The smartphone wars: a global perspective

While doing some research to check an assertion made by a commenter on More dispatches from the smartphone front I ran across excerpts from a very interesting report from investment analysts at Piper Jaffray on global smartphone market share and how they expect it to change in the next few years.

I can’t get at the full report, as Piper Jaffray has it paywalled and sells it to customers. But I have to say that for bankers the excerpts I can see make them look pretty smart to me. They’ve suggested a scenario for a radical change in market conditions that I actually think is quite plausible.

Before I get into that, let’s consider what they have to say about current conditions. For months I’ve been hearing a story from commenters overseas that my prediction of near-term and irreversible Android dominance was founded on a U.S.-centric view that ignores structural differences in foreign markets. According to this apologetic, Apple’s U.S. market-share crash since 4Q2009 is really a result of it being bundled with AT&T’s crappy network. Overseas, I was told, where unlocked phones are normal and carriers hawk SIM cards instead of tied hardware, Apple’s iOS is doing much better and still leads Android.

Of course, the prediction that goes with this that as soon as AT&T’s U.S. exclusive ends in 2012 and iOS goes multicarrier, we will all learn that every Android customer in the known universe really wanted an iPhone and Android’s market share will collapse like a pricked balloon. (Bear in mind this prediction has been enthusiastically seconded by the same fanboys that said Apple’s falloff in Q1 was due to customers holding off purchases until the iPhone4 came out and surely, surely Android’s market share would then collapse like a pricked balloon. Well, we know how that rosy fantasy died. iPhone4 didn’t even slow the rate of decline noticeably.)

Here’s another dose of cold hard reality: the belief that Apple is beating Android overseas is (according to Piper Jaffray) mythical. They’re expecting final market share figures for 2010 to show iOS up by just 1% at 15.9% to 14.9%. Backing this up is a report from Canalys that puts iOS’s 2Q2010 global share at 13% and, though it doesn’t give a figure, says the momentum is all with Android. According to IDC, Apple itself only claims 16%. And Gartner Research says that as of August Android actually leads at 17% to Apple’s 14.2%!

The picture is clear: (1) iOS and Android are now at or near a statistical dead heat in global market share, and (2) every analyst expects Android to gain dramatically in the near future. What I notice in the data that they’re not saying is this: the world seems only to be lagging smartphone-OS trends in the U.S. by four to six months, suggesting that carrier bundling in the U.S. is having relatively little effect on those trends.

This puzzled me when I first tried to understand it, but I think I get it now. The data only make sense if the rate at which U.S. consumers replace their phones (and move to Android) is substantially faster than their contract-expiration cycle. The typical contract term in the U.S. is two years; if the rates match we’d expect to see a trend lag of a half cycle, about twelve months. This is clearly not what’s happening. The U.S. market looks more like the (unbundled) worldwide smartphone market than one might expect.

I can only guess why U.S. customers are replacing their phones that fast, but the simplest hypothesis seems also the most likely: a truly massive move is on from dumb “feature” phones to smartphones (and Piper Jaffray’s report agrees with my guess). I think Android isn’t mainly stealing present customers from Apple; instead it’s doing hugely better among customers up-migrating from hardware like my old Samsung VT660. And I don’t think we need to look much further than price for a reason; Android phones are cheaper and Apple just isn’t justifying its price premium for most. I also think the most plausible model is one in which once users land on a smartphone their choice is fairly stable, though with erosion towards Android continuing at a low rate.

That’s what’s happening now. What comes next?

The consensus scenario among most of these analysts seems to be that Android clobbers the living snot out of iOS in 2011, with its growth being especially strong in Asia. Canalys is emphatic on this last point, and it makes sense; those customers are more price-sensitive than Americans. But Piper Jaffray thinks something even more dramatic is going to happen.

Piper Jaffray thinks that Nokia and RIM are hardware companies in their DNA, and are soon going to be forced to face up to the fact that they’re just not very good at software – not good enough to compete with Google, anyway. It’s expecting one or both of them to fold and jump to Android in 2011. The ZDNet report on the Piper Jaffray study says there’s already been one bloody internal battle at RIM about this.

On due consideration, I think this is plausible and that I really should have spotted this possibility before Piper Jaffray did. It’s following out the same pro-Android logic I did many months ago about software development being a cost sink for handset manufacturers that they’d best get rid of. Probably the only thing stopping Nokia is the amount of money and prestige they’ve sunk into Symbian, but it has to be dawning on them now that Android has Symbian seriously outgunned. (I actually thought the Nokia acquisition of Symbian smelled faintly of doom back in 2008 before Android was a real factor. I wasn’t sure why, there was just that odor about it to me.)

So the international question reduces to this: is Android going to swamp iOS quickly on more less smooth continuation of present trends, or are Nokia and RIM going to fold and shuffle their userbases into the Android column even faster? Not a lot of good news on the Rialto for Apple fans, and I suspect it’s going to get worse before it gets worse.

121 comments

  1. The US market is different than the international market in other ways. Apple is a stronger brand here, particularly when compared to Asia. I do think as soon as there is a Verizon iPhone, a certain percentage of the Android buyers will switch ASAP. I have, however, observed about half of my friends on Verizon that intended to get an Android device to “hold them over” wound up falling in love with Android. Who knows how big the fallout really is going to be.

    Android only holds price advantage on low-end devices. The problem is that the low-end Android devices aren’t very good. That’s going to change in a year or two, where Nexus One grade hardware is the baseline for “free” phones, and newer high-end phones get dual cortex A9 CPU’s and better GPU’s (Samsung Orion, Nvidia Tegra 2, TI OMAP 4), >= 1GiB RAM, and >300 dpi screens. Prices of high-end Android phones, on subsidy, are similar to the iPhone — all of them on Verizon are $200 (Droid 2, Droid Incredible, Droid X, Samsung Fascinate), on ATT and Sprint are >= $200 (Samsung Captivate $200, Dell Streak $300, HTC Evo $200, Samsung Epic $250). iPhone is at 3 price points: $100 for 3GS (low end), $200 for 16GB, $300 for 32GB. The cheapest Android phones are in the $100 range on the above carriers. So, for now, price is comparable.

    Even after iPhone is on Verizon, Android is going to win because there will be (eventually) good devices at the “free” level, and the fact that there are multiple form factors possible on Android. Samsung is also talking Android in TV’s. I could see Android being used in cars for control of stereo, GPS, etc. Android will proliferate in ways no one company or one person can predict. iOS can only do so in ways Apple can think of. Given, they are a very creative company, but they can’t beat everybody else in the world combined.

  2. There are virtually no high-end android handsets available in Europe. There is no GSM version of the DroidX, EVO 4G, or the EPIC. I am holding back purchasing an Android device until Samsung or HTC put something out that is worthy of the label iPhone 4 killer. Though I totally share ESR’s long-term perspective on Android, namely that it will eventually be the global dominant player, I think that Apple and Blackberry are still going very strong in Europe.

    Please give me a GSM version of the Epic 4G, DroidX or the Evo 4G and I’ll stop complaining. Most people don’t realize I guess that Motorola’s reputation is very weak in Europe and that most shops don’t even carry the brand anymore. This leaves you with Samsung and HTC to carry out the Android battle in Europe… and so far they have sitting on their asses.

  3. The only thing that worries me about your conclusion is that it is so dependent on that consensus among analysts. My internal cynic places about as much credibility on industry analysts as it does on CC Fundamentalists.

  4. Umm all this thing about non US carriers not selling contracts is only partially true. In some of Europe (certainly the UK, France, Germany and IIRC Switzerland) people regularly buy phone plus contract and here in France (I don’t have enough recent info about other markets) the carriers give you loyalty points that mean that once you’ve been a customer for a few years you can get your next phone for free if its an older model or heavily reduced if its a new sexy one.

    Now you can also buy unlocked phones too. But there are plenty of phones sold with a contract

    Nokia has, unfortunately, become a dysfunctional company in its handset division and no longer seems able to develop interesting stuff. I hope that this will change because there are a lot of smart people in Nokia and in Symbian and other companies that Nokia has bought recently (Qt) and their hardware is of excellent quality. My experience with an HTC Hero last year did not impress me with the HTC quality. The “smart” parts worked OK but the device kind of failed at the more basic “phone” parts – it felt plasticky and brittle so I was worried abotu dropping it and its camera was crap. Nokia on the other hand produce great hardware – battery life excellent, call quality great, none of this “death grip” thing to get the antenna to work etc. and so on. If Nokia do bite the bullet and go android they will almost certainly have the best android handsets around and android should get a huge lift from that. If I were google I would be seriously looking at ways to get Nokia to use Android (maybe an agreement to share stuff with the Intel/Nokia linux OS?)

  5. BTW I could certainly understand it if RIM adopt android. RIM’s OS and basic system services are buggy and do basic no nos like drop/corrupt packets under load leading to retransmits. Even better as I understand it, RIMs upper level API is basically Java so porting RIMs apps to Android should be relatively painless

  6. The smart phone take over rings true to me. Readership here is likely higher end, but we all have low-tech parents or non-tech friends that just want a phone. Have you seen the options for these users? They are awful. The Nokia brick of old no longer exists. A pity. Sometimes a phone just needs to be a phone.

    Once forced into the mid-tier, clunky, “feature phone” it’s a no brainer to spend $50 more for a computer in your hand with consistent UI and user control over apps.

    The deciding factor is then data plan costs. $35 a month more for internet on your phone is tough to swallow. Once taken, it’s hard to go back, but it is a significant transaction friction. If you could buy a WiFi only smart phone in the US, the uptake would swing exponential. My old Treo700p didn’t require a data plan an it was my red pill down the smart phone rabbit hole.

  7. To RIM or Nokia going Android, this makes me wonder who the third OS will be. Cringley made a point a while back that tech generally spirals down to threes: Windows, OS-X and the UNIXes for example. His prediction in mobile was RIM, iOs and Android (mapped to the above by my personal feelings rather than market share).

    Nobody I know wants a WinMobile phone. Palm? I can’t even remember who bought them after their HUGE marketing push a few months back.

    I prefer a three share market to a two share one so I hope Cringley is right in, at least, his assessment of shares. Interesting times.

  8. I think that corporate sales of Blackberry’s would have to crater before RIM would jump to Android. I suppose another possibility is they could launch a “non-corporate” handset and use Android on that in an effort to supplement their corporate domination. Still, I think that RIM’s value-add of having a reputation for being able to integrate a phone into a corporate setting means they won’t be going anywhere soon.

    As for Nokia, they’ve just switched to that MeeGo OS and have yet to release any phones with it. Supposedly there’s an N9 coming that looks interesting- though still no word on a release date. Given that they’ve partnered with Intel on MeeGo, I think they’re going to stick with their current path for the time being. I’ll hedge by stating it’s possible that internally they already have misgivings about MeeGo, but I haven’t seen any rumblings of that nature.

    So I’ll go with Android continuing on it’s current trend. I think if Nokia or RIM switch, the dust will have long settled in favor of Android.

