The Smartphone Wars: Expectation and Surprise

If style is truly the contrast of expectation and surprise, it’s been a very stylin’ couple of weeks in the smartphone world while I’ve been on vacation.

Expectation: Nokia’s sales in China dropped 41% year-over-year in the last quarter. This is worse than it looks because China and the the Pacific Rim were Nokia’s big hope for volume sales; the brand has disappeared from Europe and never had much presence in the U.S.

Nokia’s strategy, insofar as it’s actually had one since the tie-up with Microsoft, has been to hang in there on volume sales of dumbphones until it could deliver world-beating WP7 handsets. The major risk here, other than the wild unlikelihood of WP7 ever becoming anything consumers actually want, was that the Chinese electronics industry would undercut them on price-performance. I predicted this would happen, it is happening, and Nokia’s disastrous July earnings call is the result.

Stick a fork in Nokia, they’re done. Specifically, their recovery strategy is busted. I’m now projecting that the tattered remnants of this once-proud company will be on the acquisition block within 18 months.

Surprise: A new study says that worldwide, Android has 30% of the tablet market (with more detailed figures here). It appears that Android tablets have been doing far better than I knew in price-sensitive overseas markets.

I’ve taken a lot of flak for noting that Android seems to be executing a classic technology disruption on iOS, especially in view of Apple’s record quarter. But this is pretty strong evidence that even where Android has looked weakest in the past it is fast undercutting Apple’s “premium” positioning.

I’m not surprised this is happening, but I’m surprised it’s happening this fast. Android tablets have on the whole been disappointing to me so far, most of them timidly-designed me-too products that are way too overpriced to compete effectively against Apple’s brand strength.

But…if they can cop 30% global market share with the indifferent products they have now, they should rapidly be able to double that when the vendor tier gets its act together. Apple, watch out! Because there isn’t going to be much warning before the critical price-performance threshold gets passed. There never is; it tends to come as a shock to both disruptee and disruptor.

348 comments

  1. I think Android tablets are still very much in a shakeout period – if you search for “Android tablet” on Amazon, for instance, you still get a dizzying profusion of mostly no-name brands with user reviews all over the map. I wound up buying an iPad in March because, although my first choice was an Android table, I didn’t feel comfortable committing money to any of them, not even after playing with a few of them in stores. I can’t seem to find figures for the market share or sales totals of individual Android tablet models, which is very disappointing…

  2. >I’m now projecting that the tattered remnants of this once-proud company will be on the acquisition block within 18 months.

    A good discipline with these sorts of prediction is to estimate what the market capitalisation will then be. They are worth a little over $20 billion right now, which is within range of the larger private equity deals.

    1. >A good discipline with these sorts of prediction is to estimate what the market capitalisation will then be. They are worth a little over $20 billion right now, which is within range of the larger private equity deals.

      Good point. On current trends, I think we’re looking at market cap of $8 billion or less a year out. Maybe a lot less.

    1. >Why aren’t you reporting the other big sales change in China?

      Because I didn’t know of it. Now that I have, the report looks pretty meaningless. It’s all about growth rates with no clue about baseline of achieved levels in actual use numbers or share. This omission causes me to suspect that we’re seeing jackrabbit growth from a tiny base to share that is still quite small.

  3. I got a VIewSonic gTablet just recently for a bargain. I ordered it on e-bay but it’s not yet delivered so I have yet to experience the Android on this or any other medium. I was wary of the no-name China brands, but most Android tablets from brands are just too expensive.

    But I do get the impression, reading reviews, that these Android tablets (or most of them anyway) need to be rooted to get the most out of them and the default ROM sucks most of the time. I guess that’s one of the temptations of Android – the ability to tweak and customize to our desire.

  4. The Tohsiba Thrive which I plan to get next week is the first one that isn’t a me-too – removable battery that charges in 1 hour (with a real charger, not USB), real ports including SDXC capability and real USB so you can plug in a hard drive. Most of the others were trying to be an iPad in the worst way. The iPad is at best a niche device. You are starting to get differentiation in the tablet space just as there is in the phone. You have an iPhone, or you have dozens of different Android phones offering different features, form factors, and capabilities. GM beat Ford “any color as long as it is black” 100 years ago.

    Meanwhile Microsoft is trying to do a heterogeneous ecosystem. The XBox, WinP7, and Windows-pc are all different platforms – even processors! – with different learning curves, barriers to the marketplace, etc. iOS and Android are basically the same OS for the devices (TV), phones, media players, and tablets. So I can write one app and with at most some optimization hit the entire ecosystem for apple and . With Microsoft, although your zunepass and XBox Live stuff might let you cross-over, the developers have to write multiple apps. This won’t create a working ecosystem, merely islands with scheduled cargo ships.

    Apple is profitable and has lots of cash, but that is not unlike the mining ghost towns where some big lode was discovered. Gaining and sustaining are different things. The cash pile will collapse if their margins shrink. They are also placing a huge bet that in a few years bandwidth will drop in price – Lion is a 4G download only update. They are relying on iCloud to do sync. This works for small things but not gigabyte video. Netflix has already had pushback (via level3 v. comcast) and Apple plans on doing even more and is burning the bridges – no blu-ray media, and dvd only via external drive. I think in many places there will be a lot of bandwidth, but not everywhere in the immediate future, and they are foreclosing themselves to those markets. Worse, that is LANDLINE broadband. Wireless is going to be worse. If you depend on bandwidth for your mobile device, you might have problems.

  5. @tz:

    I’ve been thinking about the Thrive, as well. Might test-drive one at Fry’s.

    Personally, I’m not sure that Apple relying on ever-increasing bandwidth is a huge problem, for the same reason that Microsoft made a killing in the 90s by relying on ever-increasing compute and memory.

    Sure, not every potential customer will have the bandwidth easily accessible, but Apple has to look at the big picture. For example, in China, if you want a real Apple, your best bet is probably to first try to find a real Apple store (apparently harder than you would think — there are only 4 of those, but a zillion knock-offs). Once you’re at the real Apple store, you can use their free wifi for an upgrade, and they can try to sell you more stuff.

    In any case, if you happen to be an Apple customer without bandwidth, you’re probably not as valuable to them because you won’t be buying as much stuff anyway. For two reasons: (1) you can’t download other stuff they try to sell you if you don’t have the bandwidth; and (2) area bandwidth availability is increasingly a better proxy for average income.

    So (especially if Apple is capacity limited) making their products more appealing to those with more bandwidth might be a good thing from their perspective.

  6. Isn’t wireless bandwidth limited? So having an increasing number of users using an increasing amount of data, you are stuck in a zero-sum, degrading everyones usage abilities?

  7. Bryant is correct: tablets shipped is not “market share.” Check this link, which I got from Daring Fireball: if Honeycomb is in use on .9% of Android devices, and there are now 135 million Android devices, that’s nowhere near 30% of the 28.73 million iPads sold. Is there really a giant wave of Android 2.x tablets to make up the difference?

  8. Good point. On current trends, I think we’re looking at market cap of $8 billion or less a year out. Maybe a lot less.

    That’s a fairly bold prediction. It assumes a lot. for one that Nokia’s feature phones won’t continue to sell well, that Symbian-related sales will continue to tank at currently projected rates and that Nokia won’t find other ways to stay afloat. It also assumes that the market will behave rationally.

    Note to Apple worshippers: Market capitalization is simply computed by taking the stock price and multiplying it by the number of outstanding shares. They’re not as useful as a basis for comparison between two companies that they are a basis for a comparison of the same company over time.

  9. The article that Bryant mentioned had a reference to Analysys International. I found an article in reasonably passable Mandarish on their site:

    http://english.analysys.com.cn/article.php?aid=105770

    Note: This article was recently published, but appears to be about Q1. It won’t match up with more recent data

    There are several insights to be teased out of this article. For a start, it makes Nokia’s prospects look even grimmer on mainland China. The total number of handsets sold is increasing quarter on quarter, so (a) it’s not like a temporary market decline — people are buying handsets, just not Nokia’s. Next, only 30% of the handsets sold in the quarter were smartphones, so (b) featurephones are still selling like gangbusters (just not Nokia’s), and (c) there is a lot of room for conversion to smartphones (which Nokia hasn’t managed to capitalize on).

    If I read the article correctly, iOS’s installed base increased from 5.4% of the smartphone market to 6.1%.

    Symbian’s installed base of the smartphone market decreased by 0.3% to less than 60%. Interestingly Nokia’s smartphone sales were up 5% quarter on quarter. I guess this means that where Nokia is really taking it in the shorts is on margin and on declining featurephone sales.

    RIM is outperformed the overall smartphone market with a 9.2% sales increase.

    It does look like Apple is doing well in China, with its Q1 installed base increase apparently almost double that of Android’s installed base increase. Unfortunately, for Android, only the 0.4% increase was mentioned, so we have no real idea of Android’s penetration in the Chinese market.

    Finally, the real wildcard in this report: “19.91 million smartphones have been sold in Q1, 2011 (excluding smuggled and copycat mobile phones)

  10. @Jeremiah Shepherd

    Isn’t wireless bandwidth limited? So having an increasing number of users using an increasing amount of data, you are stuck in a zero-sum, degrading everyones usage abilities?

    Not necessarily, no. The biggest limitations on wireless data transmission rates are related to range and geography.

  11. @Jeremiah Shepherd:

    So having an increasing number of users using an increasing amount of data, you are stuck in a zero-sum, degrading everyones usage abilities?

    Short answer: No. Keywords for research include MIMO, beam-forming, and microcells. (That last one assumes wires to take the data closer to the mobile device.)

    @PapayaSF:

    Bryant is correct: tablets shipped is not “market share.”

    Many shipped Android tablets don’t access the Android Market. Your daring fireball link won’t take those into account. It is an open question how that affects the “ecosystem,” but it certainly seems to put to rest the myth that only Apple can make a sellable tablet because nobody else has the economy of scale necessary to do so.

  12. > The Android tablet study has the same tricky flaw that those original Samsung sales reports had — it’s reporting tablets shipped rather than tablets sold.

    Absolutely correct.

    If we look at actual units sold, Apple has sold very nearly 29 million iPads. We can estimate Android tablets sold since we know the total number of Android devices in use (135 million) and the percentage of those units running honeycomb (0.9%). This means that we are looking at a number somewhere between 1 and 2 million total Android tablets sold – and probably towards the lower end of that range. Based on these numbers (and I am assuming that webOS and Blackberry tablet sales are negligible – probably in the hundreds of thousands) Android has less than 5% of the tablet market. This is an estimate, and it is hard to know the exact numbers, but there is absolutely no way that Android has anything like 30% of the market.

  13. @Tom:

    This means that we are looking at a number somewhere between 1 and 2 million total Android tablets sold – and probably towards the lower end of that range.

    Complete, utter, bullshit. Barnes & Noble alone have sold over 3 million Android tablets.

  14. @Patrick Maupin:

    Short answer: No. Keywords for research include MIMO, beam-forming, and microcells. (That last one assumes wires to take the data closer to the mobile device.)

    The more fixed infrastructure we have, the better our wireless infrastructure will be. Seems counterintuitive, but I had the same questions Jeremiah had and this was the conclusion I came to. :)

  15. @Morgan:

    > The more fixed infrastructure we have, the better our wireless infrastructure will be.

    Absolutely correct. Also, in my quick typing, I forgot one searchable keyword that is just the logical extension of one of the others: in some quarters, “femtocell” is the new (possibly smaller) “microcell”.

  16. @Patrick

    >Complete, utter, bullshit. Barnes & Noble alone have sold over 3 million Android tablets.

    Those are ebook readers. They’re not running honeycomb, and they’re not in the same market as full fledged tablets. They happen to be based on Android, yes, but that doesn’t put them in the same space as devices like the iPad and the Galaxy Tab.

  17. Tom — the wild card is the Nook. I don’t think it runs Honeycomb, and it’s had some degree of success. Although at that point I’d argue there are two “tablet” markets; the Nook isn’t competing that directly with the iPad or other Android tablets. Either way, we don’t know if the original study includes the Nook or not.

    Patrick — nice find. There’s something wrong somewhere, though. The original article I linked to cites Analysys as follows: “At the end of 2010, Apple had 8.3 percent of China’s smartphone market, an increase from the 5.4 percent it held at the beginning of the year, according to Beijing-based research firm Analysys International.” But your link has it as 6.1% in Q1 2011? Possibly the original is getting the dates wrong, and they should have said that iPhone was at 5.4% at the end of the year. On the other hand, no idea where the 8.3% came from in that case.

    If iOS was at 5.4% in Q1 2010, and they had 250% growth since that time, and (to be conservative) 100% of existing customers upgraded from a 3GS, that’d put Apple at 8.1% now. I think I’m making too many assumptions there to be comfortable with that number, though.

  18. Both the Nook numbers and the ASUS numbers are shipments, not sales. I wouldn’t be surprised if the Nook had a really high sell-through, but the distinction is still worth making.

  19. i think Meego was the last chance for nokia for standing strong as an independent company ( at least in Asia market ). but now they are in microsoft’s Jail. and the future is so clear … !

  20. @Tom:

    > Those are ebook readers. They’re not running honeycomb, and they’re not in the same market as full fledged tablets.

    I must be weird, then. I’ve been considering whether I want to buy a small tablet like the Nook Color, or a bigger one like the Samsung…

  21. @Bryant:

    Yes, those analysys numbers look confusing, but…

    > If iOS was at 5.4% in Q1 2010, and they had 250% growth since that time,

    Remember, that 250% was for “greater China”. We’re comparing Apples and potatoes here…

  22. @Tom, Bryant:

    There’s been plenty of anecdotal evidence ASUS has had trouble meeting demand. The initial launch in the UK sold out almost instantly. In the US, Phandroid was treating Transformer sightings like Wii sightings from a few years back. For weeks, they’d barely get done posting that someone had the Transformer before they’d have to post the update that it was sold out again.

    In my (Asian) neck of the woods, I personally had to go to 10+ stores to find one that had the Transformer in stock. Even then they had to send me to a different branch to actually get my tablet (the last one) and I had to special-order the keyboard dock.

    Beyond that, the article also says that revenues at ASUS’s assembler for the Transformer (Pegatron) are up 51.7% (attributed to strong Transformer sales) and that ASUS is increasing their orders and targets for the rest of the year. Would they really be stupid enough to do that if the Transformer weren’t selling?

  23. @Patrick:

    You’re not weird. You can’t call the Nook Color just an ebook reader now that it can play Angry Birds (and that’s without hacking it at all).

  24. @Patrick

    >The Apple number is also shipments.

    The 29M number I cite does not come from the report. Apple actually releases official quarterly sales numbers. If you add them up you get to 28.73M. The reason this number looks the same as the shipments number is that for Apple there is no difference – they’re selling them as fast as they can make them.

    >What’s your point?

    My point is that share of shipments doesn’t mean a thing. Just because Android tablet manufacturers are making a lot of tablets it doesn’t mean they are going to sell those tablets. In share of units *sold* Apple is still totally dominant.

    > I must be weird, then. I’ve been considering whether I want to buy a small tablet like the Nook Color, or a bigger one like the Samsung…

    That’s fine. You might also be conflicted whether you should buy a laptop or a tablet, or you might not be sure whether a bicycle is a better fit for your transport needs over a truck. Doesn’t mean they are in the same product category.

    And, as Bryant points out, even the 3M number for Nooks is *shipments*, not units sold.

  25. @Ravi

    >There’s been plenty of anecdotal evidence…

    All very interesting (sincerely) but nothing you have said is terribly solid. One nice thing about Apple is that they actually give us real sales numbers. The Transformer has been out since March; why no official numbers from Asus?

  26. @Tom:

    > And, as Bryant points out, even the 3M number for Nooks is *shipments*, not units sold.

    That was as of March, btw. And it was a total reflecting ever-increasing shipments. Is B&N really in it to see how many they can stock in each store, or is it just possible that they are increasing shipments because they are increasing sales. Do you really think that B&N hasn’t easily sold more than your laughable estimate of only 1 million Android tablets?

    > One nice thing about Apple is that they actually give us real sales numbers.

    Aside from those units sold at Apple stores, do you have any evidence whatsoever that Apple’s sales/shipment numbers are calculated any differently than anybody else’s? Look, 4.6 million Android tablets were shipped last quarter. According to those who, unlike me, segregate the market into tablets and eReaders. Do you really think that out of that and all previous shipments, only 1 million have made it to consumers?. You make this stupid claim, and then dance around it. Dance, monkey, dance!

    > That’s fine. You might also be conflicted whether you should buy a laptop or a tablet, or you might not be sure whether a bicycle is a better fit for your transport needs over a truck. Doesn’t mean they are in the same product category.

    What can I do on a “tablet” that I can’t do on a Nook Color? I like Ravi’s Angry Birds benchmark, which, btw, is something I don’t expect to see on any e-ink product any time soon.

  27. I actually pre-ordered my Thrive, and I’ve been playing with it for a few days now. Here are my thoughts:

    I purchased a tablet at all because I’m a technical consultant and SaaS entrepreneur who’s likely to be working on phone/tablet apps in the near future, and I thought it’d be a good idea to get a feel for something besides my beloved Pre, which is starting to wobble when it goes uphill. Between Android and iOS, Android was the obvious choice because it’s somewhat more open and because the SDK is something I’m fairly familiar with.

    I purchased the Thrive over other Android tablets because it’s running Honeycomb, it has full-sized HDMI and USB host ports, and Toshiba’s at least trying to send the “open platform” message by such routes as putting the user-replaceable battery in the big-font feature list. Not having WWAN support isn’t a big deal for me; if I decide that it’s worth paying for, I’ll add a MyFi and get data for my laptop at the same time. If I really need cell data in a pinch, well, the great thing about webOS is that tethering’s always possible–if nothing else, I can fire up Terminal and ‘echo 1 > /proc/sys/net/ipv4/ip_forward’! Additionally, after learning from my experience with an early-early ebook reader, I always try to purchase devices with standard charging ports, and the Thrive uses a run-of-the-mill pin-barrel connector (at 19V, but that appears to be common for laptops).

    My initial impression is that the screen is about equivalent to the iPad’s: I actually prefer the iPad’s 4:3 aspect ratio, but it makes sense that a 16:9ish screen is going to be more popular with people who want to watch movies. The vibration feature is crisp and not overdone, and the audio system is surprisingly good for a tablet. I think I would be interested in an Eee Transformer-style keyboard attachment. Connecting a USB keyboard or mouse(!) Just Works, and Honeycomb theoretically permits userland drivers for things that the distro doesn’t support itself. I’m thinking a USB-MIDI 25-key keyboard could be interesting with a tablet.

    The one major valid complaint I’ve seen against Android tablets generally is the lack of tablet-specific apps. This is true (though, of course, small-screen apps run fine on it), and I place the blame squarely on Google for taking such an insane amount of time to get 3.x released for phones: Writing tablet apps cleanly requires you to use Fragments, an SDK feature that isn’t available on 2.x, meaning modern-style tablet apps aren’t backward-compatible.

  28. Archos sold around a million tablets in Q4 2010, and has supposedly seen sales rise. They aren’t a huge presence in the US market, but I love my Archos 7 that I gave to my daughter (as I’m going to buy one of their new 8″ ones in Sept. that have faster processors than any other Android tablet OR iPad).

    They already have ‘real’ usb ports you can hook up to external drives, keyboards, etc (it’s a checkbox: “host mode” or not). “Real” external charger. Better price/performance than the other “name brand” android tablets, solid hardware, good looking, and you can get quarter terabyte HD’s in them instead of flash storage if you want. They just aren’t big in the US. But if I can’t (or don’t want) to afford Steve Job’s design aesthetic, French style is second best imho.

    I don’t think they’ll ever be the force in tablets that some of the other bigger players could be, but it’s nice gear and they’ve been in the portable media player space for so long that they know how to make durable stuff that looks good too. So I’m waiting with bated breath to spend my Q2 bonus money from work on an Archos 8 G9 in Sept (dual core cpu faster than those tegra 2’s, HDMI out, optional USB 3G GSM wireless adapter that has an interal slot to stick it in and also works on PC’s and Mac’s….still pondering if I want the 16GB flash or the 250GB HD).

    Regarding the Nook, if you can’t manage to download an sd card image and properly format the card and drop it on there, tons of people on ebay will sell you one that will either 1) multi-boot your Nook into any one of a number of android versions/builds or 2) just plain wipe the sucker with the new modded android. Rooting other Android tablets is fairly trivial, as is adding the Market to ones that don’t ship with it (all of them aside from the Samsung and Motorola ones to date, the new Archos line will have it legally in Sept too). Just flashed my Archos 7 with Honeycomb and the Market before I gave it to my daughter, fastest way to take all my business related stuff off of it and give it an upgrade at the same time (she’s 3 and had, thank god, only MADE and not deleted tasks in my task manager. She can’t use a mouse well yet but has NO problem with a touchscreen).

    I think that the whole android tablet market is completely a mystery so far as sales numbers go, as all the chinese generics are not getting reported in any analysts numbers, as they come right out of the factory to the wholesaler (webstore, ebay seller, whatever), and may or may not have the Market “sideloaded” onto them. The Amazon App Store has made lots of sellers of generic Android tabs care a lot less about hacking google’s Market onto them now too. It’s be interesting to see how many devices that come up as “unknown XXX” have shown up on either of them, which is what non-mainstream tabs show up as.

  29. @Patrick

    >Do you really think that B&N hasn’t easily sold more than your laughable estimate of only 1 million Android tablets?

    Ok, I said 1-2 million, but whatever.

    That estimate was based on a calculation of tablets running Honeycomb, which is why the Nook was not included. I guess it is a fine line. There is no clear definition of what constitutes a tablet, so I guess we could argue about this all day. Is the iPod Touch a tablet? I don’t think so, but maybe somebody will disagree and ask ‘what can the touch do that a tablet can’t’.

    Let’s admit that the Nook is a tablet for the sake of the argument. How many have been sold? We don’t know. How many have been shipped? I am not even sure of this, since a lot of the reports I see have cited this 3M number as an ‘estimate’ from ‘sources’.

    There is a lot of hand waving and obfuscation around these numbers. Why can’t these companies just report their sales numbers?

    >Aside from those units sold at Apple stores, do you have any evidence whatsoever that Apple’s sales/shipment numbers are calculated any differently than anybody else’s?

    Not sales/shipments: SALES! That is my whole point.

    If a company tells me ‘we have sold x units of such-and-such a product’ I take it at face value. I guess we could start wondering if they are lying, but at that point we might as well just end the discussion.

    Apple tells us they have *sold* 28.73M iPads (here is the latest quarterly announcement – http://www.apple.com/pr/library/2011/07/19Apple-Reports-Third-Quarter-Results.html).

    I would be really happy if somebody could point me towards a comparable announcement from an Android manufacturer.

    My only point in all this is that Eric seems to have misinterpreted the report he cites. Easily done, I might add. The report is *not* evidence that Android has 30% of the tablet market share.

  30. @Christopher Smith: Writing tablet apps cleanly requires you to use Fragments, an SDK feature that isn’t available on 2.x, meaning modern-style tablet apps aren’t backward-compatible.

    This is not really true. Fragments are backward-compatible all the way back to Android 1.6. I think the problem is different: most tablets sold are cheap “unofficial” clones without Android market and this means there are no money in tablet-optimized Apps for Android.

  31. > This is not really true. Fragments are backward-compatible all the way back to Android 1.6.

    Thanks for the link. I hadn’t seen it, and the API docs don’t mention the backport.

  32. @Tom:

    That estimate was based on a calculation of tablets running Honeycomb, which is why the Nook was not included.

    Not just the Nook. There are a lot of Android tablets sold as tablets rather than ereaders, which haven’t yet upgraded to Android 3.x. And those are real tablets by most people’s definition.

    Ok, I said 1-2 million, but whatever.

    You’re forgetting “and probably towards the lower end of that range” but, as you say, whatever.

    Not sales/shipments: SALES! That is my whole point.

    Apple tells us they have *sold* 28.73M iPads (here is the latest quarterly announcement – http://www.apple.com/pr/library/2011/07/19Apple-Reports-Third-Quarter-Results.html).

    Yes, but some of those could be sitting at third party stores. IOW, as I pointed out, there is no real difference between Apple numbers and others’ numbers, except for those sold direct to customers from Apple (a large number to be sure).

    If a company tells me ‘we have sold x units of such-and-such a product’ I take it at face value. I guess we could start wondering if they are lying, but at that point we might as well just end the discussion.

    Most companies consider things sold when they ship. Accrual accounting.

    I would be really happy if somebody could point me towards a comparable announcement from an Android manufacturer.

    Well, this should make you happy: (btw, it wasn’t all that hard to find)

    http://www.bgr.com/2011/03/23/samsung-again-forced-to-defend-tablet-sales/

    OK, so did Samsung “sell” over 3 million Android tablets, or are total Android tablet sales “probably” at the low end of the 1-2 million range?

    My only point in all this is that Eric seems to have misinterpreted the report he cites. Easily done, I might add. The report is *not* evidence that Android has 30% of the tablet market share.

    Any tablet that was shipped but not yet sold to a consumer is sitting as inventory somebody’s books — either the manufacturer’s or the store’s. Accountants hate that, even in an era of cheap money, especially in the deflationary electronics market where you can sell something next month for 95% of what you could have sold it for this month.

    Computers and computer-to-computer B2B communication over the internet make forecasting adjustments much easier all up and down the supply chain than even 5 years ago. You can bet that companies aren’t just building these things for the sake of letting digitimes or some other third party say that they built them.

  33. @Patrick Maupin: Yes, but some of those could be sitting at third party stores. IOW, as I pointed out, there is no real difference between Apple numbers and others’ numbers, except for those sold direct to customers from Apple (a large number to be sure).

    Even these are not guaranteed. For example it was easy to buy iPad2 in Russia way before May 31 (official start of sales). Most shops even had choice: US model, UK model, DE model, etc.

    Do you really believe these shops sent mails personally to Jobs and asked him to adjust statistics?

    Shipments are easy to count, sales are not. Apple has it’s own stores so it’s easier for them to estimate sales – but even these are estimates. I think ESR have another step to add to his article

  34. @Patrick Maupin

    “Yes, but some of those could be sitting at third party stores”

    Here is what Apple said on the iPad (http://seekingalpha.com/article/280344-apple-management-discusses-q3-2011-results-earnings-call-transcript):

    “Turning to iPad. We continue to be thrilled with its incredible momentum. We sold a record 9.2 million iPads during the June quarter compared to 3.3 million in the year ago quarter, an increase of 183%. We were able to increase production sequentially by over 4.5 million and we sold every iPad we can make. We launched iPad 2 in an additional 36 countries. And combining the original iPad and iPad 2, we ended the quarter with distribution in a total of 64 countries. Recognized revenue from sales of iPad and iPad accessories during the quarter was over 6 billion, compared to 2.2 billion in the year ago quarter, an increase of 179%. We ended the quarter with about 1.05 million iPads in channel inventory, a sequential increase of about 200,000, which was well below our target range of 4 to 6 weeks. Today, 86% of the Fortune 500 are deploying or testing iPad within their enterprises, up from 75% last quarter.”

    Is there any question about whether they are able to sell the iPads or if they sit on the shelf for more than a brief period of time? For all intents and purposes iPads shipped = iPads sold.

    1. >For all intents and purposes iPads shipped = iPads sold.

      I agree this is probably true. What I don’t get is the implied assumption that the sell-through rate on Android tablets isn’t just as good. As Patrick pointed out, nobody wants to carry inventory and retailers aren’t buying tablets for their health.

  35. @patrick

    >Not just the Nook. There are a lot of Android tablets sold as tablets rather than ereaders, which haven’t yet upgraded to Android 3.x. And those are real tablets by most people’s definition.

    Yes, the problem is I just don’t know of a way to estimate pre-3.0 sales. I was under the impression that there was a consensus that anything non-Honeycomb (perhaps excepting the Nook, which has a specialised UI) was a pretty terrible tablet user-experience and therefore was a sales flop, since it was designed for phones and not tablets. Perhaps I am wrong, but, again, I just don’t have any numbers to indicate otherwise.

