anything that doesn’t support their hegemony on this old, fading technology. Funny!
@Patrick Maupin, of course I should have noticed that 25 million activations is higher than 17 million shipped. Taking iOS activations to be 17 millions, then 14% x 55/17 = 45%.
@Nigel, afair those numbers for RIM/BB (and HTC) were USA only. Android 10 million non-activated seems to be about the ratio I would expect for now, I hadn’t noticed that 64.4 million number. My intuitive expectation is it will become a larger percentage of shipments.
Global Share Guesstimate
—————-
Android: 55%
iPhone: 14%
Nokia: 14%
RIM/BB: 9%
Other: 6%
————
98%
Thanks.
@Jeff Read
“Get it right, keep it straight.”
How do you interpret “[Apple] can handle changes in course.”?
@Nigel
“@winter that courier was cancelled is indicative of poor vision by Gates and Ballmer. To extrapolate this to Microsoft is going to fall apart next year comes under the heading of “wishful thinking”.”
Courier and Bing and… Just another symptom of their upcoming demise. I expected MS leadership to fail on any new development. And they never cease to oblige me. Just another example in a long list.
I know that MS are failing because they were hemorrhaging talent a few years back and they never were able to stop the outflow. And whenever they succeed in attracting new talent, they chase it away within a short time.
Without good developers but with lousy leadership, what can MS do but fail?
Winter,
You wrote that “Apple can handle changes in course” as if they were reacting to paradigm shifts in the business. That would be great in its own right, well beyond the capability of many companies. But Apple isn’t reacting to paradigm shifts; it’s causing them. Just about every major disruption in personal computing originated at Apple. That makes them one of a kind.
Does anybody know anyone that actually gives two fucks about Bada?
Indeed. Microsoft handles changes in course by forcing their way into a #1 spot, either by sweetening the pot through payouts (“money hats”) to key strategic partners or acquiring them outright. Underestimating them is almost as foolhardy as underestimating Apple, though with Ballmer in the lead they’ve lost some of that Gatesian cunning.
Nigel, here are Apple’s revenue numbers for iPhone and iPad, pulled from their 10Qs:
Total revenue 4Q11 (ended September 24, 2011) ($M): 28,270
iPhone and Related Products and Services ($M): 10,980 = 38.84%
iPad and Related Products and Services ($M): 6,868 = 24.29%
Total revenue 3Q11 (ended June 25, 2011) ($M): 28,571
iPhone and Related Products and Services ($M): 13,311 = 46.59%
iPad and Related Products and Services ($M): 6,046 = 21.16%
Total revenue 2Q11 (ended March 26, 2011) ($M): 24,667
iPhone and Related Products and Services ($M): 12,298 = 49.85%
iPad and Related Products and Services ($M): 2,836 = 11.50%
>Q will probably see Apple around 18-20% and Android not grow as much as in 3Q if Apple does ship the 28M smartphones some analysts thinks it will.
Apple estimates total revenues for 1Q12 at $37 billion, representing a 30% jump over the just reported quarter. If the Q4 percentages hold (and remembering that Q4 was a down quarter): this yields 22.345M iPhones. I don’t see where 28M is likely.
Also, and this is important:
This is about 1/2 of the current Android activation rate. But if that’s true, and Samsung is larger than Apple, then Samsung is selling over 50% of all Android phones.
Does anyone here really think that Samsung sells 50% of all Android phones?
The only fair conclusion is that there is a lot of fiction^Wguessing out there in the published numbers.
More data leading one to question the conclusion
http://arstechnica.com/microsoft/news/2011/11/the-end-of-an-era-internet-explorer-drops-below-50-percent-of-web-usage.ars
Quoting: “Safari’s long-term dominance in mobile is clear. Also clear is that Android’s sales growth isn’t at all reflected in its Web usage.”
> I think these are reasonable, if slightly conservative, extrapolations.
I think, given all the above, (which you have yet to acknowledge or address), they’re troll-bait.
FWIW I wasn’t deliberately trolling Eric in those posts. :)
I have never disagreed that Android was doing well, was easily going to hit 50% share, etc. I find 90% share possible but not inevitable. 70% seems probable. The things I think that’ll keep Android from 90% are WP7 and linux.
Which is surprising to me given I had figured that Android decisively killed Linux phones. It seems that the lighter weight linux phones are still better for low end smart phones than Android. Given where Samsung sits Bada could become quite relevant to the discussion. Some folks believe they shipped 4.5M Bada units in 2Q. How many they shipped 3Q should be interesting to conjecture about.
I can envision Samsung eventually shipping more bada phones than RIM ships anything. Shame about WebOS. It’s too bad for both RIM and WebOS that they screwed the bid and let HP win.
@life That 28M number was pulled out from thin air by some analyst. I forget which. Then they nearly doubled that imaginary number for their most recent estimate.
One thing to note in your revenue analysis is that iPhone ASPs will drop due to 3GS and 4 price drops. Given that AT&T is sold out of the 3GS and the 4 appears to also be selling well too I would assume that demand and production is high enough to push the average sale price down overall. Depending on how you fudge that ratio I would guess that 28M is plausible. Especially if you assume that Apple is still guiding conservatively. Which will be remarkable given the $37B guidance.
Yes, there are a lot of questionable numbers out there. While I wasn’t trolling Eric I don’t mind tweaking him about comscore. But he consistently uses the same metric so it’s probably off by about the same amount every time. Good enough.
The only numbers that I believe have some vague resemblance to reality are those produced by Google, Samsung, Apple, RIM, MS, etc in their financials. Even then, they are obviously going to get spun to put the company in the best light.