  9. Eric,

    Quick comment about the “cycle”. You commented on the cycle being 2 years (contract length), but MOST carriers will give you a good deal (if not the same deal) on another phone half way through the contract if you’ll re-up for another 2 year contract. I know tons of people who do this (in fact my wife just did it). So, the cycle is effectively 1 year on phone refreshes, and your half-cycle then is the 6-months that matches your observation.

  10. Gerry: Sprint is bundling MoxyMail with it’s Android phones, and this app supports MS Exchanges PUSH mail service, so it effectively replaces BES and kills RIM’s advantage in the corporate email space. My (small) company is currently moving to an all Android 2.2 phone base for the technical people because we have need for connectivity everywhere and the wireless AP function in the 2.2 devices saves us money on not having to buy wireless cards and pay for that additional service.

    If Sprint is an acceptable network for you, then RIM has no further advantages beyond momentum.

  11. aczarnowski> … this makes me wonder who the third OS will be.

    Likely WebOS. With the HP purchase, I think it’s going to get at least a breath if new life and that plus a modern core OS, puts it in the running with iOS and Android. It’s the only logical option I see, unless MS wants to sink a couple hundred million into WinMo to try to bring it back from the brink… again.

  12. “If Sprint is an acceptable network for you, then RIM has no further advantages beyond momentum.”

    Sprint, T-Mobile and ClearSomething seem to be credible players in the big metro markets. But for the other 99% of the geography of the US it’s either AT&T or Verizon or nothing. Like Democrats/Republicans or DSL/Cable, it seems to be a law of markets that two supposed competitors are essentially the equivalent of a monopoly … that only with an introduction of a third+ player does actual competition begin. Someone should analyze it and explain why.

  13. Michael,

    Someone should analyze it and explain why.

    I believe Schumpeter has done just that, with his theory of imperfect competition.

  14. MS *is* sinking quite a bit of money into Windows Mobile right now, and say they’ll launch the new version in February. Their new version features a good deal of eye candy and multi-touch and better APIs and all that, and the specimens I’ve seen look pretty neat, though I can’t say how it really works day to day. But even if it is good, I kind of doubt anyone will care–the ship has sailed on mobile OSes and they missed it (esp. by Feb), and their record in mobile (Zune anyone?) is pretty dismal.

    But: they’re going to try, and MS is certainly a persistent company. Maybe they’ll capture a few percent share.

  15. I don’t believe Windows Mobile (or any of its past names) has ever held a significant market — for a long time it was pretty much thrashed by BlackBerry, and since iPhone and Android have appeared, it has pretty much disappeared. I know of a couple “trendy” Windows Mobile phones launched in the past year that have totally been discontinued from the market — nobody wanted them. Why would they? It’s an unstable mobile OS (think Windows 98/Me levels of stability), difficult to develop for (they pretty much incorrectly made it almost like writing Win32 applications, nevermind that Win32 is a terrible API; also last time I looked, the free versions of Visual Studio didn’t support it and MinGW-CE was broken for the latest versions of Windows Mobile), difficult to use (the UI is probably on par with BlackBerry, which is not a good thing), and now practically unsupported amongst any of the hobbyist developers now going for iPhone and Android.

    I believe melee already said it correctly, Microsoft missed the boat on mobile OSes and will probably never “recover” (the word recover used somewhat loosely, as it normally implies they had a strong position in the first place).

  16. Probably the only thing stopping Nokia is the amount of money and prestige they’ve sunk into Symbian, but it has to be dawning on them now that Android has Symbian seriously outgunned. (I actually thought the Nokia acquisition of Symbian smelled faintly of doom back in 2008 before Android was a real factor. I wasn’t sure why, there was just that odor about it to me.)

    It’s not just prestige. They don’t want to lose control over the software. Nokia absolutely does not want to become just a hardware manufacturer. Even Jorma Ollila stated many years ago that his job was to turn Nokia into a software company, and he was the previous CEO. I think Kallasvuo has been keenly aware of this from the start of his term. For whatever reason, Nokia’s software seems to always be lackluster even if the managers say that it’s a priority.

    A lot of developers inside Nokia have thought that Symbian was dangerously outdated for many years. The purchase did smell of doom; it was Nokia’s move to make sure that they could keep Symbian alive, because they were dependent on it at a time when the other users of it were doing very badly. Nokia was just too slow in getting onto Linux. They had Maemo and the N770 in 2005 already, but they should have been pushing it aggressively onto the phones and not just internet tablets. Maemo was a small project inside Nokia and apparently it didn’t have that much of an effect on the phone-making for a long time. Now they’re scrambling to get MeeGo together. It’s very late, in some respects, but if anyone can pull off another phone OS at this point, it would be Nokia. On the other hand, the whole smartphone thing is still new and another Android-size disruption might come along at some point.

  17. Reporting from the Vienna subway: the young and hip all have iPhones, it is as much a fashion or coolness must-have for the disco-going types as hair gel, (and this sort of hipness is unfortunately popular here, my local supermarket began selling Ed Hardy _wine glasses_ fer crying out loud…). Even if it is two generations old an has a broken display, but the brand name must be iPhone.

    Actually it is surprisingly often two generations old and broken… could it be that these are mostly lower-class folks who are typically imitating the middle-class fashion N years ago? There are some signs that seem to point into this direction. Middle and upper class fashionism tends to focus less on brand name and more on quality: f.e. for clothes anything from any designer brand will do if it looks good. Lower-class imitation tends to be strictly brand-name based: it may be anything, it may look anyhow, but it must have DKNY or D&G printed on it in large letters. The stuff I see happening with iPhones here seems to be tending towards this direction.

    Oh and I never ever saw anyone owning an iPad here. People play around with it in the electronics stop, say that it’s real cool and everything and then go on buying a netbook instead.

    Folks over 30-35 and middle-classish tend not to have iPhones, whatever else they have is a bit hard to tell without invading one’s private sphere but it is phones big, smart-looking and having modern conductive touchscreens, so it is most likely either Android or Symbian. Probably more often Symbian, because since about 1999 and the lovely, sexy, cult-creating 3210 Nokia is the No. 1 brand in the CEE region.

    Convince Nokia to adopt the Android and you instantly won over every place between, say, Hamburg and Athens. Given that Nokia is a Finnish company… hmmm… Android -> Linux -> Linus… it might not be impossibly hard. Chances are they have many influential tech leads who are really into Linux.

  18. Shenpen:

    Remember that Nokia/Intel’s Meego is also a Linux kernel, and in fact the userspace looks a lot more like a normal distribution than Android does. It’s probably also secure against the whole Oracle/Java lawsuit fiasco, if that’s worth anything.

  19. Michael,

    “There are virtually no high-end android handsets available in Europe.”

    Actually I’m considering getting either the HTC Legend or the Samsung Galaxy S offered by A1.net – could you point me to any online article describing in what way they are inferior to the high-end Android products or the iPhone? I tried grilling the sales clerk but you know… I just got the usual low signal-to-bullshit ratio sales speech.

    Personally… maybe I am wrong but I think I’m living in an age where online apps can do almost anything one needs to get done, so exactly what killer features can a smartphone have which is not substituted adequately by a good browser and cheap enough internet connection? I don’t find it convincing that once we are talking about phones we must go back to the nineties and focus on “desktop” apps instead of web apps… of course there can always be some special cases.

  20. “Like Democrats/Republicans or DSL/Cable, it seems to be a law of markets that two supposed competitors are essentially the equivalent of a monopoly … that only with an introduction of a third+ player does actual competition begin. Someone should analyze it and explain why.”

    One part of the explanation could be “choosing against” or “voting against”. Look at which one sucks more and choose the other one – rational for the consumer/voter, but if both competitors suck pretty much X will pick up the people disappointed with Y, Y will pick up the people disappointed with X.

    To put it in other words, it is pretty much part of human nature to see a group of people or an organization as the bad one, th enemy, the one I am against, and in a duopolic market X is monopolistic with regard to those who make it a point to hate Y.

    In a triopoly things change, because X and Z must actually compete for all those people who hate Y. But isn’t it the same problem one level higher: X being a monopoly with regard to people who hate both Y and Z? Well, no, this is something about human nature: we tend to choose one enemy of one kind, and focus on it, and see the other ones as at worst uncomfortable allies, who if are more than one, must compete.

  21. I am betting on Meego. It is a superior architecture though it is a long way behind.

    I wouldn’t. OTOH, I wouldn’t necessarily bet against it, either. The main trouble with Meego is that, at the moment, no one can buy a Meego device in the general market.

  22. One part of the explanation could be “choosing against” or “voting against”.

    Ah, right. I should have cited Albert Hirschman instead of Schumpeter. He illustrates the point in Exit, Voice and Loyalty by talking about people who switch from one car company to another, when all of them are of similarly poor quality.

    Theoretically, the same condition of low competition could hold with three or even more actors, it just becomes less stable as the number of actors grows.

  23. > I cited figures from three different market analyses (every one I could find).

    Lets see the statistics.

    Because what you didn’t quote was (in part) this:

    Now it’s not all about market share. Apple will command half the profits of the smartphone industry.

    and (quoting again):
    For Piper Jaffray’s projections to play out there is one significant wild card: Nokia and RIM have to cave and go to Android.

    So *both* Nokia and RIM have to switch to Android for the PJ prediction of 50% market penetration to come true, and Apple would still take fully *half* the profits in this space.

    The linux ‘community’ are all wired into a survival trip now. No more of the speed that fueled that 00’s. That was the fatal flaw your trip. You crashed around America selling “open source” without ever giving a thought to the grim meat-hook realities that were lying in wait for all the people who took you seriously… All those pathetically eager acid freaks who thought they could buy Freedom and Prosperity for three bucks and a copy of the Perl ‘Camel’ book. But you loss and failure is ours too. What you will take down with you is the central illusion of a whole life-style that you helped create… a generation of permanent cripples, failed seekers, who will never understand the essential old-mystic fallacy of the Hacker Culture: the desperate assumption that somebody will ignore the body odor and allow them to breed.

    A man like you who has blown all his options can’t afford the luxury of changing his ways. He has to capitalize on whatever he has left, and he can’t afford to admit — no matter how often he’s reminded of it — that every day of his life takes him farther and farther down a blind alley… Very few toads in this world are Prince Charmings in disguise. Most are simply toads… and they are going to stay that way… Toads don’t make laws or change any basic structures, but one or two rooty insights can work powerful changes in the way they get through life. A toad who believes he got a raw deal before he even knew who was dealing will usually be sympathetic to the mean, vindictive ignorance that colors this blog.

  24. Symbian will continue for a while. It does something the iPhone and Android still can’t; it has an app that can record your phone calls. Without third party hardware or call forwarding services. For business people taking orders, or people being harassed, this is a must-have feature. The Android bug report has been open for ages. And there are some theories as to why it isn’t getting fixed, but no real answers from the Android team. Some say the firmware of a lot of android phones doesn’t support it. Eric, can you publicise this issue?

  25. Some Other Guy,

    Very evocative, but I still don’t see your specific complaint. Is it that there are less profits in open-source, and therefore less salary for coders?

    If so, you are focusing on the wrong issue. Profits are only one measure of power. And salary, after a certain point, is less important than a meaningful life under your own command.