    > Yes, but some of those could be sitting at third party stores. IOW, as I pointed out, there is no real difference between Apple numbers and others’ numbers, except for those sold direct to customers from Apple (a large number to be sure).

    It’s certainly true that Apple does have a reseller programme, but the majority of their sales are direct. I admit I don’t have concrete numbers to hand, but in the past I have seen reports indicating that about 3/4 of their hardware revenues come from direct (i.e. web store and their retail stores) channels.

    >OK, so did Samsung “sell” over 3 million Android tablets, or are total Android tablet sales “probably” at the low end of the 1-2 million range?

    Sorry, but I honestly don’t see the sales announcement from Samsung. I tried to look up the release they reference in the article, but could not find it. I believe that the Tab runs a pre-Honeycomb version of the OS (correct me if I am wrong). If this is the case then these sales would not be included in my 1-2M number. I know I wasn’t clear initially, so let me be now: that estimate was for Honeycomb tablets only. I honestly don’t know how to estimate pre-3.0 sales given the lack of official numbers.

  36. > For all intents and purposes iPads shipped = iPads sold.

    I can believe that. But I can also believe the same of the Asus eee transformer. Tom apparently can’t, because all that is anecdotal, so he should ignore the Apple anecdotes as well.

  37. @Patrick

    >Not just the Nook. There are a lot of Android tablets sold as tablets rather than ereaders, which haven’t yet upgraded to Android 3.x. And those are real tablets by most people’s definition.

    Right. The only problem is that I don’t know of any way to estimate sales for these devices, given the lack of official sales numbers.

    >Yes, but some of those could be sitting at third party stores. IOW, as I pointed out, there is no real difference between Apple numbers and others’ numbers, except for those sold direct to customers from Apple (a large number to be sure).

    It is certainly true that Apple does have a reseller programme. However, the majority of their sales come through direct (i.e. web store and their retail stores) channels. I admit I do not have concrete numbers to hand, but I seem to recall seeing reports in the past that indicate on the order of 3/4+ of their hardware revenues coming from direct sales rather than third-party channels. And most of that is from the web, which is by far their biggest single channel. I think that alone is more than 60%.

    >Well, this should make you happy: (btw, it wasn’t all that hard to find)

    Sorry, I’m not trying to be difficult, but I honestly do not see the sales announcement from Samsung here. I tried to look it up on their PR site but could not find it. The article references a ‘statement’, but they don’t quote any numbers or give a link.

    >OK, so did Samsung “sell” over 3 million Android tablets, or are total Android tablet sales “probably” at the low end of the 1-2 million range?

    I believe that the tab runs 2.x Android (perhaps it is upgradeable now, I am not sure) and therefore these sales (whatever they may be) will not be included in my 1-2M estimate. Just to be clear (I know I wasn’t initially; sorry): I estimate 1-2M *honeycomb* tablets sold. I just don’t know how to estimate pre-3.x sales, given the lack of official sales numbers. The reason I initially said Android tablets (without specifying honeycomb) is that I honestly thought that the whole pre-Honeycomb Android tablet thing was a total flop, but maybe I am wrong. After all, 2.x wasn’t even designed as a tablet OS, it was for phones, so I wouldn’t have expected much success.

    Honeycomb actually looks like a reasonable attempt, so I am surprised we have not seen much uptake there.

  38. I just found this page, which may provide some useful data:

    http://developer.android.com/resources/dashboard/screens.html

    Here we have active Android devices by screen size. I looked it up, and the ‘xlarge’ category is 7-10 inches, which is what you would expect for a tablet. The table tells us that 0.9% of active android devices are tablet size, which corresponds *exactly* with their number for active Honeycomb devices.

    I am not sure how to interpret this, since there must indeed be a large number of pre-Honeycomb tablet-sized devices out there. Perhaps devices like the Nook are not ‘activated’ in the same way that normal Android devices are, since there is (I believe) a custom UI layer on top of Android on the Nook.

  39. @Tom:

    Note: This data is based on the number of Android devices that have accessed Android Market within a 7-day period ending on the data collection date noted below.

    I don’t think the Nooks have access to the Market by default.

  40. @Christopher

    Ah, interesting. I was trying to see how they gathered the data.

    This may explain the discrepancy. The only apps for Android designed specifically for tablets are for Honeycomb. Therefore I suppose that pre-Honeycomb tablet users would have little reason to access the store. And, as you point out, Nooks cannot access the store.

    I would have thought we would see some access from pre-Honeycomers though, as I understand that you *can* run non-customised apps on these devices, even though the experience will be sub-par.

    Perhaps it is the case that not all Honeycomb users access the store, and that only a very small number of pre-Honeycomb users do so.

    Still, what we see here is that there are only about 1.2M Android tablet users out there who are accessing the app store on a regular basis. That alone is a pretty damning statistic.

  41. @Tom:

    Sorry, I’m not trying to be difficult, but I honestly do not see the sales announcement from Samsung here.

    Apparently, that sort of thing normally happens during their earnings call. See, e.g.:

    http://www.pcmag.com/article2/0,2817,2376893,00.asp

    (Be sure to check the update at the bottom of the article)

    I honestly thought that the whole pre-Honeycomb Android tablet thing was a total flop, but maybe I am wrong.

    I think that maybe you are partly right, but 7 or 8 months out of date. I would have liked for you to be completely wrong about this at this point in time, but google’s still sitting on the Honeycomb source code, and presumably aren’t giving it to all comers. Theoretically, we should see source for Ice Cream Sandwich in Q4. I have to admit though, I’m getting a bit annoyed at the delay myself. I think we’ll see a lot more market activity when source is available, but obviously google has experience with fragmentation they’re trying to avoid. I sure hope they’re using the intervening time to future-proof the software so that upgrades are so painless that no sane vendor would ever balk at one.

    Just to be clear (I know I wasn’t initially; sorry): I estimate 1-2M *honeycomb* tablets sold.

    That’s actually a reasonable guess. I personally think 2M is at the low end of what has shipped, but as you point out, hard data is hard to come by. But in any case, I think Honeycomb shipments are ramping quickly.

    Still, what we see here is that there are only about 1.2M Android tablet users out there who are accessing the app store on a regular basis. That alone is a pretty damning statistic.

    Do you really think that everybody who buys a tablet will access the market more than once a week, especially during the summer? In any case, the data was collected during a 7-day period ending on July 1, 2011, which partly explains why I think your estimate is low. Asus has probably sold 350K tablets since then, for a start, and Samsung’s 10 inch tablet is out now, and Vizio’s got a new tablet, etc.

  42. @Patrick

    >Apparently, that sort of thing normally happens during their earnings call. See, e.g.:

    If this 2 million number is correct, then that is quite impressive. However, I really question why Samsung would refuse to confirm it, and why, if PCMag has seen a transcript of the call as they say, they can only confirm that sales were ‘discussed’ rather than confirming the number specifically. Also, Apple puts recordings of all its earnings calls on its website, so I don’t see why Samsung would not do the same.

    >Do you really think that everybody who buys a tablet will access the market more than once a week, especially during the summer?

    There was some analysis done at the start of this year that indicated an average of 60 apps downloaded per iOS device (iPad, iPhone, iPod Touch) and I am sure it cannot be that different for Android. I don’t have that many apps, and I am constantly connecting to the store just to get updates to the apps I have, much less get new ones. Anybody with a significant number of apps will be connecting to the store quite regularly just to keep up to date.

    But, regardless, what we are looking at with those breakdown numbers from google is the *proportion* of device sizes, not the absolute number, and that is what I am basing this analysis on: 0.9% x 135M total devices = 1.2M.

  43. @Tom:

    If this 2 million number is correct, then that is quite impressive. However, I really question why Samsung would refuse to confirm it… Also, Apple puts recordings of all its earnings calls on its website, so I don’t see why Samsung would not do the same.

    Apparently the shipments were 2 million even earlier. They probably have confirmed it; I don’t speak Korean.

    http://blogs.wsj.com/digits/2011/01/31/samsung-galaxy-tab-sales-actually-quite-small/

    There was some analysis done at the start of this year that indicated an average of 60 apps downloaded per iOS device (iPad, iPhone, iPod Touch) and I am sure it cannot be that different for Android… Anybody with a significant number of apps will be connecting to the store quite regularly just to keep up to date.

    I thought there were only about 60 tablet-optimized apps for Android?

    0.9% x 135M total devices = 1.2M.

    That makes a bit more sense, but I’m still not sure that’s all that valid. Tablets and phones have different use cases; a lot more casual surfing gets done on tablets, for example. And as we have seen with the Nook Color (and a lot of Chinese knock-offs, I’m sure) not every Android has market access. Also, notice the overlap at 7″ between “large” and “xlarge” screens:

    http://developer.android.com/guide/practices/screens_support.html

    Also note that, starting with Android 3.2, these screen size groupings are being deprecated, so will become rather useless. In any case, I’m not sure how good the data is — would developers who don’t read English all that well who are rushing out to ship a product notice if they classified the screen wrongly?

    In other news, most public companies struggling with sell-through will make that information public (or worry about shareholder lawsuits). Here is an example where that is happening:

    http://www.bgr.com/2011/06/24/acer-cuts-2011-tablet-sales-forecast-in-half-to-2-5m-units-citing-poor-sell-through-competition/

    I don’t see this sort of announcement from, e.g. Asustek or Samsung right now…

  44. @Patrick

    The upshot of all this is that we lack hard data on Android tablet sales, which has kind of been my point from the start, and I think you have recognised this.

    We can try to piece a rough picture together from scraps here and there, but you can slice it a number of different ways. The reason I really think it is different for Apple’s numbers is that we have regular, official sales (and they explicitly use the word sales, rather than something like ‘shipments’ or ‘activations’) reports from Apple posted on their website directly. I trust these numbers because Apple controls their distribution chain to a degree that no other manufacturer can match, since most of their sales are through their own channels. I still haven’t seen an official statement from any of the Android companies saying ‘ we sold x million units to consumers over x period of time’.

    Reading second-, third-, fourth-hand reports on blogs and news sites makes it even more obvious to me that nobody really has solid sources for the Android numbers we are seeing. An anonymous tip crops up on one site, and pretty soon the whole web is reporting it as fact. I see the terms ‘sales’, ‘shipments’, ‘activations’ used interchangeably and without regard to their meaning. I have also seen different reports giving conflicting data, for example I saw one article reporting 400k shipments a month for the Asus tab, and then another report quoting a source from Asus saying that they shipped 400k for the months of March and April combined. I am afraid that I just don’t trust these reports anymore; it is clear that the people writing these things just don’t care.

    That is why I keep asking for official numbers straight from the source, but I haven’t seen any. They must be out there, but I just can’t find them for all the noise.

  45. “Those are ebook readers. They’re not running honeycomb, and they’re not in the same market as full fledged tablets. They happen to be based on Android, yes, but that doesn’t put them in the same space as devices like the iPad and the Galaxy Tab.”

    True, but they are the ‘foot in the door’ that will help lead to general, non-geeky public demand for tablets. People will replace their old ereaders and find that the new ones have more features for less money than the old one cost, and maybe they use the new features and like them and ….

    It’s the same thing that happened to cell phones and is another path that will lead to Android dominance of the field. I, for one, welcome our new B&N overlords.

  46. LS: “True, but they are the ‘foot in the door’ that will help lead to general, non-geeky public demand for tablets. ”

    Cool, does that mean that Apple gets to include the Ipod Touch in its Ipad figures?

  47. @LS, Michael Hipp:

    Tom’s already shifted from “only 1-2 million (probably lower end)” to “Apple’s the only real tablet company because nobody else will give us real numbers, and the only real number I believe is a pretty damning statistic.” I completely disagree with that assessment, but he has fallen back to a position based on opinion rather than purported fact, and since I fully expect the facts that will destroy his position (though probably not his opinion) will be made completely clear in a couple of months, my work here is done for now.

    @SHG:

    > Cool, does that mean that Apple gets to include the Ipod Touch in its Ipad figures?

    Everybody else was discussing Android in the context of smartphones, until Steve Jobs and his minions said you have to consider the ecosystem, and that means that iOS is kicking butt because of all the iPods. It makes sense that as iPods fade into the distance, you wankers will concentrate on size again, but it doesn’t really matter because no matter how you slice it, iOS has lost in smartphones and, unless things change drastically, will continue to lose in smartphones, and will, in less than a year, be losing dramatically in whatever handheld rectangular objects you wish to discuss.

    In other news, Android is kicking some serious smartphone butt in the UK:

    http://www.engadget.com/2011/07/22/comscore-androids-uk-market-share-explodes-as-apple-overtake/

    BTW, O2 had an exclusive iPhone distribution agreement, but it apparently ended in late 2009. You can now get the iPhone subsidized from 6 different carriers in the UK, or buy it unlocked full-price from the Apple store. So don’t give me crap about how Apple always wins hands-down head-to-head if they’re not hamstrung by an exclusive carrier deal.

  48. @ Maupin

    >It makes sense that as iPods fade into the distance, you wankers will concentrate on size again, but it doesn’t really matter because no matter how you slice it, iOS has lost in smartphones and, unless things change drastically, will continue to lose in smartphones, and will, in less than a year, be losing dramatically in whatever handheld rectangular objects you wish to discuss.

    You seriously think that in less than a year, Apple will have “dramatically” lost its marketshare lead in tablets? Wow… I can’t wait to revisit here in twelve months. I’m sure you’ll change the goalposts buy then. How will you measure this “loss?” Marketshare? Profit? Let’s be explicit here so that you don’t waffle in a year.

    1. >You seriously think that in less than a year, Apple will have “dramatically” lost its marketshare lead in tablets? Wow… I can’t wait to revisit here in twelve months. I’m sure you’ll change the goalposts buy then. How will you measure this “loss?” Marketshare? Profit? Let’s be explicit here so that you don’t waffle in a year.

      Marketshare. Patrick and I expect that, having already run a successful technology disruption in the smartphone marked, Android is going to do the same thing in the tablet space.

  49. Noticed a link at the bottom of that last article to the comscore original:

    http://www.comscoredatamine.com/2011/07/google%E2%80%99s-uk-smartphone-audience-grew-by-634-percent-since-may-2010/?utm_source=feedburner&utm_medium=%24{datamine}&utm_campaign=Feed%3A%20comscoredatagems%20%28The%20comScore%20Data%20Mine%29

    And when I realized that comscore has an entire new internet domain for data snippets like this, I found they stole (and extended) one of esr’s charts:

    http://www.comscoredatamine.com/2011/06/u-s-trend-in-smartphone-subscribers-by-platform/

    Well, OK, they probably did it all on their own. But still…

    And I found this, which shows that Samsung doesn’t do nearly as well in smartphones in the US as they do globally:

    http://www.comscoredatamine.com/2011/07/top-smartphone-oems/

    I expect a major push from them now that AT&T is not the only iPhone carrier and Verizon isn’t relying merely on Motorola and HTC to counter. They’ve been holding back (e.g. not even releasing the Galaxy S II yet here), but they’re probably planning on unleashing a lot of stuff this quarter before the iPhone 5 hits.

    1. >And when I realized that comscore has an entire new internet domain for data snippets like this, I found they stole (and extended) one of esr’s charts:

      Oh, good. I’m going to link to this, which they generated using more data than is in their press releases.

  50. @Patrick

    “lost in smartphones” “lose in smartphones” “losing dramatically”

    That root word you keep using. I do not think it means what you think it means.

  51. @daz

    >“lost in smartphones” “lose in smartphones” “losing dramatically”
    >That root word you keep using. I do not think it means what you think it means.

    Yes, unfortunately Eric and Patrick have the idea that ‘winning’ in technology means achieving some kind of dominance over the entire population. Apple, on the other hand, only cares about two things:

    1. Making money.
    2. Making great products.

    Everything else is just noise.

    1. >Yes, unfortunately Eric and Patrick have the idea that ‘winning’ in technology means achieving some kind of dominance over the entire population.

      Don’t be any more of an ass than you can help. The whole point of Android is to deny anybody “domination”. This, as I have explained repeatedly, is Google’s grand strategy. Patrick and I are in favor.

  52. Upgrading many “non-google approved” android tablets to have access to the Market is fairly trivial, and much easier than applying a new flash image to most phones. You just plug it into your computer’s usb port and drop the update.img file into the root of your sdcard (or internal storage that masquerades as such), disconnect from the computer (windows, linux, mac, it doesn’t matter) and android detects it and pops up a dialog box saying that it has detected an upgrade and asks if you want to apply it.

    Most currently shipping tablets (all of them right now aside from samsung and motorola last I checked) don’t have market access or google apps installed, and it can take quite a lot of hacking to make an image for them that includes them. Archos made it trivial: they added linux symlinks from the read-only directory where those apps must be with the correct filenames to a writeable part of the filesystem, so you just have to get a copy of the right files (the google framework and the apps you want), drop them in there and sure as bob’s yer uncle you’re in google land.

    Fortunately all that google requires to use the Market, as they don’t check devices ID’s when you connect to it, is a copy of the software and a valid google ID.

    Oh and contrary to many media reports, the VAST majority of android apps look and work just fine on Eclair (2.1) and Froyo (2.2)…the whole ‘not tablet optimized’ business IMHO is merely FUD. I’ve had more trouble with apps on my smaller than typical resolution Huawei Comet phone running running Froyo than I have with my Archos 7, running either eclair or froyo.

  53. @daz:

    >That root word you keep using. I do not think it means what you think it means.

    It does mean what I think it means when it comes to marketshare. If you had been following this blog, you would realize that I honestly don’t give a rat’s ass if Apple sucks all the profit out of the market as long as Android beats it in marketshare.

    More Android phones have been shipped worldwide than Apple phones. And when Android takes off in a market, it usually does so with a vengeance. For example, the marketshare numbers in that UK chart I just posted showed that over the last year, the increase in the number of people who carry Android phones was 2.7 times as big as the increase in the number of people who carry Apple phones.

    @tom:

    Yes, unfortunately Eric and Patrick have the idea that ‘winning’ in technology means achieving some kind of dominance over the entire population.

    No, Android “winning” means to help deny Apple dominance over a really large population, and to use competition to reduce the ill effects of the dominance they have over Apple iPhone customers. If it weren’t for Android, you’d see Apple being a lot more evil than they are already.

    Apple, on the other hand, only cares about two things:

    No, they just care about the “making money” part, but have decided the best way to do that involves a serious amount of control over the customer.

  54. @esr

    >Don’t be any more of an ass than you can help. The whole point of Android is to deny anybody “domination”. This, as I have explained repeatedly, is Google’s grand strategy. Patrick and I are in favour.

    Google’s strategy is based on their recognition that as personal internet access moves to mobile devices their existing advertising channels will no longer be effective. Android is an attempt to correct that by creating a direct channel to the mobile space that they control. It’s a smart move, but it has nothing whatever to do with protecting the people from ‘evil’.

    1. >Android is an attempt to correct that by creating a direct channel to the mobile space that they control. It’s a smart move, but it has nothing whatever to do with protecting the people from ‘evil’.

      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; their goal is to prevent anyone else from getting in their way, which is a different matter.

      This business goal has the effect of defending the users from domination, an effect which Google’s founders have repeatedly said they value. On the most cynical possible interpretation, this is just a play to win positive brand strength for Google. But it doesn’t take much study of Google’s culture or people to see that Google would want this outcome even if there weren’t a solid business case for it.

  55. @Tom:

    Android is an attempt to correct that by creating a direct channel to the mobile space that they control. It’s a smart move, but it has nothing whatever to do with protecting the people from ‘evil’.

    You’re conflating motivation with action. By giving away code and releasing source code, google is relinquishing control. Same as they do with their “data liberation front.” Compare and contrast with Facebook blocking every attempt at easing data out of their systems. Now, it may very well be that google’s non-evilness is situational and that when they start losing badly in search, things will change. But that doesn’t alter the fact that, whatever their motivations, Android is a force for good in the world.

  56. The question repeatedly comes up: Who is killing RIM? Who is attracting dumbphone users?

    Well, in the UK, it would appear that Apple is not doing a very good job at either of those. In May 2010, there were 3.78 million iPhone users, 2.88 million RIM users, and slightly less than 0.75 million Android users.

    Android added 4.7 million users, Apple added 1.74 million users, and RIM, starting from a position a million users down from Apple, added 1.70 million users, a rounding error away from the number Apple added. The UK smartphone market obviously isn’t saturated yet (42% of handsets), but Nokia and Microsoft had small (probably accelerating) declines. Judging from the one-year snapshot, RIM’s still doing OK in the UK, and is as popular with new smartphone users as Apple.

  57. @Patrick

    > Android is a force for good in the world.

    Completely. I have always liked Android, and I am glad that there is at least one strong competitor out there to take on Apple, although I personally prefer iOS. What I object to is this idea that Google has a ‘grand strategy’ to save the world from Apple’s supposed ‘evil’. It’s a fantasy.

    Both of these companies are in the business of maximising their profits. To suggest that Apple is ‘losing’ this battle is insane.

    1. >What I object to is this idea that Google has a ‘grand strategy’ to save the world from Apple’s supposed ‘evil’.

      No fantasy. They do, though it’s not specifically Apple they’re trying to discombobulate. The kind of evil Apple is pushing is hostile to Google’s long-term business model.

  58. Patrick Maupin Says:
    “Now, it may very well be that google’s non-evilness is situational and that when they start losing badly in search, things will change. But that doesn’t alter the fact that, whatever their motivations, Android is a force for good in the world.”

    The result of this, for me, is that Google seems to be the one single large company I can think of where their own interest seem to frequently align with mine. Contrast with, say AT&T, where my ability to do business with them is generally in spite of all their efforts to chase me away.

    I hope the situation with Google continues tho I fear it won’t. But it has at least suggested the possibility that the free market can survive the capitalists.

  59. @Tom:

    Well, I think Apple is evil (in quite a few things they do, though certainly not everything they do), and would be much more evil if Android weren’t around as a competitor. YMMV.

    Both of these companies are in the business of maximising their profits. To suggest that Apple is ‘losing’ this battle is insane.

    There are lots of ways to make money. Some are better for society than others. Apple suing Samsung for a tablet “look and feel” that first showed up in 1973 just doesn’t feel like it’s good for society to me. Again, YMMV.

  60. @Michael Hipp:

    But it has at least suggested the possibility that the free market can survive the capitalists rent-seeking, bogus patent-wielding, would-be monopolists.

    FTFY. Other than that, I completely agree with the sentiment.

  61. Well, I tried to FTFY, but I guess I don’t know how to do a strikeout in wordpress…

  62. >Er, and they “control” it how? Android phones are general-purpose devices, customizable by their users. Android is open source.

    That’s all great, but Google runs the development of Android. Certainly anybody may take the code and do whatever they like with it, but in reality Google knows that only a tiny number of people will actually do so. In reality, users run the code that is given to them. As demonstrated by Honeycomb, Google controls the development process and the release of new code. This gives them a huge amount of influence over the way that users – in practice – will use the OS. Not to mention the fact that they created and continue to lead the Open Handset Alliance.

    I think it is cool that they open-sourced the code, and they have picked up a lot of points with the free software movement, but their motives are to secure and extend their dominant position in online advertising.

    Providing services for free is also classic Google, and it makes perfect sense given their business model. They can do it because they only care about spreading their reach as far as possible. They don’t make their money from selling products and services to their users; they make money by selling their users to their customers.

    I’m not saying they are not also motivated to create great products – they are – but their business is advertising, not saving the downtrodden people of Earth from Apple.

  63. @Tom (and whoever else)

    As an illustration of why ‘losing’ is such a strange frame to apply to this kind of discussion, consider the Mac. I assume our gracious host and like-minded souls would say that Apple “lost” to Wintel, or to Microsoft (or Dell or whomever the box-maker du jour is/was). Yet there are more Macs being sold now than ever before, with growth rates higher than the ‘Personal Computer’ market as a whole, and delivering billions to Apple Inc and its owners, with the attendant supporting industries of developers etc. I’m not sure in what way Apple “lost”, with the implication that the race is over and the medals awarded, or is ‘losing’ with the medals to come at some point in the near future.

    All that being said, I’d be very happy indeed if another platform or platforms sustained sufficient marketshare of the right kind to sustain competitive pressure on Apple in the touch/handheld market. In practice that looks like it means Android or maybe HP/WebOS. Android is obviously going well in phones, but I’m not convinced that is happening in ‘tablets’, at least not yet. I also have my doubts about Eric’s conception of a ‘classic technology disruption’ as he is applying it to phones. Clearly there is a disruption, and there are multiple ‘disruptors’, but I’m far from convinced this is going to play out in the conceptually orderly way Eric seems to be envisioning.

    d (your friendly neighbourhood Apple fanboy)

  64. Patrick Maupin Says:
    “Well, I tried to FTFY, but I guess I don’t know how to do a strikeout in wordpress.”

    :) I still think my version is more correct, if only for its simplicity. In the wider scheme I fear Goldman-Sachs far more than Apple and as far as I know they haven’t sued anyone on a look-and-feel patent, yet.

    I still don’t know how to do *anything* in WordPress. Where’s the bloody Help button?

  65. daz Says:
    “As an illustration of why ‘losing’ is such a strange frame to apply to this kind of discussion,”

    I don’t think it’s strange at all. Here’s how I might phrase it: “Apple has lost the ability to lock us up in their exquisitely decorated jail.”

    Apple does some good work and they have pushed this industry in some good directions (some bad ones too). I don’t particularly want to see their marketshare crash to zero in any category. But there is a line of argument that it might serve them right.

  66. @Patrick

    >There are lots of ways to make money. Some are better for society than others. Apple suing Samsung for a tablet “look and feel” that first showed up in 1973 just doesn’t feel like it’s good for society to me. Again, YMMV.

    Unfortunately everybody plays the lawsuit game these days. It’s particularly offensive when patents are involved, and the system needs serious overhaul. Google tends to get involved less than others, but it’s certainly not immune.

    Apple chooses to make money by selling products and services to its customers. Google makes money by selling your attention to advertisers. Both of them engage in lawsuits, but it’s not a revenue stream so much as a strategic side game.

  67. @Patrick

    What’s the magic about marketshare? I have been following the blog for a goodly time, although often not the comments, and I am not much of a commenter myself. I mean, obviously I understand the attendant economic benefits, but you’re much less concerned about that, judging from your comment that you don’t care if Android vendors can make any money out of their devices. I understand the emotional satisfaction one gets from seeing one’s preferences shared by others (I was an Apple fan boy in the 90s ;^) but I sense there’s something else for you that I’m not getting.

    d

  68. @Patrick

    “Apple has lost the ability to lock us up in their exquisitely decorated jail.”

    Oh good grief. Can you dial down the hyperbole to eleven? I switched from an iPhone to a HTC Desire HD earlier this year. Hardly a jail. I’ll be switching back I think, as I prefer the amenities and service of the Apple holiday resort.

    “Apple did lose. And almost died. …[then] they staged a comeback”

    I don’t think you can stage a comeback after you’ve *lost*. That’s kind of what ‘losing’ means. Or should we be on the look out for the Detroit Tigers making a comeback in the 2006 World Series?

    d

  69. @Patrick

    >Apple did lose.

    Wait… What?! Apple just made more than $5B selling Macs in a single quarter. Sign me up for losing right now if that’s what it entails!

  70. @esr

    >The kind of evil Apple is pushing is hostile to Google’s long-term business model.

    Not sure what the ‘evil’ is you are talking about, but you are dead right that Apple’s products are hostile to Google’s business model. The reason is that by providing great products to their customers in exchange for money, they weaken Google’s ability to provide free products in exchange for showing you ads.

  71. @daz:

    judging from your comment that you don’t care if Android vendors can make any money out of their devices.

    I care deeply that at least some Android vendors make a good profit, because that is absolutely necessary for the platform to grow. But it doesn’t have to be nearly as big as Apple’s profit to be a good profit by almost anybody else’s metrics.

    Take the current PC market for example — Apple has over 90% marketshare of the > $1000 PC. I don’t mind if they keep that, as long as I can go down to Fry’s and buy a nice $399 laptop whenever mine dies. That means that I care that some company be able to make and sell a decent laptop at the $400 price point. That happens because of stiff competition throughout the entire product chain, from CPUs and DRAM and hard drives on up.

