The Smartphone Wars: Not the expected surprise

I have plotted the May 2011 comScore data. There are two conspicuous things to notice about it. One is that recent rumors of an Android stallout seem utterly bogus, and the other is that Apple appears to be actually gaining some share rather than simply bouncing around 25% in a random way.

In my last post, I took apart some tendentious claims that Android’s market share had peaked in March. We now have two months of share data collected with consistent methodology past March and I think we can say this claim is busted. Whether we interpret market share as share of total userbase or of recent adopters, there’s no stallout showing here; after slowing down a bit from the holiday season Android growth has been remarkably steady. Over-50% share still looks like happening in October.

On the other hand, Apple may be climbing out of a rut. While 1.3% share growth over 18 months is nothing to brag about (especially when compared to Android’s 32.9% gain) it looks to me like during the last 4 months Apple’s competitive position has been improving just enough to be noticeable. Given the timing, this probably is due to Apple going multicarrier in the U.S. – though if I were Apple I’d certainly have been hoping for a more dramatic improvement. Apple’s longer-term problem is that Android continues to gain about two users to every one of Apple’s.

RIM, Microsoft, and HP continue to tank. Microsoft may be falling a little faster recently. Overall, there are no surprises here.

UPDATE: Actually, we do get a surprise today. Ars Technica says there will be an iPhone 4S, except it calls it a 5S, with a ship date of September.

185 comments

  1. Actually, there are at least 3 conspicuous things to notice about it.

    The third one is that Samsung’s share of total handsets stopped declining and actually reversed from last month (24.5% to 24.8%), a probable indication that the impact of the iPhone 4V on them was short-lived.

    The thing is, in the US, Samsung phones are carried by 2.85 times as many people as Apple. Of course, they’re mostly featurephones, and Samsung has heroically been slowly increasing their total marketshare at the same time as people have been abandoning featurephones in droves.

    I have seen reports that Samsung made its target of 19 million smartphone handsets for last quarter (more than Apple shipped in Q1), but nobody said how much they beat the target by. Before the beginning of the year, Samsung set an aggressive goal to ship 330 million total handsets this year, an 18% increase over last year.

    We all know how rapidly the dumbphone business is declining. I don’t know whether Samsung is going to make their goal or not, but certainly more than compensating for the bleeding in the dumbphone business and ramping total handset penetration is an excellent sign for them.

  2. Apple’s marketshare since Jan shows some growth (24.7 -> 26.6) => roughly 2 percentage points.
    It no longer can be dismissed as statistical noise. A trend is there, but a very modest one. A fraction of these 2 percentage points is what the Verizon move has netted Apple.

  3. This is not really on subject but the continuing strength of Android phone sales also brings the continuing Microsoft presence in “IP” licensing. Now Chrome is getting mentioned. Here is today’s announcement:
    http://www.microsoft.com/Presspass/press/2011/jul11/07-05WistronPR.mspx

    The one thing I notice is they are almost all with companies I have never heard of like Itronix and Velocity Micro, HTC being the exception. Seems odd to me that MS is licensing software to folks whose primary contribution is hardware. I can’t help but think they are trying to somehow ‘surround’ Google rather than deal direct. And why nothing from Google on this? You would think if they thought they needed a license they would have it. And if not, they would be ‘reinforcing the perimeter’ against this sort of thing.

    Seems to be getting harder to have a MS free computing experience or to contribute to their bottom line. Love to hear any thoughts on this.

  4. Y’know — I may be all wet, but…

    I’m wondering if comscore’s spreadsheet/script/man-behind-the-curtain/whatever has a mistake and doesn’t low-pass the RIM data.

    It would be interesting to see that chart you just added the total to with 3 month averaging for RIM. Of course, you’d have to make up data for the first two months.

    1. >It would be interesting to see that chart you just added the total to with 3 month averaging for RIM. Of course, you’d have to make up data for the first two months.

      I’m not inclined to torture the data that much. But you know where to download my source code and data :-)

  5. @PhilK:

    > Love to hear any thoughts on this.

    Any lawyer worth his salt is going to tell a small-time hardware company with a low volume high margin business to just roll over and pay the Windows tax, until some big company whacks Microsoft sideways. You can always hope that the judge will give a ruling that the patents were egregious, and give you some ammunition to go sue MS to get your money back if you care at that point.

  6. @esr:

    That’s fine. It’s just that, looking at the rounded shapes around Xmas and then earlier this year for Apple and Android, it appears that RIM would follow the same shape, just a bit lower. I was thinking you could tease the seasonality out a bit better. One of these days I’ll set up a website…

    @PhilK:

    One more thought. Obviously, once the cost of defending a patent relative to the size of your business drops below some threshold, it’s better to fight. This happens both with big companies, and with little companies with scrappy in-house counsel. One of my favorite posted-on-the-internet letters of all time is from a CEO who used to be a patent litigator:

    http://www.audioholics.com/news/industry-news/blue-jeans-strikes-back

    Of course (wandering a bit farther afield), a close second in the intellectual property category will always be this one:

    http://drdobbs.com/184404796

  7. @Patrick Mauphin

    > Any lawyer worth his salt is going to tell a small-time hardware company with a low volume high margin business to just roll over and pay the Windows tax, until some big company whacks Microsoft sideways.

    True enough and that could well be the advice these companies are getting. That doesn’t really address Google’s lack of response. Unless there really is MS IP in Android.

  8. @esr:

    Did you notice this line in that Ars Technica article?

    Pegatron overhauled its entire factory setup to satisfy a 10-million-iPhone CDMA iPhone 4 order, but less than four million of those have shipped.

    @PhilK:

    > That doesn’t really address Google’s lack of response. Unless there really is MS IP in Android.

    Google probably ignored it when it was HTC, because HTC already had a relationship with MS. It is quite plausible that the public $5.00 number is mostly masked by some other funny money going the other direction, or at least, a reduction in some other money going the same direction. (HTC still sells MS phones.)

    The other agreements were just struck. If google is going to do anything about this, it’s still early days. But it might prove difficult for google, in any case. The best strategy of which legal battles to pick is not always obvious.

    1. >Pegatron overhauled its entire factory setup to satisfy a 10-million-iPhone CDMA iPhone 4 order, but less than four million of those have shipped.

      I did notice that and meant to comment on it. It probably means that demand for the 4V is way lower than Apple was expecting. And it’s not easy to see how, if they’ve only shipped < 4M in 5 months, they're going to clear the other 6M before the 4S/5S comes out in September.

  9. > And it’s not easy to see how they’re going to clear the other 6M …

    I don’t know that they will. The way the article was worded, they were ready and willing to build 10 million, but that didn’t happen.

    I don’t know if Apple had other companies building the 4V. If so, that could explain this in a way that doesn’t look quite so bad. If not, well that makes Apple’s problems look even worse than at first glance — they were supposed to ship those CDMA phones internationally, as well. 4M minus 2.2 M the first quarter would mean only 1.8 M the second quarter, or 600K/month.

    Of course, some of the components Apple contracted for will be non-returnable, which partly explains some of the discounting going on:

    http://news.cnet.com/8301-13506_3-20076811-17/verizon-lining-up-iphone-4-discount/

    Other components, such as the screen and battery, will be usable in the much wider 3G market. An interesting optimization problem to figure out what to push back to the vendor, what to scrap, and what to reuse…

  10. I’m still trying to figure out why Google is letting Microsoft get away with forcing manufacturers to pay for using Android. Consider that Microsoft can now make Windows 7 less expensive than Android for those makers…

    1. >Consider that Microsoft can now make Windows 7 less expensive than Android for those makers…

      Which would be more interesting if WP7 had a market share higher than statistical noise. The comScore figures obscure this because they lump in WP7 with pre-WP7 WinCE phones that are actually doing better. The reality is there is zero consumer demand for WP7 and the handset makers know it.

  11. I’m still trying to figure out why Google is letting Microsoft get away with forcing manufacturers to pay for using Android. Consider that Microsoft can now make Windows 7 less expensive than Android for those makers…

    I haven’t heard anything about Samsung, Huawei, or ZTE paying the MS tax. HTC is a special case, and very few other manufacturers probably matter yet.

    I suppose it’s arguable that Google expected they would take home a few patents from Nortel, and now they have to come up with a new strategy. I, for one, would like to see them make an SEC filing that said they set aside $4.5 billion for legal fees in the coming year.

  12. The carriers are probably still suspicious of microsoft, and nearly no one is wandering into the carrier store(s) and asking for a windows phone.

  13. @esr:

    Ars was using iPhone 5s as the plural form of iPhone 5, not claiming that the next iPhone is going to be called the iPhone 5S.

    That being said, based on the specs that have leaked, this phone looks like an iPhone 4S, not an iPhone 5. Which is trouble for Apple because the flagship Android “iPhone 4S” competitors have already launched (Samsung Galaxy S 2, HTC Sensation, LG Optimus 2X/G2x, Motorola Atrix [with the Droid 3 / Milestone 3 arriving very soon based on the reports from Verizon and China]). Even worse, some of these are LTE phones or will have LTE variants soon, leaving Apple’s competitors free to move onto the next thing.

    1. >That being said, based on the specs that have leaked, this phone looks like an iPhone 4S, not an iPhone 5.

      I agree, and see the same problems ahead for Apple that you do.

  14. Interesting article here about the future of RIM. It caught my eye because I was pondering the outcome of the Nortel patent auction and RIM’s future and came up with a similar scenario.

    http://crackberry.com/10-reasons-why-google-will-buy-research-motion

    There’s a lot of reasons why it wouldn’t make sense which the author kind of glosses over, especially the part about Google not wanting to be in the hardware business competing with its current partners. But I could see a spinoff of the hardware business while retaining the IP and OS elements. Maybe. Interesting to speculate on.

  15. I’m not convinced that Google will buy any handset manufacturer, least of all RIM. I really get the feeling that Google knows what its strategy is. Unlike several of the players like MS and Nokia who I think really don’t have a clue what their own strategy is. Something stupid like buying a handset manufacturer is something I thought MS would do … and am surprised that they seem to have changed course from what I guessed was their course (unless I was really wrong – hey it happens). But I think Google has a clear strategy set out for itself and is moving firmly and focused on executing it.

    I’m not saying its an optimal strategy necessarily, I’m not that smart. But the signs are there that Google is clear on what they are doing.

    Apple too, really knows what their strategy is. I am not as convinced as some that the Apple strategy has a likelihood of failure but the future path of iPhone 4S / iPhone 5 and suing their main partner, Samsung, and lastly the weak “iCloud” silliness makes it look like they are losing focus and getting a bit lost.

    1. >I’m not convinced that Google will buy any handset manufacturer, least of all RIM.

      I’m not either; it would be too threatening to their relationships with the rest of the Android Army. Acquiring defensive patents is the only good reason I can think of to do it. The least implausible scenario is one in which Google waits for RIM’s market cap to drop to on-the-cheap levels, buys the company, keeps the patent portfolio, and sells off the rest of the assets.

  16. Apple can’t get left behind in 4G and still maintain the cachet of being the premium brand.

    1. >Apple can’t get left behind in 4G and still maintain the cachet of being the premium brand.