  26. Where are all these “fanboys” making those predictions, Eric?

    I’ve watched the smartphone threads here reasonably well, and the pro-Apple (or I think more accurately anti-anti-Apple) commenters here… don’t seem to have made those predictions about Android’s marketshare being temporary or a fluke or about to burst.

    (Maybe they did and I missed it? I suppose that’s possible, though I usually read pretty well. Certainly none of the ones I noticed, let alone the comments I made myself, suggested any of those points.)

    Is that just a convenient pseudo-strawman (aka the famous MacMac)? Or, granted that some idiot somewhere on the interwebs really did predict that, does it matter? All sorts of idiots predict all sorts of stupid things… but that doesn’t make ’em relevant or worth even mentioning.

    Seriously, can’t Android do well without having to inflate an Apple-fan target to Take Down?

    (Can’t Android and iOS both be great successes? That sure looks like how it’s going down, and in line with everything I’ve said in all the threads about it here.

    There is no “winner take all” here, and without iOS Android seems like it’d be mired and bogged with nothing to really push against, and even more provider-trapped than it is now. Look at 2007-era Android screenshots for an example of what iOS phones did for it.

    Likewise, Android gives Apple real competition in a way RIM and Symbian aren’t very good at providing.

    Seems a lot more like synergy than anything else, and good for consumers all around.

    I guarantee you that if all there was was Android we’d see utter stagnation again as the carriers (who are in practice in control of the Android Experience for something rounding to 100% of Android users) turned it into a prettier and slightly more useful version of Exactly The Same Phone.

    iOS is not “the enemy” of phone innovation and choice. The carriers, with their lockdowns and bundles are.*

    *Yes, iOS devices are all “locked”. But at the same time they all actually have upgrade paths, and everyone buying one is AWARE it’s locked. Not so much with Android phones that aren’t the now-discontinued Nexus. Likewise, Apple does bundle things… but now they’re all easily hidden, if not removed, and at least they’re not active pains in the ass, random third party “demos”, or other horrible vileness.)

    1. >iOS is not “the enemy” of phone innovation and choice.

      Why yes, yes it is, as long as it’s not open source.

      I look forward to iOS failing because that will put another nail in the coffin of closed-source development. It’s not really about Apple or about Google, it’s about whether users will actually have control of everything in their device. Including the software.

  27. Sigivald:
    > There is no “winner take all” here

    I would posit that, yeah, there kind of is. If Apple takes majority control of the market, such that the phone that ‘everybody’ has becomes some version of an iPhone (or some device running iOS), then we end up in a heinously vendor-locked environment wherein users are required to pay a not-all-that-small toll just to install apps of their own choice.

    I for one don’t want that to happen. I don’t even personally care for Android much, in its current state of vendor appropriation. Honestly I’d be much happier with a world where none of the various products take majority control.

  28. “I am betting on Meego. It is a superior architecture though it is a long way behind.”

    Betting on it to do what, be the next VisiCalc?

    1. >Betting on it [MeeGo] do what, be the next VisiCalc?

      Sadly, I have to concur that I doubt MeeGo is going anywhere. In an ideal world I’d prefer it to Android – open source, and closer to a stock Linux distribution. I just don’t see it having enough win over Android to make up for Google’s branding and financial mass. MeeGo folks, I’d love to be wrong about this. Make me look stupid for not believing. Please!

  29. >Actually it is surprisingly often two generations old and broken… could it be that these are mostly lower-class folks who
    >are typically imitating the middle-class fashion N years ago?

    It’s probably a lot simpler than that. Even a current Android phone offers very little in terms of improvement over the 1st gen iPhone and until the recent iOS 4, newer iPhones didn’t offer much more unless you were a “latest and greatest” freak. I’ve looked and given honest tries, I really have. I have no contract tying me to my carrier and only a preference to remain on a GSM network. Yet all of the android offerings I’ve seen are at best as good as my current, old, worn out 1st gen iPhone. And that to me is truly a saddening discovery.

  30. >>iOS is not “the enemy” of phone innovation and choice.

    >Why yes, yes it is, as long as it’s not open source.

    So, “open source” is the engine of innovation, eh? Why is Android such a pallid imitation of the iPhone then? Why are the various GUIs on X windows such shabby knock-offs of NeXTSTEP and MS Windows? Why is Open Office the poster child for hideous amateur efforts that can’t even match Microsoft for ease of use or functionality?

    I don’t suppose you have anything to say about the way that GCC basically killed off a vibrant business in compilers, which had dozens of vendors actively advancing the state of the art? I remember Aztec C, Lightspeed C, Mix C, Turbo Pascal, back before we got tipped into about a decade and a half of stagnation in compiler development. Thank goodness Apple got sick of the FSF and decided to sponsor the LLVM project.

    BTW, “Some Other Guy”: maybe we should just take numbers? How about I start as “Some Guy 42”?

  31. Seriously, can’t Android do well without having to inflate an Apple-fan target to Take Down?

    If Android had something worthwhile to offer, it could. Trouble is, it’s crap, so all they can muster to drive any interest is to beat the “apple is eeeevil!” drum.

  32. > So, “open source” is the engine of innovation, eh? Why is Android such a pallid imitation of the iPhone then? Why are the various GUIs on X windows such shabby knock-offs of NeXTSTEP and MS Windows? Why is Open Office the poster child for hideous amateur efforts that can’t even match Microsoft for ease of use or functionality?

    > I don’t suppose you have anything to say about the way that GCC basically killed off a vibrant business in compilers, which had dozens of vendors actively advancing the state of the art? I remember Aztec C, Lightspeed C, Mix C, Turbo Pascal, back before we got tipped into about a decade and a half of stagnation in compiler development. Thank goodness Apple got sick of the FSF and decided to sponsor the LLVM project.

    So, which is it?
    Is OSS so inept that it can’t compete with commercial offerings or is it evil because it so successfully competes with commercial offerings?

    1. >Is OSS so inept that it can’t compete with commercial offerings or is it evil because it so successfully competes with commercial offerings?

      Both. Open-source methodology is so powerful that projects using it routinely out-innovate closed-source competition in any software area that is neither (a) heavily dependent on UI design, nor (b) so locked down by patents that open source basically can’t play there. And our weakness in UI isn’t a consequence of open-source methodology but of cultural/historical baggage. I’ve done my share and more to try to fix that, and at least we can muster minimal competence on most projects these days.

      Even after admitting that weakness, though, claiming that Android has a crap UI is just nuts. It’s a very good fit to function, and non-technical end users are happily adopting it in droves.

  33. Yet all of the android offerings I’ve seen are at best as good as my current, old, worn out 1st gen iPhone.

    Depends on what you mean by “as good as”. Just staying in the iphone world(to compare apples to apples so to speak), at the very least the 3GS is a 150% performance upgrade(on cpu alone) compared to the original iPhone. This means that apps built to run on a 3GS will either not run or (much more likely based on anecdotal evidence) run much more slowly.

    To put this into another perspective, it would be like saying that the low end release of the pentium 3 chipset was as good as the high end fully clocked release. While in terms of potential features the two chips were identical, in terms of what you could actually run in the day and (more importantly) how well you could run them, the difference was night and day. People voided warranties regularly for less than a 150% improvement in processor speed.

    You’d have to find someone different to talk about actual feature upgrades, but wikipedia lists the features that Iphone3G added as “Assisted GPS, 3G data, and Quad-band UMTS/HSDPA” which to me is a compelling feature set.

  34. Open Source software does a number of things brilliantly. It’s designed around peer review processes.

    The drawback of this methodology is that the only things that will be found are the things the peer reviewers find important enough to fix…and user interfaces tend to be low on the list, as are usability studies.

    There are exceptions to this – but most of the ones that show up are ones like Google or Canonical, who realize that OSs aren’t religions; the OS should largely fade into the background and provide a consistent user experience.

    And there are design decisions made in the core of Unixes – for sound reasons – that mess up UI experiences.

  35. Is OSS so inept that it can’t compete with commercial offerings

    I didn’t say it couldn’t compete. If you’re giving away McDonald’s burgers, you’re going to make a serious dent in the local restaurant market. That doesn’t mean that that what you’re offering is any more than barely adequate. It’s kind of like public schooling.

    In the case of GCC, it basically sucked all the air out of the room for the compiler vendors, so we had a Dark Age of compilers until LLVM came along.

    Even after admitting that weakness, though, claiming that Android has a crap UI is just nuts.

    You’ve just demonstrated your lack of taste.

  36. claiming that Android has a crap UI is just nuts.

    I think the problem that people talking about iOS vs Android UIs don’t get (and indeed most fandom reguarding UIs in general don’t get) is that saying UI A is better than UI B is much the same kind of argument as Programming Language A is better than Programming Language B but without the objective measure of “feature sets” to separate them. Whereas we can objectively say that Lisp is better than VB because Lisp has a richer feature set, GUI “feature sets” are highly subjective.

    In a UI, any given feature is probably either illusory (e.g. I once had a constructive(i.e. not an internet argument of two people talking past each other) discussion about using multiple programs at once on both MacOS and Windows where we both thought the other OS was poorly designed for doing multiple things at once. The obvious implication being we were both wrong, they were designed to do multiple things at once in different, non-compatible ways) or highly subjective (e.g. i’m sure someone out there likes the dock menu and it’s zoom feature, i think it’s a poor substitute for the task bar on Windows XP with entirely the wrong informational cues). The thing is we become use to the idiosyncrasies of our chosen interface and any other interface decision is _by definition_ poorly designed (for us).

    There are some things which have become universals, like pressing enter/return at the end if you’re filling in a form should submit the form(though it wasn’t necessarily always so). There are also some good rules of thumb which you should only break if you’ve got a good reason (e.g. always allow undo, preferably arbitrarily long). Anything else is vanity.

  37. @ESR: I copletely agree with your argument about the superiority of Open Source. I just don’t agree with your prediction that Android is going to eat Apple’s lunch.
    Success in the marketplace is rarely based on technical superiority. See Windows as a case in point. What matters is applications. Functionality.
    Currently there are about 6 iPhone users in Europe for every Android user. They are not going to switch if that means losing some of their favorite Apps.
    Also you’re missing the elephant in the room: Symbian. Currently owns about half the smartphone market, and is Open Source too. Symbian users aren’t likely to jump ship (and lose their free mapping and navigation app) too.

    1. >Currently there are about 6 iPhone users in Europe for every Android user.

      [citation needed]

      Every source I can find says the userbases are at parity. If you have evidence otherwise, I’d like to hear it.

  38. You’ve just demonstrated your lack of taste.

    The operating system can be as well designed as can be, but it would be a total flop in the mass market if it has a crap UI. Why then, is Android not flopping (rather the opposite, everybody else is)?

    I’ve got an Android phone now and it is wonderful. Please tell me where the problem areas are so I can see them for myself. :-)

  39. I didn’t say it couldn’t compete.

    If you’re giving away McDonald’s burgers, you’re going to make a serious dent in the local restaurant market.

    Or in other words, “If you’re giving away McDonald’s burgers, it’s hard for local restaurants to compete”.

    You’ve just demonstrated your lack of taste.

    You prefer your home screen to be reminiscent to a prettier version of windows 3.1 run with program manager maximised to something that is uncomfortably similar to the dashboard on osX? That would make me question your sanity.