    Marketshare goes hand in hand with network effects. Microsoft showed how you could use those network effects to create lock-in, and Apple was a quick study, and even took it farther than Microsoft could ever have dreamed with their 30% cut of all apps and content on their curated marketplace.

    A high enough marketshare makes Android “good enough” for more users, which attracts more developers, which attracts more users, etc. Now, it’s possible that something like WebOS could come from behind (especially if HP frees it), or even Microsoft with Windows Phone, or RIM, but those scenarios seem increasingly unlikely. In any case, a duopoly is always preferable to a monopoly, but a duopoly where one of the vendors gives away source code for free is a dream come true.

    There are currently no killer apps on iOS for Apple to leverage into a bigger monopoly, and any such apps would probably be replicated pretty quickly onto Android. But if Android hadn’t ramped quickly enough, that might not be the story. It might have been that the only really viable smartphone platform was Apple (which actually seemed likely for awhile).

    But now Apple has to treat their own users and developers a bit nicer than they would have otherwise, or risk losing them. And anybody who really detests Apple has a viable option. And that option is likely to stay viable for the simple reason that everybody agrees it’s a viable option. Such are network effects.

    Oh good grief. Can you dial down the hyperbole to eleven? I switched from an iPhone to a HTC Desire HD earlier this year. Hardly a jail. I’ll be switching back I think, as I prefer the amenities and service of the Apple holiday resort.

    I didn’t write what you’re responding to here. But if you wanted to write an app and put it on a website for your friends, you might realize that, like a Club Med on Haiti, things look a bit different from different perspectives.

    I don’t think you can stage a comeback after you’ve *lost*. That’s kind of what ‘losing’ means. Or should we be on the look out for the Detroit Tigers making a comeback in the 2006 World Series?

    That’s silly. The Detroit Tigers lost, but they didn’t go out of business, and they can try again later. Same thing with Apple. BTW, their PowerPC Mac lost. They spent a lot of time and money building chips, etc. working with IBM, and then they threw in the towel and decided to build Windows compatible PCs and put Unix on them, and steal an open source browser. Then they started to get some market traction. They decided for sentimental reasons to keep the Macintosh name, but it’s not even really clear to me if that helped them or hindered them. But the incrementally enhanced 68K -> Power PC mac is gone. So yeah, lost.

    @Tom:

    >$5B selling Macs

    Oops, sorry. That was revenue, not profit. But I think the point stands :)

    The point that you’re a troll who can’t understand, despite the fact that I’ve written it multiple times, that I really don’t care about Apple’s profits but care deeply that they don’t own all the smartphone users in the world? Yes, that point absolutely stands.

    The point that you’re a troll who started this whole argument off with the ludicrous claim that under 2M Android tablets had shipped, and then came up with other opinion-based arguments after you lost all the factual ones? That point stands too.

    The point that you can’t understand that google’s disintermediation efforts benefit the entire web, and not just google, that point definitely stands, and takes you from just trolling to being an imbecile.

  72. @Tom:

    Just noticed this bit of trolling:

    Google makes money by selling your attention to advertisers. Both of them engage in lawsuits, but it’s not a revenue stream so much as a strategic side game.

    WTF??!? WHO THE HELL DID GOOGLE EVER SUE?

  73. Ah, wait. Never mind. I remember — they sued the Department of Interior for rigging a bid so only Microsoft could win it. Hmm, tell me again how that’s not about revenue?

  74. @patrick

    >But if you wanted to write an app and put it on a website for your friends, you might realize that, like a Club Med on Haiti, things look a bit different from different perspectives.

    Could you explain this? Apple provides completely standards-compliant web browsers for both its platforms. In fact, it uses the same open-source rendering engine as Google’s Chrome browser. There is a decent case to be made that Apple is one of – if not *the* – leading corporate advocate of web standards today. If you want to go out and make a web app (and I do, frequently) you don’t need to worry about Apple. It’s Microsoft you want to be worried about.

  75. >Ah, wait. Never mind. I remember — they sued the Department of Interior for rigging a bid so only Microsoft could win it. Hmm, tell me again how that’s not about revenue?

    Actually I had their suit against Microsoft in mind, but like I said google does play this game a bit less than most others. They still play it though.

    And by the way, I am not interested in portraying Apple as some sort of saint. They are as self-interested as any other big company. I do object to this constant accusation of ‘evil though. I have never understood that.

  76. @Tom:

    But if you wanted to write an app and put it on a website for your friends…

    Could you explain this? Apple provides completely standards-compliant web browsers for both its platforms.

    I mean an app, you know, like you can download from the app store (wanna-be TM). On Android, those can be “side-loaded.” On Apple, not so much unless everybody involved pays Apple to be a developer, or one of them pays Apple to be a developer and the rest all work for the same company.

    Actually I had their suit against Microsoft in mind…

    The only one I know of is where Microsoft sued them first. Is there another?

  77. @patrick

    >I mean an app, you know, like you can download from the app store (wanna-be TM). On Android, those can be “side-loaded.” On Apple, not so much unless everybody involved pays Apple to be a developer, or one of them pays Apple to be a developer and the rest all work for the same company.

    Oh, sorry, it sounded like you were talking about web apps.

    Well sure you can make apps! If you want to put it up on the store to make it super-easy for everybody it costs you like £50 to become a developer and get access to the store and everything. If not, and if it really is just for a couple of friends (and they’re technical) you do have the ability to do what we call ‘ad hoc’ distribution, which I have used for testing apps in the past. Really, though, if it is for more than a few people the store is really the way to go. Free hosting, free distribution. Yes, you have to pay a small amount to become part of Apple’s developer programme, but it is a pretty negligible amount for what you get.

    If you really don’t like any of that the you can always make a web app and remain completely unregulated.

    However, if you really want to be able to mess about with your phone, customise the hell out of it, and put any program you like on it then the iPhone is not for you. Go and buy an Android phone. The iPhone is for the 99.9% of people who are not hackers and who just want to get stuff done with their phones. If making something that is designed for mere mortals to use easily and safely is ‘evil’ then I guess Apple is evil.

  78. >The only one I know of is where Microsoft sued them first. Is there another?

    I seem to recall that around 2006/7 they sued them (or it may have been an antitrust complaint or something) for their integration of Live Search into Windows.

    Who really cares though? I wish all these stupid lawsuits would go away just as much as anybody.

  79. @Patrick, again, sorry about the misattribution.

    you did write this though: “I honestly don’t give a rat’s ass if Apple sucks all the profit out of the market”, which is a silly overstatement. That’s OK, these are blog comments, everyone gets a bit carried away sometimes. I get where you’re coming from now. Kind of.

    On the “lost” thing, I’m still thinking you don;t quite understand what the word means as verb without an object. The Tigers lost the 2006 world series. They cannot try to win it again, the game is finished. They can try again but it will be in a different competition, not the 2006 one. Markets don;t usually work that way.

    BTW Apple put Unix on their PowerPC Macs in 2000, then transitioned to Intel in 2006, and your apparent definition of ‘stealing’ is even more idiosyncratic than your one for ‘losing’, particularly as applied to an Open Source layout engine, which in Apple’s hands has become an open source layout engine. You must have pretty harsh words for all those phone and tablet vendors that have ‘stolen’ Android. ;^)

    Come to think of it, isn;t webkit the Android browser’s layout engine?

    d

  80. However, if you really want to be able to mess about with your phone, customise the hell out of it, and put any program you like on it then the iPhone is not for you.

    That’s because Apple pretends like they sell you the phone, but then they maintain control. At least they learned they would get in hot water and the world was watching when they went so far as to brick rooted phones. Personally, I think that when you sell somebody something, you should relinquish control. I’m funny that way.

    Who really cares though? I wish all these stupid lawsuits would go away just as much as anybody.

    Yet another wild-assed assertion, and then when you’re called on it, it’s “who really cares?” Lucky for you, google’s really not litigious, so feel free to defame them to your heart’s content.

  81. On the “lost” thing, I’m still thinking you don;t quite understand what the word means as verb without an object.

    OK, they “lost” the huge marketshare lead they had over Android, and they are still, at this very moment, “losing” marketshare to Android. Feel better now?

    BTW Apple put Unix on their PowerPC Macs in 2000, then transitioned to Intel in 2006, and your apparent definition of ‘stealing’ is even more idiosyncratic than your one for ‘losing’, particularly as applied to an Open Source layout engine, which in Apple’s hands has become an open source layout engine.

    They gave it back under duress, a long time after they were shipping object code, in a big steaming pile of unmergeable patches. Go ask the konqueror guys what they thought at the time. It wasn’t until the dispute became very public and Apple realized that they were pissing off exactly the kind of people they might like to have working for them that they put the code in a more easily accessible public repository. And, btw, they’re still at it with the copyright violations:

    http://www.osnews.com/comments/24717

    You must have pretty harsh words for all those phone and tablet vendors that have ‘stolen’ Android. ;^)

    Only the ones who are violating copyright by, e.g., not making available any (L)GPL changes. And especially the ones who violate others’ copyrights while simultaneously using copyright law as a club to tie their hardware and software together.

  82. @daz:

    you did write this though: “I honestly don’t give a rat’s ass if Apple sucks all the profit out of the market”, which is a silly overstatement. That’s OK, these are blog comments, everyone gets a bit carried away sometimes. I get where you’re coming from now. Kind of.

    It’s too much for me to ask you to read the entire history of posts and comments on this subject on this blog. But likewise, if I make a hyperbolic statement like this, it just might be because I’ve explained the same distinction between marketshare and profit 300 times. This blog is constantly inundated with Apple fanbois (e.g. Tom) who think that profits are the only thing that matter and we shouldn’t be discussing these other things. There are plenty of other places on the web for him to go and gloat with his buddies and taunt his enemies about how well Apple is doing. I don’t care about that, and I don’t think esr does either. I care about whether handheld devices serve those who paid for them and hold them, or serve some other master. Google seems to be on my side in this; Apple seems to think that when they sell you something, it’s still theirs.

  83. @Tom:

    The iPhone isn’t for 99.9% of the people. Let’s go down some of the list of who the iPhone isn’t for:

    – people who want physical keyboards
    – people who want a phone bigger or smaller than Apple’s standard (e.g. people who want bigger onscreen keyboards to avoid clumsiness and/or bigger screens to avoid eyestrain)
    – people who want a removable / replaceable / upgradeable battery (which includes people trying to stay off the upgrade treadmill and people who want their phone to have a better chance of surviving if dropped in water)
    – people who want HSPA+ or LTE or WiMax
    – people who want to sync their music to clouds other than Apple’s
    – (most recently) people who want to buy eBooks from anyone other than Apple (which includes anyone who ever wants to read an eBook on anything other than an iPhone or iPad. Amazingly, Apple hasn’t even bothered to release iBooks for Mac yet)
    – … (I’m sure other people have plenty more examples)

  84. Upon re-reading, “for 99.9% of the people” should be quoted. I specifically do *not* mean to imply that the universe of people that the iPhone isn’t for includes 99.9% of the market. I’m just saying that 99.9% is way too high a bar for the percentage of the market Apple is addressing (or intends to address, I suspect).

  85. The table tells us that 0.9% of active android devices are tablet size, which corresponds *exactly* with their number for active Honeycomb devices.

    I can’t prove this so take with a grain of salt but a plausible explanation would be that xlarge came in as a standard with honeycomb. Thus pre-honeycomb devices would be incapable of declaring themselves xlarge. Bear in mind that Honeycomb was specifically (re-)designed for tablets and the stated reason for not completely open-sourcing the code was because it wasn’t ready for prime time on non-tablets. So it wouldn’t surprise me that the Venn diagrams for android tablet devices and android honeycomb devices would be mostly the same thing.

    The closest I can find to support this is the “supports-screens” xml node of AndroidManifest.xml which gained “android:xLargeScreens” in API Level 9 (Gingerbread) so any tableet device running Froyo or earlier certainly wouldn’t understand that property.

  86. @Patrick

    “Feel better now?”

    Greater clarity is always welcome, but I’ve not felt too bad about you making very little sense. :^)

    “Go ask the konqueror guys what they thought at the time.”

    I’m unlikely to go find a konqueror person of any gender, and at any rate, what they thought ‘at the time’ is less interesting then what they think now, with the benefit of hindsight. It seems to me there were problems, people got together and fixed ’em. I prefer not to speculate on what motivated people to fix the problems, although my general disposition is to assume ‘good will’. (And my general assumption about the cause of problems is ‘incompetence’.) If Apple is still having problems getting the source out, well, that sucks and they should lift their game.

    “It’s too much for me to ask you to read the entire history of posts and comments on this subject on this blog.”

    Thanks for your patience and consideration. From my PoV, I have no easy way of determining if you’re an articulate nutcake or what. No trying to be snarky there, just lettin’ you know. And really, to be fair, I’m pretty sure Eric is trying to attract idiot Apple fan boys to these threads. I mean, posts where he writes ‘I’ve laid a trap for you…’ what do you think is going to happen ;^)

  87. @Tom: I do object to this constant accusation of ‘evil though.

    You are either new or ignorant. Jobs said it best himself: we want to provide the most advanced and innovative platform to our developers, and we want them to stand directly on the shoulders of this platform and create the best apps the world has ever seen. Note “our developers” passage there. For any other company “our developers” mean “developers we pay” (company own developers, may be contractors, etc). You can ask such people to just stop using something – and they will for “he who pays the piper calls the tune”. For Jobs “our developers” are “everyone who wants to develop for our platform”.

    Apple (at least “Apple with Jobs”) wants to control everything and everyone. Yes, sometimes they relent and temporary permit some transgressions (for example Jobs hate Flash and thinks it’s not important enough, but Unity3D is not such a big deal, so for now you can have it), but this is at their whim: they act line an overlords who own all developers and users, not like mere suppliers of hardware and software. Apple and Apple alone can decide who’ll live and die – and when.

    @Tom: Well sure you can make apps! If you want to put it up on the store to make it super-easy for everybody it costs you like £50 to become a developer and get access to the store and everything.

    No. It does not cost you £50. It cost you your soul. Well, that’s hyperbole, of course, but you must sign “top-sikrit” agreement (which forbids you to use GPL, by the way) and only then you can develop anything. Feels more like joining some kind of sect rather then normal EULA signing (which is evil too, but compared to what Apple does to “their” developers… ugh…).

    A lot of companies try to control you (Nintendo is actually more evil then Apple), but they don’t have a chance to control significant portions of our life (I have a DS, but I don’t even remember the time when I’ve last turned it on… but my phone is always with me) – thus they are not as problematic.

  88. @daz:

    Greater clarity is always welcome, but I’ve not felt too bad about you making very little sense. :^)

    I think it would have been pretty obvious to any sentient being what I was getting at. I thought you were just playing, but now you’re either trolling or an imbecile. Maybe both. Your pick.

    Although my general disposition is to assume ‘good will’.

    Fool me once, shame on me. Etc. The one thing Tom got right is that Apple’s in it for the money. If delaying giving stuff back to their competitors helps in that, they’re happy to do it.

    (And my general assumption about the cause of problems is ‘incompetence’.)

    After everything you’ve written, do you really expect me to believe that you think Apple is incompetent? Or are you just projecting your own imbecility again?

    ‘I’ve laid a trap for you…’ what do you think is going to happen ;^)

    There’s a difference between dealing with the vermin that are already present, and attracting new ones. He said “trap,” not “bait”. In any case, he didn’t do that this time, so any vermin who wandered in previously should already have understood by now that the smartphone discussion here isn’t about profit and share price. Hit the Yahoo! board for that stuff.

  89. @JonB:

    I think you’re definitely on to something. See:


    Android 2.3 and above supports the “xlarge” resource qualifier that is activated for tablets.

    And:


    Unfortunately when I add :

    android:xlargeScreens=”true”

    To my Manifest, it gives me this error :

    error:No resource identifier found for attribute “xlargeScreens” in package ‘android’

    So, for Android < 2.3 (e.g. versions used by the bulk of the shipped tablets) it would have been impossible to classify a tablet as "xlarge".

  90. @khim:

    > Nintendo is actually more evil then Apple

    And Sony has been desperately trying to be the most evilest of all. Stiff competition.

    BTW (speaking of evil Sony, who is now shipping Android), lots of manufacturers and carriers do obnoxious things with Android, too. Google uses the carrots of the market and maybe being able to partner for first development on the next version to try to corral them.

  91. @daz:

    After re-reading your last post, I realized that you’re one of those people who think it’s OK to gratuitously insult others as long as you smile. Sorry, I didn’t play the game right, so I’ll make up for it now.

    What an asshole you are! :^) :^) :^) :^) :^) :^) :^) :^) :^) :^) :^) :^) :^) :^) :^)

    (Hopefully that was enough to make up for any insults you consider to be gratuitous.)

  92. >but you must sign “top-sikrit” agreement (which forbids you to use GPL, by the way) and only then you can develop anything

    Oh good lord. Can we have anymore hyperbole here? Yes, the developer agreement is so top secret that you have to *GASP* go to the public apple developer web site and click the links to begin enrolling in the developer program. Hell, you might even simply be able to get it by downloading the PDF: http://developer.apple.com/programs/terms/ios/standard/ios_program_standard_agreement_20110713.pdf (though you might need a free registration for that). So sekrit!

    It’s also worth noting you can download the dev tools and SDK without agreeing to that agreement.

  93. @Patrick

    I’m very sorry you’ve taken such offence at what I’ve written and I apologise if I’ve been out of line. Trolling? No, I was tweaking your apparent pedantry a bit at first, then having a bit of a proper chat. An imbecile? Possibly, I expect there’s at least a few people that think that. I don’t agree, but hey, what would I know?

    Again, I apologise for my ill-chosen words, and I hope the unpleasantness they’ve caused you in the morning does not ruin your day.

    d

  94. Incidentally, the forbidding of the GPL is a result of the terms of the GPL. You can use other FOSS software and license your software under those licenses on the apple store, but the GPL imposes additional conditions on Apple and is likely unable to be linked with the Apple libraries anyway, exactly as RMS intended. Of course we know the GPL isn’t real freedom anyway so it’s not like you’re losing out on anything.

  95. Here’s a good example of Apple evil. In order to get the 30% cut of transactions, they’re systematically crippling and excluding apps that point outside their system.

    This sort of thing is what Android is designed to end. It is exactly Patrick’s “elaborately decorated jail”. And it is why iOS – and closed-source platforms in general – must die.

  96. Patrick Maupin,

    This thread is unusually cranky for you. Normally you are more patient. I hope things are going well for you. I’m kind of worried about the incompetence of our current leaders, many of whom do not realize the ship is going down. They are expecting to it keep cruising for a number of trips, instead to calling for the crew to prepare the lifeboats. Bet on the passengers to suffer.

    daz appears to be discussing in good faith, and really, so does Tom. daz’s presumption of good will and incompetence is a capital way to approach the world. Similarly, it’s hard to competently lead when a political system that has worked for almost 150 years (going back to Bismark) falls apart.

    Yours,
    Tom

  97. @tmoney Yes, the developer agreement is so top secret that you have to *GASP* go to the public apple developer web site and click the links to begin enrolling in the developer program.

    Not all that public. You must give your personal data (address and phone number) before you’ll even see said agreement. I don’t know why Apple decided to play this game hide it from archive.org and similar sites, but yes, Apple pretends it’s some kind of secret you are not entitled to know unless you already decided to be a slave.

    @tmoney It’s also worth noting you can download the dev tools and SDK without agreeing to that agreement.

    Link, please? All “front page” links return me back to sell us your soul, then you’ll get the goodies page.

    @tmoney Of course we know the GPL isn’t real freedom anyway so it’s not like you’re losing out on anything.

    Puhlease. GPL is not perfect, but the only real problem WRT to Apple’s Store is that said store limits the user and forbids any and all forms of redistribution. Sure you may say that the ability to sell your soul to one particular overlord is important enough freedom and that makes GPL non-free, but the very fact that you need to sell you soul to ever do anything with iOS is much, much worser…

  98. @Tom DeGisi:

    You’re right. I’ve been cranky for reasons having to do with people procrastinating until it’s too late to do things the right way, so then they have to put a bandage on it. This is happening to me at all levels, currently infecting the country, my personal life, and my professional life. What’s worse, it’s contagious. (Both the procrastinating and the crankiness.)

    Sorry, Tom and daz — some of what I wrote was over the top.

    @esr:

    I can’t take credit for Michael Hipp’s jail metaphor.

    @tmoney:

    I absolutely agree that GPLed software is not “free” and I used to abhor it in all cases. I still have no use for the viewpoint that all software ought to be GPLed — I will use GPLed programs quite happily but not, in general, GPLed libraries, because a library is something that I add to my mental “toolkit” which I want to be able to be general purpose and not constrained by arbitrary licensing issues.

    But after the SCO case and the Apple WebKit and VLC cases and other similar cases, I have come to the conclusion that the GPL is a valuable part of the ecosystem, because it helps us find out which companies have a total, utter, disregard for other peoples’ copyrights, yet who think their own copyrights are the most precious things in the world and should be wielded as potent weapons. As khim points out, it also lets us ponder when and where the GPL is causing problems (quite a few cases IMO) or simply acting like the canary in the coal mine and showing where other problems exist, like in Apple’s app store.

  99. @esr ” And it is why iOS – and closed-source platforms in general – must die.”

    I hope you keep in mind that what you WANT to happen has no relation to what WILL happen.

    1. >I hope you keep in mind that what you WANT to happen has no relation to what WILL happen.

      On this particular topic, that’s a pretty hilarious thing for anyone to say to me. I’ve already blown up the software industry once, and Google is executing on multiple strategies that I have good reason to believe I taught Larry and Sergei how to think about. What will happen is already pretty well tied to what I want, even if Android fails to gain another point of market share. But it won’t stop there.

  100. @esr

    “>For all intents and purposes iPads shipped = iPads sold.

    I agree this is probably true. What I don’t get is the implied assumption that the sell-through rate on Android tablets isn’t just as good. As Patrick pointed out, nobody wants to carry inventory and retailers aren’t buying tablets for their health.”

    It’s hard to say. If you run this query it doesn’t sound good for the Xoom: http://www.google.com/webhp?sourceid=chrome-instant&ie=UTF-8&ion=1&nord=1#hl=en&cp=17&gs_id=b&xhr=t&q=motorola+xoom+sales
    Anecdotally, I haven’t seen hardly any Android tablets (or BB/Palm) tablets in the wild so far, maybe 2 or 3?

    But on the other hand, I see ads and product in the stores all over the place. So you’d think the product would get pulled if it wasn’t selling.

    Seems like we need Google to release a tablet activation # to really know….

  101. @esr

    >Here’s a good example of Apple evil. In order to get the 30% cut of transactions, they’re systematically crippling and excluding apps that point outside their system.

    I think this is a mistake on Apple’s part. I believe it will ultimately hurt both them and their customers. And I think they will realise this and change their policy, but I could be wrong. Is it ‘evil’? No, I think that description is a bit silly. Apple’s policy is misguided. It doesn’t come from a malicious place. Apple has no motivation to hurt their users. Sometimes they just make mistakes, and this is one of them.

    Not everything has to be at one extreme of either ‘evil’ or ‘grand strategy to free the people’. The smartphone industry isn’t a Tolkein novel.

    >This sort of thing is what Android is designed to end. It is exactly Patrick’s “elaborately decorated jail”. And it is why iOS – and closed-source platforms in general – must die.

    Isn’t it enough just for there to be alternatives? Why does it have to ‘die’?

    1. >Isn’t it enough just for there to be alternatives? Why does it have to ‘die’?

      As an object lesson to anyone who even dreams of reviving this sort of shit.

  102. >Not all that public. You must give your personal data (address and phone number) before you’ll
    >even see said agreement.

    Or if you’re really that interested, you could put some fake information into the fields. I’m sure John Smith of 6 Anystreet, Anytown NY 12345 won’t mind at all. Hell you can even fill in 555-1212 for the phone number and it will take it. As for why, well honestly, because the only people that have any particular need for the contract are the potential developers and Apple.

    >Link, please? All “front page” links return me back to sell us your soul, then you’ll get the goodies page.

    At the bottom of that page you will note this link: http://developer.apple.com/programs/register/

    To which you can register as a free developer and access the SDK and dev tools, which has its own general agreement and not the same as the iOS or Mac Developer agreements.

    >GPL is not perfect, but the only real problem WRT to Apple’s Store is that said store limits the user
    >and forbids any and all forms of redistribution.

    Citation please? The problem with the GPL is that it requires a) that any linked software be GPLed, a likely impossibility for anything that uses the Apple APIs, and b) requires the distributor to offer the source code, which then puts a burden on Apple to be a source code distributor since they distribute the app.

    >but the very fact that you need to sell you soul to ever do anything with iOS is much, much worser…

    As I said, hyperbole, and one of the reasons that despite the fact that I like what the EFF does from time to time, and like Patrick I can see the uses of GPL, I find myself unwilling to commit to either cause.

  103. @Tom:

    It is very clear the ebook nonsense is coming from an evil place. They’re trying to kneecap their competitors, users be damned. There was nothing unreasonable or unfair about the previous status quo, but they decided to upend it anyway. And they’re not even trying to pretend they’re competing on the merits with iBooks. They don’t even have iBooks for Mac. For *Apple*, that’s beyond bush league.

  104. @tmoney:

    Citation please? The problem with the GPL is that it requires a) that any linked software be GPLed, a likely impossibility for anything that uses the Apple APIs, and b) requires the distributor to offer the source code, which then puts a burden on Apple to be a source code distributor since they distribute the app.

    No, the linking with system libraries is not a problem. No, the giving source code is not a problem — a lot of vendors are more than willing to do that. Here’s some background on the problem:

    http://www.macstories.net/news/vlc-for-ios-may-soon-be-gone-from-the-app-store-due-to-videolans-complaint/

  105. @tmoney:

    With the free SDK, how hard is it for a coder to, e.g. email an app he builds to a few of his non-coder friends and for them to install it on their phones?

  106. @Ravi/esr

    I was just reading a bit into the background on this, to see if we could shed any light on Apple’s motivations, when I found that they have actually already reversed their stance. The app store guidelines now state:

    “Apps can read or play approved content (specifically magazines, newspapers, books, audio, music, and video) that is subscribed to or purchased outside of the app, as long as there is no button or external link in the app to purchase the approved content. Apple will not receive any portion of the revenues for approved content that is subscribed to or purchased outside of the app.”

    So I guess they did realised their mistake.

    >They don’t even have iBooks for Mac. For *Apple*, that’s beyond bush league.

    I think the reason for this is just that not many people want to read books on PCs. It is a pretty horrible experience. That is what tablets are for!

  107. @ Ravi,

    Actually, there’s a perfectly non evil (but still IMO misguided) reason for such a change. Consistency. By requiring that every app using in app purchasing use the apple API, they also guarantee that every in app purchase is billed to the user’s itunes account, and all the ups (and downs) that come with that, such as refund capabilities.

  108. @tmoney Citation please?

    FSF have pretty good article, for example.

    @tmoney The problem with the GPL is that it requires a) that any linked software be GPLed, a likely impossibility for anything that uses the Apple APIs,

    Have you ever read the GPLv2? As a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable. GPLv3? The “Corresponding Source” for a work in object code form means all the source code needed to generate, install, and (for an executable work) run the object code and to modify the work, including scripts to control those activities. However, it does not include the work’s System Libraries, or general-purpose tools or generally available free programs which are used unmodified in performing those activities but which are not part of the work.

    @tmoney and b) requires the distributor to offer the source code, which then puts a burden on Apple to be a source code distributor since they distribute the app.

    It’s not like it’s hard to them to organize this: just ask for the source code from an author. GPL is quite explicit here: You are not responsible for enforcing compliance by third parties to this License. Somehow it’s not problem for Google, Nokia, and others, but oh-so-problematic for Apple.

  109. @ patrick

    >No, the linking with system libraries is not a problem. No, the giving source code is not a problem —
    >a lot of vendors are more than willing to do that. Here’s some background on the problem:

    Right sorry, I forgot c) the new GPL doesn’t let you use any form of DRM.

    Obviously if it were only about the rights that the GPL is supposed to protect, the availability of the iOS VLC source code with the reminder that it is distributed under the GPL would be sufficient don’t you think?