—
Folks will notice I confusingly mix calendar and fiscal qtrs by not specifying which. I think most folks will figure it out through context but mostly I mean CY. Except when I’m not. :)
@Jeff Read
Microsoft has tried all their tricks to get Bing at #1. They wasted $6B in the process and are still way below Google. The same story for WP7.
What is different now is that they seem to be unable to secure a controlling position where they can consolidate their 80% margin. Without this margin their shareholders have no reason to keep money in MS coffers. So they will call for a payout.
@Nigel:
> Given where Samsung sits Bada could become quite relevant to the discussion.
I agree. I don’t think Bada is relevant in the states, but it certainly is globally. And in Samsung’s position, with the expertise and willingness to juggle many SKUs at a time, they’d be stupid to put all their eggs in google’s basket, given that (a) they’re selling more than enough bada phones to support development; (b) google’s agreement may be fine now, but who knows exactly what the future will bring, especially with integration of MMI; and (c) Samsung does enough business with Microsoft to probably not be bothered too much by them about shipping an unassuming OS like bada.
@Life as we know it:
As Nigel points out, we don’t know exactly how much bada there is out there. I remember reading a few quarters ago it was 2 million, I think. Probably not growing as fast as Android. But if Samsung shipped 28 million smartphones last quarter, and 3 million of them were bada, then they are close to 50%.
I guess the other thing I don’t know is whether bada gets classified as a smartphone or as a featurephone. If 28 million is the number of Android phones shipped by Samsung, they they are right at 50%…
@Patrick Maupin It’s one of the difficulties of the less detailed reporting from Samsung. Others were closer to reading that Samsung was close to 10 million Bada phone sales around now (counting about 5 mill in 2010 and so far in 2011) after failing to meet the goal of 10 million shipped in the first half of ’11 they foolishly set at the start of the year, than this article’s pessimism: http://www.unwiredview.com/2011/09/26/samsung-bada-is-a-dud-less-then-10-million-shipped-sales-50-below-expectations-and-falling/ . But who knows? Maybe they are pumping out Waves. Anyway, it’s probably 2-3 million per quarter now. Hopefully (for Samsung), it’s growing finally, but it doesn’t look good. I’d also assume Samsung is producing several hundred million or so Windows Phones as well. Who knows? We also got Tizen coming up (I haven’t actually used an N9 yet so I’m unsure if Meego really has potential; either way, the quick addition of Tizen is not a good sign for Bada). Android is certainly the largest share of their smartphones, but Samsung is definitely not making it easy to know their platform mix.
Samsung, after doing lots of crowing for a while, has gotten very quiet and opaque about its mix of feature and smartphones, platforms, ASPs, etc this year.
Is Apple evil?
http://fosspatents.blogspot.com/2011/11/apple-loses-ipad-design-lawsuit-against.html
Should have quoted from the article that points to:
http://fosspatents.blogspot.com/2011/11/apple-loses-ipad-design-lawsuit-against.html
“I think, given all the above, (which you have yet to acknowledge or address), they’re troll-bait.”
14% x 64.4M / 17.1M = 53%
I should have mentioned that consistency with Nigel’s estimate, in my prior reply to Patrick.
About the global prospects of iPhone sales vs Android sales. Note that the iPhone 3 costs five times the cheapest Android. No points for guessing how this will develop.
Google bets big on mobile internet in India
http://timesofindia.indiatimes.com/tech/news/internet/Google-bets-big-on-mobile-internet-in-India/articleshow/10590524.cms
We are talking about close to 900 million mobile phone customers in India (even with double or triple phone owners, that is a lot). The one thing that holds smartphones back in India is the price of 3G.
Just for comparison, a related story from China:
From September:
Eight Alternative Android App Stores From China
http://paidcontent.co.uk/article/419-eight-alternative-android-app-stores-from-china/
From AppleInsider (October 31):
Little money made off Android in China during current ‘chaotic phase,’ says VC
http://www.appleinsider.com/articles/11/10/31/little_money_made_off_android_in_china_during_current_chaotic_phase_says_vc.html
I see Eric’s vision coming true: Cheap Android is railroading through the mobile world.
My post with a link to “Google bets big on mobile internet in India” is awaiting moderation.
Update:
Google: Android 4.0 to be open sourced in “coming weeks”
http://www.zdnet.com/blog/open-source/google-android-40-to-be-open-sourced-in-coming-weeks/9852
@winter You seem to think that folks disagree that Android will do well in India and China. The only points of contention is that both RIM and Nokia do well there and view it as one of their last few markets. RIM’s OS is old but runs on some really cheap hardware. Nokia is replacing their S40 OS with their lighter than meego linux variant.
Android is pretty heavyweight so I can see Meltemi and Bada having legs in the under $100 smartphones. App stores are a key element but with the situation in China the game is still wide open. Im surprised that Baidu hasn’t tried to dominate that. Given that one of the larger segments in China is games it’s going to be a lot of OpenGL coding anyway. For that many folks will look to Unity and other cross platform engines. I’ve written a little OpenGL ES code on Android. It works okay in Java but I’d assume most game devs will want/need to use an engine and NDK.
It would be smart for both Samsung and Nokia to court (aka pay) these cross platform development tools to support Bada and Meltemi.
I think it’s a little hopeless for RIM but QNX is a solid kernel. They’re pretty screwed without WebOS though and now WebOS is damaged goods. I’m not going to say they have no chance but I don’t see anything exciting happening with their current leadership.