  40. In the case of GCC, it basically sucked all the air out of the room for the compiler vendors, so we had a Dark Age of compilers until LLVM came along.

    When, exactly, was this dark age that you talk about? There have been who knows how many widely used competitive compilers all along. See Wikipedia’s list. If proprietary development is so wonderful, I’m sure that Microsoft or Sun with their resources were perfectly capable of making a superior compiler, air or not. Would you like to complain about Linux sucking the life out of the Unix vendors next?

  41. According to Comscore the figures were 9,4 million iPhone users, and 1,6 million Android users (and 33 million Symbian users) in the 5 biggest European countries combined, per end of march 2010. I don’t know what the figures are right now, but Id’ be very surprised if Android had come even.
    The comscore site is not direct link friendly, and most of their data is only accessible via secondary sources, like this one:
    http://www.allaboutsymbian.com/news/item/11342_comScore_data_shows_smartphone.php
    This data shows how big the gulf is between Europe and the US. Symbian is virtually non-existent in the US, but commands more than half the market in Europe. That is why I think that trying to extrapolate the global future of a particular Smartphone ecosystem based on what you see in the US is a bit risky.

    Comscore is a company that measures phone usage by actually mining the phone companies own traffic data. They periodically publishes a “state of the mobile EU”, which you have to buy, but you can look at a free presentation that gives some data:
    Just go here and download the slides (figures are on the third last slide):
    http://www.comscore.com/layout/set/popup/Press_Events/Presentations_Whitepapers/2010/The_State_of_Mobile_EU5_Mobile_Media_Landscape_and_Trends

    1. >I don’t know what the figures are right now, but Id’ be very surprised if Android had come even.

      Well, Canalys was recording 866% Android growth over a single quarter in 2010, so I think I at least see a consistent set of claims here. Assuming nobody’s just pulling numbers out of their butt, all the growth has been in the last four months. Your Comscore info is correct but out of date.

  42. > Actually I’m considering getting either the HTC Legend or the Samsung Galaxy S offered by A1.net – could you point me to any online article describing in what way they are inferior to the high-end Android products or the iPhone? I tried grilling the sales clerk but you know… I just got the usual low signal-to-bullshit ratio sales speech.

    I personally think the Samsung Galaxy S is the best Android handset you can buy in Europe at the moment. Though it ain’t exactly fair to compare the two since the iPhone is twice as expensive as the Galaxy S, let me do it nonetheless. The iPhone 4G offers sturdy build quality, an excellent highres display, a great camera and perhaps most importantly: a very, very smooth user experience. Check this for yourself at your local T-Mobile shop. Then again, the Samsung has a more plastic feel to it, the display is less, and the user experience is slightly limited because Samsung puts “touch whizz” on top of Android, and finally the camera is less great.

    Admittedly, I am being highly critical here and the Samsung is not a bad phone, especially considering it’s price point. If you like Android and are willing to flash roms this could perhaps be the perfect phone. I wouldn’t want to consider the Legend because of it’s outdated specs, there’s probably something in the pipeline with HTC (think the quad-band slider as seen on Engadget) so I’d just wait a bit if like HTC.

    http://www.engadget.com/2010/06/24/iphone-4-retina-display-vs-galaxy-s-super-amoled-fight/

    1. >Though it ain’t exactly fair to compare the two since the iPhone is twice as expensive as the Galaxy S

      And that price difference would neatly explain the crazy worldwide growth rates Canalys reports.

      It’s all fitting together pretty well, I think. The iPhone is being savagely undercut on price, and a lot of people with better things to spend money on than a status toy have gone Android in just the last four months. Expect this to continue.

  43. @shenpen:

    If you are living in Vienna, maybe you’re interested in a cup of coffee sometime?

  44. Some Guy:
    > I didn’t say it couldn’t compete. If you’re giving away McDonald’s burgers, you’re going to make a serious dent in the local restaurant market. That doesn’t mean that that what you’re offering is any more than barely adequate. It’s kind of like public schooling.

    > In the case of GCC, it basically sucked all the air out of the room for the compiler vendors, so we had a Dark Age of compilers until LLVM came along.

    In my city there are soup kitchens giving away plenty of McDonald’s quality burgers and the the restaurants seem to be doing just fine. :)

    I haven’t worked much with C/C++ compilers over the last 15 years but in the arena where I do work (web application development) I’m seeing the exact opposite of your argument. Many of the companies I’ve worked for or consulted with are paying large sums of money for commercial app servers and database management systems when there are perfectly good open source ones like Jboss, Tomcat, Resin, and Jetty available. When asked why they will say it’s for the support, security, scalability, features, etc. When presented with numbers or tests that show the open source ones are as capable of performing the task at hand (and often better) their answers become more vague. What often turns out to be the case is that the people choosing the technology are not the ones paying for it and they’re are much more interested in loading up their own resumes with names like Oracle, WebSphere, WebLogic, and MS-SQL Server than with choosing the best tool for the job.

    After seeing this, over and over, I find it very hard to believe (especially with the price of C/C++ developers today) that companies have been dropping good and innovative compilers and forcing their developers to use something barley adequate to get the job done just to save a few bucks on compiler fees.

    BTW: This trend seems to be turning around. In the last few years I’m seeing more and more companies looking for people who have experience with open source products and languages.

  45. Re: UIs:

    The benchmark for a UI is not “well, it ‘feels’ more polished” or “it ‘seems’ more integrated” or whatever other non-quantifiable rubbish I’ve seen coming from fanboys. The benchmark for a UI is “Can a non-technical user figure out how to work this thing without reading a damned thing about it or getting any prodding from a more experienced user?” Another one (which is often ignored by Microsoft) is “Does the UI get in the way of getting stuff done?”

    If the answers aren’t “yes” and “no”, respectively, the UI is crap.

    1. >If the answers aren’t “yes” and “no”, respectively, the UI is crap

      And what makes dismissing the Android UI as crap ridiculous is that 866% worldwide growth figure. Lots and lot of ordinary end-users are picking up Android phones and finding them easy to use without reading a manual. That’s design success, right there; the UI is doing exactly what it’s supposed to.

  46. > Lots and lot of ordinary end-users are picking up Android phones and finding them easy to use without reading a manual.

    That was my experience back in May.

    I had promised (read, caved) to buy my wife an iPhone; though I was angling for an Android device since I want to develop apps for a smartphone and refuse to pay the Apple premium required to develop for their platforms. We waited over 2 months for AT&T to come out with a non-crippled Android device (“It will be here sometime, soon!”) with no luck. So, one day after lunch I took her to the local Verizon store so that I could play around with the Droid Incredible to see if it met my expectations (Yep). At this time, I was fully resolved to the prospect of our becoming a two-carrier family and willing to pay a premium to do so. But, she got bored watching me work one of the display devices and picked up the old model Motorola Droid device at the next booth over and started playing with it. I didn’t even realize that she had done so until she spoke these words “You know, this is just as easy to use as the iPhones that I’ve seen, plus it’s got a real keyboard. I think I like it better.”

    We eventually chose Sprint rather than Verizone, but we’re both carrying around Android phones today. And she’s been very happy with hers.

    Anecdotal for sure, but illustrative nonetheless.

  47. esr: Doesn’t follow.

    I said phone innovation and choice, not software development.

    Open source is not required for innovation in UI (in fact, historically, I’m not sure any of the open source UI projects have ever innovated anything, after PARC – I’m willing to be corrected, but it sure seems like the big guns in UI innovation are entirely outside the open source arena, for perfectly explicable reasons; the same ones that explain why the OSS community is excellent at making servers and compilers and terrible and making desktop environments that aren’t just half-assed copies of Windows).

    (And yes, some of that is cultural baggage -but I submit that part of it is also the endless forking and the difficulty of finding someone to set a project’s UI standard, set it to something good, and enforce it.

    Look at Android, in fact.

    Google made a perfectly workable UI… but not an especially innovative one [compare 2007 Android, again, with post-iPhone Android, and notice how what they’re copying changes] and Google’s team there doing the UI was not open to contributions from outside, now, was it?

    Google isn’t accepting UI patches for Android, are they? It might be open source, but in practice it’s “open” to the carriers and nobody else, especially if you can’t find a way to root your phone [ie, if your OS is actually secure, AND the carrier doesn’t want you replacing the OS].)

    A non-open platform doesn’t reduce choice by providing competition for the non-open platform. It increases it. A world consisting of Android and Meego is less choice than one consisting of Android, Meego, and iOS. Adding another option never reduces consumer choice. Saying “open source” doesn’t change that. (Now, if somehow Apple could “own the market” and make all the others marginal, you might ahve a point. But outside of the nonexistent fanboys, nobody thinks Apple’s going to own 75% of the market.)

    Sure seems undeniable that non-open iPhones have provided phone OS innovation. Look at before and after, the entire state of the smartphone OS and UI market. The change is unmistakable, and it’s unmistakably an improvement in user experience.

    Likewise, “choice” has not diminished.

  48. “And our weakness in UI isn’t a consequence of open-source methodology but of cultural/historical baggage.”

    Or rather an inside -> outside approach, which is the opposite of the outside -> inside “rapid prototyping” approach of commerical closed source. Hackers seem to focus on getting the entrails right first and worry about the UI later, while commerical closed source wants something ASAP that’s demoable and sellable, nice screenshots and all, and usable, and worry about the dodgy entrails later. I know it hands-on: I am a Navision ERP consultant (Dynamics-NAV is the new name), they boasted having a sexy MFC UI even back in the nineties and how much cooler it is than old green-screen COBOL ERP apps, yet they did not get something as basic as inventory valuation right until 2005, V4, I’m not shitting you, in 2002 I was chasing bugs that posted a “rounding” entry with an amount in the millions… which means between 1989 and 2004 users could never be sure what their COGS is, which means they never knew how much money they actually made, which means why the fuck did they have an ERP system for if it could not reliably supply the most basic and most important data? But in the sales demos it always looked very sex and yes it was very usable too. Same stuff for most closed-source commercial software.

    This theory predicts the following. Closed source commerical software from the POV of the average users is like “OK first version is neat, but buggy… second is better… third even better” – a _linear_ progression. Open source from the POV of the average users is like “First version… it is a joke, do they really mean me to use it? Second version… the same… third version, OK now I might try using it but really it sucks… fourth version: HOLY SHIT this is awesome!!!” The progression from the end-user POV is not linear but like a log curve, one that stays real low for a long time and then just shoots up like crazy.

    Case in point: VLC VideoLan in Windows. I tried to use it on Linux when the UI sucked i.e. when in a typical early-OSS way library parameters were exposed as configuration parameters. Now VLC is beloved by average Windows users because they can just double-click any kind of video they downloaded from the net and it will just shut up and play it and not bother them with stuff like codecs like the Windows Media Player does… it had good entrails for a long time but a sucky UI, user experience was real bad, and when thex fixed the UI and introduced some automatisms it shot up real fast, overtaking the closed-source commercial competitor.

  49. > Anecdotal for sure, but illustrative nonetheless.