    It’s also worth noting that the iOS developer agreement does not explicitly forbid you from using the GPL. It just requires that you be able to comply with all the terms of whatever license you use, or the underlying software you distribute uses. Seems like a perfectly reasonable requirement. That most GPL software is incompatible with being distributed by the app store is not evidence of the evilness of apple, unless of course you believe that anything not GPL compatible is evil.

    As to ad hoc distribution, my understanding (having not done it myself) is the official ad hoc method is onerous on the distributor. Your friends must provide you with their devices UUID, which you then have to enter into the dev system as an authorized beta tester, and you are restricted to 100 testers. That said, of course you can always go the jailbreak route, and then it’s just a matter of jailbreaking the device you want it installed on.

  110. >FSF have pretty good article, for example.

    And yet, there’s this paragraph, from their own static page under the App Store section:

    Your license to each Product that you obtain through the Service is subject to the Licensed Application End User License Agreement set forth below, and you agree that the terms of the Licensed Application End User License Agreement will apply to each Apple Product and to each Third Party Product that you license through the Service, unless the Product is covered by a valid end user license agreement entered into between you and the licensor of the Product (the “Licensor”), in which case the Licensor’s end user license agreement will apply to that Product. The Licensor reserves all rights in and to the Product not expressly granted to you

    >It’s not like it’s hard to them to organize this: just ask for the source code from an author. GPL is
    >quite explicit here: You are not responsible for enforcing compliance by third parties to this License.

    But they are responsible for themselves, and since they are the ones distributing the software, they would be responsible for providing the source. If I were running an app store like that, I’d drop any GPL software I got a complaint about too. Too much hassle.

    Also if I may snark for a moment, we already know that Google doesn’t concern itself with distributing source when it’s inconvenient to them, so that they’re hands off with their app store as well is no surprise.

  111. @tmoney:

    Obviously if it were only about the rights that the GPL is supposed to protect, the availability of the iOS VLC source code with the reminder that it is distributed under the GPL would be sufficient don’t you think?

    As to ad hoc distribution, my understanding (having not done it myself) is the official ad hoc method is onerous on the distributor. Your friends must provide you with their devices UUID, which you then have to enter into the dev system as an authorized beta tester, and you are restricted to 100 testers. That said, of course you can always go the jailbreak route, and then it’s just a matter of jailbreaking the device you want it installed on.

    The GPL means different things to different people. Many think the GPL overreaches, GPL v3 even more so. Nonetheless, it’s hard for me to imagine that anybody who chooses to use the GPL would think that it’s “just” about the source. It’s obviously about being able to let your downstream users make use of both the source and the objects they create from the source. They ought to freely be able to give them to whoever they want. And you just can’t guarantee that will be happening with the Apple app store.

    I didn’t (and still don’t, really) think that what Tivo did was a huge problem, primarly because the TiVo was not marketed as, and is seldom, if ever used as, a general purpose computing platform. But the whole point of a smartphone is that, in addition to being a phone, it is a general purpose computing/communication platform. That’s why it matters.

    That said, of course you can always go the jailbreak route, and then it’s just a matter of jailbreaking the device you want it installed on.

    I’m sure you see the problem with this. Give a program to your best friend, and all he has to do is endanger all his data and void his warranty to run it.

  112. >It’s obviously about being able to let your downstream users make use of both the source and the
    >objects they create from the source. They ought to freely be able to give them to whoever they want.
    >And you just can’t guarantee that will be happening with the Apple app store.

    The problem as I see it though is at what point is it good enough and the downstream users have to do some things for themselves? If you have the source as a downstream user, you can do what you want with it, Apple’s app store be damned. Of course you have to be able to compile that code and install it, but if you can’t do that, what good is the GPL to you anyway? And how much of a right does anyone have to require that Apple or anyone else enable you to use that source code? If the fact that you have to compile and install it yourself is enough to be incompatible with the GPL, well there’s a whole raft of OSS software that’s incompatible.

    Ultimately, I think that the legal text of the GPL could have been satisfied in the VLC and the GNU Go examples, but that the political spirit would not be, and neither the FSF nor Apple were interested in actually working a solution out. The GPL is a political weapon and about more than just writing your own printer drivers these days.

    >I’m sure you see the problem with this. Give a program to your best friend, and all he has to do is
    >endanger all his data and void his warranty to run it.

    You endanger your data any time you install an application from an untrusted source. As to the warranty voiding, we’ve been down this road before, and you and I just disagree. If you want to use your device outside the supported designs of the manufacturer, you should be perfectly free to do so, and no law should prevent you. But equally, the manufacturer should be free to deny you warranty support for that. It’s a trade off.

    Besides, I still have to do the equivalent of a jail break to update my “modern” android phone to the latest software, so the whole jailbreak = warranty void thing doesn’t phase me, nor does it to me count as a strike against the iPhone. Again though, on this point you and I differ.

  113. @tmoney:

    Kindle, Kobo et al didn’t use in-app purchasing and this was completely clear. They just provided a link to their website that (effectively) said if you want to buy our stuff go over here. Now they can’t even mention they have a website (not even for account creation – so if you don’t know how to use the app downloading it isn’t going to enlighten you). They can’t even mention in the app that they sell stuff or even why the bookstore link was removed. That goes way, way beyond any desire for “consistency”.

    Kobo, in particular, is getting roasted in App Store reviews for removing the bookstore link (down from 4.5 stars to 2.5 stars in only a few days). According to Kobo’s lead iOS developer on Twitter, they predicted this would happen and tried to get permission for all sorts of ways to explain what was going on or at least smooth things a little for their customers. Apple’s response was (effectively) “tough”.

    Moreover these policies have been discussed (and protested) for months (since the Sony Reader app rejection), so this isn’t a simple mistake or misunderstanding. This was deliberate and considered and therefore, in my opinion, evil.

  114. @tmoney:

    Of course you have to be able to compile that code and install it, but if you can’t do that, what good is the GPL to you anyway? And how much of a right does anyone have to require that Apple or anyone else enable you to use that source code? If the fact that you have to compile and install it yourself is enough to be incompatible with the GPL, well there’s a whole raft of OSS software that’s incompatible.

    At the end of the day, the GPL guarantees that you are allowed to pay somebody to compile and install software for you, but that’s no reason for Apple to make it that difficult.

  115. Patrick Maupin,

    > You’re right. I’ve been cranky for reasons having to do with people procrastinating until it’s too late to do things the right way, so then they have to put a bandage on it. This is happening to me at all levels, currently infecting the country, my personal life, and my professional life. What’s worse, it’s contagious. (Both the procrastinating and the crankiness.)

    I’m sorry to hear that. It’s a bad thing when you are crankier with those you disagree with on this site than I am. Although I really think Kevin S. Van Horn is producing tendentious claptrap, he almost certainly deserves more respect than my personal crankiness was allowing him.

    Yours,
    Tom

  116. @Tom:

    Apple may have changed their policy, but they haven’t realized their mistake. They just got too much blowback from the initial version, so they decided to offer their competitors the ogre’s choice: “Die fast or die slow.” Look at what is happening to the Kobo app in Apple’s App Store. Think about the incentives Apple is creating. It is clear that the end goal hasn’t changed at all, just the path Apple is taking.

    As for the people don’t want to read ebooks on their computers, that’s just wrong. Plenty of people want to read ebooks on computers. I was reading ebooks on my PC (usually PDF or encrypted PDF) years before the Kindle or the iPad came along. At least one significant publisher (Ellora’s Cave), started with just ebooks and kept growing to the point where it made sense to add some paper books, too. Every major ebook vendor but Apple (Amazon, B&N, Kobo, Sony and even Google) offers a desktop reading option (mostly apps with the exception of Google’s browser-based reader). With the possible exception of Kobo (I didn’t see it on their site, but I only took a quick glance), everyone else offers a Mac option. They’re not making these applications for the fun of it (IIRC, Amazon was practically badgered into Kindle for Mac). They’re making them because a significant slice of ebook customers want them. This is a slice Apple is content to ignore, which is a clear sign they’re not seriously trying to succeed with ebooks.

  117. Ravi is right. Reading ebooks on either Windows computers or Android phones is not horrible. It is pleasurable. I’m sure there are tablets which improve the experience. We have had thousands of years to improve books and tens of years to improve ebooks.

    Yours,
    Tom

  118. @esr “On this particular topic, that’s a pretty hilarious thing for anyone to say to me. I’ve already blown up the software industry once, and Google is executing on multiple strategies that I have good reason to believe I taught Larry and Sergei how to think about. What will happen is already pretty well tied to what I want, even if Android fails to gain another point of market share. But it won’t stop there.”

    So because of you all software will end up being free and open. You should take up world peace next!

    1. >So because of you all software will end up being free and open.

      Well, that’s a little more than I actually hope for, especially since I’ve made an economic case for the rationality of proprietary code in some unusual circumstances.

      >You should take up world peace next!

      No. I was able to exert the kind of tremendous leverage that I did only because of a combination of domain knowledge and skills that wouldn’t transfer well to any other cause.

      I did at one point consider trying to do for the firearms-rights movement what I did for open source, but that problem got solved by other people before I made up my mind to take a serious swing at it.

  119. @Ravi

    >Apple may have changed their policy, but they haven’t realized their mistake. They just got too much blowback from the initial version

    I really don’t understand this. They listened to what people were saying and changed their policy. Why is that bad? Managing something like the app store, and making it into a place that people trust, and love to use, and it profitable for developers is a really tough thing to do. Just look at all the other attempts. Mistakes are going to be made, people are going to get angry when they don’t like decisions. Not everybody is going to come out ahead. But there is really no reason to think that Apple’s motivated by anything other than making the best experience for their users.\

    >As for the people don’t want to read ebooks on their computers, that’s just wrong. Plenty of people want to read ebooks on computers.

    I don’t deny that there is certainly a sizeable minority of people who might be interested in such an activity, but that is not Apple’s vision for ebooks. Their vision is tablet-based, and to me that makes a lot of sense. It is a far superior reading experience. I certainly don’t want to read a novel while staring at a computer screen. Again, that won’t satisfy 100% of people, but it will provide the best experience for the majority. I am quite familiar with the way Apple operates, and they are not interested in filling every niche and pleasing every last person. The are lead by a strong vision of The One True Way. Some see that as arrogant – and no doubt there is some truth to that- but like it or not, that is the way they work. And that’s not to say that at some point down the road they won’t produce a desktop iBooks client, but it’s not a priority.

  120. Well that’s what I get for not updating my browser before posting. The RIM layoff looks to be 10% of their workforce.

  121. I don’t deny that there is certainly a sizeable minority of people who might be interested in such an activity, but that is not Apple’s vision for ebooks.

    You don’t even notice the irony in your statement, do you Tom?

  122. fake account, did you not read p. 15 yourself?

    “However, Android’s share of mobile devices is growing rapidly, doubling from fall 2010 to spring 2011”

    What am I supposed to think about people who find some “devastating” quotation but don’t even read the very next line?

  123. >You don’t even notice the irony in your statement, do you Tom?

    The point I was contending was that Apple is not making a serious attempt to make a good ebooks solution, and that we know this because they have not created a Mac client. My counter was that this is because Apple’s vision for ebooks is tablet-based, not desktop-based. It isn’t that they aren’t trying to make a good solution, it’s that they disagree with Ravi on what that good solution looks like.

  124. And my point, Tom, is that you seem oblivious to how this statement of yours reinforces others’ point that Apple’s a well-decorated jail. If you don’t agree with their “vision”, you don’t get what you want.

  125. >And my point, Tom, is that you seem oblivious to how this statement of yours reinforces others’ point that Apple’s a well-decorated jail. If you don’t agree with their “vision”, you don’t get what you want.

    You could say that of any product you don’t like. “This car only has 3 doors! I need 5 doors. Gaaah, I am stuck in a well-decorated jail! What will I do?!”

    Crazy.

    If you don’t like the way Apple has made its product, use a different one. Like I say, Apple creates the best product *as they see it*, they don’t try to fill every niche or appease every potential consumer.

    1. >You could say that of any product you don’t like.

      Trivially, yes. But the analogy only has teeth when the product is in a category with strong winner-take-all effects. Cars aren’t like that. Software platforms are.

  126. >Trivially, yes. But the analogy only has teeth when the product is in a category with strong winner-take-all effects. Cars aren’t like that. Software platforms are.

    I was just talking about iBooks, but since you bring it up, software platforms obviously do *not* have strong winner-take-all effects. That is demonstrated by the last 30 years of personal computing. One platform has a huge chunk of the market, but there has been a strong second-place in Mac OS and a long tail of Linux, Unix, and other niche systems. If you don’t like Windows you have plenty of options. The same is true with smartphones, and ebook readers for that matter. If you really have to read on the desktop just use Amazon’s Kindle system; it has iPhone, iPad, and Mac versions, and it’s great.

    1. >I was just talking about iBooks, but since you bring it up, software platforms obviously do *not* have strong winner-take-all effects.

      *boggle*

      What drugs were you smoking when you wrote that? The runaway-leader effect in these markets is quite well documented.

  127. @PapayaSF:

    I agree that RIM in a world of hurt. And I think part of the disconnect is that in this fast-moving business looking at installed base snapshot from two points a year apart tells you nothing about what happened in the last 3 months of that year…

    @Tom:

    You could say that of any product you don’t like. “This car only has 3 doors! I need 5 doors. Gaaah, I am stuck in a well-decorated jail! What will I do?!”

    No, the car analogy would be: “I don’t get it. Yesterday, this thing let me go to my favorite corner store and buy a cup of coffee and my favorite magazine, but today it completely shuts down every time I get within half a mile of the store, and the onboard GPS navigation map no longer even shows me the store, but I know it’s there, because I parked and walked over to it.”

    1984 indeed.

    @fake account:

    I don’t know what’s supposed to be discussed. As the father of a liberal, vegan, Macbook- and iPhone-carrying senior at that very university, none of that data looks unexpected to me. As SPQR says, Android is apparently finally starting to catch on a bit, and, given what I think you came here to taunt us over, I think I know why you didn’t point us at the last line of page 14: “Currently, only one third of the campus population is utilizing mobile devices on wireless leaving room for dramatic growth.” Now, that is curious and newsworthy. I wouldn’t at all be surprised to see that 1/3 number grow by a factor of two this coming fall semester, fueled by cheap Android offerings.

    Education has traditionally been a stronghold for Apple, and it’s only natural they’ve driven this up to the University level, especially with their sexy Macbook offerings. The student population at UT is reasonably affluent, and a computer is a necessary tool to go to school these days. If you connect the dots between the charts on that page using the accompanying text, basically 1/3 of the parents bought their kids Macbooks, 1/3 bought them Windows laptops, and of the people who couldn’t/wouldn’t afford laptops, the majority (unknown size) bought Windows.

    As a dad of a university student, I was essentially faced with a Hobson’s choice: Windows or OSX. Sure, most things work under Linux, but you have to pick and choose your battles, and some of those are unwinnable. It’s heartening to see that 14% of wired departmental computers are now Linux — that should slowly translate into better support for students using Linux. Faced with the Windows/OSX choice, and with sufficient funding, and with a student who is going to be clicking on who-knows-what sites and isn’t running a computer under your direct control, the answer is almost a no-brainer. In retrospect, the absolute best part of the Macbook is the magnetic power connector[1], second-best being the low weight of the platform. (Other manufacturers of similarly designed high-end laptops make exactly the same mistake as the other manufacturers of high-end tablets in pricing their offerings at the same price as Apple.)

    Apple has been making corporate inroads with the iPad and iPhone, and with new grads wanting to use their Macs. But this trend line isn’t as steep as the Android trend line. It will be interesting to see what happens when these seriously collide. Be sure and point us at the updated report this time next year.

    [1] The older daughter bought a fairly high-end HP laptop when she started school 6 years ago. I have replaced the power connector on the motherboard twice on that laptop. If you ever have the same issue, and are savvy enough to figure out the part number and that you don’t really want to buy 3000 of them for 35 cents apiece, then you might prefer to find a reseller that buys them in bulk and will sell you one for $6 or $7.

    On a side note, this 6 year old laptop, which has seen much better days, was (as Windows laptops are wont to do) getting slower and slower, to the point where it became truly unusable. It is now running Natty Narwhal, and my daughter has yet to find anything she needs to do at medical school or in her lab that it is unsuitable for. I know that every year is going to be the year of Linux, but the data points I see are getting increasingly heartening for this prospect. Ironically, when Linux finally hits the big time, it will probably owe a lot of its success to Apple — the same people who view Android as a poor man’s iOS will probably start to view Linux as a poor man’s OSX. I can’t begin to predict when this will happen seriously, but I think it will, and that the hockey stick will eventually take off a la Android. Microsoft has zero cachet any more, and is unlikely to gain it back.

  128. @Patrick Maupin:

    If Linux (or Ubuntu or whatever) does begin to take off that way (and some people consider it “a poor man’s OS X”) the same people who call Android an iPhone knockoff will say the same about Linux. They will again be completely oblivious to (or studiously ignore) the colossal, ahistorical irony of the charge.

  129. @Patrick

    >No, the car analogy would be:

    My analogy was to the idea that if a product does not meet with your satisfaction, ownership of such a product would be in some way like being in a ‘well-decorated jail’. In this context the analogy is perfectly sound.

    You are making a different analogy. I assume you are making an argument that Apple might suddenly put out an iOS software update, or alter its server software, in such as way as to hinder your ability to use your device. First, of all, let’s just note that this is a completely separate argument, but let’s run with it.

    Well, it’s certainly technically possible that they could do such a thing. It’s also possible that Google or one of the carriers could do the same for Android.

    Why on Earth would they? It makes no sense whatsoever. Apple’s business is selling these devices to people. In order to do that people have to trust them. Breaching that trust would be catastrophic for them, and what would they achieve? Nothing. You always put some degree of trust in your platform vendors, no matter how technically proficient you are.

    I know what you are going to say: “The difference is that with an Android phone we could replace the compromised vendor software with a clean version of the OS”. Absolutely true … for about 0.001% of the people using the phones. For almost everybody there is no effective difference in the level of ‘openness’ between Android and iOS. I have made this point on this blog before. So, let’s just dispense with the idea that, on a societal level, Android provides any more control for the user than iOS.

    Even if you are one of the few people with the knowledge and inclination to replace the OS on your phone, you can still never be completely sure that the low level firmware and hardware in your devices is under your control. We all have to trust our vendors, even Android users.

  130. > You could say that of any product you don’t like. “This car only has 3 doors! I need 5 doors. Gaaah, I am stuck in a well-decorated jail! What will I do?!”

    I may have misunderstood, but don’t books purchased from the iBookstore only work with iBooks? If so that car analogy would be more like having to sign up to a contract that you will never use a car from another manufacturer and then never producing any cars that have more than three doors. It’s not the fact that iBook isn’t available for any desktops that makes it a “well-decorated jail”, but rather the restriction that you have to use their software, their way, on their device, the no desktops attitude just makes it obvious. (Many other eBook readers are just as bad of course)

  131. @TimP

    >I may have misunderstood, but don’t books purchased from the iBookstore only work with iBooks? If so that car analogy would be more like having to sign up to a contract that you will never use a car from another manufacturer and then never producing any cars that have more than three doors. It’s not the fact that iBook isn’t available for any desktops that makes it a “well-decorated jail”.

    Ok. I was in fact responding to that exact argument, put forth by SPQR, which is why I came up with the analogy I did.

    You are now making a separate argument, which is that the iBook system is a jail because of the DRM attached to some books sold through the system. If your point is that you can’t read some iBooks on non-iBook software then you are absolutely correct. And you’ll struggle to find any commercial ebooks (i.e. not the free stuff from gutenberg etc) available without DRM. I don’t like it, you don’t like it, but the reality of the ebooks industry right now is that publishers insist on these protections. Hopefully in time DRM for ebooks will start to fall away, as it is with music, but for now we are stuck with it.

    I don’t think it is reasonable to describe Apple as ‘evil’ for this, or to describe the entire iOS platform as a ‘jail’.

  132. @Ravi:

    They will again be completely oblivious to (or studiously ignore) the colossal, ahistorical irony of the charge.

    Exactly. Some of them are paid reactionaries, and some of them just don’t like change or think it can’t possibly happen. But the intertubes make it much easier for people to figure out the real capabilities and downsides of products before they buy, and (for example) right now the ASUS Eee Pad Transformer is #1 in both the Computer & Accessories and the Electronics/Computers & Add-Ons/Tablets categories at Amazon. There is no reason why the same thing couldn’t happen with a preinstalled Linux computer within the next couple of years.

    @Tom:

    You are making a different analogy.

    Yes. Your analogy was about not buying a car with three wheels. My analogy is about not buying a car from a manufacturer who can (and will!) sneak up in the middle of the night and remove one of the wheels that was affixed to the car when you bought it.

    I assume you are making an argument that Apple might suddenly put out an iOS software update, or alter its server software, in such as way as to hinder your ability to use your device.

    Or might just stop you from using some of the apps that were the reason you bought the device. Arbitrary and capricious updates to developer/content producer agreements are nothing new to Apple, and the latest round (started in February, slightly revised to make it not quite so onerous last month) has started having its effects:

    http://www.tomsguide.com/us/Kindle-Kobo-NOOK-App-Store-e-Reader,news-11971.html
    http://www.slashgear.com/kindle-wsj-spotify-ios-apps-more-lose-store-links-25167338/

  133. Books from IBookstore use the ePub file format, which is free and open. Apple does add FairPlay DRM (supposedly at the insistence of publishers) which is only mildly restrictive and apparently easily defeatable.

  134. @Tom:

    So, let’s just dispense with the idea that, on a societal level, Android provides any more control for the user than iOS.

    It’s coming. When Samsung gives out pre-release phones to CyanogenMod developers, it becomes increasingly difficult to pretend that manufacturers (besides Apple) are not seeing and responding to increasing user demands for control. One of the marketing checkboxes for Android phones will soon be (already probably is, informally) a “rock-solid, unbrickable bootloader that always lets you get back to default configuration.”

  135. @TimP:

    At this point in time *every* other notable ebook vendor is far better than Apple in terms of device restrictions. They all have a Windows reader (or browser option in the case of Google). They all have an Android app. Most have a Mac reader as well. When permitted (see Sony), they all have an iOS app. Some even go beyond that (Amazon is on WP7 and the TouchPad and Google claims support for every browser with JavaScript [though they probably only mean desktop browsers]).

  136. @Tom:

    > about 0.001% of the people using [Android] phones [would be able to root them]

    Do you really believe that only only 0.001% of people could follow the instructions for rooting their phone? In some cases these instructions are as simple as “Download file to memory card; turn of phone; insert memory card; turn phone back on; follow prompts”. Sure it’s probably not going to be the majority, but I’d be shocked if it was under 1% of the population, and wouldn’t be at all surprised if it was more like 10-20%, and then when you add in people who know someone who can root their phone (or in an extremely bad case are willing to pay to have their phone rooted) you’re probably looking at closer to 50-90% of people who could get their phone rooted if they wanted. Also if what Google hypothetically did was bad enough lots of people would be demanding non-Google versions of Android on their phones, and there is nothing stopping the hardware vendors from giving it to them, so you’d see a flood of non-Google supported phones on the market.

    > I don’t think it is reasonable to describe Apple as ‘evil’ for this [DRM in iBooks], or to describe the entire iOS platform as a ‘jail’.

    That’s correct: describing the entire iOS platform as a jail [for this reason] is unfair, but describing iBooks itself as a “well decorated jail” is not.

    > … you’ll struggle to find any commercial ebooks (i.e. not the free stuff from gutenberg etc) available without DRM.

    That’s true (though I know at least Baen sells eBooks without DRM), and as Ravi mentioned most other ebook vendors have a much wider variety of hardware that supported their DRM, which doesn’t make it less of a jail, but does make it a less restrictive jail.

  137. You could say that of any product you don’t like. “This car only has 3 doors! I need 5 doors. Gaaah, I am stuck in a well-decorated jail! What will I do?!”

    This metaphor is incomplete. You breeze past the point of it being a Jail. For it to be complete you’d have to add “And now that i’ve put my groceries in this 3 door foo-car I can’t change to a 5 door bar-car without completely re-buying them”.

  138. @Patrick

    Nobody knows what a bootloader is or why they would want one. Nobody knows what CyanogenMod is. Nobody understands why they would ever want to ‘change the OS’ on their phone. Most people don’t even know what an OS is. This stuff is all well and good for hackers, and I am pleased it is there, but it’s irrelevant to almost everybody who uses phones.

    > the latest round (started in February, slightly revised to make it not quite so onerous last month) has started having its effects:

    There is almost no impact on users resulting from this. The worst result is that for some apps, instead of having direct hyperlinks in apps to external stores, people will have to directly go to those stores themselves in the browser. It’s silly, I agree, but the user has only lost a small amount of convenience, and no functionality. It’s a VERY long way from this to your ‘1984’ scenario.

  139. @TimP

    >Do you really believe that only only 0.001% of people could follow the instructions for rooting their phone? In some cases these instructions are as simple as “Download file to memory card; turn of phone; insert memory card; turn phone back on; follow prompts”.

    I can tell you have not spent much time with novice users :)

    I once worked in a small company where I became the de facto tech support guy (even though that wasn’t my actual job) and I spent some time dealing with people who were not tech-savvy. Before I had that experience I would probably have agreed with you, but what I learned during that time shocked me. It will surprise you, but there is a significant number of people out there who don’t understand what a ‘file’ is. I am not kidding.

    And it is not just a case of knowledge. First you have to even *want* to root the phone, and for most people that will never happen. They don’t know what such a thing means, or why they would ever want it. You could try to explain it to them: ‘it means replacing the software on the phone’. ‘Software? Isn’t that a computer thing?’.

    If a normal person has their phone malfunction to such a degree that rooting might help they will just buy a new one, or take it back to the shop for repair.

    My 0.001% number is not people who might be capable of following instructions to root the phone. That number will be higher. What I mean is that 0.001% of people ever *would* root the phone, in reality. And that is what matters.

    >That’s correct: describing the entire iOS platform as a jail [for this reason] is unfair, but describing iBooks itself as a “well decorated jail” is not.

    I think that is fair. Any system that involves downloading DRM’d content is a ‘jail’, if you want to describe it like that. And there is plenty of that available on Android too by the way.

  140. @esr

    >The runaway-leader effect in these markets is quite well documented.

    ‘Runaway-leader’ is a bit different from ‘winner-takes-all’. You might well end up with single platform taking a big majority of the market, but you will still have strong alternative options. As long as you have the option to choose another product, which you clearly do, why are you worried?

  141. @Tom:

    I think you’re missing the point on bootloaders. It is true people have to trust their vendors regardless, but… If I don’t trust Red Hat as a Linux vendor (or if I just change my mind), I can switch to CentOS or Ubuntu or Debian or … (on the same hardware). If I change my mind about Apple I can what? Buy a new phone / tablet / laptop / etc? That is qualitatively different.

    You’re also not taking into account the automatic check the possibility of loading a third-party ROM like CyanogenMod provides. Just as side-loading means that there are all sorts of things that are more difficult or impossible for Google to do by leveraging their control of the Android Market, third-party ROMs and open bootloaders create all sorts of limits to control at the OS level as well.

    And while it is true that the typical user isn’t going to root their phone and change their ROM on a whim… Do you think ordinary users won’t understand “I’ll upgrade your phone and you’ll get XYZ feature you want”? or (even better) “you’ll get ABC feature that the latest OS update took away”? Do you really think people wouldn’t spring up to do that for them if it became important enough? We live in a world where kiosks in malls, vendors on eBay and others are willing to say that they will unlock / jailbreak your iPhone for you, after all. And that’s with Apple willing and able to claim it is illegal (not to mention actively fighting jailbreaking in OS updates). Installing a third-party ROM on a phone with an open bootloader is far easier and much less risky (for a vendor).

  142. @Tom:

    Nobody knows what a bootloader is or why they would want one. Nobody knows what CyanogenMod is. Nobody understands why they would ever want to ‘change the OS’ on their phone. Most people don’t even know what an OS is.

    You may not underestimate some of the populace, but you certainly underestimate a large portion of it.