With respect to Apple in China and India I think they’re happy with the way things are going regardless of who dominates the share game. In terms of mind share Apple isn’t lacking anything in China and with more retail stores opening there they’ll continue to do well. They won’t lack for Chinese devs either as long as they continue to sell millions of units there each qtr and their app store actually generates profits for their devs.
With respect to developers, China matters but it’s a major porting effort from the perspective of language. You’ll see the big dev shops either have Chinese teams or hire a reputable local company to do a port. But indie developers are likely going to have to wait till they hit the big times to make that jump.
Going from English Android to Chinese Android will be harder for them than hitting both Android and iOS. If you don’t do a good job you’ll end up as a Chinese internet meme with “All your bases are belong to us” level of translation. Then again, there might be good business for companies doing Chinese language ports. I’d pick a company in the EU or the states though even if it is more expensive. At least there’d be someone to sue if they steal your code.
Going the other way is easier but I doubt you’ll see a lot of apps going from China to the west either. Their local market is big enough that catering to that well is probably a better ROI in effort. There are a lot of asian games and such that never make it to the EU or US.
Pretty much only the big guys will cater globally.
@Nigel
“The only points of contention is that both RIM and Nokia do well there and view it as one of their last few markets. ”
Both are named as doomed in these articles.
@Nigel
“App stores are a key element but with the situation in China the game is still wide open.”
With 8 different Android App stores? That seems to be ideal for a Survival-of-the-Fittest. With the Survivor quite good.
Chinese for Android is already available:
http://www.pinyinjoe.com/other-os/android-chinese-support.htm
But I assume you mean China specific apps.
@Nigel:
Short term, sure. Longer term, bets against Moore’s Law never seem to turn out well.
Windows is pretty heavyweight, so I can see DOS and DesqView having legs in the under $500 computers.
@winter Yes, just supporting the character set is just the first step to make a chinese (or china specific) app.
I couldn’t write a Chinese android app without an exceptional amount of effort…and I took mandarin in college. I could probably do a better job with German which I’ve never learned. Okay, I can probably manage a flashlight app. On/Off is pretty easy. WTF I’d do with chinese bug reports I dunno but a flashlight app is pretty hard to screw up so I’d be safe there as well.
Regarding Nokia and RIM, for a while it seemed like Intel couldn’t do anything right and AMD was going to eat their lunch. Then they started executing well again. Nokia is a big player and if they can start executing well I think they can pull through.
RIM…meh.
I have to say that I’m a bit biased regarding Nokia. I’ve always generally liked Nokia hardware (not that weren’t some stinkers now and again) and am partial to “plucky” Finland. If it weren’t for the fact I’ve got so many iOS apps I’d get a Nokia WP7 next refresh. I might get one anyway to play with development wise.
BB phones I’ve always disliked. I try to give RIM the benefit of the doubt simply because I know I don’t like their stuff much.
Nokia’s biggest problem is this:
http://scobleizer.com/2011/06/09/why-nokias-elop-is-wrong-about-mobile-sales-users-arent-idiots/
If you can’t get Finnish startups to care about Nokia then it is pretty bleak. Nothing a hundred million dollars wouldn’t solve though in VC money. Something both Nokia and MS can afford. Not sure why Nokia and MS hasn’t gone in on a $200M iFund like pool.
WP7 didn’t get developers on board because its API sucked. “Mango” alleviates some of those problems, with real sockets and other things that allow the development of richer applications. We’ll see how that goes in the coming months, but if Microsoft were smart they’d be pushing the message that says “Hey devs, our APIs aren’t shit anymore. You can do everything that can be done on Android or iOS, and use the best-of-breed Visual Studio tool suite to do it!” Granted, they may have done irreparable damage to their platform by releasing a crippled API in the first place, but it’s worth a shot if they still want to compete. Also, money hats to key development partners would probably go a long way.
I do know that major game developers are targeting WP7, which they’re still largely not doing for Android. However, these days iOS is still the premier mobile game platform, even above and beyond the (3)DS and PSP.
@nigel
It seems that Nokia has to do something drastic to have a future. If your own employees do not believe in your future, it is bleak indeed.
MS have to chase taillights. WP7 has nothing that makes it better than Android. So why would people buy it?
@eric
Perhaps an interesting question for you: Are network effects predominantly global or local? As in how coupled are the Chinese and US app markets from a developer perspective in terms of network effects?
Take for example that RIM is currently doing really well in India. A local developer might target BB first and other phones second simply because of the local market regardless of overall global platform shares. Even accounting for the probability that BB is going down…to maximize 2011/2012 revenue I’d have to look at BB first.
As an enterprise app developer I’m sheltered from all this but I doubt as an independent app developer I’d target the Chinese market. To me, that market is mostly invisible. I mostly care about the US, UK, Canada and Australia (aka English speaking nations) market. So local network effects are more important to me than global effects. Maybe.
Perhaps you see this differently.
>Perhaps you see this differently.
Yes. I don’t think you’re completely mistaken about language barriers localizing network effects, but I think this attenuation is relatively weak. Here’s why…
Global sales volume predicts that rate at which hardware vendors amortize their NRE and other fixed costs – thus, ultimately, the price trajectory of the hardware. If I’m an app developer looking for a volume market, I’m going to strongly prefer the platform that’s likely to get cheapest fastest.
@Jeff – Hey remember the thing about messing with user data? You can say nice things about Apple again. :)
http://www.macrumors.com/2011/11/02/apple-posts-ios-5-0-1-beta-for-developers/
“iOS 5.0.1 beta introduces a new way for developers to specify files that should remain on device, even in low storage situations.”
It was a bug and it’s being fixed in 5.0.1.