    Yes. Some Guy is busy castigating Eric for Eric’s tendency towards chest thumping rhetoric, while engaging in the same sort of chest thumping rhetoric himself. But Eric’s chest thumping rhetoric is better supported by the numbers, while Some Guy’s chest thumping rhetoric is mostly supported by Some Guy’s personal attachment to the iPhone UI and the iPhone’s physical polish.

    Myself, I usually avoid chest thumping rhetoric, on the theory that I will then have to back it up, and I always like to believe I win all my arguments (mainly by choosing my battles carefully). Of course the reality is almost certainly that I win almost none of my arguments but I’m stunningly good at convincing myself I’m right, whether or not I actually am.

    That said, engaging in chest thumping rhetoric is a lot of fun, isn’t it?

    I am willing to stipulate that Some Guy’s personal attachment to the iPhone UI and disrespect of the Android UI is actually backed up by his personal experience. However I believe that he hates the Android UI for the same reason I hate the default bash UI. I’m used to the ksh UI with vi mappings. My fingers have them mapped down into the bone by now. When I use the bash UI it never works the way I want it to. This does not mean the default bash UI is rotten. I actually expect that the bash UI with emacs mappings is almost certainly more capable. But right now I don’t need to teach it to my fingers.

    I suspect the iPhone’s physical polish is pretty good, but not perfect and often matched. It’s not like Jobs has been kidnapping all the great industrial designers and squirreling them away in his secret hideout beneath an otherwise non-descript seamount.

    Yours,
    Tom

  50. @Kurt absolutely, at any time, m/i/k/l/o/s dot h/o/l/l/e/n/d/e/r at gmail.com, without the slashes.

    @MF: thanks for the info, what I was actually looking for is more info supporting the claim that there are high-end Android devices and say the Galaxy S is not one of them.

  51. Nexus One I believe is still the more powerful phone, but to claim that Galaxy S isn’t a high-end phone because of one (or even more perhaps, I’m not totally familiar with the Android landscape) example is just downright silly.

  52. @Mike Swanson – Nexus one has a slightly slower CPU, and MUCH slower GPU than Galaxy S. The Galaxy S series stutters a bit when there’s a lot of reading from the internal memory, because for some God-forsaken reason, Samsung decided to use their own home-brewed filesystem, and it sucks. The Super AMOLED screen on the Galaxy S is the best on the market, excepting possibly the iPhone 4’s (the latter is better for reading text, the former better for everything else). Frankly, I’m hoping that Apple’s push for > 300dpi resolutions on phones catches on with all the Android devices next year…

  53. And thus are things playing out as I was predicting them previously. Android’s eating the low-end of the cell market as the market transitions from smartphones being the premium niche to a purely smartphone market. Apple appears to be retaining their position as the premium choice with solid sales but overall sales numbers are shifting to Android based heavily on the low and mid-range handsets which are killing off the market for regular cellphones.

    Note that Android could have a ‘crap’ UI and still sell very well, as long as its UI is better than Symbian, WinMobile or BlackBerry and the hardware is more reliable than the webOS handsets. And IMHO, that is the case, Android’s UI isn’t any great shakes, but only webOS and iOS are arguably better and webOS is crippled by an earned reputation for unreliable hardware(I’m one of many Pre owners to have had hardware failures) and limited carrier options.

    1. >And thus are things playing out as I was predicting them previously. Android’s eating the low-end of the cell market as the market transitions from smartphones being the premium niche to a purely smartphone market. Apple appears to be retaining their position as the premium choice with solid sales but overall sales numbers are shifting to Android based heavily on the low and mid-range handsets which are killing off the market for regular cellphones.

      No offense, but this was hardly a difficult prediction for anyone not blinded by the Apple mystique.

  54. And in more news, the company formerly known as Deutsche Telekomm has announced the G-2.

    I’m a T-Mobile customer in a metro area, and this puppy is thus a bit tempting even though I have a Nexus One. Android 2.2, HDSPA nearly-4G data speeds, and a slide-out keyboard – what’s not to like?

    For my needs, an iPhone 4 couldn’t even compete with this, let alone best it. The difference in data speed alone would be decisive.

  55. Tom,

    I categorically deny having any “personal attachment” to the iPhone’s UI. Show me something better, and I’ll switch to it. I’ve switched computing platforms several times in my career, and I will be very happy when there’s a better product available. I really liked my Sony-Ericsson phone before the iPhone hit the market, but today I don’t miss it one bit.

    As for your accusation that I “hate” Android, that’s just stupid. There are all kinds of inferior products on the market, and none of them are worth the energy to hate. As far as I’m concerned, Android is interchangeable with WinMo 7: If I were forced to use it, I might hate it, but I’m not, so I don’t.

  56. How would you maximize Apple’s business?

    I can understand that Jobs may not want to allow third party manufacturers to build iOS devices because this might reduce the carriers’ incentive to carry his preferred iPhone configuration, which he may think reduces his ability to drive user satisfaction and market share. I can not understand what is gained for Apple from limiting application developers and the types of applications (e.g. Flash) users may choose. If security and user satisfaction were the goal, simply providing users with the choice to opt-out of Apple’s application “firewall” should suffice. Rather it must be that Jobs wants API lockin, but certainly he must see that this is impossible. The market will adopt that APIs that are best fit, and will open source them. There is no way to go back to the Ballmer model. Jobs should focus on maximizing market share by forsaking the unrealistic dream of API lockin, and retain his integration leverage (without forcing unnecessary integration lockin) to drive profit margins and carrier support. I thought facing his death bed, he would grasp the urgency of life and the utter stupidity/futility of retarding humanity’s race towards maximum interoperability. Can anyone offer a plausible explanation for how the current iPhone strategy is best for anything other than short-run maximizing the stock price (profits, perceived monopoly) at the cost of market share holding power?

    Some Guy(s), open source did not cannibalize for-profit software innovation, it shifted and improved the for-profit priorities towards *SPEED* of integration. Lockin business models are slow to innovate “WAN” integration, rather they prioritize innovation within a private space, e.g. how much C compiler innovation did we really need in the early exponential information sharing explosion priorities of the Web1.0/2.0. As we move to Web3.0 (web applications, fat clients, code mashup interoperability), I expect there will be a renewed priority on compiled (static typed) languages. The key point you may be missing is that 80/20 rule applies to UI (preference) and any other private space, i.e. declining marginal utility, and that wide area integration (humanity’s economy-of-scale, because we don’t live forever) trumps private space order because it has probably constant or even *increasing* marginal utility[1] (Mr. Raymond noted the only known positive scaling law of software engineering). I am 45, and never again do I want to throw away years of my effort (physical degradation of my eye sight, and motor function in my lower extremities, etc) writing code that dies because it no longer inter-opts with the new trend. There are too many other interesting things I would rather be doing than coding, so it darn well better have a long half-life utility.

    [1] wide area integration is really just about maximizing freedom, because when maximum independent actors are able to inter-opt freely, that is on trend with the 1856 second law of thermodynamics, which states that the universe (a closed system by definition) is trending to maximum disorder (maximum uncorrelated possibilities).

  57. Nokia just replaced it’s CEO with Stephen Elop, late of Microsoft.

    I’d say things are about to get worse for anyone who had hope for Symbian, or that Nokia would replace Symbian with Android.

    Which is, overall, a blow for Android.

  58. Some Guy,

    > I categorically deny having any “personal attachment” to the iPhone’s UI.

    Well then stop talking about how wonderful it is all the time. Given what you personally say repeatedly and at length, I categorically deny that your categorical denial has any actual meaning.

    > I’ve switched computing platforms several times in my career, and I will be very happy when there’s a better product available.

    OK. I don’t see how my comment says otherwise.

    > I really liked my Sony-Ericsson phone before the iPhone hit the market, but today I don’t miss it one bit.

    “I really like” and “personal attachment” mean the same thing. You were personally attached to your Sony-Ericsson phone. You gave up your attachment when something better came along. Happens all the time.

    > As for your accusation that I “hate” Android, that’s just stupid. There are all kinds of inferior products on the market, and none of them are worth the energy to hate.

    This is either a reading comprehesion problem on your part, or a clarity problem on mine. I said “However I believe that he hates the Android UI for the same reason I hate the default bash UI.” Subsequently I said “This does not mean the default bash UI is rotten. I actually expect that the bash UI with emacs mappings is almost certainly more capable. But right now I don’t need to teach it to my fingers.” There is no energy behind the word “hate” in my paragraph. I used it in the “don’t you just hate that” sense.

    Try again when you can identify something that’s actually stupid. I wrote a very common usage of the word “hate”. That you read a different common usage does not qualify as stupidity on my part.

    Yours,
    Tom

  59. Only 4.5% of Android users are running 2.2?

    Want to tell me again how “open” Android is, Eric?

    That has nothing to do with the “Openness” of the platform. It has more to do with the vast differences in hardware between devices. One can’t just download stock Android and put it on any hardware without customization. Many of the device drivers on the given devices are closed source, and owned by the manufacturers. Also, the individual manufacturers build software UI stuff that runs on top, and that has to be ported.

    This is why devices with the most vanilla hardware/software configuration are the first to get any new version working, such as the Nexus One — HTC’s hardware is all incredibly similar, and the N1 is NOT a Sense UI (HTC’s proprietary skin) device. Motorola, Samsung, etc. use different hardware configurations that require drivers to be ported. Lord knows I can’t wait for Samsung to release Froyo for the Captivate. I could go the Cyanogen route, but am a little too chicken, as I don’t really have a backup device. When I have a second device in a year or so, I’ll likely be more likely to experiment.

    Also, many older/cheaper devices don’t get OTA updates to the newer OS’s, for performance reasons mostly. The G1, for instance, runs like crap on 2.0 and 2.1. It becomes acceptable with 2.2 because of the introduction of the JIT, but there’s still no “official” upgrade path, though Cyanogen 6 is popular among the types that own one. This is analagous to Apple not allowing iOS 4 run on the original iPhone — it just doesn’t have the hardware to support the new OS features.

    Apple skirts this issue altogether by only having one new device per year. It doesn’t thwart the problem altogether; they’ve had issues with iOS 4 running on the iPhone 3GS, (mostly performance related) for instance.

    1. >Also, the individual manufacturers build software UI stuff that runs on top, and that has to be ported.

      I wonder how long this will last. The carrier UIs hurt time-to-market, and I don’t see any indication that they’re acting as successful differentiators. They seem to exist mainly to cover up value subtractions like advertising crapware and various kinds of feature lockout, and as word gets around about that consumers will push back.

      I think this is worth a blog post, actually.

  60. Ah, speak of the devil: Nokia is giving the sack to Olli-Pekka Kallasvuo and hiring Microsoft’s Stephen Elop as their new CEO. No doubt this is a direct result from the failure to compete in the US smartphone market. Nokia will start to sell the new N8 next week. It runs the new Symbian 3, which is apparently not all that much of an improvement. N8 is supposed to have the best camera in the business.

  61. Some Guy,

    > Tom, I’m sure you enjoy the armchair shrink game, but don’t give up your day job.