    As long as you have the option to choose another product, which you clearly do, why are you worried?

    We’re not worried. Because Android is doing quite well. We would be worried if it wasn’t. Because network effects keep the small players small. And network effects combined with the kind of evilness Apple learned from Microsoft can make medium players small and then kill them. See OS/2.

    There is almost no impact on users resulting from this. The worst result is that for some apps, instead of having direct hyperlinks in apps to external stores, people will have to directly go to those stores themselves in the browser.

    I can’t find the link now, but I did see one app vendor complaining that they were not even allowed to tell their customers that they had a website where purchases could be made from inside the app any more.

    My 0.001% number is not people who might be capable of following instructions to root the phone. That number will be higher. What I mean is that 0.001% of people ever *would* root the phone, in reality. And that is what matters.

    That’s frickin’ awsome! If I simply take your 0.001% number to represent the number of people using CyanogenMod, then there must be 50 billion Android devices on the planet!.

    In reality, we have evidence that 0.4% of the Android-using devices on the planet are running CyanogenMod. That’s actually a pretty healthy base to grow from, and farther along on the growth curve than I would have thought they would be by now.

  143. > What am I supposed to think about people who find some “devastating” quotation but don’t even read the very next line?

    I’m not here to tell you what to think. Only Eric does that.

    I did read the quote in question.

    My only directive was (and I quote): “discuss”.

    The report is interesting in that, for the subject population, Android isn’t anywhere near as successful as other data would indicate. Android is rising rapidly, but we don’t know if these are replacement units, or if they’re due to Android becoming “more affordable” to a given segment of that population.

  144. >That’s frickin’ awsome! If I simply take your 0.001% number to represent the number of people using CyanogenMod, then there must be 50 billion Android devices on the planet!.

    500k *downloads* not active users. I would be surprised if one in ten of those downloads represents an active user, but we don’t know. And whatever the percentage is now, that number is likely to start going down as the smartphone market grows to embrace less technical users. Right now the Android installed based is disproportionately geeky compared with the general population. Even if that 0.4% number holds steady, that is still insignificant for strategic purposes.

  145. @Tom:

    To respond to a couple of other points:

    @Ravi

    >Apple may have changed their policy, but they haven’t realized their mistake. They just got too much blowback from the initial version

    I really don’t understand this. They listened to what people were saying and changed their policy.

    Changing the means and not the end is not learning from their mistake, in my opinion. I think their mistake is treating partners / developers as disposable nuisances when they decide to change direction. Far from fixing the problem by changing the policy, their policy change has underlined it. “Don’t mention your website in your app”? “Don’t tell users the reason for the change”? And so on? How is that showing these developers even minimal respect?

    Longer-term, what do you think is going to happen to those ebook apps? For every single one of them, they are (at best) only going to get a few new users via iOS. Non-iOS users (particularly Android users) are unquestionably easier to acquire and more valuable once you have them. Where are they going to invest their time and money from now on? Do they have any incentive to so anything with their iOS apps but make sure they are at least minimum-functionality readers for their content? On this path, what do things look like in a year or two? How is this anything but the ogre’s choice “die fast or die slow”?

    There is almost no impact on users resulting from this.

    The users filing one-star reviews for the Kobo and Kindle apps would disagree. Last I heard, Kobo had gone from 4.5 stars to 2.5 stars in a few days. Mathematically (assuming there’s no time-weighting or the like), that suggests Kobo nearly doubled their pre-update review total in a few days and they’re overwhelmingly one-star reviews of the new app.

  146. @Tom:

    Based on the previously limited availability of Gingerbread and Google’s Android Market version reports, at least one person concluded that CyanogenMod downloads and users tracked reasonably well (there was a significant Gingerbread slice not explainable as the either Nexus One or Nexus S). Sadly, I don’t remember where I saw that (though it was a fair bit back since they were making an argument based on the specific Gingerbread versions available to the Nexus phones at that time).

  147. Ok. I was in fact responding to that exact argument, put forth by SPQR, which is why I came up with the analogy I did.

    Ok, fair point. But the point your metaphor brushes past is that Apple’s vision doesn’t apply just to the product I (hypothetically) bought from them, but to everything that product touches.

    That nodding dog, the particular form of leather seats, the tires, windscreen wipers and the stickers on your rear window all of these things have to be registered and authorised by AppleCar and are subjected to (arguably) capricious rejection and removal practices. AppleCar’s response of course is that if you’re not using proper AppleCar car accessories then your car experience isn’t as good as they intend it. And if you should be so bold as to use them anyway, why AppleCar’s CEO thinks your license should be revoked and the car crushed as a warning to all. At the very least you should consider that sticker on your rear window as voiding your warranty.

    There are very few other products that attempt to go this far. I have no problem with having to tick a box when you try to install a “non-jobs approved app” saying “it’s not apples fault if i make my iphone run like ass and applecare is just going to tell me is ‘hard boot your phone back to factory default’.”, that response is defensible and even reasonable. It’s the whole “not on my product except through me” attitude that is distasteful.

  148. @Ravi:

    The users filing one-star reviews for the Kobo and Kindle apps would disagree. Last I heard, Kobo had gone from 4.5 stars to 2.5 stars in a few days.

    Yes, that’s what I remembered but couldn’t find. Tom really doesn’t get it that some people might have bought iPhones because they saw something like Kobo on a friend’s phone, and now, Apple has taken it away…

    @Tom:

    500k *downloads* not active users. I would be surprised if one in ten of those downloads represents an active user, but we don’t know.

    Actually, I think they do know. My bad for picking loosely-worded article. Try these for a start:

    http://www.webestigate.com/2011/07/19/cyanogenmod-reaches-the-half-million-user-milestone/
    http://stats.cyanogenmod.com/

    Even if that 0.4% number holds steady, that is still insignificant for strategic purposes.

    I don’t know wtf you mean by that. I fully expect that number to grow, and to put significant pressure on the Android handset vendors to do the right thing and not load the phones up with so much crapware they are unusable, and to keep the updates coming. If that number stays low as you seem to think it will, that will be because the handset manufacturers are keeping the users happy.

  149. Update: There is some sort of time-weighting of recent reviews (at least based on the US iTunes web pages):

    http://itunes.apple.com/us/app/kindle/id302584613?mt=8
    http://itunes.apple.com/us/app/id301259483?mt=8
    http://itunes.apple.com/us/app/barnes-noble-nook-for-iphone/id384910586?mt=8

    Kindle, Kobo and Nook are all at 2.5 stars for the current version and 3 stars for all versions. I don’t know whether Kobo’s previous claim of 4.5 stars was a current version or all versions claim. I suspect the former, but since I don’t know the weighting system it is hard to come to a solid conclusion.

  150. I don’t know wtf you mean by that. I fully expect that number to grow,

    There is a plausible argument that would go something along the lines of, “the kind of people who would install cyanogen are also highly likely to be the kind of people who would be early adopters of android”. This would suggest that the exceptions here or there would get swamped by the explosive growth from what are likely to be dumbphone converts.

    Depends on what you think the true watermark is. If you believe it’s 0.001%(for example) then it’s not surprising that you’d believe that the current value is overly influenced by self-selected early adopters. On the flip side, it’s easy for me to see how improvements in process simplicity (and support) could result in that current 0.4% watermark going much higher as you predict.

  151. @Tom:

    And you’ll struggle to find any commercial ebooks (i.e. not the free stuff from gutenberg etc) available without DRM.

    Baen.

  152. @Tom, Jakub Marebski:

    O’Reilly also sells DRM-free ebooks today. Besides that, some obscure author is planning on selling a few DRM-free ebooks from some site called Pottermore in the not-too-distant future.

  153. Sticking this here because the G+ comments are closed, and, well that’s why I’m writing here>

    Mr. Raymond, if you wish us poor folk in the cheap seats to be able to read what you wrote and comment on it, then either you’ll have to post it here, or get a better link.

    When I click on that link I get the G+ “we’re overloaded” webpage. Which I believed 2 weeks ago, now, not so much.

  154. Mr. Raymond, if you wish us poor folk in the cheap seats to be able to read what you wrote and comment on it, then either you’ll have to post it here, or get a better link.

    Don’t worry about that particular post. I expect he’s planning to write most of his commentable posts here, but the link was discussing a specific technical/policy issue with G+.

  155. @JonB:

    There is a plausible argument that would go something along the lines of, “the kind of people who would install cyanogen are also highly likely to be the kind of people who would be early adopters of android”.

    On the flip side, it’s easy for me to see how improvements in process simplicity (and support) could result in that current 0.4% watermark going much higher as you predict.

    Right. I think that there are a lot of Android early adopters who are not actually all that thrilled about bricking their new shiny. Process simplicity and support will go a long way.

    If manufacturers make it virtually impossible to brick the phones and consumers realize that this has happened, watch out! Demographically, I break the customers into two groups: the young and the fearless who weren’t around in the bad old days, and those whom Microsoft has already trained to do a full wipe and reinstall at least monthly.

  156. >My analogy is about not buying a car from a manufacturer who can (and will!) sneak up in the middle of the night and remove
    >one of the wheels that was affixed to the car when you bought it.

    Patrick, I have a lot of respect for you. While you and I don’t see eye to eye on lots of things, your arguments are almost always well reasoned and lacking the heated emotion element that a lot of nerd battles tend to have. That said, you say something like this quite often, and I’d like to you please take it in the spirit intended when I tell you to stop with the bullshit. Apple has not once auto updated anyone’s phone, nor snuck in the middle of the night and removed someone’s app. In fact, of the two who have admitted to having a kill switch, Google is the only one who has used that kill switch on an app (for admittedly a valid reason). The iPhone doesn’t do automatic OTA updates, and in fact, not even their computers do automatic updates. So can we please stop with the “Apple is going to take away your apps and everything you’ve worked for in the middle of the night and you won’t have a choice” crap? Until they actually do such a thing, this is an unfair charge.

  157. @tmoney:

    I’d like to you please take it in the spirit intended when I tell you to stop with the bullshit.

    I’m not bullshitting.

    So can we please stop with the “Apple is going to take away your apps and everything you’ve worked for in the middle of the night and you won’t have a choice” crap? Until they actually do such a thing, this is an unfair charge.

    Can you really, truly not see that if someone bought the iPhone partly based on the availability of the Kobo app, and then Apple makes a unilateral change in the developer terms that requires Kobo to either make their app more inconvenient, or give Apple a cut their business model might not be able to afford, that the resultant loss of user functionality is directly on Apple’s shoulders?

  158. @tmoney:

    BTW, a lot of Kobo customers seem to think it’s directly on Apple’s shoulders, and they’re not all that happy about it:

    http://www.theregister.co.uk/2011/07/25/apple_bookseller_online_sales_ban/
    http://www.facebook.com/notes/kobo/changes-from-apple-affects-kobos-iphoneipad-app/10150723555980541
    http://blog.kobobooks.com/iosstore/

    Google apparently thought Apple was bluffing:

    http://www.reuters.com/article/2011/07/25/idUS348730223220110725
    http://9to5mac.com/2011/07/24/wsj-kobo-falling-in-line-in-apples-app-store-google-books-is-out/

    I will admit, however, that my “sneak up in the middle of the night and remove” quote was slightly out of line. Apple’s far too brazen for that. They explained back in February exactly what was going to happen, and only backed down in some really minor ways. And it may be that some lucky users are grandfathered into being able to buy online (although some of the quotes on the Kobo site make me wonder about that), at least until they need new hardware, or inadvertently press the upgrade button.

    Class action suit in 3…, 2…, 1…

  159. >Can you really, truly not see that if someone bought the iPhone partly based on the availability of the
    >Kobo app, and then Apple makes a unilateral change in the developer terms that requires Kobo to
    >either make their app more inconvenient, or give Apple a cut their business model might not be
    >able to afford, that the resultant loss of user functionality is directly on Apple’s shoulders?

    That loss of functionality may be on Apple’s shoulders (as I said, their policy is horribly misguided and reduces not increases the user’s experience. I can understand wanting consistency and restricting in app purchasing to the in app API, but restricting external links is stupid. But none of that is apple sneaking in in the middle of the night and taking away functionality you already bought and paid for. If you bought the phone before the change and downloaded Kobo, you have the version of Kobo that works, and apple isn’t taking that away from you. It’s backed up on your computer and there is no auto update. If you bought your phone after the fact, and then discovered the inability to buy in the app and decided this was important to you, you can take the phone back.

    I’m fine with giving apple crap for the stupid things they do. And this was a stupid decision, and they deserve plenty of crap for it. But it isn’t apple sneaking in the middle of the night and taking away things their users have already bought and paid for.

  160. fake account writes: “The report is interesting in that, for the subject population, Android isn’t anywhere near as successful as other data would indicate. “

    Amusingly, that’s just another example of you cherry picking from that data.

  161. @tmoney:

    I made another post before you posted this, which is apparently stuck in the moderation queue, which (pre-)addresses some of your points. But the elephant in the room is the slightly uninformed consumer who thinks Apple has his back.

    If you bought the phone before the change and downloaded Kobo, you have the version of Kobo that works, and apple isn’t taking that away from you. It’s backed up on your computer and there is no auto update.

    But it isn’t apple sneaking in the middle of the night and taking away things their users have already bought and paid for.

    What about if you haven’t been following the controversy and see a “new version notification” and then update? It would sure feel like Apple taking away something that you had before in that case.

  162. @fake account:

    The report is interesting in that, for the subject population, Android isn’t anywhere near as successful as other data would indicate.

    I already explained that Apple’s popularity at UT is completely unsurprising to me. The surprising thing in that report was the 1/3 mobile penetration. Android should be kicking some serious ass soon with that growth potential and lower-cost prepaid and non-contract handset offerings.

    Android is rising rapidly, but we don’t know if these are replacement units, or if they’re due to Android becoming “more affordable” to a given segment of that population.

    Say what? The comscore report is all about installed base. Which is growing.

  163. @fake account:

    There is one other possibility as well. My daughters have AT&T iPhone 4s, and very consciously keep them in WiFi-only mode for data. I know a lot of people with Android phones who have older unlimited data plans. If they never bother to stop using the carrier’s data, then UT wouldn’t even know they exist.

    That would easily explain why UT only thinks 1/3 of their students use mobile data. A lot of the rest of them are probably just not bothering with UT’s WiFi network for their mobile data needs.

  164. @tmoney

    >I’m fine with giving apple crap for the stupid things they do. And this was a stupid decision, and they deserve plenty of crap for it. But it isn’t apple sneaking in the middle of the night and taking away things their users have already bought and paid for.

    Right. And, by the way, any developer on any platform can issue an update to their software that removes functionality. Doesn’t mean you have to install it.

    @Ravi

    >O’Reilly also sells DRM-free ebooks today. Besides that, some obscure author is planning on selling a few DRM-free ebooks from some site called Pottermore in the not-too-distant future.

    What is this supposed to prove? The point remains that DRM is a fact of life in the ebook industry today. Not a good fact, but a fact nonetheless. I guarantee that if publishers didn’t insist on DRM then there would be no DRM. You will probably have noted that all the examples given of DRM free ebooks are of small or independent publishers who can control the whole process themselves.

    @Patrick

    >I don’t know wtf you mean by that. I fully expect that number to grow, and to put significant pressure on the Android handset vendors to do the right thing and not load the phones up with so much crapware they are unusable, and to keep the updates coming. If that number stays low as you seem to think it will, that will be because the handset manufacturers are keeping the users happy.

    What I mean is that for the purposes of our discussion about whether the ability to root phones is relevant for the general population, 0.4% proves that point (and I strongly suspect that the number of active c-mod users is actually far lower than even that). And you can ‘fully expect’ that number to rise if you like, but you must have noticed that the android user base is about to be swamped with millions of people who don’t know a web page from a word document.

  165. >What about if you haven’t been following the controversy and see a “new version notification” and
    >then update? It would sure feel like Apple taking away something that you had before in that case.

    They might indeed. And insofar as Apple’s policy changes forced this application change on the developer, they would be correct. But ultimately, it still lies on the user to verify that they actually want the updated software before they install it.

    As a for instance, Skype recently updated their android app. Under the old version, when I dialed a number and selected to complete the action with Skype, it would start up, sign in and complete the call. Under the new version, it starts up, signs in, and fails to complete the call because it can’t figure out the country code and doesn’t keep a default setting. It also then doesn’t quit when you want and constantly signs you in and out of the skype network every time you change from wi-fi to cell data and back again.

    Insofar as Skype made the decision to implement these features and break the functionality that I relied on, the blame for this falls on them. But in the end, I’m the dumbass who installed a new version of the software without first checking to ensure there wasn’t anything that I relied on that would break. Even though I should have no reason to expect calling functionality to break between versions (seeing as that is a major function) it’s still on me for not reading into what this update did before I made the choice to install it.

  166. @Tom:

    Right. And, by the way, any developer on any platform can issue an update to their software that removes functionality.

    Yes, and any developer on any other platform can issue an update to their software that doesn’t remove functionality. But apparently that’s not always allowed here in paradise.

    Doesn’t mean you have to install it.

    What if the user wasn’t paying attention and doesn’t know that a large chunk of functionality is about to go missing? People have been trained to hit the upgrade button. What you say made more sense back in the days of 3.5″ floppies or CD-ROMs or DVDs, or when working with reasonable companies that let you download older software versions.

    0.4% proves that point

    There you go again. First it was 0.001%. Now a number that’s 400 times that big proves your point. Of course, you don’t say how. It’s still early days for CyanogenMod.

    (and I strongly suspect that the number of active c-mod users is actually far lower than even that).

    Then you still aren’t paying attention. Those numbers apparently come from that subset of users that says it’s OK for their phones to tell CyanogenMod that they are using it.

    you must have noticed that the android user base is about to be swamped with millions of people who don’t know a web page from a word document.

    Yes, but collectively people are quick learners. And CyanogenMod is just one manifestation of the openness of Android. Some manufacturers and carriers will be nicer to users than others and word will get out. I’ve also noticed that some of the phone manufacturers are starting to take CyanogenMod seriously. I’m waiting for the day when one of them starts working directly with CyanogenMod and directs their users there for upgrades.

    This whole side-tangent of CyanogenMod started when you wrote this mis-prediction:

    I know what you are going to say: “The difference is that with an Android phone we could replace the compromised vendor software with a clean version of the OS”. Absolutely true … for about 0.001% of the people using the phones.

    I’ve shown that you’re off a couple of orders of magnitude, and I’ve already explained multiple times that competition — that the mere existence of CyanogenMod — means that competitors in the Android space have to treat users a bit better. You view 0.4% as a low number that is meaningless. I view it as a nice high number that indicates that CyanogenMod will be around for a long time to help keep the more venal instincts of the vendors in check.

  167. @tmoney:

    They might indeed. And insofar as Apple’s policy changes forced this application change on the developer, they would be correct. But ultimately, it still lies on the user to verify that they actually want the updated software before they install it.

    Words fail me. I’m going to need somebody else to translate for you why I think this is terribly wrong, because we’re obviously from different planets.

  168. @Patrick

    >What if the user wasn’t paying attention and doesn’t know that a large chunk of functionality is about to go missing? People have been trained to hit the upgrade button. What you say made more sense back in the days of 3.5? floppies or CD-ROMs or DVDs, or when working with reasonable companies that let you download older software versions.

    But, again, this is just as true on Android or any other platform. And @tmoney has given us the concrete example of Skype on Android.

    >There you go again. First it was 0.001%. Now a number that’s 400 times that big proves your point. Of course, you don’t say how. It’s still early days for CyanogenMod.

    By 0.001% I just mean ‘an insignificantly small number’. The point is that it is not something normal people are going to do.

    >I’ve shown that you’re off a couple of orders of magnitude, and I’ve already explained multiple times that competition — that the mere existence of CyanogenMod — means that competitors in the Android space have to treat users a bit better. You view 0.4% as a low number that is meaningless. I view it as a nice high number that indicates that CyanogenMod will be around for a long time to help keep the more venal instincts of the vendors in check.

    But that’s *not* the point I am contesting. The point I am contesting is the idea that a significant number of real people will ever in reality root their phones. All phone manufacturers have pressure on them to make sure their users have a good experience, it’s called the free market.

  169. >Words fail me. I’m going to need somebody else to translate for you why I think this is terribly wrong,
    >because we’re obviously from different planets.

    Sorry, but it seems perfectly reasonable to me. If you are going to insist that you be in control of your device, then you are responsible for the software that is installed on that device. The only responsibility that Apple bears in this matter is their policies which mandated the update. But until Apple is forcibly pushing updates on their users, it is still the user’s responsibility to decide whether they want an update or not.

  170. @PapayaSF:

    Good smear article. Mentions Samsung by name and then uses an unnamed source to say that some manufacturers are having 40% return rates. Apple got their money’s worth out of that one.

  171. @papayaSF

    >Speaking of shipped vs. sold: Android’s Dirty Secret: Shipping Numbers Are Strong But Returns Are 30-40%.

    Interesting, but the article is weakly sourced.

  172. @Tom:

    The point I am contesting is the idea that a significant number of real people will ever in reality root their phones.

    And maybe they won’t if the vendors relent and make it unnecessary.

    All phone manufacturers have pressure on them to make sure their users have a good experience, it’s called the free market.

    Sure, but this is balanced by the pressure to make money via alternate revenue streams. (See Apple.) CyanogenMod helps move the equilibrium point for Android phones toward less obnoxious preloaded bloatware, and it does this even if a very small percentage actually use it.

  173. >Sure, but this is balanced by the pressure to make money via alternate revenue streams. (See Apple.)

    If you look at Apple’s most recent earnings report (I linked to it earlier) you will see that Apple’s revenue overwhelmingly comes from hardware sales. iTunes and associated services are insignificant. Apple has said in the past that they break even on running the iTunes store. I suspect that this is no longer the case for apps and in-app stuff, but it is certainly nowhere near being a significant earner for them. They are never going to jeopardise their massive hardware earnings to make a few bucks on ‘alternate revenue streams’.

  174. One of the points to the bootloader, thence the ability to load cyanogenmod, or rom of your choice, is that the fact you can do it if you get annoyed enough limits the degree to which the vendors will be willing to annoy their customers.

    Yes, very very few of the phone and tablet users will ever do it. But the knowledge on the part of the vendors that they could will be a strong reason for those vendors to not get too obnoxious about what they put on the phones or tablets as the standard load.

    True, if right now I set out to install cyanogenmod on my nexus s, it would take me hours, and probably a significant risk if bricking it. But if T-mobile gets too hard to get along with, I could.

    If they do that to very many people, pretty soon someone like Patrick or Eric or … will do first a step by step paper on how to do it — may already exist. Then someone will do a script that will take all the bits and load them, then someone will set up a web site with all the bits and the scripts together in a package that you can load onto a usb stick and install with one click.

    Much of the above may already exist — I haven’t dug into it much at all — because T-mobile hasn’t irritated me enough to make me want to go to the trouble. Actually, aside from coverage and lack of co-operation with other carriers when outside of the main areas in the pacific northwest, t-mobile and the nexus s haven’t limited me for anything that I wanted to do.

    The fact that all the above could happen is much of what limits how hard the vendors will push to control the environment.

  175. @Tom:

    If you look at Apple’s most recent earnings report (I linked to it earlier) you will see that Apple’s revenue overwhelmingly comes from hardware sales.

    This is true now. It may remain true if Apple decides that being a niche player is OK. But if Apple decides to get down in the mud and try for larger market share, their hardware margins are headed south, and they will be looking for replacement revenue.

    They are never going to jeopardise their massive hardware earnings to make a few bucks on ‘alternate revenue streams’.

    Then why, pray tell, are they fucking with Kobo?

  176. @Tom:

    The Pottermore reference was intended to demonstrate that DRM is not inevitable and that DRM-free ebooks are going to be exponentially more mainstream a year from now than they are today. That will have good consequences.

  177. @Jim Hurlburt:

    One of the points to the bootloader, thence the ability to load cyanogenmod, or rom of your choice, is that the fact you can do it if you get annoyed enough limits the degree to which the vendors will be willing to annoy their customers.

    That’s one manifestation of a simplified model. One of the reasons for the success of the PC industry is that it’s easy to get started (the PC usually comes with an OS), but the OS is not an integral part of the PC — just bits on the hard disk. The BIOS loads the OS. Everybody knows you can replace the OS without damaging the machine. The BIOS itself is a bit trickier — you can often replace it, but you might run a risk of bricking the device.

    With some phones, the bootloader and OS reside in the same flash. This makes brickage more likely. But there are ways around this, like being able to lock sectors in the flash.

    I think that, as the phone becomes a more indispensable computing device, the PC model is more likely to take root. Samsung, for example, sells phones with at least 3 different operating systems. There’s no reason why those phones have to have different hardware. Just as we’re slowly disintermediating the carriers from phone sales, I think we’ll slowly see handset vendors decide that google and CyanogenMod are doing a good enough job on the software, and that they don’t really want to be in the software business after all (except for device-specific drivers).

    One of the few things that keeps them in the software business is the incremental revenue for installing crapware. But it’s my understanding that google’s offering a carrot to reduce that in the form of better ad revenue sharing. It will be interesting to see what happens.

  178. @Patrick

    >This is true now. It may remain true if Apple decides that being a niche player is OK. But if Apple decides to get down in the mud and try for larger market share, their hardware margins are headed south, and they will be looking for replacement revenue.

    It’s always been true. It’s how Apple operates; they don’t do low margin. But I suppose the remote contingency exists that they could abandon their massively successful model, start selling low-margin shit hardware and pivot their business to make money selling ebooks. Well, there is also the remote possibility that the Android manufacturers could band together and decide to lock down their phones to make c-mod an impossibility so they can load it up with more crapware. I don’t think that is remotely likely, but it’s possible.

    We all have to trust our hardware vendors not to suddenly start behaving erratically. Let’s just deal with reality until such a thing happens.

    >Then why, pray tell, are they fucking with Kobo?

    I know you will find this hard to accept, but Apple’s rationale behind this is that they are trying to create an app ecosystem that people trust. They want their users to spend freely on the app store, safe in the knowledge that their credit card info isn’t going to be stolen, or that they will have to go through an annoying registration process to buy things. To that end, they have created a centralised purchase system, so that everything you buy through the store, or through an app, goes through Apple’s trusted one-click process.

    @Ravi

    >The Pottermore reference was intended to demonstrate that DRM is not inevitable and that DRM-free ebooks are going to be exponentially more mainstream a year from now than they are today. That will have good consequences.

    I agree! But the fact remains that TODAY, if you want to create an ebooks store offering books from major publishers you are going to have to deal with DRM.

  179. >Good smear article. Mentions Samsung by name and then uses an unnamed source to say that some manufacturers are having 40% return rates. Apple got their money’s worth out of that one.

    Right. This was linked on HackerNews early this morning and even the commentariat there noticed that it was inconsistent and evidence free.

  180. @tmoney:

    If you are going to insist that you be in control of your device, then you are responsible for the software that is installed on that device.

    But with other devices from other vendors, if you make a mistake, you can usually rectify it one way or another.

    The only responsibility that Apple bears in this matter is their policies which mandated the update.

    And we strongly disagree about the morality of this policy.

    But until Apple is forcibly pushing updates on their users, it is still the user’s responsibility to decide whether they want an update or not.

    Consumers are effectively being taught to rely on the cloud for data storage, and to not even bother with physical copies of apps, because they can always be re-downloaded. Apple itself is complicit in cloud-training. This exception to the trend will be a rude shock for many of their customers.

    @Tom:

    Apple’s rationale behind this is that they are trying to create an app ecosystem that people trust.

    That’s hilarious. If it were true, they’d be taking care of these customers, instead of wasting time chasing down vendors who have a separate arrangement with their customers.

    No, I’m afraid that Occam’s Razor shows that requiring all vendors to give it 30% and making it way too easy to buy stuff means that it’s all about the incremental revenue.

  181. @Patrick

    >That’s hilarious. If it were true, they’d be taking care of these customers

    You’ve got to be kidding me. Just don’t give the kid your password. Apple deliberately makes it so you have to enter your password before buying anything. That’s the craziest lawsuit I’ve seen for a while.