@esr I can buy that. But do you think there are a lot of hardware cost differences between a WP7 and Android handset though? Personally, I see linux (or vxworks, FreeBSD, whatever) kernel + UI layer still being more thrifty hardware wise. No VM layer.
Apple will never be cheapest* and iOS will never run on anything but an Apple device. Therefore Apple will never be THE volume market unless they manage to corner a market like they did with iPods. On the other hand, it is arguable they may have a very lucrative (for the app developer) market based on demographics that even with small share they are still very relevant in terms of developer mindshare. That’s probably something that you don’t much care if it happens or not. Or maybe you do care given your predictions. I think that’s where you and I disagree on the most.
—
* I think Tim Cook can continue to optimize Apple’s supply chain for Apple to be very price competitive on the high end and still maintain high margins. They’ll just wont have a $80 unsubsidized iPhone anytime soon, perhaps ever.
>But do you think there are a lot of hardware cost differences between a WP7 and Android handset though?
No. But observably there is a difference in retail unit cost – and there has to be, because Microsoft’s business model requires it to collect a license fee. So does Apple’s. Google’s does not. That translates directly to lower unit prices, and indirectly to a larger expected size of market. This effect is independent of language barriers.
>That’s probably something that you don’t much care if it happens or not.
Correct. It is not required that Apple die, only that the closed-source, walled-garden model is seen to be a dead end and abandoned.
8.1M Bada devices to date. 110M app downloads to date. 13,000 english apps.
http://www.mobile-ent.biz/news/read/110m-bada-app-downloads-to-date/016040
That’s more apps than I expected though I bet most are filler type apps (wallpapers, etc). They probably aren’t going to hit 10M by 2011 given the new Waves are a little late I think.
Eh, maybe, Samsung ships a heck of a lot of phones so what’s 1.9M more in a qtr? :)
@esr mmmm…I’m not opposed to the closed-source walled-garden as much as you are so long as there is choice in the market for both open and closed approaches. I’m definitely on the far right in the pragmatist school of open source.
I’m not certain that from a purist standpoint that 200,000+ closed source apps really adds much to the concept that closed-souce is a dead end. And given a number of significant Android app stores (like Amazon and Barnes and Noble’s Nook store) may be walled-gardens I don’t see that Android necessarily ends the walled-garden approach even for app markets unless you count side-loading and on some devices, rooting (messy for average users IMHO).
I doubt the Chinese android app markets will be very open/free either. If anything apps will have to pass self-censoring on the part of the app developers (to avoid bad personal outcomes), app review by the market owner (so the owners avoid bad outcomes) AND passing the official censors (who metes out bad outcomes to the folks they don’t like).
If I were an independent chinese app developer I’d stay the hell away from building social apps. Unless I were a chinese app developer not actually in the PRC.
A GPL’d Android would have enforced your desired outcome perhaps. On the other hand, I’m of the opinion that a GPL’d Android would have been DOA. The license doesn’t preclude walled gardens anyway.
@patrick “Short term, sure. Longer term, bets against Moore’s Law never seem to turn out well.”
Sure. But that assumes that Android isn’t going to add more core features to take advantage of quad core A15s next year too.
Software doesn’t take more cycles because modern developers suck. Software takes more cycles because it does more stuff. That and we trade cycles for productivity (aka abstraction).
WRT to Windows…it’s not exactly snappy on netbooks. Hence Splashtop OS. And Chrome OS for that matter. Yeah, I think those might have legs. Not as much now that tablets are the rage…
@Nigel:
Yes, the creepage will always be there. But the nice thing about Android is that older versions will be available, and any manufacturer who cares can do some tuning. Whether it’s called “Android” or not, or can use the app store or not, right now the Android software is the primary driver of compatibility of phones all up and down the performance spectrum.
This I know implicitly. But it still plays into the Moore’s law conundrum. Much effort spent optimizing stuff for today’s hardware/OS is wasted tomorrow. Granted, there are tradeoffs to be made about exactly when tomorrow comes, and how many phones you can ship today, and there is some money to be made all up and down the spectrum, but a long-term strategy will account for ever-increasing hardware capabilities.
But for a new OS to gain traction, there has to be some sort of quasi-standardized hardware for it to run on. I think Android is providing a valuable service here, much as MS-DOS did a couple of decades ago, in helping to get the chicken and egg cycle going.
@Nigel:
I don’t think closed-source is a dead-end. However, I do think that the larger the software market, the more likely that open source is to be able to win, and you can’t get much bigger than the market that Android is playing in.
For one thing, I count side-loading as a huge deal. For another thing, a zillion walled gardens is infinitely preferable to one big one.
It will be interesting to see when the Chinese government eventually makes google block access to its own app store from inside China.
I can’t speak for Eric, but that wouldn’t have come anywhere near my desired outcome.
Agreed on both counts. And I will agree that, to some extent, we have conflated two separate but somewhat related issues here. Apple is in a position to have, not only a walled garden, but _the_ walled garden, for iOS, by dint of the control they exercise over the OS.
A Microsoft desktop OS style strategy would have eschewed control over any walled garden for a cut of all the revenue in the ecosystem. (They may be repeating that strategy somewhat successfully, but I’ll reserve judgement on that until the MMI and B&N lawsuits are over with.)
Google has turned the walled garden on its head. Rather than make the walled garden the only place you can go to get software, the walled garden is the best place you can go to connect with the customers (but you’re free to try on your own, if you’d like).
http://www.guardian.co.uk/technology/2011/nov/03/lg-rights-issue-android-smartphone-funding
….. Android is Winning.