      This is a problem Apple already has, of course. The 4S release in September imples that there’s no plausible release date for a 4G-capable iPhone 5 before 2012, which means we’re looking at 6 to 8 months during which Android phones will be making all the running in 4G/LTE.

  17. The least implausible scenario is one in which Google waits for RIM’s market cap to drop to on-the-cheap levels, buys the company, keeps the patent portfolio, and sells off the rest of the assets.

    If someone else (HTC or Motorola perhaps) is talking about consuming RIM i could see Google putting some money into the kitty in return for the rights to the patents (with a guaranteed license back of course).

    Personally i found the reasons a little bit unlikely. If QNX conversion is “measured in weeks” then why has RIM been osbourning blackberry consistently for months now. As you say, the patent portfolio is really the only decent reason there.

  18. “Microsoft Corp has demanded that Samsung Electronics Co Ltd pay $15 for each smartphone handset it makes based on Google Inc’s Android operating system as the software giant has a wide range of patents used in the mobile platform, local media reported on Wednesday.

    Samsung would likely seek to lower the payment to about $10 in exchange for a deeper alliance with Microsoft for the U.S. company’s Windows platform, the Maeil Business Newspaper quoted unnamed industry officials as saying.”

    http://www.reuters.com/article/2011/07/06/us-samsung-microsoft-idUSTRE7651DB20110706

  19. Microsoft is doing their level best to try to sue Linux and Android into oblivion. This is a direct result of their utter failure in the marketplace. The funny thing is, they saw it coming years ago.

    1. >Microsoft is doing their level best to try to sue Linux and Android into oblivion.

      No, I think Microsoft is doing its best to be a patent parasite, sucking the blood of other companies without actually killing them.

  20. More specific writeup here: http://fosspatents.blogspot.com/2011/06/android-dispute-microsoft-thumps.html

    The ones involving FAT are a given, I guess. They keep coming up, and I wish someone would hack up a way around them. Others seem pretty weak, and I can’t believe that Microsoft doesn’t have vulnerabilities with prior art. I was hoping that someone would put together a Groklaw-like effort to take these patents off the table. Does that mean there’s merit in what Microsoft is saying? Can Google not step in and put pressure on Microsoft to stop this? On the other hand, it doesn’t seem that this threat is slowing Android down any… ;-) I guess it’s just another “Microsoft tax.”

  21. @esr:

    No, I think Microsoft is doing its best to be a patent parasite, sucking the blood of other companies without actually killing them.

    I think they'[re going for the “death of a thousand cuts” plan. Instead of suing for massive damages, they’re trying to get as many companies as possible to settle out of court. It generates FUD and keeps revenue flowing to Microsoft. The idea being that once “conventional wisdom” comes around to say “yeah, Linux violates Microsoft patents, so we need to protect ourselves,” maybe enough companies will be sufficiently annoyed to not bother it.

    @David:

    The ones involving FAT are a given, I guess. They keep coming up, and I wish someone would hack up a way around them.

    There is a hack around them using random characters for the 8.3 filename. It’s simply not enabled in the kernel that ships with Android or, indeed, most Linux distributions.

    1. >I know you guys hate this guy….

      I don’t hate him. I never noticed him before. He’s not hateful, just kind of tendentious.

      Tendentious because his definition of “multiplatform” seems to function as more of a rhetorical device than anything else. There’s no principled difference between the distribution of developer attention in notionally non-multiplatform era of the PC (Windows + Mac + Linuxes) and the notionally “multiplatform” distribution of developer attention smartphones seem to be heading for (Android + iOS + WP7 and other bit players). That is, in both you see a couple of whales and a bunch of minnows. As I’ve written before, this is exactly the result to expect in any technology market with strong network effects.

  22. Actually, you know what? The threat of these patents is really serious, because, in the future, Android won’t just be tied to a handset, where the cost is absorbed. Eventually, someone will generic-ize Android, and make a “proper” distribution out of it, to be put on things like laptops or wall-wart computers or routers. Who pays the “tax” then? “Linux” — in Android — is finally set to take over in a role at the forefront of the user’s experience. (“Desktop” isn’t the right word here.) These patents would seem set to prevent that.

  23. @phil:

    I think Horace Dediu is pretty sharp. He’s been better than all the professional analysts at predicting Apple shipments/profit. He’s really good at the short term game. Having said that, he’s human, and like all humans, he misses stuff sometimes.

    In this case, for a start, he’s missed the fact that there are a lot of programs out there that let you easily build apps for multiple platforms.

    On phones, I think you’ll find these levels of apps:

    1) Web apps (rapidly rising in quality as HTML5 and javascript interpreters get better)
    2) Generic apps, written using appcelerator or something like that, maybe with minor platform tweaks
    3) Lovingly handcrafted apps that seamlessly feel like part of the platform

    Apple is great at #3, and Android is slowly catching up. Most of the apps in the windows store are going to be #2.

    Where I think Horace goes wrong is in conflating all these categories. What is the point of a platform if it costs more than Android and doesn’t provide the user with lots of apps in the #3 category?

    Unfortunately for Microsoft, they are being watched closely by antitrust regulators all over the world, and it is going to be very hard for them to pursue a strategy with big manufacturers that makes the cost of Android more, or even anywhere near, the cost of Windows mobile. That’s not the way patents work if the defendant has good lawyers on his side.

    @David:

    > The threat of these patents is really serious

    That’s exactly what Microsoft wants you to think. Don’t buy into it. Yes, patents are the sand in the wheels of commerce. No, the grains aren’t that big and the gears aren’t that fine. It’s not going to grind to a halt.

  24. HTC is buying S3. Hadn’t heard of S3 in awhile, but it makes great sense. S3 has some graphics patents that, e.g. Apple might not have licenses to…

  25. @Patrick Maupin

    “there are a lot of programs out there that let you easily build apps for multiple platforms.”

    This popped into my head immediately too…… but we have no data on how many apps are just exports from multi-platform frameworks. I really have no idea on this. And multi-platform apps tend to suck from what I’ve seen.

    However, to his point – it’s much easier to develop a mobile app than an old school desktop app. Your distribution problem is completely solved…. and they tend to be simple focused apps.

  26. phil,

    Except he’s already wrong. The power law of distribution is already affecting the mobile space. In gaming, a year ago anyone could make an iOS game and have a shot at the brass ring. These days, you better have an Angry Birds-like phenomenon or the support of a major publisher.

    That said, that there’s such a disproportionate app-to-user ratio for Windows Phone bodes well for that platform. Ease of development is EVERYTHING, and given a gneration or two Microsoft may well catch up.

    I just bought an Xbox 360. Hardware wise, compared to its console brethren from Sony and Nintendo, it’s a pile of fucking shit. Not well designed or built. But since the Wii’s gimmick is all tapped out, the 360 emerged as the leader of the 7th console generation. Why? Ease of development. Developers can use industry-standard DirectX APIs to write games for it just like they would on a PC. Faster turnaround time, more games, more options for consumers = boom, winning. This is what Microsoft does best.

  27. >There’s no principled difference between the distribution of developer attention in notionally
    >non-multiplatform era of the PC (Windows + Mac + Linuxes) and the notionally “multiplatform”
    > distribution of developer attention smartphones seem to be heading for (Android + iOS + WP7 and
    >other bit players).

    Sure there is. Remember that up until about 2001, there was no Windows + Mac + Linux, there was just Windows. Hell, a developer shipping just Windows + Mac was a huge deal (see Bungie or even Microsoft) let alone all 3. These days, multi platform developers are not so surprising. The costs and barriers of multi platform development have gone down and investing in development on one platform doesn’t mean excluding all the others. There may still be whales, but the whales of 2010 are smaller than the single whale of 1995.

    1. >Remember that up until about 2001, there was no Windows + Mac + Linux, there was just Windows.

      This claim just demonstrates that you weren’t paying attention.

  28. >>Remember that up until about 2001, there was no Windows + Mac + Linux, there was just Windows.

    >This claim just demonstrates that you weren’t paying attention.

    Ok, you’ve piqued my curiosity. Name three industry-standard software projects that were available on Windows, Mac and Linux in 1999 or 2000. My curiosity won’t be satisfied if your list includes only text-interface applications like Nethack or POVray (which is not to say that POVray was industry-standard in 2000).

  29. > Name three industry-standard software projects that were available on Windows, Mac and
    > Linux in 1999 or 2000.

    How about games? Loki Software was releasing Linux ports of games in 1999 and 2000. Bungie’s Myth II: Soulblighter was then available on Windows, Mac, and Linux. I don’t know for sure if the other games Loki ported were also available on Mac, so I can’t speak to them.

  30. I was paying attention. And if you were a consumer in 1999, you consumed Windows, or for a very small minority Mac. Sure Red Hat was going to take the world by storm, indeed 2000 would be the year of the linux desktop, hell even Best Buy got in on the action and carried a few boxed copies of Linux. And the only people who bought them were computer geeks in the first place. The year of the linux desktop came and went, and there were no linux desktops, unless you counted the sooper sekret config option, buried 10 pages deep on Dell’s website.

    If you can’t see the differences between software development and distribution today vs the mid 90’s, I have to wonder if you were paying attention.

    1. >If you can’t see the differences between software development and distribution today vs the mid 90?s, I have to wonder if you were paying attention.

      Of course there are differences; distribution costs are way lower, for one. But none that support Dediu’s basic claim. In the 1990s, major application got ported to Windows and Mac, with an exceptional few also getting ported to Linux. In the developing smartphone world, major apps get ported to Android and iOS with an exceptional few also getting ported to WP7 and elsewhere. The more things change, the more they stay the same.

  31. tmoney,

    Before their acquisition by Microsoft, Bungie was a NOTORIOUS Mac fanboy game developer. They were committed to shipping Mac only until they caved to market pressure and released a PC version of Marathon 2. That there are Mac versions of pre-Halo Bungie games is therefore extremely unsurprising.

    jsk,

    And Loki was shut down because there’s no money in Linux gaming. There’s no money in it for a number of reasons: small user base, smaller user base willing to pay, library hell, etc.

  32. @ jsk

    Sure there were minor excepts, I already mentioned Bungie. Remember you’re talking to a mac guy here, so what cross platform development there was, I was well aware of. But Bungie and Loki were the exceptions, not the rules (indeed, after Bungie was acquired by Microsoft, they stopped cross platform development). But this was the era when even Adobe was considering halting their mac development. The era when a multi million dollar deal was necessary to ensure the continued production of Microsoft Office on the mac.

  33. @Jeff Read
    > And Loki was shut down because there’s no money in Linux gaming. There’s no money in it for a number of
    > reasons: small user base, smaller user base willing to pay, library hell, etc.

    Agreed, there was not much money in it _at the time_.

    Are you familiar with the Humble Bundle? http://www.humblebundle.com Occasional releases of indy games packs at pay-what-you-want prices, DRM-free, and you get access to versions for all three platforms. There have been three iterations so far, and each one saw Linux getting a 25% share of purchases, and each time Linux had the highest average purchase price. Windows is still gaming king, but the idea that there’s no money in Linux? Pshaw.