    It’s not head-shrinking, it’s reading what you write. There’s plenty of material on multiple threads. If you want to avoid giving the impression that you like the iPhone, don’t praise it repeatedly. Similarly if you want to avoid giving the impression that you don’t like Android, don’t criticize it repeatedly. And if your main argument does not hang on the superiority of the Apple product (which is clearly subjective), stop saying it’s all about the superiority of the Apple product.

    What’s more likely here is that I haven’t written enough or well enough on this subject here for you to have a sense of where I’m coming from, which is making it hard for you reply to me.

    Yours,
    Tom

    1. >It’s not head-shrinking, it’s reading what you write.

      Indeed it is. He’s got you nailed, Some Guy.

      I’m not as nice as Tom is, so I’ll put the case more directly: if you want not to be taken for a fanboy, stop frothing at the mouth.

  62. I wonder how long this will last. The carrier UIs hurt time-to-market, and I don’t see any indication that they’re acting as successful differentiators. They seem to exist mainly to cover up value subtractions like advertising crapware and various kinds of feature lockout, and as word gets around about that consumers will push back.

    I think this is worth a blog post, actually.

    Yeah… I think HTC, Motorola, Samsung, et al, over-estimate greatly the value of their skins. Pretty much everyone who’s used stock Android hates TouchWiz (Samsung) and Blur (Motorola). I know a few who really like Sense (HTC), but even more who don’t. Given, the stock Android launcher/home screen is kinda crummy compared to a lot of after-market ones (I’m a big fan of Launcher Pro), but all of the menu tweaks and additional little features make things worse, in my opinion, if for no other reason because they make different Android devices so… different out of the box, and slow the time to market of updated ROMs.

  63. Nokia certainly needs to sort out its smartphone strategy ASAP, but it will be hard to write off the sunk costs in Symbian – I wonder if appointing an outsider points to that?

    I hadn’t realized Nokia had leading market share in emerging-market/3rd-World phones, though. I can imagine a nice scenario for them where Android takes over in the developed world (with iOS devices as a boutique brand), and Meego phones, tablets, and netbooks take over elsewhere as those places skip the Windows PC entirely; it seems to me that Meego actually has some advantages over Android if you want to fill both roles with one OS.

  64. >>And thus are things playing out as I was predicting them previously. Android’s eating the low-end of the cell market as the market transitions from smartphones being the premium niche to a purely smartphone market. Apple appears to be retaining their position as the premium choice with solid sales but overall sales numbers are shifting to Android based heavily on the low and mid-range handsets which are killing off the market for regular cellphones.

    >No offense, but this was hardly a difficult prediction for anyone not blinded by the Apple mystique.

    Go look at the responses to me saying exactly this in previous comment threads. Quite a number of people kept saying this wasn’t the case. It’s the people blinded by the Android’s iPhone killer mystique who were primarily arguing against my position, not the Apple fanboys (although a few of them did argue against it as well).

  65. I live in the Philippines, the world’s largest SMS market by quantity of sent messages (well at least it was before Twitter), because for one thing Filipinos hate to be alone and up until recently, call time was too expensive for the poor masses. The smartphones are not mainstream because they cost too much relative to average income. Nokia apparently rules the non-smartphone market because their basic phones are the most reliable, performing, and best value. My guess is Samsung is #2 here. The used cell phones from USA are very popular here (see my Sprint clam shell phone from 2001 here). Some cheap Chinese smartphones are getting some limited adoption. On the high-end, afair from late last year (when I purchased a Nokia 2690 while I am waiting for some clarity in smartphones), an unlocked iPhone cost $400 – $500 here at a few selected importers. Since Nokia has the widespread dealer network for its mainstream phones and brand reputation, it apparently wins in smartphones too. Honda enjoys a similar brand reputation in small motorcycles, because Honda Dreams from 1990s are still running reliably. My old Nokia phones still work even the keypad and screen covers have been replaced several times. Given the resale market for used phones here (native in the mountain can obtain a used cell phone for $10), that really matters.

    Nokia’s next move (or absence of a strategy change) is thus critical. Nokia is afaics in the volume hardware business and they don’t make money on the service plans in the Philippines at least, as it is mostly pre-paid here. I reason that Nokia wants to have tight control over the quality of the integration of the software and the hardware in order to differentiate and meet their profit margins with quality control, so I would expect they will not go Android, except for the urgency which they may not be able to realistically overcome in any way. If someone had the appropriate micro-kernel OS ready today, it would probably be a better fit for Nokia and they would have a realistic chance to trump Android. Also I can say that the UIs on Nokias non-smartphones are much simpler for the average uneducated (non-high school graduate, even high school here is only 10 years because there is no 7th and 8th grade) person in the developing world to use.

    The USA-centric view may not be the model that is adopted by the the developing world. There are factors that have to be considered.

  66. ESR, the Android UI is crap. The UI creates expecations and then breaks them all the time. Simple example: if you have music loaded with multiple albums by the same band, click on the band. The screen opens up the list of albums. Now you want to get back to the previous screen, so you hit back. Oh, FAIL, it throws you out into the home screen.

    Same thing with Twitroyd. You hit the speech bubble to type something, then decide you want to abandon it, so you hit back. Oh, FAIL, it throws you out into the home screen.

    Same thing in the phone app. You make a call to somebody and it throws you into the Call Log afterwards. Want to go back and make another call? Hit back! Oh, FAIL, it throws you out into the home screen.

    There’s three ways (in Froyo) to get from home screen to home screen. You can do a side-to-side slide (which used to work better on 2.0), or you can press one of the dotted buttons in the lower corners, or you can hold down the applications button until all of your home screens are displayed and then you can pick one. Unfortunately, they’re all SLOW. I’d prefer that they have spent the time to make just one be fast. Why do you need three, anyway? So that people can find the feature more easily?? But they only need to find it until they remember it; why use up the screen space?

    It’s as if they never did a usability study.

    1. >Same thing with Twitroyd. You hit the speech bubble to type something, then decide you want to abandon it, so you hit back. Oh, FAIL, it throws you out into the home screen.

      Is this an Android fail or an app designer fail? Granted we should expect the dialer widget to have consistent back behavior, it’s unclear how the Android guys could make third-party apps adhere to any one convention about this.

      (I never noticed the phone log thing, BTW. Am you too fussy, or am I not fussy enough? Serious question.)

  67. Is this an Android fail or an app designer fail? Granted we should expect the dialer widget to have consistent back behavior, it’s unclear how the Android guys could make third-party apps adhere to any one convention about this.

    Apple rejects apps which do not conform to the company’s conventions for what the buttons should do.

    Much like Unix and X11, the extreme openness of Android is what makes its usability so horribly broken.

    1. >Much like Unix and X11, the extreme openness of Android is what makes its usability so horribly broken.

      You can tell me Android’s usability is broken when it’s not pounding the living shit out of every other smartphone UI in sales growth among nontechnical end users, quarter after quarter after quarter, even on hardware where carriers have done their best to hobble it. Until then, I’ll just laugh. Because in the end, it’s not the esthetic judgments of people like you or me or Russ that matter.

  68. > You can tell me Android’s usability is broken when it’s not pounding the living shit out of every other smartphone UI in sales growth among nontechnical end users, quarter after quarter after quarter, even on hardware where carriers have done their best to hobble it. Until then, I’ll just laugh. Because in the end, it’s not the esthetic judgments of people like you or me or Russ that matter.

    This is really the point I was trying to make when talking about bash above. The comments Jeff, Russ and Some Guy make excoriating the Android UI are wildly exaggerated. The bash UI isn’t horribly broken either, but I’m just not used to it so things I expect to work don’t and I get frustrated. It’s the same way when I go up to the absolutely gorgeous Macs in the store and try to play with them. I can tell they are fantastic, but they don’t work the way Windows does, and I get frustrated. It isn’t worth figuring it out because my wife and children are very very happy with the Windows apps – like kids games – we own.

    Non-technical users are obviously liking their Android phones, just like my wife like her Palm. Android is not horribly broken, it just frustrates you because you know how is could be better.

    For some reason I always think the phone wars, the computer language wars, the OS wars, the computer maker wars, and for that matter the Glock vs. 1911 wars are all pretty silly. They are all just tools, and what mostly matters is the muscle memory that you are used to.

    I’m not married to my phone, Perl, Sun OS, Compaq or my Makarov.

    Yours,
    Tom

  69. > If you want to avoid giving the impression that you like the iPhone, don’t praise it repeatedly.

    When did I ever say that I didn’t like the iPhone? I said that I’m not emotionally attached to it, and it’s rather presumptuous of you to claim to know my emotions. As I’ve said, show me something better and I’ll buy it. I’ve done so before, and I’ll enjoy doing so in the future. As for Android, you claimed that I “hate” it, which is nonsense.

  70. Simple example: if you have music loaded with multiple albums by the same band, click on the band. The screen opens up the list of albums. Now you want to get back to the previous screen, so you hit back. Oh, FAIL, it throws you out into the home screen.

    Huh… That’s rather worse than I would have guessed. It sounds like Google’s UI designers never even used a web browser before.

  71. Is this an Android fail or an app designer fail?

    Way to miss the point, Eric. Don’t ever change, the world needs apologists for mediocrity, too.

  72. if you want not to be taken for a fanboy, stop frothing at the mouth.

    Oh, I know you’re an expert at frothing at the mouth, Eric. Whenever your name comes up in conversation, the word “loon” follows within a few seconds (and not from me, I might add.) I often am put in the position of defending you, even though as a lIbertarian and a second-amendment advocate myself, I often wish you’d STFU.

    Anyhow, I defy you to show any example of “frothing” on my part. I don’t go off into hyperbole the way you do.

    1. >I often am put in the position of defending you, even though as a lIbertarian and a second-amendment advocate myself, I often wish you’d STFU.

      Your credibility went from zero to negative with that statement. This is a classic variation on concern trolling, and is even more ludicrous than your denial that you’re a raving Apple fanboy on a vendetta. My commenters will make their own judgments, but from now on I will simply write you off as a troll who tells unusually elaborate lies and refrain from responding.

  73. Some Guy,

    > When did I ever say that I didn’t like the iPhone? I said that I’m not emotionally attached to it, and it’s rather presumptuous of you to claim to know my emotions. As I’ve said, show me something better and I’ll buy it. I’ve done so before, and I’ll enjoy doing so in the future.

    I’m not claiming to know your emotions either. I’m still just reading what you write.

    > As for Android, you claimed that I “hate” it, which is nonsense.

    Are you reading what I write? I think you missed one of my comments.

    Given the wide range of feelings encompassed by the word hate, my claim that you hate Android matched what you write, but your claim that you do not hate it does not match what you write.

    Are you reading what you write? I think you missed all of your comments.

    Yours,
    Tom

  74. Some Guy, Mr. Raymond doesn’t need defending (the wise mirror of Matthew 7 applies to those call him or any other person a loon) and any truly anarchist libertarian (free market individualism), myself included, would not identify himself with a group to the extent of being responsible (surety) for the reputation of a member. As what I’ve read, I don’t think anyone here is overly attacking you (the “fanboy” may be accurate because I thought you were already coming around to the point in next paragraph) and let us remain engaged in the rational discussion.