“Profitability is the canary in the coal mine. It causes a brand to be tarnished in the eyes of distributors who, because of sales cycle times, are extremely sensitive to obsolete inventory. A loss maker is seen as a maker of damaged goods. It then turns off the tap of incentives, promotions and hence visibility in the eyes of consumers. It’s a vicious cycle from which few (if any) can recover. Past greatness offers no succour.”
@Nigel, the “walled garden” applies to the OS, not just app stores. Android is open source, and thus not a walled garden. This means any one can make an Android device, etc..
There has never been a mainstream open source OS. This year that changed with global share for Android surpassing 50%. I am surprised that no one in the world has noted this monumental first-ever event horizon for open source.
Closed source is a dead-end. No one ever again be able to make a closed source OS and get any momentum.
Apple’s margins and volume will get torn to shreds as the global smartphone and tablet markets saturate and open source is tweaked by many independent entities to address every nook & cranny of the markets.
Smaller companies will be more nimble than Apple at advancing the state-of-the-art, one the OS has stabilized. It was the need for an OS-scale integration that afforded Apple the ability to wall off a section of market and win on their state-of-the-art vertical integration.
@shelby
As stated above, source license is orthogonal to do with being a walled garden. I can create a walled garden using Apache or GPL license code. In fact it’s easier with GPL licensed code since if I am the primary code producer I’m the only one that can sell a product with any secret sauce and control access to the ecosystem. The best anyone else can do is fork and attempt a matching garden (walled or otherwise)…but even then they can’t have any non-GPL’d secret sauce to make their garden more compelling than mine.
Linux is a mainstream open source OS with greater than 50% share in certain server segments.
That android is the second open source OS to do that is cool but doesn’t preclude closed source operating systems from gaining traction. The key to remember here is that Android is open source AND viable simply because Google can afford to spend billions to make it so. You need both components to make a viable mainstream open source OS. This was true even for linux. IBM and HP sunk billions into Linux as well for strategic reasons. So the business model for mainstream open source operating systems is fairly narrow and only one of many viable ones.
Android can go closed at any time. It can do so at any point a major developer thinks it see closed source market advantage outweighs community involvement. *cough*Amazon*cough* This is either a feature or a bug depending on your perspective. As a fan of permissive licenses I think it’s a feature. Many would vehemently disagree.
Apple, with an open source XNU mach kernel and BSD userland (together known as Darwin) and many open source components (webkit, bonjour, CUPS, LLVM/Clang, etc) is a hybrid of open and closed source. It’s a very pragmatic approach that is superior, IMHO to being slavish to any particular licensing ideology. One that Google has adopted at a fundamental level. Their closed source layer is in their ecosystem services (vs UI and API layers). One they protect quite diligently and very astutely. Even the FSF is unwilling to challenge Google in that area or it would have closed the SaaS loophole in GPL v3.
Apple has also been very agile. Whether post-Jobs Apple will remain agile remains to be seen. Saturation of the smartphone market is sometime off in the future and something that Apple will have to deal with someday. However, Apple is able to maintain both large margins and millions of units shipped for their Macintosh computers. The computer market can be safely considered mature and what you have described has not happened to Apple there. Margins are the best in the business and apple is increasing computer share today.
What is also clear is that Apple’s primary competitors are not small nimble companies but very large multi-national corporations with billions in revenues and billions in ready cash. Google, Samsung, Microsoft, etc are all multi-billion dollar companies with deep pools of talent and money. They all play to win. Smaller companies that advance the state of the art tend to get bought by Google, Apple, etc. Few will ever grow to challenge them. Facebook being one of the few examples where this occurs. Amazon being another.
Besides, Apple doesn’t so much advance the state of the art as advances the state of the practice. They are most adept at mainstreaming state of the art over creation of the state of the art. In a many arenas they actually have advanced the SOA but equally many of their market advantages have come from acquisitions (Fingerworks, Siri, Intrinsity, PA Semi, etc).
I’m just going to skip over the whole patent issue with software here. That’s a whole other can of worms that deeply conflicts with the notion that Android is open and therefore closed source operating systems have no future. It would require another huge post all its own. Source license, again, is just one piece of the IP puzzle.
On a shorter note, the Nook 2 looks pretty cool too. Now I have to decide on which to buy. Kindle Fire or Nook 2. I’ll probably go Kindle Fire because I’m a Amazon Prime member…if I can also sideload Netflix on it. I like the Nook’s design better though. I find the Kindle Fire a little fugly.
esr Says:
> because Microsoft’s business model requires it to collect a license fee.
Correct, back when they were a software company. Now they can add patent licensing revenue to s/w licensing revenue.
From: http://www.businessweek.com/magazine/content/10_04/b4164051608050.htm
IBM’s patents produce considerable income: Fees from licensing and custom-developing intellectual property for other companies through Sept. 30 were on track to top $1.1 billion in 2009.
Nonetheless, a study conducted for Bloomberg BusinessWeek by Ocean Tomo, a Chicago intellectual property consulting firm, concludes that IBM’s collection of U.S. patents over the past five years ranks only eighth in value. No. 1 is Microsoft, which ranked third [in quantity], with 2,906 patents issued last year.
Not an original thought, but what if Winter’s crash and burn scenario comes true (unlikely, IMHO), and they choose to become a patent troll instead? Then instead of cross-licensing / Windows subsidies / Cooperative Marketing, they simply demand cash or else?
@Marco
It already happened. Nathan Myhrvold started his venture from and with MS. IIRC he started with a portfolio from MS. And MS are trolling their patents for all their worth.
My prediction of MS crashing is based on the unsustainability of their 80% margin in the face of competition. As they have milked office productivity increases due to the introduction of desktop computers for a living, they are out of income when productivity does not increase anymore. That happened somewhere after XP SP2/3.