  182. Tom Says:
    > @Patrick
    >> the mere existence of CyanogenMod — means that competitors in the Android space have to treat users a bit better.
    > But that’s *not* the point I am contesting.
    > The point I am contesting is the idea that a significant number of real people will ever in reality root their phones.
    > All phone manufacturers have pressure on them to make sure their users have a good experience, it’s called the free market.

    I gotta go with Patrick here.
    Before CyanogenMod, the phone manufacturers were subject to far more pressure from the telopolies, than from end-users.
    Now.. there is a limit to the evil the telopolies can impose on us.

  183. @Tom:

    Just don’t give the kid your password.

    It is my understanding that if you set it up where a password is required, that password is required even for free apps, or for upgrades to existing apps.

  184. Apple has made Kindle, etc remove the links to their bookstores for one simple reason: Apple wants 30%. Any talk of “consistency” or any other BS is pure spin. From a user standpoint, this policy is a negative. I can’t believe anyone is arguing otherwise.

    “Plenty of people want to read ebooks on computers.”

    The only people who read ebooks on computers are a very niche nerd audience. It’s a horrible experience compared to paper or a tablet.

  185. @Patrick

    >It is my understanding that if you set it up where a password is required

    There’s no other way to set it up.

    >that password is required even for free apps, or for upgrades to existing apps.

    Correct. But you know up front that if you give somebody your password then they have the ability to do anything with your account, including spend money.

    Under those circumstances I would not think it advisable to give the password to your children, and if you do then it’s your problem.

    If you want to give your children autonomy then they need to have their own device with their own account. If you want them to use your device then you are going to have to control app installations.

    I am not clear what Apple is supposed to have done wrong here.

  186. >But with other devices from other vendors, if you make a mistake, you can usually rectify it
    >one way or another.

    And where shall I go for the older Skype app for my Android phone to restore lost functionality? (Seriously if you can point me somewhere that would be awesome)

    >And we strongly disagree about the morality of this policy.

    You misunderstand me. I do not argue that the policy is stupid and misguided (immoral? perhaps, depending on it’s purpose, which we do not know and can only guess at). I am arguing that in implementing this policy Apple has not come like a thief in the night to steal from their paying customers. They have not pushed this update to customers without choice, they have not disabled the application by way of kill switch. They have not hacked into your computer and removed the app, and they haven’t deleted it from your phone. They haven’t pushed a new software update to the phone that kills the old app. They have implemented a stupid policy, and they are wrong in that. But this constant fear that you have (and keep spouting whenever such topics come up) that Apple is sneaking in and deleting your stuff just isn’t true and it’s bullshit, and until Apple does something like that (unlike Google which again, already has) the charge is unfair and distracting from the real point.

    >Consumers are effectively being taught to rely on the cloud for data storage, and to not even
    >bother with physical copies of apps, because they can always be re-downloaded. Apple itself is
    >complicit in cloud-training. This exception to the trend will be a rude shock for many of their customers.

    Indeed it will. And I hope that consumers learn this lesson well, because trusting your data to “the cloud” is in my mind, even more dangerous than trusting it to a proprietary app. If you thought that getting your data out of an old unsupported app sucked, just wait until you have to get your data out of someone elses hard drive.

    >That’s hilarious. If it were true, they’d be taking care of these customers, instead of wasting time
    >chasing down vendors who have a separate arrangement with their customers.

    Come on, this is stupid. If you give your amazon account password to your kids, they can buy stuff with your credit card. Is that Amazon’s fault?

    >Apple has made Kindle, etc remove the links to their bookstores for one simple reason: Apple
    >wants 30%. Any talk of “consistency” or any other BS is pure spin. From a user standpoint, this
    >policy is a negative. I can’t believe anyone is arguing otherwise.

    I don’t think anyone here has argued the policy isn’t a net negative. That doesn’t mean that there aren’t perfectly reasonable non money-grubbing evil reasons for such a policy. If it were all about 30%, then the obvious solution is to reject all apps that allow any sort of outside purchasing and don’t allow in app using the APIs as well.

  187. Incidentally, that link to splatf is right on the money. The policy is wrong, and the end result is a crappier experience for the users. My only objection in all of this has been the characterization of this as sneaking in the night and taking away that which you already have.

  188. @Phil

    >Perfect take on Apple’s banning links to buy shit in Apps:

    It is perfectly true that the user experience in some of these apps is now worse. But there are other factors to the decision. Apple also has to take into account the overall user experience for its app ecosystem, and it is in everybody’s best interest that users feel safe and secure when buying things within that ecosystem. It’s not black and white. Creating a good user experience in this sort of complex system is a balancing act.

    Let’s separate the rule that apps have to use in-app purchase from the 30% cut, because they *are* different. I think the 30% is a bit crazy, and I hope Apple changes its mind on this. I think they will recognise that the 30% rule for in-app purchases will have a negative net impact on the user experience for the store.

    If their plan was to steal 30% of Amazon’s revenues, then I think you will agree that their plan has spectacularly failed.

    None of this amounts to Apple being evil or evidence that Apple is going to reach into your phone and start randomly wrecking your device.

  189. @Tom:

    Under those circumstances I would not think it advisable to give the password to your children, and if you do then it’s your problem.

    @phil:

    ” making it way too easy to buy stuff ”

    What do you mean by this?

    If your child pesters you into buying them an iPhone, then it may be reasonable to let them download and play with free apps to their heart’s content. It also may be reasonable to let them update apps they have previously purchased. Depending on the child, it may not be reasonable to let them purchase apps and/or purchase in-app stuff.

    Requiring the same password for all of these functions makes it difficult for parents, and should probably be changed. However, it is in Apple’s interest to make “buying” a free app take exactly the same steps as buying a paid app — that removes one more psychological impediment to spending money.

  190. @Tom:

    If their plan was to steal 30% of Amazon’s revenues, then I think you will agree that their plan has spectacularly failed.

    Obviously, Amazon wasn’t going to give them 30%, so obviously that wasn’t the plan. In fact, I can easily believe that the plan involved making it harder to buy stuff from Amazon (a direct competitor in some segments now, probably more later) and that the plan succeeded as expected.

  191. @Patrick

    >If your child pesters you into buying them an iPhone, then it may be reasonable to let them download and play with free apps to their heart’s content. It also may be reasonable to let them update apps they have previously purchased. Depending on the child, it may not be reasonable to let them purchase apps and/or purchase in-app stuff.

    Right. It might be a nice feature if Apple added an option not to require the password for free apps and updates, or if they added a bit more granularity to their system.

    But usually we don’t sue companies because their products don’t have all the features we want. It’s perfectly clear to everybody that if somebody has your password then they have the ability to buy things with your account.

    And, if the child has their own iPhone then just give them their own account, either without a credit card attached to it, or with a limited pre-paid card that you can control.

  192. >If your child pesters you into buying them an iPhone, then it may be reasonable to let them download
    >and play with free apps to their heart’s content. It also may be reasonable to let them update apps they
    >have previously purchased.

    So why then isn’t it also reasonable to either them their own account with spending limits?

  193. @Tom:

    I’ll throw in another dimension where in-app purchase is problematic. If you buy a book from Amazon via Apple’s IAP using your iTunes account is the purchase correlated with your Amazon account or not? If so, one of the big arguments for IAP has disappeared (privacy), plus you have the extra inconvenience of making the cross-account connection. If not, now you have a Kindle book you can only use in Apple’s ecosystem (since Amazon doesn’t know you’re allowed to have it outside). Neat trick. Similar considerations apply to newspaper, magazine, streaming music/video and other subscriptions, of course.

    That actually dovetails nicely with my theory about why Apple is doing all of this. I don’t think they care about the money nearly as much as they care about the lock-in. The more Apple/iOS-only services and content each customer is using, the more likely Apple will be able to hold onto that customer when the disruptive wave they know is coming finally hits. Of course, this sword cuts both ways, which is why Apple’s behavior had been puzzling me. In this view, Apple might be willing to let go of some customers in order to tighten their grip on the rest.

  194. @tom:

    But usually we don’t sue companies because their products don’t have all the features we want. It’s perfectly clear to everybody that if somebody has your password then they have the ability to buy things with your account.

    I agree that Apple went a long way towards fixing the problem when they started requiring the password for in-app purchases. But I think they created the legal and moral problem for themselves by being so cavalier as to not require that password for the longest time. And even with the password, a lot of parents are going to view the availability of juicy in-app content as a “moral hazard” and of course the class-action lawyers salivate at the prospect of getting the right jury. I’ve been a juror on cases like that and it sucks. There are no heroes here.

    @tmoney:

    > So why then isn’t it also reasonable to either them their own account with spending limits?

    How easy is it to do that on the app store? I have no idea. I did spend a bit of time searching for it, and found a way to set up an account without a credit card, but that seems a bit problematic for when you do want to spend money. Apparently, you can also use multiple accounts on a single device, but sometimes you get scary-looking messages when switching from account to account.

  195. >How easy is it to do that on the app store?

    See setting up an allowance: http://support.apple.com/kb/ht2105
    Or simply set up an account without a payment option, which you can do as simply as choosing “none” for the payment option: http://gigaom.com/apple/ios-101-manage-your-familys-itunes-store-spending/

    >but that seems a bit problematic for when you do want to spend money.

    Give your kid a gift card (available at a retailer near you!) and let him redeem it. Problem solved.

    >Apparently, you can also use multiple accounts on a single device, but sometimes you get
    >scary-looking messages when switching from account to account.

    The only “scary” message I ever got was asking me to authenticate on the account when attempting to load media.

  196. @Ravi

    > If you buy a book from Amazon via Apple’s IAP using your iTunes account is the purchase correlated with your Amazon account or not?

    Obviously it is up to Amazon how they do it, but it would be crazy if any future Kindle IAP implementation did not allow books purchased in an iOS app to be accessed on non-iOS devices. I am not familiar with the technical details of IAP, but I am sure this is quite possible. There’s no privacy violation that I can see, since Amazon will not be given your credit card details, or any other personal details that they don’t already have. All they are going to get is ‘user x has purchased product y’ and then some money sent to them by Apple.

    @Patrick

    >And even with the password, a lot of parents are going to view the availability of juicy in-app content as a “moral hazard”

    So now you are arguing against the very concept of in-app purchases? I don’t quite follow this.

    If parents want to view in-app purchases as a “moral hazard” then that is their prerogative. Apple can’t be held responsible for insanity.

  197. Ravi:

    The more you tighten, the more star systems yadda yadda.

    I thought Apple learned this lesson. The thing that brought the Mac back from the brink of extinction wasn’t the Intel hardware, it was Mac OS X. Developing for the Mac used to be an expensive proposition; the developer docs, Inside Macintosh, were costly and cumbersome to read in their own right, not counting the added cost of IDEs and development tools. With Mac OS X came free tools, free docs, familiar Unix underpinnings, an open source core, and the greatest GUI layer ever made (Cocoa).

    Early in the smartphone races it seemed Apple was willing to strike a balance between central control of The Apple Experience(tm) and enough openness to make getting started with development easy, fun, and profitable. Now it looks like they’re willing to play a hard game. It will cost them if they keep it up. Apple is headed toward nineties-style marginalization if they think the loyalty of their hardcore fans is enough.

  198. @Tom:

    >>And even with the password, a lot of parents are going to view the availability of juicy in-app content as a “moral hazard”

    >So now you are arguing against the very concept of in-app purchases? I don’t quite follow this.

    Sorry, a bit unclear:

    (1) Apple apparently didn’t have any passwords on in-app purchases. Even they eventually realized this was bad (possibly only after the FTC expressed an interest…)

    (2) Apple’s response was to put in-app purchases under the same password as other purchases.

    (3) (Apparently) hence the instant lawsuit, which perhaps correctly views in-app purchases as way more tempting and costly than average purchases of apps.

    I appreciate the education from you and tmoney on the various ways to fund a minor without going too far in the hole. However, I have to say, I did spend some (but not a lot of) time on Apple’s site and other sites trying to find similar material, and it was difficult to piece together, and it still seems a bit problematic. It will be interesting to follow the lawsuit and see whether it simply gets tossed, or how Apple approaches it.

    @Jeff Read:

    > The thing that brought the Mac back from the brink of extinction wasn’t the Intel hardware, it was Mac OS X.

    Some of both. I think that Intel hardware greatly improved their margins, and also gave customers the ability to easily run other stuff if they decided they didn’t like OS X. I know several people who run Windows or Linux on MacBook Pros. Also, there could be a lag, but sales really picked up after the transition to Intel, more so than the transition to OS X 3 years earlier.

  199. >>The only people who read ebooks on computers are a very niche nerd audience. It’s a horrible experience compared to paper or a tablet.

    I guess that makes me a very rich nerd then — and my daughters too. Gee, I hadn’t noticed all that wealth till now.

    I’ve been buying Baen books since shortly after they started the webscriptions. Their books read quite nicely in html.

    I will grant that when I got the nook a year or so ago, I read on the computer much less. But that’s because I can sit in an armchair with the nook. And take it to work, thence to lunch. The kids read from laptops, their phones and a nook.

    Perhaps our tolerance of horror is higher than yours?

  200. @Tom:

    It is up to Apple what Amazon would or wouldn’t learn about their customers via IAP, not Amazon. Newspapers and magazines certainly had significant issues with demographic information they’d ordinarily get through a subscription being opt-in via IAP. What if the Amazon account connection were opt-in? There is, after all, somce privacy component to the connection. Apple could also argue that users shouldn’t be required to have an external account to buy anything via IAP. What if purchases had to be synced? Either or both would still be a notable “speed bump” for non-iOS use of iOS purchased ebooks.

    You may think it would be crazy for there to be interoperability barriers between iOS and non-iOS books in the iOS Kindle app, but that very possibility was suggested more than once over the past several months. One reason was so Amazon could add an iOS “delivery fee” to cover Apple’s 30% cut, but another was exactly the sort of account issues I raise here.

    In any case, we’ve drifted from the original point which is that this entire debacle is a clear and unmistakable example of Apple being evil with respect to their governance of their app store. Even John Gruber (Daring Fireball) now admits that no one benefits from this nonsense but Apple, not even users. Isn’t that a classic, succint statement of evil selfishness?

  201. >I guess that makes me a very rich nerd then — and my daughters too. Gee, I hadn’t noticed all that
    >wealth till now.

    Niche my friend, not rich.

  202. @Jeff Read:

    There’s plenty of other evidence that Apple hasn’t learned this particular lesson. Consider the requirements to develop an iOS application: First, you have to buy a Mac… Plenty of people never get to the next step. For Apple’s purposes, enough do, but that doesn’t help the many who don’t.

  203. @Jim

    >I guess that makes me a very rich nerd then — and my daughters too. Gee, I hadn’t noticed all that wealth till now.

    What’s this about wealth? I don’t quite follow what you are saying.

    >I will grant that when I got the nook a year or so ago, I read on the computer much less. But that’s because I can sit in an armchair with the nook. And take it to work, thence to lunch.

    That’s the whole benefit of the tablet form factor! You can sit back in a chair, hold it in your hands, and read comfortably.

  204. @Jim Hurlburt

    “I guess that makes me a very rich nerd then — and my daughters too. Gee, I hadn’t noticed all that wealth till now.”

    I said “niche” not “rich”

    I think it might be useful to have your e-book content available on your PC once in a while. But I think almost everyone will end up choosing their tablet/e-reader instead…. especially for extended reading.

  205. Ravi,

    There’s plenty of other evidence that Apple hasn’t learned this particular lesson. Consider the requirements to develop an iOS application: First, you have to buy a Mac… Plenty of people never get to the next step. For Apple’s purposes, enough do, but that doesn’t help the many who don’t.

    In theory you’re right; in practice this obstacle is not so insurmountable as you might otherwise assume: a significant chunk — probably most — of the world’s top-tier developers already own a Mac.

  206. @Jeff

    >In theory you’re right; in practice this obstacle is not so insurmountable as you might otherwise assume: a significant chunk — probably most — of the world’s top-tier developers already own a Mac.

    Right. I think pretty much anybody who wants to develop for iOS of OS X has a Mac already. Also, Apple’s development tools and documentation is completely free. You do have to pay a small fee (somewhere around £50/60) to get access to their stores, but it’s a small price to pay for free distribution, hosting, payment processing, and exposure. In fact there is a decent case that Apple’s platforms have a lower barrier to entry for developing graphical user-facing apps than any other significant platform in the world except for the web.

  207. >>I said “niche” not “rich”
    Sorry I miss read.

    I think it might be useful to have your e-book content available on your PC once in a while. But I think almost everyone will end up choosing their tablet/e-reader instead…. especially for extended reading.

    Up untill I had the tablet, I read Baen from the brower fairly often. The nook is more convenient, but not hugely so — for books from Baen that I havn’t gone throught the hoops of downloading and transfering yet, I still to. Also, the search capabilites of Baen’s site are better than the nook when I’m hunting for something that I don’t remember exactly.

    I do have over 750 books on my nook to paw through, the bulk of them ebooks from Baen. Probably B&N is less than 100 so far along with a smattering of glutenberg and others. I did find the ‘walter bupp’ tales from Galaxy in the sixties or so. Cool.

    The kids add to the stack and I don’t prune it as a normal thing.

    for me, 12 years or so of Baen webscriptions, 1 year of nook. I would not weep if I lost the nook, whimper a bit, but not weep. The computer is a pretty good grade of OK — Nook is just some better for most purposes.

    From my perspective, one of the coolest things about Baen and webscriptions is that *every* one of the ebooks they have published is still up and for sale. Not a one of them (at least that I’ve noticed) is out of print for this purpose. And, no drm. I probably would not have bought nearly as many of them if I couldn’t store them on my computer/nook and let my kids read them.

  208. @Tom, Jeff Read:

    I’m not talking about the world’s top-tier developers. They have or can get anything they want. I’m talking about the other end of the spectrum – the student trying to use their education, the hobbyist tinkering at night, the developer from the third world who is just trying to join the party. Sure, even many of them will have Macs, but many won’t and for them it will be an insurmountable obstacle that Apple could easily remove. Plenty of people who entered Google’s Android Developer Challenge (including at least one winner, IIRC) couldn’t have easily entered the same sort of contest for the iPhone.

    In a practical sense, does that affect the quality of applications developed for the iPhone? Probably only marginally. But it is certainly evidence that Apple hasn’t really learned the lesson, deep down.

  209. >one of the coolest things about Baen and webscriptions is that *every* one of the ebooks they have published is still up and for sale.

    Holly Lisle’s books vanished between May and now (except for the Free Library books, and I can still download all those that I’ve bought).
    I don’t know what actually happened; she just says “the rights reverted back to me”.

  210. @Ravi

    >I’m not talking about the world’s top-tier developers. They have or can get anything they want. I’m talking about the other end of the spectrum – the student trying to use their education, the hobbyist tinkering at night, the developer from the third world who is just trying to join the party.

    Come on, this is a bit silly. Practically every student I have met over the last couple of years who is into computers has a Mac already. And those who don’t are in the ‘Apple is evil’ ideological camp and wouldn’t touch iOS development with a bargepole. In any case, you can get a brand new Macintosh for as little as £529, and if that is too much you can pick up second hand models for next to nothing.

    >Sure, even many of them will have Macs, but many won’t and for them it will be an insurmountable obstacle that Apple could easily remove.

    Easily? Porting your entire suite of development tools to another platform is *easy*? And supporting it? No; it would be a big hassle that Apple just doesn’t need. Its app ecosystem is thriving beyond anything comparable on the planet right now. They have clearly got it right.

  211. More info on Android ‘xlarge’ vs. ‘large’:

    http://www.betanews.com/article/7-inch-Android-tablets-will-get-Honeycomb-thanks-to-new-layout-system-in-32/1311795955

    Not only are (most?) 7″ tablets not xlarge, but their dimensions mean that they weren’t really supported properly by Honeycomb even in the ‘large’ mode.

    But Honeycomb 3.2 will support these tablets with a new layout system.

    So it never made sense to try to tease out the total number of tablets by looking at either Honeycomb activations or xlarge activations, and the whole screen size system is changing now anyway.

  212. @Patrick

    >So it never made sense to try to tease out the total number of tablets by looking at either Honeycomb activations or xlarge activations, and the whole screen size system is changing now anyway.

    I guess we are back at square one.

    One thought that occurred to me was whether there might be some data out there from a big website on the relative numbers of tablet browser headers they are seeing. That might be a decent way of estimating the share.

  213. > Again, not conclusive, and it’s a couple of months out of date, but I guess we have to take what we can get.

    Historically, browser share statistics don’t correlate well with installed base. I’m not sure about all the reasons why, but you can search back on this blog for a lot of examples. In general, we don’t give much credence to those.

    There is a report which gives an extreme lower bound on the number of Android devices being activated:

    http://www.good.com/resources/Good_Data_Q2_2011.pdf

    Good technologies is used by big businesses. If you look at the report, most of the activations were in the finance industry, where cost is not that big of an issue, style probably is a big issue, and nobody cares about being an open source nerd. 3.1% Android vs. 95+% iPad.

    The Apple bias in these markets is reflected in the phone activations, which show iPhone beating Android 2:1. Good has iPad activations in their market beating Android smartphone activations, so it’s a very skewed market.

  214. @patrick

    >Historically, browser share statistics don’t correlate well with installed base. I’m not sure about all the reasons why, but you can search back on this blog for a lot of examples. In general, we don’t give much credence to those.

    Any change you could give me a pointer to an explanation of this? I tried searching the blog but I don’t think I have hit on the magic search term.

    I suppose one argument might be that the iPad somehow encourages more use of the browser than Android? Maybe Android users just aren’t using their tablets?

    Okay, that was a little snarky :)

    >3.1% Android vs. 95+% iPad.

    I am quite surprised that the iPad is doing even *better* in the business market. Traditionally this has been quite a hard market for Apple to break into, but recently they have focused a lot on making iOS work well with business stuff like Exchange Server so perhaps they are finally making inroads there.

  215. @Tom:

    Any change you could give me a pointer to an explanation of this?

    I think the last big discussion (threaded between other sub-discussions, as usual) was here. (But before you bother going there, keep reading here :-)

    I suppose one argument might be that the iPad somehow encourages more use of the browser than Android? Maybe Android users just aren’t using their tablets?

    I would say that my main contribution to that previous discussion was the posting of this link. I would be interested in your honest opinion about whether you think the issues discussed there could possibly help provide an answer to those burning questions :-)

    I am quite surprised that the iPad is doing even *better* in the business market. Traditionally this has been quite a hard market for Apple to break into, but recently they have focused a lot on making iOS work well with business stuff like Exchange Server so perhaps they are finally making inroads there.

    If you go look at what Good Technologies sells, you might realize that it is no accident that a huge portion of their customers are in the financial industries. I think their services and iWhatever are attractive to the same sorts of companies.

  216. > I see it’s just more browser statistics. We’ve been through this before. Historically, those are completely meaningless.

    There doesn’t seem any reason why iPad users would be heavier browsers than Android users, especially given the argument that Android tablets can do Flash where the iPad can’t (or won’t).

  217. @Tom, Jeff Read:

    I think I’ve chosen bad examples and / or I’m not articulating them well, so let’s try this from a more abstract angle.

    I’m going to start with the assumption that “Apple has not learned their lesson about being exclusionary with developers” if their Mac-only restriction for iPhone development excludes more than 20% of the addressable developer community (feel free to adjust the number up or down, with a cap at 50% for obvious reasons, it doesn’t affect the argument much).

    The best number I can find for worldwide Mac OS X market share is roughly 6.27% from StatCounter. Their methodology is pretty bad (as we appear to have just discussed), so let’s round that up to 10%. Assume the Windows market share is 80% (StatCounter’s is 91.58%). Yes, I know this implies a Linux/other market share of 10%, but I’m trying to stack the deck against Windows developers as much as I can without being unreasonable.

    Assume the fraction of Mac users addressable for iPhone development is m and the fraction of Windows users addressable for iPhone development is w. Assume the maximum “exclusion fraction” is f.

    The raw number of Mac users addressable for iPhone development is 0.1m and the raw number of Windows users addressable is 0.8w.

    The excluded fraction is: 0.8 w / (0.8 w + 0.1 m), or 8w / (8w +m).

    We want 8w / (8w + m) < f. This implies 8w < 8wf + mf or w/m < (1/8) * (f / 1 – f).

    Plugging in f = 0.2 we get w / m < 1 / 32 or, in other words, Mac users have to be 32x as interested in iPhone development for Apple to be excluding less than 20% of the potential addressable developers. Even if you set the "exclusionary fraction" to the extreme 50% boundary, you get a threshold of Mac users have to be 8x more interested in iPhone development.

    I think both thresholds are absurd. I don't think you get that sort of interest ratio today even given that Apple has released Mac-only development tools (and has shown zero interest in budging).

    One thing I didn't account for above is that many developers have access to multiple OSes. So let's assume that half of the potential addressable "Windows" developers (i.e. in terms of their primary computer that counts towards OS market share) have access to Macs for development anyway. We're still looking at interest ratios of 16x or 4x. 4x is maybe just getting barely plausible, in terms of an interest level ratio after the reality of Mac-only tools, but the relevant interest fraction is the one that exists before Apple chooses available platforms for their tools, not after.

    In other words, even given many Apple-friendly assumptions, by choosing Mac only tools Apple is pretty-much guaranteed to be excluding at least 20% of their addressable developer community for iPhone development and has, at best, an outside shot at of excluding less than 50% of the available developer community.

    Does this affect the App Store or their iPhone business? Probably only marginally, at best (as noted above). But if you're assessing whether or not Apple has learned their lesson about developer exclusion. I think the case for Apple not having learned their lesson is open-and-shut.

  218. @fake account

    Android users are more likely to use a wider variety of browsers than iPad users (if only because of availability), and they’re more likely to engage in user-agent spoofing (since Android came to market after the iPhone and after a lot of websites had been “iPhone-optimized”).

    To take a concrete example, the vast majority of my Android browsing is done using Firefox. How many counters do you think get Firefox on Android right (instead of miscounting it as Linux or Windows or whatever)? Ditto for Opera Mobile (Opera Mini is probably a wash since it is available on iOS and goes through Opera’s servers anyway) and Dolphin and …

  219. @patrick

    >I would be interested in your honest opinion about whether you think the issues discussed there could possibly help provide an answer to those burning questions :-)

    Very interesting. If I read this correctly, the conclusion is basically that the iOS browser has a lower limit for the size of components it will cache. Therefore the iOS browsers will be making network requests more often, and therefore iOS devices will seem unfairly to have higher page views than Android devices.

    Do I have that right?

    I must say that at first I found this quite compelling. However, it got me thinking about the methodology that was used to establish the browser share stats I linked to. If they are purely just counting pageviews then your argument would hold. However, if they are counting uniques, then not so much (although there might still be a small effect).

    After hunting around a bit I found this:

    http://www.netmarketshare.com/faq.aspx

    “instead of counting pageviews we count daily unique visitors. A daily unique visitor is counted only once per day per website we track, regardless of the number of pageviews the visitor has. While this may seem to greatly reduce our sample size from the billions of monthly pageviews we process to only the daily unique visitors, we do so to provide a more accurate picture of market share showing the number of users of a technology instead of the number of clicks. Counting unique visitors also renders bots designed to influence market share harmless. Counting pageviews for market share reports would be susceptible to bot attacks and inexplicable jumps in market share that don’t represent the true nature of the market. ”

    So I am not really convinced that browser share is meaningless to our conversation. Obviously it will not be completely accurate, but I think it gives us a decent ballpark. Unless I have missed something.

  220. On the same note, trying to use browser statistics to tease out Android smartphones vs Android tablets? I might almost call that provably impossible (given that the variation within the Android smartphone and tablet pools is of the same order-of-magnitude as the cross-pool variation). It is certainly practically impossible given the existing quality of browser statistics that are actually gathered.