    And, the only problems I have with these games on Linux are due to the fact that I mostly run pure 64bit and not everyone provides 64bit binaries.

  34. @Jeff Read:

    Ease of development is EVERYTHING, and given a gneration or two Microsoft may well catch up.

    I just bought an Xbox 360. … Ease of development. … Developers can use industry-standard DirectX APIs to write games for it just like they would on a PC. Faster turnaround time, more games, more options for consumers = boom, winning. This is what Microsoft does best.

    Unfortunately for Microsoft, I think that this is going to be harder to leverage in the future, especially for cellphones. The thing is that the tools for generating platform-independent software are getting much better and games, to an extent not found with other software, get away with unique app-dependent user interfaces, so native look and feel doesn’t seem to be as important.

    I could be wrong, and Microsoft could come up with some difficult-to-clone must-have games for the smartphone, but I’m not seeing it.

  35. Another difference between then and now:

    1999: People who needed a powerful Unix workstation could either shell out for an expensive unit from Sun or SGI, or join the Linux community and convert an old PC.

    2011: People who need a powerful Unix workstation just buy a Mac.

  36. I could be wrong, and Microsoft could come up with some difficult-to-clone must-have games for the smartphone, but I’m not seeing it.

    If they seed the clouds a little bit the same way they did for the Xbox — trade money hats for exclusivity deals — then it becomes hella feasible.

  37. jsk,

    The Humble Bundle is a product of garage developers without much to lose by experimenting with different platforms. We had those in 1999 too; I distinctly remember a particularly good pinball game called Rollemup being released this way. Try convincing the AAA publishers to port to Linux. You can’t; even id Software is considering abandoning the Linux platform. In part this is because Microsoft’s DirectX libraries and dev tools are so much better than anything available on Linux that Windows exclusivity gets you enormous gains in developer productivity with a sacrifice of virtually none of the profit.

  38. @Jeff Read:

    > If they seed the clouds a little bit the same way they did for the Xbox

    They can probably buy their way into a certain percentage of the smartphone market with games. But, almost by definition, that has to be a lower percentage of the smartphone market than they have acquired of the gaming console market. The question is, can they buy market share with anything besides games. Could be, but I’m not seeing it.

  39. > Try convincing the AAA publishers to port to Linux.

    The “AAA publishers” can stay where they are. The Humble Bundle profits show that there IS a market available to be tapped, with very little competition, to anyone who is willing to try.

  40. there’s no money in Linux gaming

    I saw an interview with a game developer who said even if he never sold a single copy of the Linux port of his software, its existence made his code better, because it forced him to separate out the platform-dependent parts of his code. The goal is to make those parts as small as possible, leaving the vast majority the same for all platforms. And if he decides to port to a console, all he has to do is rewrite those small parts. If he does it correctly, each platform’s specific code is built as a library that can be called by the various programs, so that future games can be written to use the same libs.

    This is not unlike what the Linux kernel went through when they added support for the Alpha processor, and had to go through all the code and clean up the i86 assumptions. At the time, it was a painful process, but the result is a far more maintanable codebase. I’m sure that if you asked Linus, he’d agree that it doesn’t matter if anyone uses Linux on Alpha; going through that exercise to make portable code made it easier to do even the things that only affect one architecture.

  41. @phil: “WSJ on next gen iPhone:”
    Apple preparing for new iPhone in 3Q

    The interesting thing about this article is that it *never* refers to the new iPhone as the iPhone 5. I find that very telling. If Apple had ever referred to the upcoming phone as a 5, I think it’s near-certain that the WSJ would have reported this. Are they already trying to lower expectations?

    1. >Are they already trying to lower expectations?

      No. It’s pretty clear this isn’t the 5 at all, but the rumored “4S” referesh.

  42. @tmoney

    Ok, you’ve piqued my curiosity. Name three industry-standard software projects that were available on Windows, Mac and Linux in 1999 or 2000.

    – Netscape Navigator
    – StarOffice (okay, maybe not “industry standard” then, but it is now)
    – Sun’s various incarnations of Java
    – Adobe Acrobat (reader, at least)
    – Apache Web Server (on Mac OS X Server)
    – Tomcat (on Mac OS X Server)
    – The GNU Compiler Collection (on Mac OS X Server 1.0)
    – GNU Emacs

    Okay, so that was more than 3.

  43. > Try convincing the AAA publishers to port to Linux.

    @jsk: “The ‘AAA publishers’ can stay where they are. The Humble Bundle profits show that there IS a market available to be tapped, with very little competition, to anyone who is willing to try.”

    Sorry, jsk, but he’s right. I love Linux and would nothing better than to see it displace Microsoft as the dominant PC platform, but it isn’t going to happen soon, if ever. Looking back, I wish I had realized sooner that it would take a technology disruption to a new hardware platform to offer an alternative to the lock that Windows has on the PC. I originally thought that Linux itself was sufficient disruption, but it hasn’t proved so. (It’s interesting to speculate whether it would have been sufficient in the absence of Microsoft’s illegal, unethical monopolist tactics.)

    The reality is that there isn’t enough money in Linux PC games to build a new Electronic Arts, Inc. or to get EA to support the platform. Sure, hobbyists can make a few bucks on the side with Linux games, but there isn’t enough revenue there to grow a new generation of AAA gaming firms with a Linux focus.

  44. “However, to his point – it’s much easier to develop a mobile app than an old school desktop app. Your distribution problem is completely solved…. and they tend to be simple focused apps.

    Well, the desktop market used to be full of simple focused apps, but then Microsoft sucked all of those profits from the industry by sucking the functionality into the bloated O/S. From one point of view, the user gained value from getting bundled functionality but from another point of view innovation was beaten to death with a stick. I miss those days myself.

  45. @esr: “No, I think Microsoft is doing its best to be a patent parasite, sucking the blood of other companies without actually killing them.”

    They can go either way. I think the ideal outcome for Apple and Microsoft is to kill Android and establish a duopoly, where they can both use control of their ecosystems to prevent any disruptive innovations until they can co-opt them. And that way they get to dodge antitrust issues since we’ll have a “choice” of which walled garden to stay in. But if that doesn’t happen then yes, Microsoft will happily be a parasite.

  46. @esr:

    ‘No. It’s pretty clear this isn’t the 5 at all, but the rumored “4S” referesh.’

    First of all, the rumors are very conflicting on the next gen iPhone.

    We can be confident on the following:
    1) better camera
    2) faster

    past that there are various rumors:
    1) NFC
    2) LTE
    3) bigger screen (3.7″)
    4) low and high end models
    5) thinner
    6) lighter
    7) CDMA/GSM in same SKU
    8) more RAM
    9) 64 Gig flash option

    What would be the thing that would push it to being a 5 vs a 4S? LTE? Most people seem to think that the next gen iPhone won’t have it, but this is all speculation. I’m sure Apple would love to release the iPhone with LTE, but unless they can get decent battery life out of it – they won’t. Of course this could be the reason for the delay until the fall – waiting for a decent LTE chip. But no one has any information on this one way or another that I’m aware of.

    One other note: I’m running iOS 5 beta now. It’s quite buggy and clearly not production ready. This tells me they probably never planned on releasing an iPhone 5 in June. The cloud stuff and wifi syncing are great though….

  47. @cathy
    “The interesting thing about this article is that it *never* refers to the new iPhone as the iPhone 5”

    Up until the last minute no one knew the name of the iPhone 3GS either…

    If they go LTE, it could also be the iPhone 4G. My money is on the name iPhone 5 though. But no telling…

    Oh, in extension of my last post the other rumors are:
    10) new radical case design with tear drop shape
    11) continuation of iPhone 4 form factor
    12) elimination of the home button
    and the stupidest rumor of all:
    13) almost no local storage (all “cloud”)

  48. OT: nice #askobama tweet, Eric. Wetting my undies watching hysterical @benshapiro and @AndrewBreitbart tweets.

  49. @ Morgan

    Of your list, Netscape, Java and Acrobat are the only 3 that would have likely been found on any consumer’s computer. The stuff on OS X server never made it to consumers, and also falls into the beginnings of 2001 at which point I said things started changing. Again, not that there weren’t exceptions, but that these exceptions were hardly the rule they are considered today.

  50. @phil:

    I’m sure Apple would love to release the iPhone with LTE, but unless they can get decent battery life out of it – they won’t.

    You can quit hammering on the “Android == bad battery life” meme, because most of the Android battery life problems were fixed in Gingerbread, with probably more battery fixes upcoming in the 3.0 line. Do try to keep up.

    And clearly some of those rumors are disinformation and some are not. Rumors we can confirm from what is known publicly:

    3) bigger screen (3.7?)
    5) thinner
    6) lighter

    Bigger screens (3.7″ and larger) seem to becoming more widely available on the Android platform, and all phones tend to be getting thinner and lighter over time, so we can basically just guess that these will happen.

    7) CDMA/GSM in same SKU

    The radio in the iPhone 4V is of the dual CDMA/GSM variety, even if the GSM part is disabled. We can safely conclude that Apple chose that chipset because they were already planning on using it in future phones.

    8) more RAM

    Moore’s Law and all, sure. Same category as the thinner/lighter

    9) 64 Gig flash option

    Maybe. 64 Gig flash chips are still pretty expensive. Maybe too expensive to make it this round. We’ll see.

  51. “That is, in both you see a couple of whales and a bunch of minnows. As I’ve written before, this is exactly the result to expect in any technology market with strong network effects.”

    But don’t you predict that network effects will always turn one of the whales into another minnow? :)

  52. Of your list, Netscape, Java and Acrobat are the only 3 that would have likely been found on any consumer’s computer.

    1) Nobody specified “consumer’s computer.” Linux wasn’t likely to be found any “consumer’s computer” unless you count me as a consumer. 2) There are plenty of other examples. Just admit that you and Bennett weren’t paying attention and move on.

  53. @Morgan

    1) I did

    “I was paying attention. And if you were a consumer in 1999, you consumed Windows, or for a very small minority Mac.”

    and the original article hinted at it when talking about retail boxed distributions (I don’t recall seeing Tomcat or GNU emacs on the shelves at CompUSA)

    2) But if we tallied the total number of examples then compared to the non cross platform vs the total number now, I would bet my life on there being far more cross platform software now than in the mid 90’s

  54. @Morgan Greywolf: “You can quit hammering on the “Android == bad battery life” meme”

    I’m hammering on the LTE == bad battery life meme. Android phones vary, some have good battery life, some bad. Best I can tell, *ALL* LTE phones have bad battery life.

    “64 Gig flash chips are still pretty expensive. Maybe too expensive to make it this round. We’ll see.”

    the iPod touch has had a 64 gig option for quite a while now. And iPad already has it.

  55. the iPod touch has had a 64 gig option for quite a while now. And iPad already has it.