With the switch to mobile(phone) computing, they lost their monopoly over personal computing and hence their 80% margin.
They’re already doing this to the Android smartphone market. And no, thanks to the Republicans, Bilski isn’t going to save Android’s ass here.
@Marco> esr Says: because Microsoft’s business model requires it to collect a license fee.
Yes, and then he said, “So does Apple’s. Google’s does not.”
Which is not accurate.
Apple’s business model doesn’t require it to collect a license fee. This fact is clear once you admit that Apple bundles its software with its hardware. There is no “iOS license fee”, because Apple only licenses its software to run on its hardware, and on iOS platforms, they don’t charge.
If you’re an Android OEM, chances are that you’re paying patent license fees Microsoft (at the very least, others are still in court) for every handset (or tablet) shipped. In a very real way, Microsoft’s (new) business model requires it to collect a license fee for every Android handset. Normally these would be paid by Google, but since Google doesn’t actually build handsets, they’ve escaped liability for this. Still, it doesn’t matter, Android is not free of license fees, as esr implied.
Back in the desktop/server world, the fees for OS X Lion are $29.99. If you bought a Mac at any point after the WWDC announcement of Lion, the upgrade from Snow Leopard is free.
If you’re on Leopard, the upgrade to Snow Leopard is $29. (Snow Leopard shipped over 2 years ago, and if you bought a Mac since then, you have Snow Leopard or Lion.)
Red Hat Enterprise Linux for Workstations (self-support) is $49, and the cost climbs from there (want to run a 4 core server with unlimited VMs? That will be $4,000.) Other “LInux vendors” have similar pricing. OS X Lion Server: $49.
And while on the subject of patents and Unix, let us remember that the setuid bit was invented by the late, great Dennis Ritchie.
His employer, AT&T, applied for a patent in 1972; the patent was granted in 1979 as patent number US 4135240 “Protection of data file contents”.
AT&T fully intended to collect a license fee for every copy of Unix. The patent was later placed in the public domain. Had AT&T decided to charge for it, linux (and the various FOSS BSDs) would have certainly infringed, and as the patent didn’t expire until 1996, linux would have been stillborn, especially if AT&T had successful used the various methods to extend a patent’s lifetime.
“AT&T distributed Unix with the understanding that a license fee would be collected if and when the setuid patent issued. When the event finally occurred, the logistical problems of retroactively collecting small fees from hundreds of licensees did not seem worth the effort, so the patent was placed in the public domain. ref Doug McIlroy
link
@Jeff Reed> … Bilski isn’t going to save Android’s ass here.
Not in the US or other signatories to the GAATT/TRIPS treaty. Android will probably survive in deep Asia and Africa, and may well still move hundreds of millions of handsets. But Android costs real money now, probably at least as much as, if not more than, the Windows Mobile license fees.
When the Oracle suit settles, Google is going to be challenged to continue to not charge for its distributions. The open question is what happens to cyanogen mod and the other Android hacking groups. If Oracle wins, they, too will be guilty of infringement (and Google won’t really be able to indemnify their activities). Takedown notices are sure to follow. Cyanogen may be able to hide overseas, but anyone on US or other soil covered by GAATT/TRIPS who so much as points to the new Cyanogen activities is risking contributory infringement at that point.
“I feel a great disturbance in the Force, as if millions of voices suddenly cried out in terror, and were suddenly silenced.”, indeed.
Microsoft tried and failed to kill Linux, but they and others may yet fatally wound Android.
A lot of people’s handsets are going to be silenced indeed, when Oracle gets that judgement against Google requiring them to remotely disable any infringing software!
Jeff, I don’t know if that’s going to happen. It could, but that’s likely just a bargaining chit for Oracle to inflict on Google. There are reasonable arguments to be made to not disable current handsets without an orderly program to get people to turn them in (say, 6 months). Anything left after that might be subject to mass disable.
Imagine, if you will the chorus of jeers.
In other news today AT&T has announced that it’s slipped the target date to close its acquisition of T-Mobile by three months.
http://online.wsj.com/article/SB10001424052970203716204577017763203498818.html
Unfortunately, the longer this drags out, the further T-Mobile falls behind – since it has no LTE strategy of its own (it lacks the spectrum to setup LTE). Consequently it becomes more likely that the authorities will see no sense in keeping the company independent when it can’t actually compete anymore.
AT&T essentially put T-Mobiles’s efforts at competing on timeout for 18 months with this purchase attempt, that shouldn’t really happen in a truly competitive industry.
Effectively, AT&T is the only end for T-Mobile, other than bankruptcy.
Whenever iPhone 5 ships, iPhone 4/4S will likely be the new $0 iPhone. With T-Mobile either absorbed into AT&T or on life-support due to a lack of LTE, will become of Android in the US?
My comment is awaiting moderation. Geez, ’cause I put in one link?
“A lot of people’s handsets are going to be silenced indeed, when Oracle gets that judgement against Google requiring them to remotely disable any infringing software!”
Isn’t going to happen.
Most likely case is Oracle asks for a very large check, Google provides a very large check. It isn’t like Google doesn’t have the cash.
Worst case is that Oracle gains control over Android as part of Java and starts collecting licensing fees. That doesn’t seem likely but possible if Oracle can compel Google to stop using any Java IP in Android and it therefore holds the only valid Android fork since only they hold Java copyright and patents AND given Android is Apache licensed. Which it will then license to Samsung, LG, HTC, etc.
Google would have to restart Android with different syntax and API structures…kinda like MS did with C#. Heh, I guess they could go the Mono route. Frankly, this is what they SHOULD have done in the first place (not Mono, rolled their own Java like language).