    BTW, for those keeping score at home, here is plenty of (admittedly anecdotal) evidence ASUS Transformer is still having supply issues even *after* ASUS has ramped up to 400k / month:

    – The Transformer is out-of-stock at Staples (both 16 GB and 32 GB), home of the “$100 off a tablet” coupon (other Android tablets like the Toshiba Thrive and Acer Iconia are available)
    – Office Depot’s stocks of the Transformer are tight enough that they’re not honoring the Staples coupon on the Transformer (even though they will for other Android tablets)
    – Best Buy doesn’t currently have the 16 GB model in stores (even though it is the more common model and isn’t supposed to be online-only). They also only let you order 1 at a time online for the 16 GB model. The 32 GB model, meanwhile, is only available at half of the Best Buy stores in the Greater Boston area (which is what I’d typically check) and I’d bet stock / store is low (which is what usua
    – Amazon and Best Buy’s “sales” on the tablet aren’t meaningful – they’re the “put it in your cart and you’ll get $5-$10 off” mind game
    – … there’s probably more, but I don’t feel like spending more time to dig it up

  221. @Ravi

    What is an ‘interest ratio’? What does it mean to be ’32x as interested’ in something?

    The bottom line is this: Apple has the best and the biggest mobile app store in the world. Their strategy for attraction developers is obviously working very nicely.

    >Does this affect the App Store or their iPhone business? Probably only marginally, at best (as noted above). But if you’re assessing whether or not Apple has learned their lesson about developer exclusion. I think the case for Apple not having learned their lesson is open-and-shut.

    But what is the lesson? That excluding developers results in either few or poor quality apps? But neither of those things is true, so what is the problem?

    Given that Apple has a lot of excellent apps, either your ‘lesson’ is wrong, or they are not being exclusionary to any significant extent.

  222. @Tom:

    A 32x interest ratio would be equivalent to saying something like the following:
    – You do a survey of 10000 “Mac” users about developing for the iPhone and you find 3200 of them are interested
    – You do an equivalent survey of 10000 “Windows” users about developing for the iPhone and you find out 100 are interested

    To me, that sort of difference in the level of development interest sounds so skewed as to be absurd. As implied above, I think 4x to be the largest plausible ratio on that front.

    But what is the lesson? That excluding developers results in either few or poor quality apps? But neither of those things is true, so what is the problem?

    I should have been more precise. The effect of excessive developer exclusion is probably marginal *today*. But that marginal difference compounds with time, especially when there is a roughly equivalent less exclusionary platform available. In other words, over the long run Apple’s platforms will suffer from developer exclusion just as they did before.

    For that matter, we may have already seen this effect in action. I’m pretty sure most people found it remarkable how quickly the Android Market started to catch up with Apple’s App Store in terms of volume and range of available applications once the platform and device pieces fell into place. How many of those applications came from “excluded developers”? I know of at least one notable case of an “excluded developer”. As I think I mentioned above, at least one winner of the first Android Developer Challenge started hacking Android because he wasn’t able to hack his iPhone (though I don’t remember the specific reason he couldn’t hack his iPhone at the time). I think there may also have been a college class that turned to Android development because what they were doing wasn’t compatible with the iPhone SDK NDA that used to be required.

    This is more speculative, but these sorts of effects might also make it more difficult monetize Android apps. If there are more developers playing, it is more likely that someone will end up releasing a “good enough” app for free or for ads only. Remember, more profits doesn’t always mean that a market is more valuable or innovative. More profits can also mean that a market just has more sustainable barriers to entry (and because of those barriers it might well be less innovative).

  223. @ravi

    >I should have been more precise. The effect of excessive developer exclusion is probably marginal *today*. But that marginal difference compounds with time, especially when there is a roughly equivalent less exclusionary platform available. In other words, over the long run Apple’s platforms will suffer from developer exclusion just as they did before.

    The app store has been around since 2008; how long do we have to wait for this effect to kick in?

    And if the effect is compounding with time why are we seeing an *increase* in the rate of growth of both apps on the store and total app downloads?

    If I were an iOS developer (oh, wait, I am!) I would be much more concerned with the $2.5BILLION that Apple has paid out to developers, rather than the lack of a Windows version of the dev tools.

    And, if I could be a little snarky for a second, if I were Apple I would probably be concerned that opening up the store to Windows developers would wreck the place.

  224. Come on, this is a bit silly. Practically every student I have met over the last couple of years who is into computers has a Mac already. And those who don’t are in the ‘Apple is evil’ ideological camp and wouldn’t touch iOS development with a bargepole. In any case, you can get a brand new Macintosh for as little as £529, and if that is too much you can pick up second hand models for next to nothing.

    Perhaps you’re strange. I know whole faculties who are windows centric and wouldn’t know what to do with a mac if you threw it at them.

    Actually i’ll go even further than that. I’ve been to 2 and work with a further 2 of the top 10 universities in Australia and in all of them (and excluding personal lecturer machines) I know of one faculty that even supported Macintoshes (and they had to hire someone internally, the standard university IT support group told them to get stuffed).

    Don’t get me wrong, it’s not like you have to look in the basement behind the rabid dog to find the mac user, there’s plenty around and they’re plenty loud about how awesome it apparantly is… but “practically every” is completely inverse to my experience.

    And btw quoting the mac mini price is a bit devious as well. Thats like saying “you can get a PC for $130, here’s your gumstix device”.

  225. I suppose one argument might be that the iPad somehow encourages more use of the browser than Android? Maybe Android users just aren’t using their tablets?

    Maybe iOs’ browser has fucked up caching and will refresh every page and every image it goes to. Considering that a large chunk of the webstat methods i’ve seen use image references, having a browserstats image on every page on a site getting re-hit every time you go to a new site would skew the stats something rediciulous.

    Perhaps the browser stat system in question tags *webkit* as being “iPad”.

    I see your snark and raise you. :P

    In all seriousness It wouldn’t matter if the web stats were that Android users are 99.9999% of people on the web. I’d still suggest they’re probably full of crap(and in fact i’m pretty sure i was the first to say so when the first comment was made). Web stats are notoriously bad and trying to use them for anything other than “these browsers are hitting my web server” is an exercise in futility.

  226. @Tom:

    I guess I didn’t explain my mental model well enough. At its most basic level, the effect of exclusionary practices would naturally be relative to how the App Store could have done in the “alternative universe” where Apple wasn’t as exclusionary (i.e. released a Windows version of the development tools). That means it isn’t directly observable, so you have to try and test the theory against inferred proxies.

    That’s why I mentioned the unexpectedly fast ramp-up of the Android Market. At bottom, the iOS and Android development dynamics are subject to the same fundamental forces, so why did Android ramp up faster (or even successfully at all)? Is it purely because iOS had blazed a trail? Why would that have been enough by itself? iOS had the edge in devices, ecosystem maturity, profit potential, … Surely all of that should have made iOS grow faster (and, beyond that, suck up the Android ecosystem’s oxygen), right? Though I’m sure you hate the comparison, that is what Windows did to the rest of the world back in the day. Why didn’t iOS do the same?

    I am hypothesizing that one reason iOS didn’t is because of Android’s less exclusive developer requirements. And now that Android and iOS are in a race, iOS’s relatively exclusive developer requirements are a relative handicap. I hope at least that is uncontroversial that there is some handicap (even though iOS has other advantages). Suppose it is a 10% “developer ecosystem headwind”. That currently gets lost in the noise of Apple’s one-year head start in the market. Let the handicap compound for a few more years and it slowly becomes a bigger and bigger issue.

    And, if I could be a little snarky for a second, if I were Apple I would probably be concerned that opening up the store to Windows developers would wreck the place.

    That is *exactly* the attitude on the part of Apple that I’m highlighting as a serious mistake. That is the lesson they have not learned. They might be right 9 out of 10 times (so this pays off in the short run). Nevertheless, they should throw the doors open as wide as they can. Otherwise, as the “10th developers” steadily accumulate, they eventually overwhelm you.

  227. @fake account:

    That does sound like a minor problem. However, that is exactly the sort of thing that illustrates why Android’s openness is a significant advantage. If that had been a problem with Apple’s App Store (and there have been many), Apple is the single point of failure. The whole ecosystem suffers until and unless Apple fixes the problem (see: in-app purchase). With Android, developers and users can vote with their feet. If the Android Market has problems, you can go to GetJar or the Amazon Appstore or, for that matter, you can sell directly. The ecosystem is distributed and robust and there aren’t as many single points of failure.

  228. @Tom:

    However, if they are counting uniques, then not so much (although there might still be a small effect).

    The issue is not what they are attempting to count. It is how they are attempting to count it. I have looked at their site before and I have never seen any information about how they do this, e.g. “we put a web bug here and it has a cache expiration date and…”. Perhaps you can find such information; I have given up.

    Even with that information, there will probably be holes. For example, is it something that AdBlock+ will block on Android Firefox? How do they determine uniqueness, and how does that interact with some devices caching better than others? (E.g. a mobile device that changes IP address 3 times during the day might be counted 3 times if not caching properly, but only once if caching properly.)

    Right now, it is impossible to say anything other than (a) I can’t know how bad their data is; and (b) in the past, when we have had similar browser based data along with other data that seemingly should correlate with it, the correlation never appeared. So I discount browser-based data.

  229. @JonB

    >Don’t get me wrong, it’s not like you have to look in the basement behind the rabid dog to find the mac user, there’s plenty around and they’re plenty loud about how awesome it apparantly is… but “practically every” is completely inverse to my experience.

    Well I don’t have any experience with Australian universities, and I am talking about students, not faculty. I wonder when the last time you had contact with your unis was? When I was a student there were few Macs around; I had the only one in my group of friends. This was long before the iPhone. But when I went back a few months ago they were *everywhere*.

    >And btw quoting the mac mini price is a bit devious as well. Thats like saying “you can get a PC for $130, here’s your gumstix device”.

    The £529 Mac Mini is actually a pretty nice machine: 2.3Ghz core i5 sandy bridge CPU, 2GB RAM, 500GB HD, dedicated graphics. Plenty for programming work. And I assume that any PC user is going to have a display already.

    >Maybe iOs’ browser has fucked up caching and will refresh every page and every image it goes to.

    If you look at the methodology page I linked to you will see that they are counting daily uniques, not page requests. I am sure that web stats aren’t going to give us a 100% accurate picture, but these guys seem to be able to distinguish not only using user agent strings, but right down to the model of device. They break down their numbers to iPad, Xoom, Galaxy Tab, PlayBook etc. It is easier to do this than you might imagine. Not only do you have the user agent string, but with javascript it is also possible to determine the exact set of browser plug-ins present, which gives us a lot of data just in itself, the screen resolution and colour depth, the system fonts, and who knows what else. With all this data it is quite possible to get a very accurate picture of the machine that is making http requests.

  230. @fake account:

    Android developers are furious about Google’s changes to the way search works in the Android application marketplace.

    The changes were put into place on 1 July, and since then, a lot of previously very popular applications have now effectively disappeared.

    The change in rankings on the Android market might be a serious problem. But I sincerely doubt it. They’re probably just the Android market store analogue of the same sort of clearing of the underbrush of crappy link farms from search that has a few companies all riled up.

    One developer said his app had been top of the list when searched for by name or by specific keywords.

    Well, obviously there’s still a list, which means that some application is on top, which means that consumers will find something to download. The consumers are happy, and the new application top dogs are probably pretty pleased, and not bothering to complain to El Reg.

    The complaint I always heard in the past was that search in the Android Market was irretrievably broken and didn’t work worth crap. See, e.g.:

    http://www.lifeofandroid.com/news_detail/why-google-must-take-out-the-android-trash/

    It stands to reason that when google starts to wield the same sort of discrimination that makes them #1 in search on their own market, there are going to be some whining losers. Really, experience has shown us that it could not possibly be otherwise.

  231. @Tom:

    We cross-posted, but:

    > If you look at the methodology page I linked to you will see that they are counting daily uniques,

    If I look at the methodology page, it describes what they are trying to do. That’s a goal. It’s not a frickin’ methodology.

  232. @Patrick

    >If I look at the methodology page, it describes what they are trying to do. That’s a goal. It’s not a frickin’ methodology.

    They don’t give a detailed technical exposition, I agree. And the results won’t 100% accurately reflect market share, I agree. However, minor effects that might be caused by things like AdBlock are not going to skew the results all that much; relatively few people will be using such things.

    Although we don’t know the exact technique used by these people we do know the following:

    1. They are apparently able to distinguish not just browsers, but also the *model* of device being used, which leads me to think they are being quite sophisticated.
    2. They are using javascript, which means that they have a pretty large amount of information from each device to work with.
    3. They are counting daily uniques, which almost completely solves the problem of variable caching.

    Also, although they don’t specify it, I would be willing to bet they use cookies as well as IPs in order to provide a more robust way to distinguish unique users. Why wouldn’t they?

    Again, they are still not going to have completely accurate information, but to dismiss this data as ‘meaningless’ is a bit much. There will certainly be error bars here, but we can still have some confidence that we are getting a reasonable ballpark figure. I think it is the best data we have seen so far.

  233. 1. They are apparently able to distinguish not just browsers, but also the *model* of device being used, which leads me to think they are being quite sophisticated.

    You’re gullible, aren’t you?

    2. They are using javascript, which means that they have a pretty large amount of information from each device to work with.

    Uh-huh.

    3. They are counting daily uniques, which almost completely solves the problem of variable caching.

    There’s no way that I know of that you can “count daily uniques”. Again, that’s a goal. You can try to count all accesses and then try to adjust correctly. Both those tasks are fraught with problems.

    Also, although they don’t specify it, I would be willing to bet they use cookies as well as IPs in order to provide a more robust way to distinguish unique users. Why wouldn’t they?

    Oh, I’m sure they try really hard to do a good job. I’m just not sure they can, and when they don’t describe what they do or any correlation against other data, or give any sort of error information, I have a feeling they know they’re generating crap, but wouldn’t want to spook the customers.

    Again, they are still not going to have completely accurate information, but to dismiss this data as ‘meaningless’ is a bit much. There will certainly be error bars here, but we can still have some confidence that we are getting a reasonable ballpark figure. I think it is the best data we have seen so far.

    Again, you’re missing the part where we’ve never seen web stats line up with other data. But feel free to continue to believe.

  234. @Patrick

    >There’s no way that I know of that you can “count daily uniques”.

    It’s quite simple:

    1. Somebody hits your site.
    2. You generate a unique fingerprint for them using any or all of: IP, user agent, plug-ins, time-zone, system time, system fonts etc etc (really, it’s an embarrassment of riches the amount of factors that are available).
    3. You check to see if the fingerprint has already been logged that day.
    4. If it has, you don’t increment your counter.
    5. If it hasn’t you increment your counter and store the fingerprint with a timestamp.

    >Again, you’re missing the part where we’ve never seen web stats line up with other data.

    I’d be interested to know what examples you have in mind.

  235. @Tom:

    2. You generate a unique fingerprint for them using any or all of: IP, user agent, plug-ins, time-zone, system time, system fonts etc etc (really, it’s an embarrassment of riches the amount of factors that are available).

    If you’re using the IP address for a mobile device, you’ve already screwed up and you might double-count it. Just as one example.

    I’d be interested to know what examples you have in mind.

    I would have to go back and find it. You probably can do that as easy as I can. Just take a look at what this same company was saying a year ago, for example, and compare against other stats available, like comscore and sales figures. Whenever I did this before on smartphone data, it didn’t correlate.

  236. @patrick

    >If you’re using the IP address for a mobile device, you’ve already screwed up and you might double-count it. Just as one example.

    Obviously you wouldn’t use IP for mobile. I was just giving you a general picture of how to do unique logging. You don’t need IP, there are plenty of factors available.

    >I would have to go back and find it. You probably can do that as easy as I can. Just take a look at what this same company was saying a year ago, for example, and compare against other stats available, like comscore and sales figures. Whenever I did this before on smartphone data, it didn’t correlate.

    I’ll see what I can find.

  237. @Tom:

    I happen to be the proud owner of an ASUS Transformer, but I’m not sure what you are going for. That being said, I did make the following observations:

    Browser
    – All of three of the User Agent options built into the stock browser (Tablet, Desktop, Mobile Phone) contain the strings “AppleWebKit” and “Safari” (“Mobile Safari” in the Mobile Phone case)
    – All three of the User Agent options built into the stock browser were considered unique (by themselves)
    – Only the Tablet and Mobile Phone strings contain “Android”. This is ironic because I find Desktop works the best for the stock browser and it is probably the default – I remember a review that specifically complimented the Transformer because out-of-the-box you got full websites and not mobile websites.
    – To my eye, it looked like the only way you’d have any clue what I was really using is if you were specifically looking for the “Transformer TF101” bit of the UA string. Otherwise it seemed far too easy to make a simple mistake (e.g. look for this bit, but forget to cross-check with that)

    Firefox
    – Firefox 5.0 for Android (where I do the majority of my browsing both on my phone and my tablet) was almost certainly being mistaken for something else. I figure they have to be resolving Firefox 5.0 out of the UA, so I find it pretty implausible that 29 other Android (or even arm) Firefox 5.0s have hit that site. 30 different Firefox 5.0 hits (including mine) seems like the most reasonable possibility.
    – Firefox didn’t mention the Android version, Transformer or anything else that would have distinguished it from Firefox on an Android phone
    – Firefox 6.0 beta was unique (like the stock browser alternatives), but I think that’s expected even if it is being confused with desktop Firefox

    I’m not sure what I get out of this except that anyone who tells you that they can reliably detect and count the Android tablets (vs everything else) that are hitting them based on browser information alone is probably deluded or lying (and we haven’t even gotten into 2.x tablets which are almost certainly harder to figure out). Before I would have any interest in someone’s numbers, I’d need to see the lab where they own one of each model they’re claiming to be able to detect, their archive of stable Android versions for each device, the list of browsers they used on each device (there’s also Dolphin and Opera Mobile and all sorts of others) documentation for the classification they do (and how they test it) and so on.

    I don’t think this confusion is deliberate. I suspect it is a consequence of the reality that Android tablets are new enough that they are almost always better off pretending to be “something else”. But that does mean that they are exceptionally hard to notice unless you know *exactly* what you’re looking for.

    1. >New US smartphone statistics from Nielsen, collected in June:

      No surprises here. Those figures are well within statistical error of what I’d expect from naive extrapolation of the comScore stats.

  238. In fact, let me make some predictions. To plus or minus 1%, here’s what I expect the share figures to look like in the June comScore stats: Android: 39.8%, RIM 23.5%, Apple 27.2%, MS 5.0%, HP 2.2%. Total smartphone userbase will be 79.0 million, plus or minus 1M.

    1. The reason I can make these predictions with confidence is that the trendlines in the comScore data look pretty stable. They use three-month smoothing, yes, but with the exception of RIM we’re seeing stability in the deltas over periods of a year or more.

  239. > No surprises here. Those figures are well within statistical error of what I’d expect from naive extrapolation of the comScore stats.

    Agreed. No surprises, just more ammunition for troll hunting.

  240. @phil:

    > My only guess is that since in flight WiFi costs $, Android users are cheap and iPhone users have more cash?

    I don’t think buying an iPhone will give you more cash. ;-)

    But yes, the demographics of those willing to splurge on in-flight data probably line up nicely with the demographics of those willing to splurge on an iPhone.
    OTOH, take a look at Ravi’s comment right before yours. I’m sure you can come up with at least one additional guess.

  241. So, if we compare these web usage statistics:

    http://www.netmarketshare.com/2011/07/01/Mobile–Tablet-Crosses-5-percent-of-All-Browsing-Globally

    with these newly released Nielsen numbers:

    http://techpinions.com/why-it-matters-that-apple-is-the-number-1-smart-phone-manufacturer/1347

    it is obvious that iOS is overrepresented in browser share (and I think Android is also, although not by anywhere near as much).

    I’m not really sure what to do with these numbers now. Perhaps one of you numerical types could try to apply these ratios to the tablet browser share stats and come to some conclusions?

  242. @Ravi

    Thanks for doing that, although I was really more interested in the non-UA info. Are you seeing things like system fonts, and plug-ins showing up? It’s that kind of granular data that can be used to construct a fairly reliable per-model fingerprint.

    >- To my eye, it looked like the only way you’d have any clue what I was really using is if you were specifically looking for the “Transformer TF101? bit of the UA string. Otherwise it seemed far too easy to make a simple mistake (e.g. look for this bit, but forget to cross-check with that)

    I am surprised that the UA actually has the model in it! That would certainly be one important element in constructing a reliable system.

  243. Plugins (or more accurately the Flash plugin) shows up… when it is enabled (for the stock browser, Firefox doesn’t currently have Flash support). No fonts showing up anywhere. Screen size shows up. It is slightly off in Firefox (reserving some space?), but otherwise accurate. Yes, Yes, No for the supercookie test (for both stock and Firefox). I think that covers everything interesting.

  244. @Patrick Maupin
    “OTOH, take a look at Ravi’s comment right before yours. I’m sure you can come up with at least one additional guess.”

    User agent string fun? Maybe, I’d have to mess around with the stock android brower to see how it behaves. All initial access would be done via a payment page, depends on the competence of the developers.

    But would android show up as mobile safari? I dunno…..

  245. @Tom:

    > It is obvious that iOS is overrepresented in browser share (and I think Android is also, although not by anywhere near as much).

    I’m going to play devil’s advocate here. It’s not at all obvious this is the case. iPhones do WiFi. Do all the other smartphones? Do people who worry about going over their limit switch to browsing on something else like a computer if they’re on a Blackberry, but keep browsing on an iPhone and just switch it to WiFi mode?

    > I’m not really sure what to do with these numbers now.

    I’ve never been sure what to do with them, except ignore them.

    > Perhaps one of you numerical types could try to apply these ratios to the tablet browser share stats and come to some conclusions?

    Even if we knew or could guess what processes skew the data for phones, I’m having a hard time imagining that the internet usage patterns for tablets are the same as for phones.

  246. @phil:

    > But would android show up as mobile safari?

    If I understood Ravi’s post correctly, it would be extremely easy for a dumb web stats app to think his Android Eee Transformer had a mobile Safari browser.

  247. @Tom:

    Yes, that’s interesting. 690K Xooms YTD, and another 610K to 810K expected for the next 6 months due to “greater competition from a range of other tablets running Google’s Honeycomb operating system and [Osborning].”

    Motorola isn’t kidding about “greater competition.” Besides Samsung’s offerings, digitimes reports that is now shipping 400K Eee Transformers a month.

    And Asus isn’t resting on its laurels. They just might single-handedly revive MeeGo. C’mon — admit it. For $200, that puppy looks pretty good…

  248. 1. Somebody hits your site.
    2. You generate a unique fingerprint for them using any or all of: IP, user agent, plug-ins, time-zone, system time, system fonts etc etc (really, it’s an embarrassment of riches the amount of factors that are available).

    Well there’s your first problem. I’m mobile, my ip is not static.

    Assuming that HSDPA is magical and can maintain the same IP no matter what cell tower i’m attached to (and I assume this because i don’t know for sure it can’t) at the very least i’ll have 4 different IPs in the same day if i have a work wireless and home wireless (which is not exactly uncommon).

    And any browser that offers information about plugins and system fonts to javascript is flat out broken. (which is not to say that android doesn’t, just that relying on it is a terrible plan)

  249. Oh and yes i know you’ve said “oh but you wouldn’t use IP on mobiles”… what else would you use?

    It’s 100% plausible that two people with stock phones from a carrier will look otherwise identical to the browser. The only feature that would differentiate them is their IPs.

    Christ you’d be better off relying on cookies… at least you avoided that obvious pitfall.

  250. >And any browser that offers information about plugins and system fonts to javascript is flat out broken.

    You might want to click that eff link Tom posed, you’d be surprised how much information your browser gives up.

  251. I’m not really sure what to do with these numbers now. Perhaps one of you numerical types could try to apply these ratios to the tablet browser share stats and come to some conclusions?

    Seriously man… Browser stats are crap data.
    Give them away.

  252. You might want to click that eff link Tom posed, you’d be surprised how much information your browser gives up.

    Not really, i’ve read the page before. My point is that relying on it to differentiate people is on one hand flawed methodology (how many system fonts have you installed on your mobile or tablet?) and on the other shows a lack of forward thinking. It might work for now but customers have a vested interest in choosing a browser that makes it not work.

  253. @JobB

    >Not really, i’ve read the page before. My point is that relying on it to differentiate people is on one hand flawed methodology (how many system fonts have you installed on your mobile or tablet?) and on the other shows a lack of forward thinking. It might work for now but customers have a vested interest in choosing a browser that makes it not work.

    I’ll take some argument on how easy it might be to reliably distinguish models of tablet (although I think it is certainly doable if you are careful) but there is absolutely no question that it is *child’s play* to very reliably and persistently uniquely identify users.

    Let’s rule out IP, since we are talking about mobile.

    First of all we have cookies. Probably the easiest method is just to set a standard http browser cookie. That will cover most people right off the bat.

    But maybe your browser doesn’t accept cookies, or you have them disabled, or you regularly flush them. Got Flash? In that case we can store a cookie in your Local Stored Objects, a so-called ‘Flash Cookie’. Not too many people know about those, and Adobe doesn’t make it obvious how to disable them. Also, the great thing about Flash Cookies is that they work cross-browser, so we can identify a single system across multiple browsers on that system.

    At this point we have covered probably 90+% of web users, but maybe you smart guys have disabled cookies and Flash Cookies. Don’t worry, we can still set cookies!

    If you have Silverlight, then that has a comparable storage facility to Flash, so we can store a cookie there.

    If you have an HTML5 capable browser then we can store data in Session, Local, Global, or Database storage that has exactly the same effect as a normal cookie.

    Maybe you don’t have an HTML5 browser or Silverlight.

    Well, as you know, CSS colours visited links differently to normal links. Javascript can query the elements in a page to determine their colour. Let’s say we want to store the cookie ‘124’. What we can do is, using scripting, force the browser to access the URL http://tracking.com/1 then the URL http://tracking.com/2 then the URL http://tracking.com/4 (obviously I am just using tracking.com as an example).

    We have now set those URLs in the browser history.

    If your browser ever sees those URLs in a page again CSS will colour those links differently.

    When the browser goes back to one of our pages we can set a link on the page ‘http://tracking.com/1’ and check the colour of it. If we see that it is the colour of visited links we know we are on the right track, so now we set another link ‘http://tracking.com/11’. We check the colour, but find this time that it has not been visited, so we move on to ‘http://tracking.com/12’. Ah! Back on track. I think you see where this goes. All this happens completely invisibly, using scripting, so the user never even knows.

    In this way we can use the web history to set cookies. It’s not even that hard to code this stuff.

    At this point you are basically screwed if you are trying to avoid cookies, although there are some other esoteric techniques that can be used (for example, etags).

    But, let us assume that we can’t use cookies of any sort. Hard to believe, but let’s just run with it as an exercise.

    First of all, let us note that this in itself gives us some useful identifying information about you. We know that you don’t allow cookies (that’s one bit of information, and with that one bit we have carved the space of all possible users in two), we know that you don’t have Flash or Silverlight (two more bits of data, and now we have chopped the userspace into 8). Remember, every bit of information halves the search space, so we can rapidly hone in on a unique person.

    We have your UA, which represents a rich data source. Typically we can get 13 or 14 bits of data from that alone. We have your HTTP_ACCEPT headers (3/4 bits), your browser plugins (19 bits), your time zone (2.32 bits), your screen size and colour depts (4.32 bits). I learned all this from the panopticlick project. You can read their full report here – https://panopticlick.eff.org/browser-uniqueness.pdf.

    This is how browsers work. It is not a case of them being ‘broken’.

    The Panopticlick people found that, using fingerprinting alone, “if we pick a browser at random, at best we expect that only one in 286,777 other browsers will share its fingerprint”.

  254. Oh, and I forgot one other cool one I heard. Apparently some web tracking companies are looking at your system clock, and then checking to see how many milliseconds it is off from their central clock. As you can imagine that provides a massive source of entropy just by itself.

  255. Or perhaps it was seconds, not milliseconds. I am trying to find the source, but I forget where I read about it. At any rate, it is certainly a nice source of additional information.

  256. @tom

    User Agent value…

    Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US) AppleWebkit/532.5 (KHTML, like Gekko) Chrome/4.1.249.1045 Safari/5.35

    On a Galaxy Tab using Dolphin!

  257. @Tom Forest

    >Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US) AppleWebkit/532.5 (KHTML, like Gekko) Chrome/4.1.249.1045 Safari/5.35

    For the purposes of uniquely identifying users this does not matter of course.