    The iPod touch and the iPad don’t have the same economics constraints as the iPhone. But I’ll concede that Apple could go that route. It isn’t like Android phones haven’t only had the option for over a year now. *laugh*

  56. ars on the WSJ article: http://arstechnica.com/apple/news/2011/07/wsj-next-iphone-to-be-thinner-and-lighter-than-iphone-4.ars

    “More interesting is the fact that one supplier claimed to have been told by Apple to be prepared to ship enough components to Apple manufacturing partner Foxconn to build 25 million iPhones by the end of the year. Apple’s additional manufacturing partner, Pegatron, reportedly received an order for as many as 15 million next-gen iPhones. If both sources are correct, Apple could be forecasting iPhone sales of 40 million.

    That number may seem high, especially given the fact that the new hardware is not expected to be available until the latter part of the third quarter. But Apple already moved nearly 19 million iPhones in the first calendar quarter of this year, and that figure is double the amount sold the same quarter the year before. With Apple expected to announce another record quarter later this month, 40 million total iPhones in four months might not seem quite so far-fetched.”

  57. @phil:

    BTW, if Apple really does ship, say 9 M iPads and 15 M iPhones in the quarter, I think it says a lot about Apple’s strategy.

    Apple’s gross margin on the iPad is, I believe, somewhat lower than its margin on the phone. That’s one reason why other manufacturers have taken awhile to step up to the tablet plate on appropriate pricing.

    Apple has done a good job lately of figuring out the next business to be in and transitioning over there. iPod -> iPhone -> iPad.

    With the iPod, they had a category killer in the combination of good design and iTunes. With the iPhone, they had a category killer based on good design and network effects from the iPod — it was really ahead of the competition when they rolled it out. With the iPad, they’ve essentially defined a category (it was there before, but not large enough to worry about) with design, price, and network effects.

    The iPad’s usability and low price allowed them to greatly magnify the network effects before there was any viable competition — it’s amazing that an analyst can predict that their iPad shipments will be 60% of their iPhone shipments for the just-ended quarter.

    Will this large and increasing lead be enough to slow the Android juggernaut’s advance into the tablet space? Stay tuned.

  58. >So now it’s the mid-90s and we’re confining ourselves to boxed software? Keep shifting your
    >position…you’re getting funnier.

    You would be less of an ass if you actually read the discussion instead of reflexively assuming anyone who disagrees with eric is wrong and/or stupid.

    I haven’t shifted my position and it isn’t now the mid 90’s. From the very first post where I disagreed:

    Remember that up until about 2001 … There may still be whales, but the whales of 2010 are smaller than the single whale of 1995.

  59. There has been some back and forth about how to treat the iPad with respect to “ecosystem” discussions. I don’t have a link and for the life of me I can’t recall where but I saw an article yesterday somewhere about a company that manufactured in dash car navigation systems going from a Windows CE (if memory serves) implementation to an Android implementation. The article had a photo of the Android based unit sitting there with all the Android apps displayed.

    Now I know that these already exist but they caused me to think afresh about the utility of a general purpose computer system in my car, that I can easily purchase additional applications for aftermarket. And it reminded me again of the utility of the ecosystem integration that Google is pushing. Reading my email in the car dash is going to really ruin my commute.

  60. @tmoney:

    I fully read your argument. I simply disagree your statement that there were no major cross-platform software applications until 2001. There was a major cross-platform movement that began in the mid-90s, with Sun’s Java flying the biggest banner. In fact, if it weren’t for that cross-platform movement, there are huge swaths of applications that wouldn’t run today … on Windows. Your problem is that you’re very narrowly focused on the mom-and-pop home user and your ignorance is showing.

  61. >I fully read your argument. I simply disagree your statement that there were no major
    >cross-platform software applications until 2001.

    And therein lies your problem. I never said that there were “no major cross-platform software applications until 2001”. In fact, in my very first post I noted that there were exceptions, and at the time, those exceptions were pretty big deals. But that is exactly what they were, exceptions to the rule.

  62. > the whales of 2010 are smaller than the single whale of 1995.

    True that.

    But that statement actually has much more to do with anticompetitive behavior than developer tools. In particular, the “you can sell DOS/windows on every machine for $50, or on a machine-by-machine basis for $100.”

    MS may pull some tricks out (patents, exclusive games, etc.) but that particular anticompetitive behavior is hard to leverage when Google says “you can sell Android on every machine for $0, or on a machine-by-machine basis for $0.”

    It turns out that Microsoft has some leverage. I’m not really sure that it’s patent leverage, and it will be hard to tell for awhile. Pretty much every single vendor who they have blackmailed or are now attempting to blackmail buys something from Microsoft.

    For precedent-setting purposes in the marketplace, it may not matter if, for example, a deal with HTC or Samsung doesn’t actually increase the revenue to Microsoft. If Samsung “buys” Android for $10, but that is mostly offset by reduced price on Windows for its laptops, well the next phone vendor won’t know about that little part of the secret.

    It will take a large phone (or tablet or laptop) vendor who has nothing to do with Microsoft and wants nothing to do with Microsoft to stop the madness.

    Enter Barnes and Noble.

    For all those wondering where google is when its customers are being sued, that might be part of the answer. If I were google, rather than attempt a broadside, I might work with one Android vendor at a time to try to whittle away at the various monopolies. Who wants to speculate on whether Google is helping B&N with their legal battle?

  63. Who wants to speculate on whether Google is helping B&N with their legal battle?

    If I were Google, I would. Very quietly and discreetly, much the same way Microsoft helped The SCO Group with their ummm….”case”.

  64. @Morgan

    Every article for every LTE phone I’ve read at ThisIsMyNext, Engadget, Boy Genius, and Anandtech all say that these LTE phones have pretty bad battery life. The only one close to tolerable seems to be the Droid Charge, but it still is far more battery-limited than a comparable 3G phone.

    1. >High-end Android tablet pricing starts to get more realistic:

      But only starts. This class of tablet needs to come down to $250 or so before it’ll have a mass market.

  65. > This class of tablet needs to come down to $250 or so before it’ll have a mass market.

    Sure. But I bet the major manufacturers will be a bit supply constrained for awhile anyway. Apple has almost made us forget the “normal” way that electronics are sold, with high initial prices and rapid reductions. But we’ll remember again when people scream to Samsung and Toshiba and Mot about the price reduction on the tablet they bought two months ago, and are politely informed that that’s life.

    Or maybe there’s a difference between Apple early adopters and non-Apple early adopters. After all, the Apple buyer knows that a bajillion people bought the same thing on the same day for the same money, but non-Apple early adopters know they paid a premium to get one early.

    On a related note, the Samsung Galaxy S 2 has sold 3 million worldwide and still isn’t available here:

    http://www.pcworld.com/article/235020/samsung_sells_3_million_galaxy_s_ii_android_phones.html

    And I’m sorta interested in Samsung’s iPod touch equivalent. It’s got nice specs:

    http://www.samsung.com/us/system/consumer/product/yp/g1/cw/ypg1cwxaa/Galaxy_Player_WiFi_Spec.pdf

    but it’s still not available here despite already being sold as a home VOIP phone:

    http://www.intomobile.com/2011/07/01/samsung-galaxy-player-pmp-offered-androidbased-home-phone-koreas-lg-u/

  66. More consensus piling on:

    http://www.reuters.com/article/2011/07/07/samsung-idUSL3E7I606Q20110707

    If the analyst consensus is correct, I think that Samsung will become both the leading smartphone vendor and the leading handset manufacturer. (Before any Apple fanbois “correct” me, I will also state that Apple will surely be, by far, the most profitable handset manufacturer this quarter.)

    I posted a link to this article a couple of weeks ago, but failed to note this paragraph at the bottom:

    Nomura raised its full year smartphone unit sales forecast in 2011 by over 6%, to 478 million sold units from its previous estimate of 452 million units. It said Nokia’s product weakness has allowed vendors of Android-based smartphones in particular to gain share.

    I found the Nomura PDF report that contains this 478 million number. The report is actually about the smartphone impact on the semiconductor industry. There are several interesting tidbits there:

    One of the key themes we see for the second half of the year is mass market Android phones exceeding expectations. Smaller android vendors in Asia such as Huawei, TCL, and ZTE have significant momentum. These vendors are primarily using Android to drive smartphone adoption at lower price points. We estimate the low-end smartphone TAM will grow from $1.5bn in 2011 to $5bn in 2014.

    TCL? I hadn’t heard of TCL. (What an unfortunate search term…)

    http://en.wikipedia.org/wiki/TCL_Corporation

    1. >One of the key themes we see for the second half of the year is mass market Android phones exceeding expectations. Smaller android vendors in Asia such as Huawei, TCL, and ZTE have significant momentum. These vendors are primarily using Android to drive smartphone adoption at lower price points.

      Pfft. If they’d been reading this blog, they might have caught a clue to that oncoming locomotive six months ago.

    2. >Samsung will become both the leading smartphone vendor and the leading handset manufacturer.

      Oh, now this is interesting.

      Samsung’s Galaxy S II has sold more than 3 million units since its debut in late April. Some analysts expect the phone running on Google’s Android platform to become Samsung’s best-ever model with potential sales of 20 million units.

      That’s twice the volume of Apple’s Pegatron order for the iPhone 4 refresh.

  67. > If they’d been reading this blog, they might have caught a clue to that oncoming locomotive six months ago.

    Well, yeah, but still…

    I mean, lots of people (even on this blog, where they should know better) are, depending on their perspective, either wringing their hands or shedding crocodile tears over what they think patents are going to do to Android, and (for example) the Microsoft and Oracle software patents aren’t going to do a damn thing in most of the world, and even most of the hardware patents aren’t going to do a damn thing in China.

    478 million smartphones this year and accelerating. That’s 6 times the number of people in this country who already own smartphones, and over 1.5 times the total population of the country. Granted, not all of the smartphone shipments are Android yet, but unless something drastically changes (WebOS for free with free porting…) the vast majority of them will be soon.

    So even if our legal system completely screws the pooch on google/Android patent liability, the horse is out of the barn, and we’ll only be hurting our own global competitiveness by shutting down everybody except Apple here. Won’t bother Android.

    1. >yet the flatline of growth per month for Android barely shows in your graph:

      A constant first derivative is not “flatlining”. And it doesn’t get any closer to being flatlining when you mislabel it as a second derivative.

      Please try not to have attacks of stupid this intense in the future. They waste my time.

  68. @Patrick Maupin

    $15/phone? (http://www.techflash.com/seattle/2011/07/microsoft-wants-15-for-every-samsung.html)

    And you were so sure of yourself. Doesn’t mean Microsoft is going to get it, of course, but if they do, why can’t Oracle get $15/phone (assuming they prevail)?

    @esr
    > Please try not to have attacks of stupid this intense in the future. They waste my time.

    So, you’re not going to address my questions, you’re just going to attack me?

    You gotta admit that the curve for android growth has flattened, while Apple is back to growing.

    1. >So, you’re not going to address my questions, you’re just going to attack me?

      Your “flatlining” has Android increasing share by 1.70% and userbase by over 2 million a month. Many product managers would kill their own relatives for this kind of growth. And yes, I will attack you when you babble like a blithering idiot, because you’re babbling like a blithering idiot.