Killing Android provides Oracle nothing as it has no competitor product. J2ME is dead. Android embraced, extended and extinguished it.
@Life as we know it, “Android is Winning.”
Winners & loses are degrees-of-freedom and fitness in a a free market.
@nigel, my reply went into the spam queue I think.
>@nigel, my reply went into the spam queue I think.
And is never coming out. Shelby, you were warned that you would be banned if you went above one post a day. You’ve been repeatedly violating this. I had been putting up with it because you’ve been having a lucid streak recently, but you’re back to semi-unhinged babbling. So you’re banned.
Your semi-psychotic word-salad prose style is quite recognizable. I won’t tolerate you returning under yet another mask. Jocelyn, or whoever you are.
Remember those 200-some patents in the Linux kernel? Near as I can tell they weren’t invalidated.
My guess is that the Android suits aren’t a last-ditch effort but rather a first-ditch one: build up a precedent that yes, these patents are legitimate IP and people do pay money on them in order to stay licensed, and then go after the data centers, the systems integrators, the large-scale users (like Google) and, if there is anyone left still using Linux on the desktop, individual users. (Menacing individuals has proven SUPER EFFECTIVE for the RIAA and MPAA, what makes you think Microsoft won’t use it?) Probably a licensing scheme will be set up wherein Windows servers will be available at a discount compared to the premium charged for licensing the Microsoft-owned technology in Linux. Oh, and you get a few bucks extra off per unit shipped if you disable booting unsigned (read: non-Windows) kernel images in your system’s UEFI layer.
Open source people have to learn that the rules that apply to everybody else also apply to them, that using other people’s IP does have a cost, and consequently, Linux isn’t free.
> Google would have to restart Android with different syntax and API structures…kinda like MS did with C#.
Google could probably leverage Chrome / NaCL and ship an ‘Son of Android’ and be back closer to it’s budda nature of only web-apps. (Chrome marketplace, anyone? Anyone? Bueller?)
I think everyone understands that the likelihood of ‘Android’ surviving this transition is low. Google agrees:
“What we’ve actually been asked to do by [Google co-founders Larry Page and Sergey Brin] is to investigate what technical alternatives exist to Java for Android and Chrome,” Lindholm wrote in the email Android chief Andy Rubin. “We’ve been over a bunch of these, and think they all suck. We conclude that we need to negotiate a license for Java under the terms we need.” <– Tim Lindholm in email to Andy Rubin in August 2010, shortly before Oracle filed suit.
Meanwhile, Larry Ellison (Mr. Oracle, in case you've forgotten) was best buddies with one Steven P. Jobs. Ellison was willing to put up the money to buy Apple for Jobs. (It's in the book.) Much later, Jobs said he would spend any amount of money to kill Android.
What if Jobs left a deal in-place to get Larry Ellison to kill Android for him? Apple can't do it (it's anti-competitive), but Oracle certainly could.
> Killing Android provides Oracle nothing as it has no competitor product.
But if Oracle assigns (transfers) the IP in-question to Apple, in-exchange for filthy lucre and a deal to get some Oracle tech into iOS?
And Apple subsequently decides to NOT LICENSE the patents / copyrights required for Android to function?
(Patents are a negative right. They allow you to prohibit others from ‘make, use or sell’ the IP in-question. You have to grant an exception (via license, normally.))
> Oh, and you get a few bucks extra off per unit shipped if you disable booting unsigned (read: non-Windows) kernel images in your system’s UEFI layer.
More likely that Microsoft comes up with a branding deal (like “Genuine Windows” or “Intel Inside”) that pays real money to the OEMS in exchange for them both disabling booting unsigned images *and* not installing non-Microsoft keys into the UEFI layer.
Presto, it’s a “Windows Experience” laptop/server, and it costs $50 less than the one that could run Linux.
@life occam’s razor. That’s pretty convoluted and while Ellison is king of oracle he does have to answer to parliament (shareholders).
I think that Ellison would do quite a bit for Jobs personally. Even put up money to buy Apple if he believed he’d make a mint on it with Jobs in charge (and man, he would have). But Killing Android for the sake of pleasing Steve? Only if there were a big payday at the end of it as well to point to for the stockholders.
See? I made this big pot o’gold with this Sun purchase you guys made fun of behind my back. AND I killed Android and made Larry my bitch. Next up, IBM. My hardware’s going to rule.
Find me a pot of gold associated with killing Android and I’ll buy that…maybe. Otherwise…not so much.
Latest comscore numbers are out.
Haven’t really analyzed them, but the easy low-hanging fruit of unhappy RIM/etc. userbase seems to be gone. Market still growing, but Android marketshare growth cut in half and Apple marketshare growth non-existent (probably due to waiting for iPhone 4s). Will be interesting to see the spike from the 4s in the next few charts.
http://www.comscore.com/Press_Events/Press_Releases/2011/11/comScore_Reports_September_2011_U.S._Mobile_Subscriber_Market_Share
Near as I can tell, Oracle initiated this suit for one reason: it’s part of their broader policy of forcing their new asset, Sun, to stop fucking around with free software, free love, free lunch — and start making money. OpenSolaris — dead. OpenOffice — Oracle is not involved anymore. OpenJDK — well, thanks to the GPL they can’t put that genie back in the bottle, not easily anyway, but they can drop the hammer on anyone else caught squirreling Oracle tech and hopefully recoup some of the money that Sun’s pro-OSS policy had cost them — hence the suits against Google over Android.