    As what I’ve read, I don’t think any rational person is saying that Android’s UI is perfected, and in some respects probably not as refined as iOS on iPhone. Apparently the point being made is analogous to that Windows kicked Mac butt even though it wasn’t as “refined”, because it was more open to competition. Android is even more open than Windows is/was in many respects. The priority of the billions person market in aggregate is not the same as your (and millions of others’) private priority on perfected UI. There are infinite things that need to be worked on right now in the computer space, and the market will choose priorities.

    Don’t worry, the best parts of the iPhone will be copied and open sourced. As I explained to you before, this doesn’t mean the for-profit opportunities are being destroyed by copying, it means the world’s priorities have shifted to *SPEED* of addressing real world needs, i.e. integration, and not getting bogged down in re-inventing the wheel from scratch every few years.

  75. @ Jocylen
    >I can not understand what is gained for Apple from limiting application developers and the types of
    >applications (e.g. Flash) users may choose. If security and user satisfaction were the goal, simply providing
    >users with the choice to opt-out of Apple’s application “firewall” should suffice.

    Ah, now this I can answer with ease, and it’s something that sadly most geeks never understand unless they are put in the position of having to deal with it. Unlike you and I who sees “Warning dong this is at your own risk” and assumes that means from here on out there be dragons and you’re on your own, the average consumer feels entitled to have everything go their way and blame the manufacturer when it doesn’t, even if the self destruct button is clearly labeled. To the average consumer, if the manufacturer puts in a switch, then they are approving and even encouraging you to throw that switch, because they wouldn’t put it there if they didn’t want you to use it. This is incidentally one of the reasons why RedHat Linux does not have a binaries/restricted repository like Ubuntu does.

    Full disclosure, I used to (but no longer) work for Apple. Let me take you back to the hey day of a mere 4 years ago. Apple introduced the intel based macs and people figured out how to install windows on it. Apple, trying to cut the damage off at the pass releases Boot Camp, with the admonishment that Apple does not and will not support Boot Camp or Windows. This admonishment is on the Boot Camp web page and the installation instructions. Not 6 months later, you could walk into any Apple store, and you would find Apple employees supporting Boot Camp and Windows. And it’s not always the customer’s direct fault either. I had a customer come in and tell me how frustrated she was with her mac and how awful it was and she had heard that things would be better and that’s why her nephew bought her one. Upon examining her mac, she had been running in windows the entire time. Her well meaning nephew had wiped all but the barest of OS X installs from the drive (enough to make bootcamp work) and then installed windows, set it as the default OS and given it to her without any explanation. Until I had shown her otherwise, she believed her mac was just a pretty windows computer, and that all the problems she was having were results of Apple failure.

    This is the whole reason why Apple doesn’t even allow you to opt into flash, because some well meaning tech guy will turn it on for his mother, and when everything goes to hell in a hand basket, despite policies and statements to the contrary, in her mind it will be Apple’s problem, and Apple will spend time and resources fixing it. And I assure you that customers do not respond kindly to “That isn’t supported, don’t do that.”

    In the end, there is an opt out method, it’s called jail breaking, and it voids your warranty. And from my perspective it works pretty well. If you are able to truly grasp what “not supported” means, you are able to jailbreak. Otherwise, you stay in the walled garden or play in a different garden.

  76. >In the end, there is an opt out method, it’s called jail breaking, and it voids your warranty. And from my perspective it works pretty well. If you are able to truly grasp what “not supported” means, you are able to jailbreak. Otherwise, you stay in the walled garden or play in a different garden.

    It doesn’t help, though, that installing any upgrade to your device wipes your jailbreak and all related files. Apple (and by extension, anyone apologising for them) can’t claim this as a viable method of escaping the walled garden until Apple is no longer actively hostile to it.

  77. @ Tom
    >It doesn’t help, though, that installing any upgrade to your device wipes your jailbreak and all related
    >files. Apple (and by extension, anyone apologising for them) can’t claim this as a viable method of
    >escaping the walled garden until Apple is no longer actively hostile to it.

    And what do you consider active hostility? Are you suggesting Apple should not close security holes and exploits (how most if not all jailbreaking occurs?). Or are you merely upset that when you alter the base system, that such alteration may disappear when you then update that base system? Do you get equally upset when you run the Ubuntu updater and it replaces your customized configuration files with new standard versions? To be fair, the Ubuntu updater asks about preserving these changes, but remember we’re talking about an unsupported use of the device here. What right do you have to demand that Apple check for and preserve your unsupported changes to the system?

    Further, it is clear that you are not very familiar with the current state of iOS jailbreaking. The process of updating my unlocked/jailbroken 1st gen iPhone was as follows:
    1) Plug into computer to sync and ensure a current backup is taken (a step one should take before any upgrade no?)
    2) Download current iPhone firmware image from apple. (a normal step in the update process)
    3) Customize the firmware with unlocking tool. (a normal step in the unlocking process)
    4) Restore my phone from the customized firmware (this is the only “imposed” step which makes this process painful, in theory it might be possible to simply update in place, but even the jailbreakers recommend this step)
    5) Restore my backup to the phone, observe that all of my jailbroken apps and changes have returned via the backup, no need to reinstall (barring necessary updates)

    Which part of that seemed actively and overtly hostile to you?

  78. As an afterthought, Tom’s post highlights exactly what I was talking about originally, where even if an option is unsupported and not encouraged or approved, people still demand that the manufacturer go out of their way to make use of that unsupported option easy and convenient.

  79. > frustrated she was with her mac and how awful it was […] she had been running in windows

    Hahaha.

    > Unlike you and I who sees “Warning dong this is at your own risk” and assumes

    I did some customer support and developer liason to customer support for a few years, so I know well to what you refer. I have probably interfaced with more than 1000 customers in my life.

    > some well meaning tech guy will turn it on for his mother

    There is a solution to this. Let the customer pay to remove the crap-wall, and they are much less like to do that and/or forget it. And there will be much fewer cases of people doing it for others.

    It can be presented as “We sell our phones at a slight discount, which we recover by having a monopoly over some aspects of your phone. We feel this also enables us to insure a quality experience with your phone. If you would like to remove the discount and end our control over your phone, please pay here”.

    If you are really paranoid, you can remind them periodically, say every 2 – 3 months.

    > In the end, there is an opt out method, it’s called jail breaking

    That is ridiculous. I doubt even I would bother with that, unless I had a very compelling reason and I wouldn’t be happy about having to maintain yet another fork of complexity in my already overloaded life.

    Perhaps you didn’t get the point that the reason users don’t have a long attention span for support details, is because their lives are already overloaded. People want simplicity and creativity. That is why the iPhone is popular.

  80. Apology for the double post, but I really need to add that for an unlocked phone, throw up a device power-on splash screen that says:

    NO LONGER SUPPORTED BY APPLE
    THIS PHONE WARRANTY WAS VOIDED BY
    TO UNDO THIS, WE WILL PAY YOU HERE.

  81. >There is a solution to this. Let the customer pay to remove the crap-wall, and they are much less like to
    >do that and/or forget it. And there will be much fewer cases of people doing it for others.

    A couple of problems. This really doesn’t solve the problem of the well meaning relative / tech guy at work. To that person’s mind, the cost of getting the full usefulness out of the phone includes the price of turning on the unsupported features, so either they will pay for it themselves if its nominal, or encourage the owner they’re assisting to pay for it. As a result, to achieve the desired effect, the cost must be high enough to discourage casual purchase ($5 would be to low), but not high enough to generate a re-evaluation of the costs associated with the device ($99 would be to high, as evidence see the gnashing of teeth over the development license).

    I can not say for certain, but I imagine that Apple receives far less negative press by simply not supporting jailbreaking than they would receive if they charged $50 / phone for it.

    It’s also worth pointing out that people find the unsubsidized price of the iPhone insane as is, it isn’t very likely that someone paying that price already would be keen on paying another fee to jailbreak the phone.

    >That is ridiculous. I doubt even I would bother with that, unless I had a very compelling reason and
    >I wouldn’t be happy about having to maintain yet another fork of complexity in my already overloaded life.
    >Perhaps you didn’t get the point that the reason users don’t have a long attention span for support
    >details, is because their lives are already overloaded. People want simplicity and creativity. That is why
    >the iPhone is popular.

    Actually, I understand this perfectly which is why I feel (and I’m sure Apple feels) that the unsupported jailbreak is the best way to handle this. By ensuring that the process is complex (or appears that way), they ensure that only the people who understand that complexity will engage in the act, thereby reducing their support requirements. It’s worth noting that this doesn’t truly solve the well meaning relative problem except in that it voids the warranty.

    To sum up, as a general rule, I think that Apple would receive less negative press for leaving the state of affairs roughly as is rather than appearing to “try and extort money out of their users for control they should already have”. And yes, I can pretty much guarantee that if Apple had a pay-to-jailbreak option, that exact phrase would be used as a headline.

  82. And what do you consider active hostility? Are you suggesting Apple should not close security holes and exploits (how most if not all jailbreaking occurs?). Or are you merely upset that when you alter the base system, that such alteration may disappear when you then update that base system? Do you get equally upset when you run the Ubuntu updater and it replaces your customized configuration files with new standard versions? To be fair, the Ubuntu updater asks about preserving these changes, but remember we’re talking about an unsupported use of the device here. What right do you have to demand that Apple check for and preserve your unsupported changes to the system?

    There’s no reason why, for instance, an upgrade of the OS should wipe the filesystem. Without taking a backup beforehand, they managed to preserve my bought applications, which implies that the upgrade isn’t fundamentally doing anything to the filesystem that would delete things (like reformatting in a different filesystem). This in turn implies that the upgrade is trawling for unofficial changes and removing them–which I would certainly call being overtly hostile. As for your concern about security holes, Apple should certainly fix things like the PDF exploit that jailbreakme.com uses. However, there’s no reason why they should obsess about the security holes in the bootcode level that allow booting unofficial code in the first place (the method I used to jailbreak). Jobs’ “game of cat and mouse” is completely inappropriate. My recommendation for Apple would be simply to allow jailbreaks without taking legal action against them, not delete unofficial filesystem changes on upgrade, and use a bootloader that will allow the phone to boot whatever the user wants. They would then, in my opinion, be in a much stronger position to claim that their walled garden is necessary from an ease-of-use perspective, and they aren’t just trying to achieve absolute control for the sake of it.

  83. >There’s no reason why, for instance, an upgrade of the OS should wipe the filesystem.

    There is no indication that it does. Once again I must assume you are speaking from a standpoint of ignorance on the world of iPhone jail breaking. Apps installed via the various jailbroken repositories are not installed as, and do not behave as a normal application. A simple for instance is that a jailbroken application can not be deleted from an iPhone by holding until the apps wiggle and pressing the big “x” button because jail broken apps don’t get the big “x” button, which means they are somehow different than a standard “app” application. It is therefore not unreasonable that when whatever changes to the system files that the jailbreak makes that allow it to operate are repaired as part of the OS upgrade, that it would forget or be unable to locate the jailbroken apps which are not installed the same way as a standard app. If for example, the jailbreak app says to the iPhone that $PATH is /bin:/usr/bin:/cydia it is entirely possible for you to “lose” all of your jailbroken apps merely by resetting $PATH to /bin:/usr/bin. And there is no reason in the world for Apple to check to see if $PATH has been modified and preserve those changes because modifications to $PATH *are not supported*

    Also, though you latch onto my use of the word backup as if it is an onerous step, it is simply the act of syncing your iPhone, which for 99% of users happens when they plug their phone into iTunes. Also, per Apple’s own documentation (http://support.apple.com/kb/ht1414) updating your device performs a backup as part of the routine process. It is not inconceivable that the update is something akin to the OS X archive and install process, preserving only certain directories and otherwise wiping everything else clean.