  69. > Your “flatlining” has Android increasing share by 1.70% and userbase by over 2 million a month. Many product managers would kill their own relatives for this kind of growth.

    Yes, but in the 4 months prior to the iPhone 4V launch, Android growth averaged 2.45%, and in the 4 months since the iPhone 4V launch, Android growth has averaged 1.725%. Android’s monthly growth fell to 70% of what it was. During the same time periods, Apple growth was 0.10% (prior to iPhone 4V launch), and since, it has averaged 0.475% 4.75X growth.

    In April, you wrote:

    Even on the very optimistic assumption that Verizon sustains its pace through Q2, Android phones are selling so much faster in aggregate (ratio of about 10:1) that iPhone 4V is barely going to budge the needle on the market share numbers (if that).

    and last week you wrote, “Android, according to comScore and Nielsen, is still gaining share at about 2% a month.” Which is true if you want to think of the difference between 2.0% and 1.725% as trivial.

    1. >Which is true if you want to think of the difference between 2.0% and 1.725% as trivial.

      Which it is. The error on this type of survey is a lot larger than that – a good rule of thumb is to treat any difference below 3% as likely statistical noise and ignore it unless it’s stable over a time period that’s long compared to your sampling rate. I was being fair about this with my 2% average; I also dismissed the maximum Android growth rate of 2.70% as an outlier in the other direction.

      Derivation amplifies noise. You’re now trying to make a big story out of tiny stochastic fluctuations amplified by derivation, which is not as blatantly idiotic as confusing your derivatives but wrongheaded just the same. The best demonstration that you’re qualifying for the wishful-thinking loony bin is the userbase numbers. These aren’t noisy and the difference is clear; Android has been growing at twice the rate of iOS since June 2010 and that growth curve has not been affected one bit by the Verizon iPhone. Look at it! If anything the slope has slightly increased recently; four months ago I thought it was basically linear, now it looks exponential.

      What actually seems to be indicated by that graph – as Patrick Maupin has previously observed – is that iOS is feasting on the corpse of RIM while Android captures most of the overall smartphone market growth (e.g. the dumbphone conversions). Whether or not that’s the case, it doesn’t get much further from “flatlining” than this.

  70. I find it totally unsurprising if iOS is taking RIM’s customers rather than dumbphone customers. The former group has already demonstrated they have money to pay for a premium product/service, while the latter has demonstrated that they aren’t.

    All of which supports my prediction that Apple is set to take one part of the market, and Android is set to take the other, larger part.

    1. >The former group has already demonstrated they have money to pay for a premium product/service, while the latter has demonstrated that they aren’t.

      Not so fast. You are correct that the RIM users have shown they have money for a premium service, but you don’t have any warrant for believing that all dumbphone users do not. Also, iPhone is not the only premium product – consider dual-core 4G/LTE Androids, the really fancy models that sell for $499 and up. I think your explanation has some validity, but beware of oversimplifying.

  71. > I think your explanation has some validity, but beware of oversimplifying.

    If you think it has some validity, it’s hard to understand why you dismissed it as fanboyish delusion earlier.

    Clearly some dumbphone users will upgrade to premium products. Clearly some android products will be perceived as premium. But if things stay as they are, Apple will capture the majority of premium customers, and android will continue to convert the majority of dumbphone users. In that scenario, you’ll see explosive growth in android share right up to the point when all the old dumbphone customers have smartphones. Let’s take a wild guess and suggest that at this point the market is 15-25% Apple, 70-80% Android, 0-10% Other. Now what happens?

    Unless Android manages to garner the perception of having the highest-quality phones as well as the cheapest ones (which would require a change) you’d expect the market to stay segmented, as it does in the automobile market.

    The point I’ve been making all along is that the analogy to the 90s desktop-OS war is very weak. By 1995, most people just didn’t see Macs as being a premium product. They were a niche product, but the niche wasn’t wealthy people or people with corporate IT accounts, it was people in certain professional fields (pro audio, desktop publishing, etc). It was a small niche, and it was always going to be hard to defend it once you could get ProTools, Quark, and Adobe software for Windows. iOS is in a completely different niche today, and a much larger one, than MacOS was in in the 90s.

    Here’s another place that the analogy breaks down: interoperability. As of right now, interoperability is not an issue with different phone OSes, partly because email is an shared, open standard, and partly because the nature of the device precludes proper device-device networking as you see on desktop networks. In the 90s, one of the major reasons to leave MacOS (even if you were a graphic designer or musician) was that you couldn’t use PC people’s files, fonts, networks, printers, or peripherals. None of that matters (much) with phones, today. In five years, if it turns out that our phones are our primary computing devices, it may turn out to be more important, but I doubt it.

    1. >If you think it has some validity, it’s hard to understand why you dismissed it as fanboyish delusion earlier.

      Probably because you loaded the last version with assumptions about Androids being an inferior alternative to Android, or some such. There is a difference between a claim about quality and a claim about positioning. Not all “premium” goods are superior, some are Veblen goods (positional display) and the iPhone has an element of this.

      >Unless Android manages to garner the perception of having the highest-quality phones as well as the cheapest ones (which would require a change) you’d expect the market to stay segmented, as it does in the automobile market.

      You’re in dreamland. Android already has the “highest-quality” tag for a significant class of users, and no I’m not speaking of geeks. Falling behind in HSDPA+, 4G/LTE, and multicore has consequences, and this is one of them.

  72. >You’re in dreamland. Android already has the “highest-quality” tag for a significant class of users, and no I’m not speaking of geeks. Falling behind in HSDPA+, 4G/LTE, and multicore has consequences, and this is one of them.

    No, *you’re* in dreamland. What do you think the market share breakdown is in the class of phones that cost the user $200 + contract?

    >Probably because you loaded the last version with assumptions about Androids being an inferior alternative to Android, or some such.

    You mean where I mentioned that my assessment of the quality of iOS vs. Android phones was different than yours? If you can point to a sentence I have written in this thread where I assert that Androids are inferior as a matter of fact, rather than as a matter of consumer perception, I would be surprised. But perhaps my memory is failing me in my old age.

    1. >What do you think the market share breakdown is in the class of phones that cost the user $200 + contract?

      I don’t know, and I’m not in the habit of claiming to know things I don’t. Point me at facts, if you think you have them.

  73. We don’t need any new facts or figures to make that conclusion. ASP for all smartphones is dropping like a stone (you have repeatedly observed and predicted this); meanwhile, ASP for iPhones is steady, much higher than the market average, and iOS market share is steady too (all these points have also been observed by you). We can divine from all this that Android ASP is significantly lower than iPhone ASP, and continuing to fall as Android userbase grows… and as you just observed, most of the growth in the Android userbase is coming from dumbphone users.

    Reasonable conclusion, based only on your own observations: most of the phones in the high-ASP bracket are being sold by Apple.

    Put it this way: Dumbphone users are not buying super-expensive smartphones, by and large. Since most of the new Android converts are (according to you) ex-dumbphone owners, Android clearly isn’t owning the expensive space in the market. But gee whiz, if it helps you to not blow a gasket, we can just say that this is clearly because wealthy people have been brainwashed by neuro-engineered apple marketing materials.

    1. >We don’t need any new facts or figures to make that conclusion.

      Is that all you have? Bollocks. You disappoint me.

      It’s easy to construct scenarios in which most of the high-ASP phones are Androids. Dead easy. Remember those $49 reconditioned 3GSes? We don’t know how many of those are out there. We don’t know what fraction of the iOS userbase they represent. A simple Google search shows that 3GSes are still for sale from multiple sources, including Best Buy and Walmart and carrier stores. Therefore we don’t know the ASP of the iPhone.

  74. @bennett:

    I find it totally unsurprising if iOS is taking RIM’s customers rather than dumbphone customers. The former group has already demonstrated they have money to pay for a premium product/service, while the latter has demonstrated that they [don’t].

    Actually, I know several people who carry iPhones who can just barely afford it. Never personally knew any RIM-toters in that category. I’m sure some ex-RIM-toters now carry iPhones, but I know others who are tightwads like me, but who merely have a functional reason to carry the RIM. Many of those are not switching phones now, and probably won’t switch to iPhone.

    I can easily afford to carry whatever I want to. And I’ll probably eventually swap the prepaid dumb phone that I never carry for a (probably prepaid) smartphone that I carry somewhat more often.

    On the corporate side of things, I can easily believe (especially given the data from good technologies that we’ve seen) that corporations are transitioning from RIM to iPhone. Lack of user control is actually a plus, and the $49 iPhone 3GS didn’t hurt at all.

    On the personal side of things, I’m not so sure. In any case, I have bemoaned in the past the lack of a survey question for the “recent phone acquirers” — “What did it replace?”

  75. @fake account:

    And you were so sure of yourself. Doesn’t mean Microsoft is going to get it

    They’re not. Or if they do, it will be part of a transaction where they pay less for windows for laptops or somesuch. Smoke and mirrors.

    of course, but if they do, why can’t Oracle get $15/phone (assuming they prevail)?

    Because even taking at face value that Oracle’s patents are valid and infringed, they are not critical to the user experience that actually nets google money. So even if Oracle walks away with anything, it won’t be anywhere near $15/phone. And even if the court system completely fails and Google owes $15/phone for past phones, you can bet that Android will exist just fine without Dalvik.

    See, for example, the Infoworld article J2ME’s future shines:

    Nevertheless, the momentum is behind J2ME. However, here’s the rub: Query any Sun executive about how big the revenue stream is (or will be) from licensing J2ME on those 1 billion handsets, and you invariably get the same response: “negligible,” “none,” or words to that effect. Nothing in the company’s annual report tells me anything different. According to Sun senior management, the company makes money by selling servers — end of story.

    Sun management believes that if J2ME wins on the handsets, then as carriers deploy more wireless services, they are more likely to choose Sun J2EE (Java 2 Enterprise Edition) servers over somebody else’s server. The same model, I am told, will work for companies deploying J2ME devices. Now I don’t have an MBA, but to me it seems that spending multimillions of dollars on Java development on mobile is the long way round to selling more boxes.

    I don’t buy it. Not that I don’t believe Sun. I just think the business model will not work in the long run. If Microsoft had 1 billion handsets running Phone Edition, you can bet your bottom dollar it would charge whatever the market will bear. Perhaps that is one good reason why J2ME is the platform of choice for most of the handset manufacturers and operators.

  76. LG is not doing so good. They were the third-largest handset manufacturer. They may still have this distinction, or if their dumbphone business is eroding as fast as everybody else’s, they may have dropped a place.

    And apropos our recent discussion about whether buying exclusives on games could keep Microsoft in the phone game, apparently part of RIM’s problem is that people actually do like to play games on their phones. Fortunately for RIM, lots of corporations and governments still mandate Blackberrys. Unfortunately that’s changing. Even the VA is considering allowing employees to use their own smartphones for work.

    That change will probably take place by October. Maybe by then those VA employees will be able to buy the bleeding edge Samsung Galaxy S 2, which we can’t get in the States yet.

  77. >Therefore we don’t know the ASP of the iPhone.

    Sure we do, they report it all the time. In Q4 2010 it was $610. In Q1 2011 it was $625.