MMI’s apparently gotten a little traction against Apple in Germany.
http://www.engadget.com/2011/11/04/german-court-grants-injunction-against-apple-for-infringement-of/
Will be interesting to see if Tim Cook wants to negotiate, or if he made a bedside promise to Jobs to salt the earth.
Schmidt joins Apple’s board on August 28, 2006.
iPhone introduced by Steve Jobs on January 9, 2007.
iPhone released June 29, 2007.
Sometime during Burning Man in 2007 (August 29 – September 5, 2007), Jobs calls Schmidt on Schmidt’s personal (prototype) iPhone. Schmidt tells his companion (mistress) “Steve was very, very upset.” and “My God, he was so angry.”
iPhone price drops by $200 on September 5, 2007.
September 28, 2007 Apple hires Daniel Cooperman as General Counsel. Cooperman was GC and Corporate Secretary at Oracle. He replaces Donald J. Rosenberg, who joined Apple a mere 10 months later. Previously Rosenberg had been at IBM for over 30 years. Qualcomm announces that they’ve hired Rosenberg the same day.
Open Handset Alliance formed November 5, 2007.
HTC Dream ships as the T-Mobile G1 for $179 in the US October 22, 2008.
Oracle buys Sun for $5.6B, net of cash and debt: April 20, 2009.
Google announces Chrome OS July 7, 2009.
Schmidt (ex-Sun VP) resigns from Apple board on August 3, 2009
October 27, 2009 US DoJ sends cable to EU suggesting that any fears about Oracle/Sun merger should be put aside.
November 25, 2009 Secretary of State Hillary Clinton also urges the EU allow the Oracle/Sun merger to proceed.
January 31, 2010 Jobs lambasts Google at an Apple “Town Hall” meeting.
“We did not enter the search business. They entered the phone business. Make no mistake: they want to kill the iPhone. We won’t let them […] This don’t be evil mantra? It’s bullshit.
January 27, 2010 Oracle acquisition of Sun complete.
March 26, 2010 Jobs and Schmidt are spotted having coffee in Palo Alto.
Jobs reportedly says to Schmidt in 2010 at a Palo Alto cafe: “”I do not want your money. If you offer me U.S. $5 billion there, I do not want it. I have a lot of money. I want you to stop using our ideas on Android, that’s what I want,”
August 12, 2010 Oracle files suit against Oracle.
June 6, 2011 Google document released that estimates Oracle’s recovery as between 1.4 and 6.1 billion dollars.
@Life as we know it:
Why didn’t you start your timeline here?
Google acquired Android Inc. in August 2005,
For similar reasons to why I didn’t start the timeline with Andy Rubin’s stint at Apple, (1989-1992), after which he went to General Magic (1992 – 1995), an Apple spin-off, where he helped develop software for a PDA-like device.
Or that he was a founder at Danger (1999 – 2003), immediately prior to co-founding Android, and that given that Danger was bought by Microsoft (and then fumbled by Microsoft), some of Danger’s patents may well be used against Android.
Apple’s $0 iPhone 3GS making inroads against Android.
http://www.sfgate.com/cgi-bin/article.cgi?f=/c/a/2011/11/05/BUNT1LQA37.DTL
With the Comscore data showing Android’s growth cut in half *before* the iPhone 4S and $0 3GS could have any effect, and the saturation marketing by all three carriers during the holidays, (the very subject of Eric’s blog post above shows just how incentivized Sprint must be), Android’s immediate future does not look bright.
Add in the legal battles that google is having to fight, the pressure to settle the Oracle suit, and its pending $12.5 billion dollar purchase of MMI, and the situation begins to look really grim.
@Patrick> Will be interesting to see if Tim Cook wants to negotiate
Unlikely. At least, “not yet”. Florian has a semi-logical analysis of what’s going on.
http://fosspatents.blogspot.com/2011/11/motorola-mobility-apparently-won.html
Patents against progress, indeed.
Still, all these scenarios of doom sound like SCO 2.0.
@Patrick> Latest comscore numbers are out.
Android seems to be falling below its trend line.
http://imgur.com/a/4c3vB#0
Apple seems to be performing well above its trend line.
http://imgur.com/a/4c3vB#1
Stacked Market Share:
http://imgur.com/a/4c3vB#2
>Apple seems to be performing well above its trend line.
All that graph actually demonstrates is that when you amplify statistical noise all you get is louder noise. See my next post.
> All that graph actually demonstrates is that when you amplify statistical noise all you get is louder noise. See my next post.
Horsefeathers.
Winter Says:
> It already happened.
> Nathan Myhrvold started his venture from and with MS. IIRC he started with a portfolio from MS.
Agreed. Nathan Myhrvold and Intellectual Ventures are patent trolls.
> And MS are trolling their patents for all their worth
While I think the facts have not been fully shared yet, I disagree that MS is a patent troll. A patent troll has no products to protect, and is uninterested in cross-licensing, or in using patents as just another negotiating point in a broader business relationship. So the frequently-quoted figure of 5 USD per copy of Android may be an gross patent license cost, but after cooperative marketing credits, Windows license discounts, etc. the net might be *MUCH* less. Barnes and Noble which is less interested in these kinds of credits has been unwilling to pay the (presumably) gross amount of the tax.
But as bad as the current scenario might be, imagine how much worse it would be for the entire computing industry if MS were to crash and burn entirely (not just lose their accustomed 80% margins, but actually declare bankruptcy) and have their patents sold to a troll (like Nathan Myhrvold). Then, there would be no joint marketing spending, no Windows license credits, etc. Just pay up or die.
Would this work on an iphone 4 running iOS five?
? And with carrier SIMs outside on the USA?