    For the purposes of identifying what tablet you are using the question is how many other tablets would share the same UA. If it is more than one, then we go to other factors such as fonts, plugins, etc, to hone in on the identity of the tablet. As somebody else noted, probably people aren’t installing fonts on their machines, so we can probably use the default font set for a tablet as one factor. After that we can take a look at the exact set of plug-ins (and also, don’t forget, it is the exact version numbers for the plug-ins as well as just the names of the plug-ins).

    Some devices are going to fall through the cracks, but I think that a determined person could do a pretty good job of identifying devices.

  258. @Tom

    “Your browser footprint appears to be unique among the 1,682,834 tested so far.”

    Also got the screen resolution wrong …

    “800x323x32”

    Actually 1024 × 600

    Totally un-modified/un-jailbroken first gen. Samsung Galaxy Tab using Dolphin and no add-ons

  259. @Tom Forest

    >“Your browser footprint appears to be unique among the 1,682,834 tested so far.”

    It’s amazing. With a totally stock configuration they are still able to uniquely ID you. There is really no way to use the web with the sure knowledge that you are not being tracked these days.

  260. > There is really no way to use the web with the sure knowledge that you are not being tracked these days.

    Well, it’s possibly a bit less useful this way, but I always surf with NoScript and AdBlock plus, and cross-site cookies disabled. I seldom enable JavaScript. It’s amazing how many sites still show you some stuff, and most of the ones that don’t simply lose me as a visitor (not that they care).

  261. @Patrick

    > I always surf with NoScript and AdBlock plus, and cross-site cookies disabled. I seldom enable JavaScript.

    I went through a period of doing the same thing, but I got tired of it. When I realised that even with no scripting and blocking cookies there were still a million ways to track me I just gave up. Unless you are willing to take extraordinary measures there is just no way to win. Of course, not having JS enabled also has security benefits, but in reality I have never seen any credible in-the-wild attacks on the Mac (or Linux) so I feel like it is too much hassle for too little benefit.

    On any site I build though I still go out of my way to make sure that there are no cookies, JS, or other gunk unless it is absolutely necessary (and even then I try to degrade gracefully) because I am scarred by the memories of what it was like to use the web and constantly be denied access to content for no good reason.

  262. When I realised that even with no scripting and blocking cookies there were still a million ways to track me I just gave up.

    If you cared really deeply about that, you’d always go through anonymizing proxies as well…

    Of course, not having JS enabled also has security benefits, but in reality I have never seen any credible in-the-wild attacks on the Mac (or Linux) so I feel like it is too much hassle for too little benefit.

    Probably low risk, I agree. But the “reward” has the potential to be extraordinarily painful.

  263. @Patrick

    >If you cared really deeply about that, you’d always go through anonymizing proxies as well…

    Oh believe me I have tried it all. I struggled with Tor for a few weeks before I got tired of waiting forever for pages to load. Add even that is not 100%. I am still not completely convinced that NSA isn’t sitting on the end nodes sucking up traffic.

    Anyway, you are right, I just don’t really care enough. It’s just a low-level annoyance I would like to go away, but the reality is that it’s not going to.

  264. I did think it interesting, though, that I couldn’t see any device specific info, e.g. ‘Android’, ‘Galaxy Tab’ etc.

  265. Samsung’s earnings are out.

    They’re doing really really well in cellphones — telecommunication business revenue rose 45% YOY, and telecommunication earnings increased rose 165% YOY on the strength of higher ASPs.

    Unfortunately, Samsung decided, for the first time ever, that the actual shipments number breakdown was sooper-sekret competitive information, so they’re only giving rough ranges, e.g. “shipment rose by high single digit percent from the previous quarter’s 70 million units.”

    Oh, well, another good data source bites the dust.

  266. @Tom Forest:

    The lack of device information seems to happen with many (most?) non-stock browsers. Given that the stock browser is one of the weakest browsers on Android, I suspect that does all sorts of “fun” things to browser-based device statistics.

    @Patrick Maupin:

    I’d read Samsung’s reporting change as, “we did great, but we didn’t beat Apple in smartphones so we’re going to leave you guessing (officially)”. Perhaps I’m too cynical, though.

  267. > I’ll probably blog on this when the next comScore figures come out.

    Yes. Unfortunately, the bulk of Samsung’s success won’t be reflected in the comscore data. For some reason, Samsung doesn’t sell in the US so much. For example, their latest hit product, the Galaxy S II, has sold 5 million units in under a quarter, but hasn’t hit the states yet. They are running neck-and-neck with Apple in global smartphone shipments, but have under 1/3 of Apple’s installed base inside the US (according to latest Nielsen data).

    OTOH, for that reason alone, Apple’s patent litigation won’t be able to hurt Samsung all that much. It’s the new trade protectionism, only affecting stuff coming into the states. And, as I’ve mentioned before, if Apple (or Oracle or whoever) gets an ITC ruling on some sort of Android-only infringement (but no Federal court ruling), then Samsung or HTC could ship the units into the states with no software, and add the software later. Or, if a court rules that’s an unacceptable workaround, then either Samsung or HTC could ship Windows software on the phones — look carefully again at the Nielsen chart — both of them already sell Windows on phones. Maybe it becomes clearer now why Samsung gave free phones to the CyanogenMod folks — if Apple won’t let ’em ship Android on their phones, they just pay the MS Windows tax and ship WinMo on phones going to the states (raising the cost of the phone here vs. elsewhere), and let people put Android on them themselves. (Assuming that Apple doesn’t shut them down at the ITC over physical look and feel.)

    I really think the right answer for phones is, as with PCs, to separate hardware from software in the mass market. Apple’s lawsuits, if at all successful, will hasten this transition, and ironically, step up the pressure on Apple itself, as the commoditization of the hardware and software will arguably occur much faster once they are separated.

  268. @Ravi:

    I’d read Samsung’s reporting change as, “we did great, but we didn’t beat Apple in smartphones so we’re going to leave you guessing (officially)”. Perhaps I’m too cynical, though.

    I don’t think the market would assign any particular value one way or another to Samsung selling 500K more or fewer smartphones than Apple.

    My internal cynic says that Samsung is going to make Apple work really, really, hard for every scrap of information they want to get in discovery over the patent stuff.

  269. @esr:

    To me, the amazing thing is that Apple doesn’t just fear Android, they seem fear Amazon, B&N, Kobo, Hulu, Rhapsody, Spotify, [Netflix except for the Apple TV deal] and on and on. Not to mention the iOS 5 announcement which left behind a football field of suddenly redundant applications (pissing off plenty of developers).

    It really feels like if you’re doing anything but writing a game (or something you’re *sure* will stay small enough to be under-the-radar) iOS is actually a very scary ecosystem to focus on for the long haul. If you manage to build something successful that Apple decides they want, they’re either going to buy you on the cheap or they’re just going to break you (because that might be easier). Even if I have a hunch about how all of this might be a strategic choice (gathering up the treasures to make the walled garden prettier and more secure), their willingness to throw out (or at least threaten to throw out) such major pieces of their ecosystem is still breathtaking.

  270. @Patrick Maupin:

    Samsung doesn’t neglect the US market frivolously. They neglect it because US carriers want so much extra nonsense and customization that it makes more sense to get their new flagship established worldwide and then follow-up with the US entry (when there’s more buzz and interest anyway). The Galaxy S II is a particularly extreme example of this. It is actually amazing that they sold 5 million of them because they did this excruciatingly slow country-by-country rollout (check out the Samsung Galaxy S 2 Release Dates blog for more). It feels like they were selling every SGS2 they had and they were looking in odd corners of the factory for ones they might have misplaced “just in case”.

  271. Before you get too carried away with the ‘Apple is scared’ mantra you might want to take a look at what counts: money. In Q2 Apple captured two thirds of the available cash in the mobile phone segment. Samsung was a distant second with 15%. This is despite the fact that Apple hasn’t released a phone in 15 months.

  272. @Tom:

    Money is a reason to be scared, not a reason to be fearless. After all, anything that threatens that big pile of money coming in is something to fear.

    Consider the late 90s Microsoft. Money coming in hand over fist, but terrified of Netscape and the Internet (and that terror ruled them long past the time it was useful for them). That’s my other common frame: “Apple = the all-powerful Microsoft of days gone by”. It certainly matches the way Apple is stomping on partners and developers: Part “we don’t need you any more” and part “squash you now before you become a problem”. How did that work out for Microsoft again?

  273. @Ravi

    Microsoft’s growth has stagnated not because of stomping on partners, but because they twice failed to adapt to shifts in the computing landscape. In the first place it was the web and now it has happened again with mobile. Compare that to apple, who invented the smartphone as we know it, and has now invented the modern tablet. Apple isn’t failing to adap to new modalities; it is leading them.

  274. @Tom:

    Microsoft’s stomping on partners was a key *driver* of the shift to the web. Don’t you remember when VCs used to ask startups how they’d deal with a “Microsoft problem”? The most common answer, by far, was to build on the web instead of Windows because on the web *Microsoft didn’t control your platform* (on either the client or server side). And the scale of the shift to the web was *not* inevitable. Mobile and Apple have shown us there is plenty of life left in native clients. So why did the shift go as far as it did? Because tech companies wanted to avoid building Windows clients as much as they could.

    Doesn’t that remind you of the growing line of companies (Readability, Financial Times, Kobo, …) getting HTML5 religion over Apple’s recent App Store moves? Do you really believe that there isn’t a similar (or even bigger) line of companies choosing to avoid native clients (or at least iOS clients) after all this?

    To take one example, SaaS vendors were spooked bigtime back in February (during the initial confusion over the new IAP policy). I wouldn’t be surprised if they are even more spooked now (especially considering the draconian restrictions related to account creation, mentioning your website and mentioning or referring to things you sell in an iOS client). On top of that, SaaS vendors have plenty of technical / infrastructural reasons to avoid native clients unless their customers absolutely demand them. How many do you think might have been thinking of an iOS client and are now telling their customers something like: “We’d love to make an iOS client, but all of Apple’s policy changes have made that too risky. We are focusing instead on HTML5 improvements to our web applications, and we you’ll be very excited by them”? I’d say more than a few.

    Personally, I think Apple is now driving a shift from native clients to HTML5 ones. Apple does control the full-featured browsers on iOS, so they have more room to frustrate that sort of shift than Microsoft did (since they couldn’t kill Firefox or other alterntaive), but it is still more difficult terrain for them. Not every detail matches, but I recognize the pattern.

    Let me throw in another question: what major tech companies are Apple’s “allies”? How does that compare to Google, Amazon, Microsoft, Facebook and so on? Does that sound sustainable? Does that sound familiar?

    Personally, I think Apple is so busy looking for enemies that it has forgotten how to make friends. That sounds familiar to me and I remember how it ended for Microsoft.

  275. @Ravi

    The alternate interpretation is that Apple (and Steve Jobs) learned their lesson from the 90’s and realized if they want to survive and they want things done right, they’re going to have to do a lot of it for themselves. Apple used to have their share of partners, Motorola, Adobe, IBM, even Microsoft, and look how that worked out for them. Then they decided they’d start taking care of themselves instead. Apple very reasonably doesn’t want to rely on “allies” because there’s no such thing in business. There’s only someone who’s financial interests align with yours temporarily. Also remember that Apple and Google used to be allies, remember that the first iPhone was full of google integration. Again, there are no allies.

  276. @tmoney:

    There’s a difference between having “allies” and relying on them. There’s also a difference between having neutrals (e.g. B&N, Kobo if you don’t grant Amazon, which I would) and turning them into enemies because you worry they might become enemies later. To be specific, why does Apple feel so strongly about forcing iBooks to be more than a floor under the ebook experience on iOS? The only answers I can come up with are lock-in and/or wanting to avoid an Adobe-like situation with the ebook “winner” (probably Amazon), when Amazon was perfectly happy to sell Kindle books to iPad customers and leave it at that. These things require a careful balance to manage the trade-offs and think that there’s an increasing amount of evidence that Apple has thrown the concept of “careful balance” on this front out the window.

  277. @ Ravi,

    Of course they want to avoid the adobe situation again. Brad Colbow hits it on the head with this http://bradcolbow.com/archive/view/the_brads_how_to_alienate_a_fanbase/ and that doesn’t even cover all of the crap. Apple absolutely wants to make sure that the success of the iPad or iPhone requires as little reliance on others as they can. It’s also, I think, about consistency. They want every purchase made in an application to work exactly the same way. That’s good for consumers. The fact that they’re going about it the wrong way is what’s bad for consumers.

  278. @Ravi:

    Good analysis on the problems (everybody already knows the advantages) of having a big pile of money, and on the continued shift to the web.

    Samsung doesn’t neglect the US market frivolously.

    I didn’t mean to suggest they did.

    They neglect it because US carriers want so much extra nonsense and customization that it makes more sense to get their new flagship established worldwide and then follow-up with the US entry (when there’s more buzz and interest anyway).

    I’d have a hard time calling that neglect. Certainly, it’s hard to imagine that a foreign corporation that has 8% of that very competitive market is “neglecting” it. They obviously carefully calculate their opportunity costs and invest accordingly.

    The Galaxy S II is a particularly extreme example of this. It is actually amazing that they sold 5 million of them because they did this excruciatingly slow country-by-country rollout (check out the Samsung Galaxy S 2 Release Dates blog for more).

    It only “feels” excruciating when you’re in it. It’s still a pretty fast ramp, just not quite of the Apple “build a zillion before you even ship” category. Also, Apple does the same thing, except they hit the US first…

    It feels like they were selling every SGS2 they had and they were looking in odd corners of the factory for ones they might have misplaced “just in case”.

    It feels like that, but I think it’s actually one level removed. I think they’ve learned from Apple that if you do it correctly, the best salesman you have is that “shiny new product” buzz that wears off after the first week it’s on the market. I could be wrong, but I haven’t really heard of any massive shortages where they’ve actually rolled it out.

    So, when you consider that the US market is really big, it makes sense to roll it out last:

    – Any design flaws can be discovered during the slower product ramp
    – Any production glitches can be slowly worked out, demand can be determined (preorders) and the production ramped appropriately

    And then it makes sense not to go into a market until you’re sure you can satisfy the demand. People who have placed a preorder are usually happy to wait as long as nobody else has one, but if other people have been carrying around the thing for a month and they still don’t have it, they get antsy.

    So I think Samsung has been ramping production like crazy, and is probably not entering the US market until they can satisfy initial demand in a way that makes people happy.

    http://www.techradar.com/news/phone-and-communications/mobile-phones/samsung-galaxy-s2-reaches-3-million-pre-orders-953697

  279. @Tom

    the only people who care about how much profit Apple makes are Apple investors.

    @Ravi

    “Apple = the all-powerful Microsoft of days gone by”

    Which is it? Android is crushing/is going to crush Apple or Apple is all powerful?

    “Personally, I think Apple is now driving a shift from native clients to HTML5 ones”

    Since the iPhone first came out, there has been a meme that the native client is dead. Just hasn’t happened. Apple has been pissing off app developers since day 1 of the iPhone’s release. First there were no apps. Then there were with a bunch of restrictions. And a 30% cut. And on the story goes, banning other browsers, banning adult content, political content, banning non-obj c code, etc.

    But actually over time they have become *less* restrictive. (still more restrictive than I would advocate tho).

    Anyway, the fact is that the app store gives developers a good path to making money. It’s almost frictionless. And they do a bunch of the infrastructure work for you. So it’s worth the 30% hit.

    There are some business models that simply don’t work with the 30% hit (like selling books, Readability, etc). But most business models work pretty well.

  280. @tmoney:

    There’s a difference between not relying on others (something Google is doing with Android by having Music, Books and Video establish a floor for the corresponding part of the Android ecosystem) and what Apple is doing: pushing apps / services that are “too successful” off of the platfor just in case they become unreliable later.

    As for the consistency argument, that’s a smokescreen. There are plenty of ways Apple could have pursued IAP consistency without putting up a “KEEP OUT!” sign directed at (among others) content retailers. Remember the months-long wait for the apparently inevitable ban? Apple followed that with a silent fine-print policy revision (so if you weren’t regularly checking the relevant policy and didn’t hear the news through other channels you wouldn’t have known anything had changed). They topped it off with capricious definition and enforcement of the new guideline. It took Kobo 2 months to comply with the new guidelines because Apple kept changing the rules (via higher-ups overriding the decisions of App Store reviewers). And to top that off, the Borders app (also made by Kobo from the same source) is allowed to have features Kobo isn’t!

    This isn’t Apple having trouble figuring out figuring out how to define and enforce certain new kinds of consistency. This is Apple using consistency as an excuse to accomplish one of their other goals: driving out or at least weakening certain competive products and services.

  281. @phil:

    I should have air-quoted ‘all-powerful’ to make my meaning clearer. Microsoft wasn’t all-powerful then any more then Apple is all powerful now. In both cases, however, their power and their even larger perceived power hid their weaknesses and masked the mistakes they were making that were exacerbating those weaknesses.

    Also, I didn’t say that the native client was dead. Instead, I specifically cited Apple and the iPhone as evidence that the native client has plenty of life in it. However, if the native client has plenty of life, that means the dramatic shift from native clients (mostly Windows clients of courses) to web applications in the 2000s was not inevitable. Both native clients and web applications have their place, but that doesn’t mean the balance between them is immutable. Microsoft encouraged the previous shift to web applications by eing an unreliable and untrustworthy platform provider.

  282. (continued – early submission)

    Apple is making the same mistake here. Apple’s level of restrictiveness fluctuates (both up and down) and their degree of control over iOS dwarfs anything Microsoft ever had over Windows. Neither of those realities is good for iOS as a platform or (over the long run) for Apple itself. Apple’s model has reduced many frictions, but added a new friction and a new risk – Apple itself.

    I’ll end with this: if Apple’s 30% cut were truly worth it most of the time, they would have no issue with alternate business models (like Kindle – we do the work and we keep the money) or (for that matter) even alternate app stores (they could even keep the review process intact by charging hourly for reviews as opposed to taking their cut later). Apple’s own behavior is revealing all of the dimensions along which they don’t believe they can compete on the merits and win.

  283. Here’s a well done graphic summarizing the current state of mobile gaming.

    Personally I distrust factoid stat sheets like that. Graphics have a tendency to hide the current state rather than show it. Or to put it a different way, it’s very easy to create a factoid stat sheet that is completely bullshit but hard to verify. It is pretty however.

  284. >As for the consistency argument, that’s a smokescreen. There are plenty of ways Apple could have pursued IAP consistency without putting
    >up a “KEEP OUT!” sign directed at (among others) content retailers.

    Again with the assumption that just because I highlight a valid reason for Apples actions that doesn’t boil down to “Apple Evil! Grrr!”, that I equally think it was the right action. Yes there are other ways they could have (and should have) done it. On the other hand, if it was all about “Keep Out” signs, Apple could have also simply said “No programs with outside stores that don’t also use the IAP”. As for waiting for an inevitable ban, no I don’t recall that. I do recall a lot of fear that the price match policy would drive out stores, and I do recall Apple relenting on that point.

    As for people not checking the fine print and news and missing out on that change, I don’t see the problem. If you aren’t interested in that news, then whether the change occurs or not doesn’t matter to you. If you are interested, you were watching and you heard about it. Case in point, you clearly have much less interest in that subject than say Amazon or B&N would, and yet even you knew about it.

    Another thought on the consistency angle. One of the proposed solutions was to reduce the IAP percentage to something that the other middle men (at lets be honest about this, B&N, Apple, Amazon and everyone of these other distributors are just that) could afford, like say 5% or so. What do you supposed happens next? Maybe developers not wanting to pay 30% start developing their applications in two parts, a free shell that you download from the App store, and then IAP content that actually provides you with the real program. Suddenly you have a bunch of developers who can manage that getting away with losing only 5% of their revenue, while other developers who don’t have that option are still eating 30%. So you know exactly what would happen next, Apple would start arbitrating whether your App abuses the IAP concept or not. Don’t you have enough problems with the arbitration Apple does now?

  285. @tmoney:

    If you are interested, you were watching and you heard about it. Case in point, you clearly have much less interest in that subject than say Amazon or B&N would, and yet even you knew about it.

    Ah, the “everybody knew it was going to happen” deductive fallacy. I can assure you it’s a fallacy. For example, my wife and I were at an actual B&N store a couple of weeks ago looking at the hardware, and the very nice lady showing it off was eager to point out that the Nook app is available for free on the iPhone.

    If I had turned to my wife and said “honey, we’ve been thinking about getting iPhones. The data plan costs a bit, but the iPhone is only $200 (less than the color Nook) — maybe we should go ahead and do that” then I could have easily been harmed by not paying attention. In that particular scenario, B&N would have been as much at fault as Apple, but I can easily envision others where that’s not true.

  286. @tmoney:

    Since you weren’t following that closely, here are some highlights of the countdown to the (canceled at the last minute) ebook apocalypse:
    http://www.pcmag.com/article2/0,2817,2379102,00.asp
    http://blog.readability.com/2011/02/an-open-letter-to-apple/
    http://www.the-digital-reader.com/2011/02/22/amazon-bn-kobo-know-their-ios-apps-will-be-pulled-in-june/
    http://www.rexblog.com/2011/05/10/23130
    http://theorangeview.net/2011/05/the-apple-ebook-apocalypse-draws-nigh/
    http://reviews.cnet.com/8301-18438_7-20062030-82.html
    http://mikecanex.wordpress.com/2011/06/06/apple-app-store-ebook-armageddon-is-now-24-days-away/
    http://mikecanex.wordpress.com/2011/06/09/apple-app-store-ebook-armageddon-is-cancelled/

    Note that the apocalypse was canceled roughly 3 weeks out after over 4 months of uncertainty, with no public comment from Apple on the cancellation.

    Following that, the uncertainty continued:
    http://mikecanex.wordpress.com/2011/07/01/chickenshit-behavior-from-corporate-giants/
    http://www.the-digital-reader.com/2011/07/01/kindle-other-reading-apps-are-live-in-itunes/

    Until:
    http://www.the-digital-reader.com/2011/07/26/ios-reading-app-roundup/
    http://quatermain.tumblr.com/post/8045895465/my-review-of-the-kobo-app

    Also note that, as of this writing, the Kindle (2.5 stars), Kobo (3.5 stars), and Nook (2.5 stars) apps have not recovered from their savage post-store-removal reviews (I believe their previous versions were all at or above 4 stars, as they all are in the Android Market). It seems that, for some reason, users of those apps didn’t appreciate the store links disappearing with little or no explanation. From other things I’ve read, I believe some of the app developers predicted this sort of reaction in advance, tried to work out something with Apple to help prevent it and Apple wasn’t interested in the problem.

    If you can look at all of that (especially the last post describing exactly how much effort Kobo had to go through to keep their app in Apple’s App Store) and say that Apple’s motives didn’t include (among other things like a desire for a more consistent user experience) some active desire to make content retailers feel unwelcome, then I don’t know what to say. I think there is no other reasonable interpretation of the whole series of events. Also, as noted in previous comments, I think that inferred desire of Apple’s says troubling things about Apple and the iOS ecosystem.

  287. Latest bizarro viewpoint from BeatWeek magazine is out.

    At least the free market hating editor-in-chief, one Bill Palmer, had the guts to sign this one:

    Apple needs indirect control over both AT&T and Verizon, so as to ensure both carriers give the iPhone 5 its due. As it stands, salesgeeks in carrier retail stores routinely tell one lie about the iPhone after another in a desperate personal bid to steer would-be iPhone buyers toward Android instead, and this goes on whether it’s the carrier’s policy to favor Android or not. What Apple needs is for these salesgeeks to be informed that they’ll be fired if such behavior continues. But a memo isn’t enough. Instead, Steve Jobs needs to be running these carriers. Because it’s not just about getting a customer to buy an iPhone 5. It’s about molding the carriers such that they offer an Apple-like experience to customers, or at least an acceptable one, as opposed to the disastrous experiences which U.S. consumers are routinely subjected to regardless of which carrier they pick their poison from.

    What Jobs really needs government regulation in the U.S. analogous to that of other civilized nations. As it stands, Verizon and AT&T can and do literally anything they want, no matter how heinous; most of it’s legal and the rest isn’t enforced. Networks are barely there because unlike other nations which regulate minimum number of towers per population density, U.S. carriers simply build a fraction of what they should. And they get away with because, despite all U.S. carriers offering a substandard experience which would be considered a joke by anyone in any other nation, Americans are almost perversely loyal to whichever carrier they’ve chosen. That leads back to them choosing to remain with Verizon and its marginally better substandard network and get stuck with an Android phone they hate rather than moving to AT&T with its marginally worse network and the iPhone they’ve wanted all along. But while Jobs can’t change that kind of mentally handicapped behavior on the part of consumers, can he change the nature of the carriers?

    I would like to believe it’s just a calculated maneuver to stir up controversy, but who knows?

  288. The importance of Series 40 and ‘mobile phone’ goes beyond serving emerging markets. It’s been widely assumed that Series 40 would gradually be eroded, to be replaced by smartphones. However, one of the lessons of the first decade of the smartphone is that it is very hard to create a single platform to cover everything from the low end to the high end. This would suggest Series 40, or at least, the space it represents, will not disappear.
    The next generation of Series 40 is set to become a quasi-smartphone platform (think Qt and WebKit); it will also draw on the user experience lessons learned from the Nokia N9 (Swipe UI) and there is a strong possibility that it will be built on top of a Linux core. This would means a renewal of the boundary line between phone-centric quasi-smartphones (today’s mobile phones) and computing-centric smartphones (today’s smartphones), rather than a subsumption of mobile phones by smartphones, is a distinct possibility. With this in mind, the most intriguing question of the next few years is whether Nokia can outperform the rest of the industry by building sustainable profits in the mobile phone segment. 

  289. @Tom Forest:

    I think they keep changing the link for some reason. Yours didn’t work either, but this one did this morning.

    http://www.beatweek.com/news/8885-cash-iphone-5-release-date-apple-dollars-signals-homegrown-network/

    If this had been posted here as a comment I’d assume it was from a iTroll, but as the comments appear to be closed at Beatweek maybe he’s serious?

    All the iPhone articles at Beatweek have the same tone, none of them allow comments, and this is the first one I’ve ever seen with a byline…

  290. Direct links don’t seem to work at all, at least for me. I had to get to it by navigating the site. He appears to be serious- as the editor in chief he can do what he wants, and he went out of his way to put his name on this one.

    Of course, he’s a loon. Totalitarian loon spouting conspiracy-theories, to boot. At the beginning of the article I just thought he was in serious need of a remedial civics lesson and a printout of the Constitution… then it got much, much worse.

  291. Yeah, that link’s dead too, now. I think that the only people who would seriously link to it are those who want to dissect it, so he’s at least smart enough to have figured that out.

  292. esr Says [formatting added]:
    > In fact, let me make some predictions.
    > To plus or minus 1%, here’s what I expect the share figures to look like in the June comScore stats:
    > – Android: 39.8%
    > – RIM 23.5%
    > – Apple 27.2%

    OK, the editorializing might be trolling, but the facts are:

    According to comScore’s June survey of the U.S. smartphone market, Android remains the leading OS, with a 40.1 percent share — up 5.4 points since March. Apple’s iOS ranks second with 26.6 percent share, and Research In Motion’s third after dropping 3.7 percent to 23.4 percent.

    http://allthingsd.com/20110804/victimized-google-has-40-percent-smartphone-share-in-u-s/

  293. @Marco:

    Yes, Apple might have slightly Osborned themselves with people waiting for the iPhone 5 — their share didn’t increase at all from May.

    BTW, the comscore data (original source) is “installed base” for US subscribers age 13 and older, derived by asking people what kind of phone they are carrying. So there is some hysteresis there vs. sales, even before comscore does their three month averaging.

    In contrast, Canalys tries to figure out how many phones were shipped each quarter, and they don’t give out US figures, but they claim that worldwide, last quarter 48% of shipped smartphones were Android and 19% were Apple.

  294. I don’t like browser based data in general, for reasons previously discussed, but this data is somewhat interesting, simply because it’s broken down geographically.

Leave a comment

Your email address will not be published. Required fields are marked *