    It’s harder to work out ASP for android given that so many companies are involved. But I encourage you to look at this graph, which gives you total ASP by company up to Q1 2011. I hope you’ll have the good grace to feel just a moment’s embarrassment.

    1. >Sure we do, they report it all the time.

      No. What we know is an approximation of the ASP of phones sold direct by Apple itself.

  78. @bennett:

    >What do you think the market share breakdown is in the class of phones that cost the user $200 + contract

    sounds like a question from an end-user perspective. Which, BTW, is the perspective esr is taking. The manufacturer’s ASP is almost completely irrelevant to the question of retail ASP, especially in a heavily subsidized market that also arguably contains a lot of refurbished phones.

    I’ve discussed in the past that, in order to maintain lock-in, even for the new (not refurbished) iPhone 3GS at $49, AT&T might be eating most of that discount itself. Maybe Apple gives them to AT&T for $50 off, and AT&T sells them for $150 off. Unless you can point to data that says otherwise, there is no known relationship between the retail ASP of a (possibly used) iPhone and Apple’s ASP.

  79. >there is no known relationship between the retail ASP of a (possibly used) iPhone and Apple’s ASP.

    But this is also true for Android phones. I think when Apple’s vendor ASP is double that of the nearest competitor, and quadruple the ASP of the majority of its competitors, the burden is on you to give evidence for your fanciful suggestion that retail ASP is no higher for iPhones than for Android phones.

    Here’s one way you could estimate it- find out how much it costs to buy a vendor-unlocked phone and a sim-only plan for two years (as you can do in civilized countries) and compare it to the cost of getting a phone+plan on a two-year contract. I just did the math for the iPhone in the UK and the answer is that it costs £131 less to get the phone on contract. Probably part of this is a vendor discount, and part of it is that they penalize you on the sim-only plan to discourage that behaviour.

    I’m not sure what the discount is on phones from HTC or Motorola, perhaps it is comparable, or perhaps not?

  80. Nobody mentions one really interesting bit of information in the graphs. The last graph – the one that shows market change in absolute numbers, says that the customer base in the US grows by about 2 million per month. Android is essentially sucking up all of this growth. The other players are swapping users between themselves. RIM and MS users go to Apple. Apple gets to show an increase in sales and can continue to to keep their heads in the walled garden.

    People are buying Android phones because they can afford them, and they provide an improved user experience over the dumbphones. The cheapest iPhones cost about 5 times of what the cheapest Android phones cost, when sold new. It doesn’t matter that the iPhone is cooler, has a slightly better user interface and a much faster processor. The cheap Android phones improve the lives of people who can’t afford the iPhone and who don’t have the type of regular income that allows you to go on an expensive price plan with Verizon or AT&T.

    Even more people will be able to afford the phones once prices drop in the Us like they have in Europe. Data plans are currently a sales driver. You get 1 GB of data traffic for free with my current Swedish operator (normally i pay $0.50 per day I use data traffic). Recently I was at Europython in Italy. The conference sold SIM cards for $30, with a one year free data plan (6 GB/month at full speed, then it slows to max 64 kbit/s). Of the $30, one third was set aside for phone calls. Ryanair is also offering SIM cards that have no roaming costs for incoming calls anywhere in Europe. I think they said their cards came from Spain Telecom.

    We are getting very close to the point where the homelsess can afford an Android phone with a data link. That is disruptive technology. The homeless will not have iPhones, because there is no way that Apple can sell their phones at 20% of todays price and make a profit. Even if they could, we haven’t seen the end of the price drops in the Android phones.

    To give you an indication, there is a Huawei phone that is a hybrid between dumbphone and smartphone – fixed keypad and small screen, but with networking, radio and stuff for sale in Sweden for about $50 including VAT. Huawei will soon stop selling it, because they will have an Android model that will be about as cheap.

    1. >Nobody mentions one really interesting bit of information in the graphs. The last graph – the one that shows market change in absolute numbers, says that the customer base in the US grows by about 2 million per month. Android is essentially sucking up all of this growth. The other players are swapping users between themselves. RIM and MS users go to Apple. Apple gets to show an increase in sales and can continue to to keep their heads in the walled garden.

      Don’t get too committed to this theory. I’ve just started plotting userbase change per month along with the platform statistics. The Android and total-userbase trends look partially but not fully correlated. Interestingly, the iOS and Android trends look more tightly correlated (though separated by about a 2x multiplier).

  81. > for your fanciful suggestion that retail ASP is no higher for iPhones than for Android phones.

    Where the fuck did I make this fanciful suggestion?

  82. @Jacob:

    Nobody mentions one really interesting bit of information in the graphs. The last graph – the one that shows market change in absolute numbers, says that the customer base in the US grows by about 2 million per month. Android is essentially sucking up all of this growth. The other players are swapping users between themselves. RIM and MS users go to Apple. Apple gets to show an increase in sales and can continue to to keep their heads in the walled garden.

    Actually, this has been discussed before. Maybe a few months ago. The good technology data seemed to show that a lot of corporate accounts moved from Blackberries to $49 iPhone 3GS. I’m sure that this theory does account for a large portion of the movement, but you will, of course have some RIM -> Android, and some dumbphone -> Apple. It’s hard to know the percentages.

  83. > Falling behind in HSDPA+, 4G/LTE, and multicore has consequences, and this is one of them.

    You’re funny when you get blustery, (and you tend to make more mistakes). For example:
    It’s HSPA+, or HSDPA, but the thing you’ve named ‘HSPDA+’ does not exist.

    T-Mobile announced HSPA+ for 50 markets last Summer, because they don’t have the spectrum or funding to get to LTE. AT&T and Verizon do. Having HSPA+ in an Android phone is only an advantage for T-Mobile subscribers, and when the AT&T / T-Mobile USA merger completes, all T-Mobile customers will be moved to AT&T phones and services such that AT&T can repurpose the CDMA/HSPA+ spectrum for its LTE efforts.

    I have to believe that customers of T-Mobile USA (already the set that produces the most churn of any set of cellular customers) are already headed for the exits.

    LTE is technically better than HSPA+ in every measure except for capex spending ( the ability to upgrade an existing WCDMA infrastructure). You’ll not see a HSPA+ phone on either Verizon or AT&T (or Sprint).

    The issue for LTE right now is battery life. Take a look: http://www.anandtech.com/show/4465/samsung-droid-charge-review-droid-goes-lte/10 Additionally, Verizon has already stated that it’s LTE footprint won’t match it’s 3G footprint before the end of 2013. AT&T just stated that it’s LTE efforts are 2-3 years behind Verizon’s.

    These are the reasons why it will be another 12 months before we see LTE on iOS devices. You can have all the early access LTE on Android that you like, but you’re going to pay for it in greatly reduced battery life, and it’s nowhere near ubiquitous. Having LTE *right now* is not (much of) an advantage.

    With reference to multicore, it’s not clear how Android’s architecture lets any app leverage the presence of multiple cores.
    Google won’t even come up with a good story on why GPU acceleration isn’t turned on:
    http://code.google.com/p/android/issues/detail?id=6914

    (and yes, I know that cyanogen builds tend to have this.)

  84. >> for your fanciful suggestion that retail ASP is no higher for iPhones than for Android phones.

    >Where the fuck did I make this fanciful suggestion?

    Well, if you agree that retail ASP is much higher for iPhones than Android phones, I’m not sure why you’d bother to bring up the difference between retail and vendor ASP.

  85. Here’s an issue I don’t think has been addressed in the various discussions of network effects and stickiness: aren’t some Android phones sold with older versions of the OS, and some also aren’t upgradeable to a newer version? I should think that would negatively impact brand/OS loyalty.

  86. @fake account:

    > Additionally, Verizon has already stated that it’s LTE footprint won’t match it’s 3G footprint before the end of 2013.

    Irrelevant red herring. They’re planning on covering where over 60% of the population lives before the end of 2011.

    > Having LTE *right now* is not (much of) an advantage.

    Doesn’t that depend on where a person lives? Believe it or not, the 60% of the population who are going to be covered by LTE before the end of this year are probably already factoring that into their decisions.

  87. @Bennett:

    Well, if you agree that retail ASP is much higher for iPhones than Android phones, I’m not sure why you’d bother to bring up the difference between retail and vendor ASP.

    This sub-thread started when Eric said “It’s easy to construct scenarios in which most of the high-ASP phones are Androids.”

    Note that “most high-ASP phones are Androids” does not implicate overall Android ASP. It’s a very simple statement that operates by model of phone, not by OS.

  88. So you’re telling me that even though Motorola, Samsung etc. have 1/4 the ASP, give or take, this is just clouding the fact that they are selling most of the high-ASP phones. Even though we know they’re outselling iPhones by only 2:1 in terms of units (from Eric’s graph in this post), and that Apple’s sales are almost entirely high-ASP models.

    Following basic arithmetic, this would require that more than 50% of Android models have higher ASP than iPhones, even though each Android company has overall 1/2 or 1/4 the average ASP of Apple in total. For that to be true, I guess the other 50% of Android phones are being given away to the vendors for free. Is that what you’re telling me? I’m not sure how robust that would be as a business model.

  89. > Following basic arithmetic, this would require that more than 50% of Android models have higher ASP than iPhones,

    This started when you said “What do you think the market share breakdown is in the class of phones that cost the user $200 + contract?” and it’s been explained to you that some unknown percentage of iPhones were sold below that number. Now you’re back acting like they’re all at that number. As esr said, bollocks.

  90. No, this started when I suggested that Apple currently dominates the high-price market, something I thought would be relatively uncontroversial, but apparently I was wrong.

    If some small percentage of iPhones are sold at a low price, and some small percentage of Androids are sold at a high price, that is really neither here nor there, unless you are a pedant. I don’t know why I’m still surprised to find pedants on the internet, decades down the line, I guess i’m just a slow learner.

    1. >If some small percentage of iPhones are sold at a low price, and some small percentage of Androids are sold at a high price, that is really neither here nor there, unless you are a pedant.

      There is a pont to this. You made an unsupported assertion about RIM refugees vs. other potential iOS or Android adopters. I agreed that the explanation might have some validity but is oversimplified, because some dumbphone users may be able to pay for a premium product and (independently) there are Android phones in the “premium product” category. You then attempted to use high ASP as a proxy for “premium”.

      Here’s the actual point: If you stick to a model in which iPhones are the only “premium” product, you will fail to predict the behavior of the market correctly.

  91. Ok, but again this applies to the Androids on that ASP chart as well. If there is a busy market in second-hand iPhones that is driving down the real ASP by a large enough margin to matter, why not also for androids?

  92. >Doesn’t that depend on where a person lives? Believe it or not, the 60% of the population who are going to
    >be covered by LTE before the end of this year are probably already factoring that into their decisions.

    Maybe, at the same time, most people have wi-fi at home, and a good chunk probably have it at work. So what does it matter whether they have 1x or LTE at home, they’re probably using the wi-fi.

  93. If there is a busy market in second-hand iPhones that is driving down the real ASP by a large enough margin to matter, why not also for androids?

    1) It’s not just second-hand Apples at $49 (some of those were at $0). New ones too. We don’t know how many. We’re going in circles again here.

    2) Apple had a head start. There are a lot more iPhone 3xx than any similarly aged Android model.

    3) A lot of iPhone 4s on AT&T were $20 early upgrades to keep iPhone 3 users locked in.

  94. @tmoney:

    Maybe, at the same time, most people have wi-fi at home, and a good chunk probably have it at work. So what does it matter whether they have 1x or LTE at home, they’re probably using the wi-fi.

    Normally, when your house is covered by LTE, so is your neighborhood. Maybe even most of your city. Maybe even most of the places where you use the data plan that you paid for.

  95. > 1) It’s not just second-hand Apples at $49 (some of those were at $0). New ones too. We don’t know how many. We’re going in circles again here.

    I don’t see how this weighs against my claim that iPhone is dominant in the high-price segment. If they are selling some percentage of new phones for cheap, that just means the high-price segment is smaller than apple would like it to be. You can bet that Steve Jobs would prefer never to sell a ‘budget’ iPhone, if he could help it.

    >2) Apple had a head start. There are a lot more iPhone 3xx than any similarly aged Android model.

    I’m sorry, maybe I’m stupid but I don’t see how this weighs against my claim that iPhone is dominant in the high-price segment.

    > 3) A lot of iPhone 4s on AT&T were $20 early upgrades to keep iPhone 3 users locked in.

    Even if that’s true, those people are still clearly in the high-price market (because they were willing to buy an iPhone 3 when it was new). AT&T had to offer a carrot here because their network is horrible, and because iPhone 3G users knew this first-hand… not because demand for the iPhone is soft. So once again, I don’t see how this weighs against my claim that iPhone is dominant in the high-price segment.

    I had an iPhone 3G on AT&T in Manhattan for two years. Their network was almost preposterously bad.

  96. @Bennett:

    If there is a busy market in second-hand iPhones that is driving down the real ASP by a large enough margin to matter, why not also for androids?

    Apple had a head start. There are a lot more iPhone 3xx than any similarly aged Android model.

    I’m sorry, maybe I’m stupid but I don’t see how this weighs against my claim that iPhone is dominant in the high-price segment.

    Do you really not think that answers the question about why there might be a bigger second-hand market for Apples than for Androids?

  97. @Bennett:

    It’s not just second-hand Apples at $49 (some of those were at $0). New ones too. We don’t know how many. We’re going in circles again here.

    I don’t see how this weighs against my claim that iPhone is dominant in the high-price segment.

    Maybe it doesn’t. But this was also in answer to your question “if there is a busy market in second-hand iPhones… why not also for androids?” I really don’t know why you asked if you didn’t want the answer, but whatever.

    If they are selling some percentage of new phones for cheap, that just means the high-price segment is smaller than apple would like it to be. You can bet that Steve Jobs would prefer never to sell a ‘budget’ iPhone, if he could help it.

    This seems a non-sequitur. I’m sure all phone manufacturers would prefer not to sell budget phones. But they all do.

  98. > Interestingly, the iOS and Android trends look more tightly correlated…

    I’m telling ya, I think that’s because comscore didn’t smooth the RIM data. All it needs is something like this before the plot:

    rim_data = [rim_data[0], rim_data[0]] + rim_data
    rim_data = [sum(rim_data[x:x+3]) / 3.0 for x in range(len(rim_data)-2)]

    1. >I’m telling ya, I think that’s because comscore didn’t smooth the RIM data.

      It’s not just RIM. The iOS and Apple deltas look well-correlated with each other but not with the other trends (RIM, HP, MS). Well, there might be some negative RIM-to-iOS correlation, especially in the latter half of the graph.

  99. @fake account
    “Having LTE *right now* is not (much of) an advantage.”

    I think it’s a differentiator. And most people won’t realize the battery life sucks until they start using it…

    So it’s good for sales, bad for customer satisfaction.

  100. Actually, it makes great sense for Apple to blow off LTE, keep the battery life up, and get the unit cost down.

    Why?

    International. Even in Q1, 70% of Apple’s sales were not in the US. They’re starting to make inroads into China and other lower-income places.

    The US market is increasingly difficult. Foreign markets have lots more dumbphones, and with Nokia imploding, Apple probably smells blood.

    In those developing markets, LTE is not such a big deal:

    http://gigaom.com/broadband/stat-shot-lte-isnt-yet-taking-over-the-mobile-world/

  101. I think it’s a differentiator. And most people won’t realize the battery life sucks until they start using it…

    You can improve battery life by turning off 4G when not in use in phones so equipped.

  102. It’s not just RIM. The iOS and Apple deltas look well-correlated with each other but not with the other trends (RIM, HP, MS). Well, there might be some negative RIM-to-iOS correlation, especially in the latter half of the graph.

    Agreed, but I think when you smooth the RIM data, you’ll see much better correlation between iOS/Android and the overall market. I was responding to this:

    The Android and total-userbase trends look partially but not fully correlated. Interestingly, the iOS and Android trends look more tightly correlated (though separated by about a 2x multiplier).

  103. > You can improve battery life by turning off 4G when not in use in phones so equipped.

    Yeah. Did anybody see this? I think it was mentioned in slashdot a few days ago:

    http://www.networkworld.com/news/2011/070111-duke-wifi.html

    Also, the Xoom apparently has another $50 knocked off of it if you use a Staples coupon:

    http://www.geek.com/articles/gadgets/geek-deals-motorola-xoom-10-1-inch-dual-core-android-tablet-449-wifi-599-4g-contract-2011077/

    That’s 25% off the starting price. Another 25% and they’ll be in business (or at least force price reductions in the Nook, etc.).

  104. > Maybe, at the same time, most people have wi-fi at home, and a good chunk probably have it at work.

    And at the coffee shop, and McDonalds, and various hotels, the airport, the grocery store, etc, etc etc.

    Which is why AT&T bought Wayport.

  105. > Also, the Xoom apparently has another $50 knocked off of it if you use a Staples coupon:

    Obviously they’re attempting to drive sales. (Chortle)

  106. > So it’s good for sales, bad for customer satisfaction.

    Which increases the return rate. You open sores whores r funny.

  107. @Patrick Maupin re: SleepWell

    That wouldn’t be too hard to implement.on something like DD-WRT or Tomato. The problem is that in order to get the power gains, it seems like all the APs within range would need to support the protocol. If enough public interest could be generated in this technology, someone could make a tidy profit upgrading existing routers/APs to use the technology.

  108. >Normally, when your house is covered by LTE, so is your neighborhood. Maybe even most of your
    >city. Maybe even most of the places where you use the data plan that you paid for.

    I don’t know. Maybe it’s just me, but the times I’m most likely to use my cell data are the times that I’m on the road. So it might cover the city I live in, and that’s great for the few times I’m where I live and work and don’t have wi-fi access, but does nothing for my trips to the beach or anywhere else.

    That said, it would be interesting to see stats on where people actually use their data (in relation to the LTE locations) and whether or not that consideration actually enters their minds when they’re making their purchase decision.

  109. > Maybe it’s just me, but the times I’m most likely to use my cell data are the times that I’m on the road.

    That’s certainly when it’s most useful, as in “Where the heck am I and what’s the best place to eat around here.”

    > That said, it would be interesting to see stats on where people actually use their data

    Agreed. Do you think you travel more than most? I used to travel a fair amount, but I don’t move around that much any more.

  110. No, I definitely don’t travel more than most, but when I’m local as I said, I’m covered by wi-fi nearly 95% of the time. When I do travel, it tends to be outside of major metro areas, and therefore outside the LTE deployments for now and into probably the next 2 or 3 years.

  111. @phil:

    claim is that the Verizon Thunderbolt costs $262 to make ($40 extra for the LTE chip)
    iPhone 4 costs $171

    Semiconductor pricing is notoriously squishy. The teardown people usually hit it in the ballpark, though.

    Notice the prices on these phones. The only best seller under $100 is the iPhone 3GS. Doesn’t this contradict the claims that 1) android is going to kill iPhone on price

    You’ll notice this market distortion is only at one carrier. AT&T is still using $50 iPhone 3GS to keep from losing iPhone customers to Verizon. Long-term, though, I think we’re going to see much more transparent phone pricing. For example, take a look at a Verizon or AT&T quarterly and see how much the pre-paid category is growing.

    Also, you’re studiously ignoring all the other phones under $200 but >= your magic $100 threshold.

    price is the primary deciding factor? This is US only of course….

    Seriously?

    The article says “The firm contacted carrier retail stores…”, not “The firm contacted WalMart, Best Buy, Target and Amazon.”

  112. @Patrick Maupin “Seriously? The article says “The firm contacted carrier retail stores…”, not “The firm contacted WalMart, Best Buy, Target and Amazon.””

    It’s just a data point. I have no idea who goes to the carrier stores. And the break out on # of sales for carrier stores vs best buy, wal mart, amazon

    Don’t forget you got on the “Thunderbolt is outselling the iPhone on Verizon” bandwagon when someone surveyed a set of Verizon stores.

  113. Actually, we do get a surprise today. Ars Technica says there will be an iPhone 4S, except it calls it a 5S, with a ship date of September

    Rumors are not surprises.

    Also, I think you misread that report a bit – “iPhone 5s” there in my reading is a plural of “iPhone 5”, not a product called “iPhone 5S”.

    Which makes sense, since the other data point for iPhone naming suggests that “S” is a suffix attached to an update of an existing phone with the same number (3G -> 3GS).

    Releasing the 5S is plausible – a revision after the 5 comes out, if they don’t want to call it a 6.

  114. @Phil:

    It’s just a data point.

    But a minute ago, it came loaded with implications:

    Doesn’t this contradict the claims that…

    But never mind that.

    Don’t forget you got on the “Thunderbolt is outselling the iPhone on Verizon” bandwagon when someone surveyed a set of Verizon stores.

    Yes, comparable handsets at the retail store. This was countering the meme that “no single Android phone can touch Apple anything, ever, in any kind of setting.” It’s interesting that you pointed out at the time that those stores weren’t the entire market.

    The thing is, those stores probably represent a market segment. Not the Apple die-hards, and maybe not the WalMart/internet cheap shoppers. So I thought it was very interesting when we had the first real evidence that when a salesman shows people several handsets, in a lot of stores, people walk out with one that is comparably prices to the iPhone but not an iPhone.

    I’m not sure how that translates to the market as a whole, but I’m very suspicious about thinking that the truly low cost segment goes to a carrier store and subjects themselves to the salesman’s pressure.

  115. @esr

    Like patrick said, being a T-Mobile phone, the HTC Sensation isn’t an LTE phone. Also the review states that the 8.3 hours talk and the 12 days standby is the claimed battery life, and wile they say it had a “remarkably stable and long battery life” they don’t provide any actual numbers for that statement.

    Interestingly, unless it’s some typos HTC’s website says the phone is 3G only http://www.htc.com/www/product/sensation/specification.html

    Is this one of these “not really 4g but we’re going to call it that because it’s a good buzz word type deals?

  116. BTW, according to wikipedia, even LTE is’t true 4G. At a glance, 4G seems to be one of those standards, like HTML5, that keeps evolving to be just a bit better than what’s out there for a really long time.

    http://en.wikipedia.org/wiki/4G

  117. News has it that 130 million android devices were sold and daily activations are at 550k / day.

    Apple claims 200M iOS devices sold. Probably means therr are currently more android than iPhones.

  118. By the way, I find it fitting that Eric used an unauthorized clone of Tetris when discussing why smartphones are so great, as Tetris is nicely symbolic of the Android situation. You see, a broad passel of IP rights relating to Tetris are asserted by The Tetris Company, LLC. While some of these rights — such as copyright over the concept of any falling-tetromino game — may be invalid, others — such as trademark of the name “Tetris” (possibly extending to anything ending in “-tris”) and trademark of the use of the Russian folk song “Korobeiniki” in a video game context — are certainly enforceable against DroidTris. And unenforceability of copyright over a game concept has not prevented The Tetris Company from convincing the ITC to ban the import of Tetris games into the U.S. on precisely those grounds.

    The way IP law works in the United States is like this: to the biggest asshole go the spoils.

    Android’s day in the sun is over. Google is up against way bigger assholes than they. And the shit hasn’t even hit the fan yet — wait till the Nortel patents start getting enforced. Apple and Microsoft may be long-time enemies, even in the smartphone race, but now they have the means — and the will — to work together to shut the biggest player out of the market entirely.

  119. @Jeff Read:

    Certainly, things are heating up and cases can move faster at the ITC than in regular court. Of course, that’s part of why HTC just bought S3 for $300 million — an ITC judge has already ruled that iEverythings infringe on some of S3’s patents.

    And that will be appealed, just like the ITC judge’s ruling that HTC products infringe Apple’s patents. And at some point, there will probably be cross-licensing.

    But you probably wouldn’t know that side of it if you get all your news from the anti-Android FUD people.

    > The way IP law works in the United States is like this: to the biggest asshole go the spoils.

    Sorry, but even if that’s true, there’s a whole world out there that doesn’t work like that that google has their eye on. And, despite where they build the stuff, Motorola and Barnes & Noble and Amazon and Dell and Viewsonic and… are all considered domestic corporations, and someone in that club might have enough clout to keep Obama from agreeing to any ITC order. Even if the patents are valid, Motorola and Dell and a few others certainly have the manufacturing clout to import non-infringing components and assemble them into infringing devices, and other companies will ramp up to do the same. So the ITC game is merely a stop-gap measure.

    BTW, the two Apple patents discussed in the article you referenced are for software. The workaround for any ITC action might be as simple as not flashing the device until it enters the United States.

    Bottom line: there will be Android tablets and phones available in the States.

  120. I predicted that iPhone going multi-carrier in the USA would be the high-water mark for Android, which prompted quite a bit of derision from ESR and a few others here. It looks like Android may have continued to gain market share for a month or two after that point, but I’m convinced that its market share in phones relative to iOS has peaked. RIM is dying, and WP7 is DOA.

    As for tablets, I’m talking to customers every week who have major iPad projects they want to do, and none of them have shown any interest in the knock-offs. For Android to get any traction at all in the tablet space, they’ve got to get over this idea that they can charge as much for a knock-off as Apple charges for the genuine article.

  121. And then we observe the next chapter of an esr prediction unfolding:

    Nokia ‘giving away phones at cost’
    With margins so thin you need a microscope to see them

    Nokia’s financial results for the second quarter of 2011 are due tomorrow, and the company has already warned investors of very bad news coming. Yesterday, it issued a peek into just how tough things have got in 2011. Nokia said its smartphone profit margins were down to 6.2 per cent in Q1 2011, with margins of 16.4 per cent on basic phones.

    In other words, Nokia’s bargain basement models, sold to emerging markets and typically making use of very old technology, make it more money than its premium “flagship” models which boast its “state of the art” features.

    “There are no very big cuts per model, but the scale – across the portfolio – is unseen for a very, very long time,” Reuters quotes one unnamed source at a European operator as saying.

    http://www.theregister.co.uk/2011/07/20/nokia_margins_need_microscope/

  122. Some here predicted that Microsoft would spend dollars supporting WM7 in the market, Eric said they were wrong.

    Watch what happens.

    I don’t think I’ve seen a bigger Android fanboy than Eric.
    His market predictive misses are legendary!

    1. >Some here predicted that Microsoft would spend dollars supporting WM7 in the market, Eric said they were wrong.

      No, I believe I said that doing so would be a stupid move for Microsoft without excluding the possibility that they’d do it anyway.

      Here’s a prediction I’d like to be wrong about but probably won’t be: the next post from “Life as we know it” will be tendentious drivel, too.

  123. @Life as we know it
    “Some here predicted that Microsoft would spend dollars supporting WM7 in the market, Eric said they were wrong. Watch what happens”

    I was one such. And eric said it would be wrong for MS to fund WP7 sales.

    We see MS spending loads of money subsidizing Bing and WP7 sales. And we see people calling for Ballmer’s head.

    I think Ballmer will be ousted early 2012 the latest. And MS starting to fall apart.

  124. @Life as you think you know it:

    Given Apple’s announced results, esr’s predictions are laughable.

    What? Android is taking over. That’s obvious.

    @Some guy:

    I predicted that iPhone going multi-carrier in the USA would be the high-water mark for Android, which prompted quite a bit of derision from ESR and a few others here.

    Well deserved, I might add.

    It looks like Android may have continued to gain market share for a month or two after that point, but I’m convinced that its market share in phones relative to iOS has peaked.

    And still well-deserved. A single Android vendor sold roughly the same number of smartphones as Apple last quarter:

    http://www.mobilebusinessbriefing.com/article/samsung-ups-smartphone-shipments-forecast

    Granted, not all Samsung’s smartphone sales were Android, but I think we will find out in a week that the vast majority of them were.

    BTW, AT&T’s results are out, so we can see how Android is catching up there:

    https://plus.google.com/100357083629018071519/posts/cQhGM6xsNXV

  125. >Android is taking over. That’s obvious.

    You’re not paying attention. Android’s share versus iOS has peaked.

  126. Your link’s broken, Patrick — I think you wanted this:

    http://www.mobilebusinessbriefing.com/articles/samsung-ups-smartphone-shipments-forecast/4623/

    Also that link doesn’t say Samsung shipped 20 million smartphones last quarter. It says that one analyst believes they shipped 20 million smartphones last quarter, which is a somewhat different thing. Given that the article also quotes the Samsung exec in charge of phones as saying they’re likely to beat 60 million smartphones in the year, 20 million in one quarter seems like an overestimate.

    The interesting thing about this last quarter, though, is that the iPhone had a record month despite the fact that it’s been over a year since the last new model introduction. This is also the first time Apple has had five quarters of sales increase in a row. The Apple Q3 iPhone sales results are generally lower than holiday season sales — not this time. It may be time for Android fans to admit that the carrier gap is relevant to sales potential, given that potential reach is the one thing that’s really changed for the iPhone over the course of the last few quarters.

  127. @Bryant:

    Also that link doesn’t say Samsung shipped 20 million smartphones last quarter. It says that one analyst believes they shipped 20 million smartphones last quarter,

    Estimates I’ve seen range from 17M to above 19M. We’ll see at the end of next week. Even 17M is in the same ballpark.

    The interesting thing about this last quarter, though, is that the iPhone had a record month despite the fact that it’s been over a year since the last new model introduction. This is also the first time Apple has had five quarters of sales increase in a row.

    Yeah, it was pretty obvious they were going to decide the quarter needed a pick-me-up.

    It may be time for Android fans to admit that the carrier gap is relevant to sales potential, given that potential reach is the one thing that’s really changed for the iPhone over the course of the last few quarters.

    It may be time for Apple fans to admit that price mattters. It will be interesting to see how far down and for how long Apple goes.

  128. Verizon earnings are out. They don’t have the iPhone 3gs to sell at $49, so even though you would think AT&T was fairly saturated with iPhone-fondlers (and even though Verizon has more total subscribers than AT&T), Verizon only activated 2.3 M iPhones, to AT&T’s 3.6M:

    http://www.appleinsider.com/articles/11/07/22/att_activated_1_3m_more_iphones_than_rival_verizon_in_q2_2011.html

    Back when AT&T had the iPhone exclusive (and before they decided to give it up), they didn’t want to piss off Apple, so AT&T Android was practically non-existent. But last quarter, AT&T sold as many Android phones as Verizon sold iPhones — 2.3 million.

    And this, on the network that all of the “I have to have an iPhone” fanbois decamped to a very long time ago. (And they’re staying there, by all accounts — AT&T’s iPhone churn is still only around 1%.) So yes, loss of carrier exclusivity is relevant to sales potential, although that’s probably not exactly the lesson you were attempting to teach.

    Bottom line: Android is currently feeding on RIM at AT&T, and Apple is currently feeding on RIM at Verizon. They’ll start gnawing seriously on each other in the coming months.

  129. Dude, you can’t attribute the sales numbers to a 1 month US Wal-Mart deal. Especially since only 5.9 million of the 20.3 million iPhones were sold domestically. The international numbers increased too. Note that this also puts a bit of a hole in the theory that it’s all refurb 3GS sales from AT&T, although I’m sure that has some effect. 1.3 million worth of effect? Well, some of that is going to be upgrades…

    Which is another important point, btw. The effect of getting another carrier is not immediate. I’d expect it to be a rolling uptick over the course of a couple of years, as existing contracts expire. And yes, this goes for Android too — but Android has fewer carriers to expand to than iOS. AT&T may have been the last big one. Apple still has China.

    I agree that the real fight comes when RIM (and perhaps Nokia) has more or less zeroed out.

    I also agree that price matters. Tim Cook does too. “We know that we need to play there [the pre-paid market] in order to have the kinds volumes we’d like to have.” I’m not sure it really counts as a gotcha when the acting Apple CEO is saying the same thing.

    Oh, and a general note: I’m not trying to teach you anything. Sometimes I think it’s worth taking a shot at some honest discussion around here; usually I’m disappointed, but hope springs eternal. I think Apple’s sales numbers this quarter really are interesting, and I don’t pretend that I know why they did what they did. Talk to me, don’t talk to the imaginary “fanboi” you think I am.

  130. > Dude, you can’t attribute the sales numbers to a 1 month US Wal-Mart deal.

    No, but I can attribute it to the same thing done all over the world. If sales were up that high anyway, why would they bother?

    >>> It may be time for Android fans to admit that the carrier gap is relevant to sales potential,
    >> It may be time for Apple fans to admit that price mattters.
    > Talk to me, don’t talk to the imaginary “fanboi” you think I am.

    I do talk to you. Sometimes I even mirror what you’re saying.

  131. It’s no big deal.

    We’re both (I think) trying to simultaneously predict the future and influence the future, using imperfect data about the present.

  132. “The Noosphere is just another name for everyone’s inner troll.”. — Jaron Lanier

Leave a Reply to tmoney Cancel reply

Your email address will not be published. Required fields are marked *