    And once again I should point out that you are asking Apple to modify what they do and how they do it to make your usage of an unsupported feature more convenient. You are asking for support for an unsupported action.

    >However, there’s no reason why they should obsess about the security holes in the bootcode level that
    >allow booting unofficial code in the first place (the method I used to jailbreak).

    So the only security Apple should be concerned with is remote security, not local security? Look I know that all bets are usually off if someone has physical access to your device, but I’m quite positive I can think of a few businesses who still appreciate the locking capabilities of the phone, and fixing bootcode exploits is part of that layer of physical security.

    >My recommendation for Apple would be simply to allow jailbreaks without taking legal action against
    >them, not delete unofficial filesystem changes on upgrade, and use a bootloader that will allow the
    >phone to boot whatever the user wants. They would then, in my opinion, be in a much stronger
    >position to claim that their walled garden is necessary from an ease-of-use perspective, and they
    >aren’t just trying to achieve absolute control for the sake of it.

    Item 1 they already do, no jailbreaker has been sued or harassed other than to have their warranty denied.

    Item 2 you are asking them to alter their code to support your unsupported actions.

    Item 3 you are asking them to approve and encourage unsupported actions.

    I have already demonstrated why and how items 2 and 3 are not options Apple is likely to take. But as I said, your complain basically boils down to you wanting Apple to support unsupported behavior.

  84. @tmoney
    > This really doesn’t solve the problem of the well meaning relative / tech guy at work
    > to discourage casual purchase ($5 would be to low), but not high enough to

    Perhaps you wouldn’t notice that cost was not the key point, because my prior comment got butchered because it thought <Customer Name> was a tag and so it didn’t appear:

    NO LONGER SUPPORTED BY APPLE
    THIS PHONE WARRANTY WAS VOIDED BY <Customer Name>
    TO UNDO THIS, WE WILL PAY YOU HERE.

    My point was that who ever pays to unlock the device, gets their name on the power-on splash screen. The payment is to establish identity (e.g. from the credit card transaction) and assign culpability, so that Apple can not be falsely blamed by the user (e.g. “Aunt Tillie”) of the device.

    > I can not say for certain, but I imagine that Apple receives far less negative press […] not supporting jailbreaking

    So you are saying user choice is not important priority to customers.

    Then it follows logically that Mr. Raymond’s theory about customer choice demand driving the end of the network access oligarchy, has no hope of succeeding?

    > unsubsidized price of the iPhone insane as is, it isn’t very likely […] would be keen on paying another fee to jailbreak

    I think the reverse logic applies. If I am going to pay the cost to be free, a small incremental cost doesn’t matter much.

    > they ensure that only the people who understand that complexity will […], thereby reducing their support requirements

    They have ZERO support requirements if the splash screen says what I propose above. Whereas, for as long as they don’t allow unlocking the phone, their support requirements will go up caused by all the people who can’t do what they want to do and get frustrated, e.g. “Apple should certainly fix things like the PDF exploit that jailbreakme.com uses” as Tom Dickson-Hunt wrote.

    Unlocking will lower their support requirements. All those who want freedom will have to agree to give up Apple support.

    > So the only security Apple should be concerned with is remote security, not local security?

    Correct, if ever they implement the correct micro-kernel. Who ever does first, is going to change the world.

  85. Oh I forgot, the likely reason the Jobs doesn’t allow unlocking has nothing to do with his stated reasons, even those he preaches to you who work(ed) for him. It is because he shares revenue with the carrier oligopoly, and they won’t allow it.

  86. I will simply write you off as a troll who tells unusually elaborate lies and refrain from responding.

    What do you think I’m lying about, Eric? That people call you a loon, that I’ve defended you from time to time, or that I occasionally wish you’d STFU lest you make yourself a handy straw-man for the statists and the gun grabbers?

  87. tmoney,

    Speaking of voiding warranties, I cracked the case of my first Mac within a month of the product’s debut, fully understanding that if I broke it, Apple had no obligation to fix it. Their policy at the time was that they didn’t want anyone opening the case because there was a CRT in there, and not everyone who wanted in would know how to safely discharge it.

    When it comes to the iPhone, Apple is trying to avoid the pitfall that they saw happen to WinMo phones, where malware could brick the phone. That’s not a very good user experience, to say the least. So, they lock it down as much as they can, a couple hundred open-source zealots denounce them for it on a regular basis, and they continue to sell all the phones they can make.

  88. Mr. Raymond doesn’t need defending

    It’s not that I want to defend him, mind you. It’s more like some statist says “esr advocates this view that you share, and esr is a loon, ergo your position is invalid”, so I briefly digress from the topic at hand and defend him in order to deprive the statist of the tactic.

  89. @Some Guy
    > they lock it down as much as they can, a couple hundred open-source zealots denounce them for it on a regular basis

    Zealots who can empower the competition. And your logic is inconsistent. If only 100s of zealots want to unlock, then what do they gain from making it difficult? Obviously it must be that more than couple 100s will unlock if they make it easier. So that will tell you they are not fulfilling all the market demand, and competition will and is.

    > and they continue to sell all the phones they can make

    Strange how Android can make phones when Apple has all the manufacturing capacity of the world fully utilized.

    > It’s not that I want to defend him, mind you. It’s more like some statist says…

    One possible anarchist response is that no two anarchists every have the exact same position. However, I find the most efficient response to statists is to hand them more rope to hang themselves. Encourage them!

  90. >My point was that who ever pays to unlock the device, gets their name on the power-on splash screen. The payment
    >is to establish identity (e.g. from the credit card transaction) and assign culpability, so that Apple can not be falsely
    >blamed by the user (e.g. “Aunt Tillie”) of the device.

    Having been a CSR before, I’m surprised you don’t already know that it doesn’t matter what the system says or who is really to blame, the user always blames the guy that took their money and sold them the device, period.

    >So you are saying user choice is not important priority to customers.

    For most customers? No. Choice to most customers in my experience takes a distance second to a product that works. The only time people want a choice is when the product doesn’t do what they want. Witness the rise and fall of IE.

    >Unlocking will lower their support requirements. All those who want freedom will have to agree to give up Apple support.

    They already do.

    >Correct, if ever they implement the correct micro-kernel. Who ever does first, is going to change the world.

    It seems to me that the correct micro-kernel is useless if the bootloader/firmware/bios still lets you chose which kernel to boot in the first place.

    >Oh I forgot, the likely reason the Jobs doesn’t allow unlocking has nothing to do with his stated reasons, even those
    >he preaches to you who work(ed) for him. It is because he shares revenue with the carrier oligopoly, and they
    >won’t allow it.

    Unlocking is different from jailbreaking. Unlocking is certainly a factor of the exclusivity contract (hence why unlocked iPhones are only sold over seas where there are no exclusivity contracts). Jailbreaking however is purely a matter of control, support and experience.

  91. @tmoney
    > Having been a CSR before, I’m surprised you don’t already know that it doesn’t matter what the system says
    > or who is really to blame, the user always blames the guy that took their money and sold them the device, period.

    Agreed, but my point is they are getting it both ways. They are just choosing to ignore the one side and favor the other. I don’t see the benefit. Jobs is in a dogfight for developer share. Never ever alienate developers. They are more important than customers, because in fact they actually do your support.

    Any way, the more salient point is that someone needs to fix the OS security/malware properly as I have suggested in the other blog threads using a provable micro-kernel with provable memory and disk protection.

    Btw, I wasn’t only a CSR, I have written several million user commercial apps, e.g. I was the 3rd programmer on Corel Painter– the 1st natural media paint program, etc.. I have also owned my own software company with confirmed 700,000+ users and 330,000 websites created with the product at any one time (AltaVista provided the counts).

    > Choice to most customers in my experience takes a distance second to a product that works

    Yeah but there are different gradients between “what works” and “what works really cool and neat features that I want and need”. Jobs is shooting himself in the foot by overstepping the principles. I literally had people returning their new Apple computers because my Cool Page software wasn’t available on Mac. Choice matters more. Choice is part of “it works”.

    > They already do.

    But it is too difficult. I will probably never buy an iPhone because of that. And btw, I use Windows, not *nix on my desktop. So I am very open to using what works with the least twiddling. And I am a developer. People wonder how I can cope with Windows and viruses. Simple, I use Baseline Shield. My computer boots to virgin. Let the virus profilerate, they don’t affect me.

    The security theater nonsense…

    > It seems to me that the correct micro-kernel is useless if the bootloader/firmware/bios still lets you chose which kernel to boot in the first place.

    Not if it requires you to press a difficult to access physical key with a paper clip.

    > Unlocking is different from jailbreaking

    Lump together, the point is give me complete control over my device. Let me decide how I can play cat & mouse with my carrier.

  92. >Lump together, the point is give me complete control over my device. Let me decide how I can play cat
    >& mouse with my carrier.

    You do have complete control over your device. No one is going to arrest you for modifying your iPhone in any way you choose (well, the FCC might have some complaints, but that’s no matter what phone you use). Nor is anyone going to force you to update the OS on your phone with a mandatory over the air update (I’m looking at you motorola droid)

    What you and others are arguing for (and the point I’ve been driving at) is for a simpler, easier and more “approved” way to do it. In other words, support for unsupported actions.

  93. > You do have complete control over your device

    Prisoners have the natural right to break the steel bars and escape, but that doesn’t mean they have the control to do so.

    Making something difficult enough, is the same as slavery.

    I am not going to provide my limited years on earth in support of any man who enslaves unnecessarily. Period.

    Do not do stupid things to alienate even a fraction of your developers if is not necessary.

    Any way, Jobs is free to do what he wants. I am not one of the people complaining. I just think he is shooting his own foot.

    But I am tired of this blog because I see how ignorant people are, they think they know math, but they don’t know how to apply it. It is best to let people make their own mistakes. Evolution determines who had the best ideas. See you in the market place.

  94. >Prisoners have the natural right to break the steel bars and escape, but that doesn’t mean they have
    >the control to do so.

    I know you probably won’t read this, but I couldn’t let this stand. A prisoner who escapes prison will find themselves hunted down, punished further and returned to jail. An iPhone user who jailbreaks or unlocks their iPhone will not. No legal authority on the earth will arrest you for jailbreaking your phone. If you happen to be dumb enough to take your jailbroken phone to Apple for service, no employee will erase it and restore it to original software, nor confiscate your phone and force you to buy a new one. There are no over the air updates forced to your phone and no matter how many times you plug it into your computer, iTunes will not detect your jailbroken phone and wipe it out.

    In short, your analogy does not at all parallel the situation.

    >Do not do stupid things to alienate even a fraction of your developers if is not necessary.

    This sounds an awful lot like “the customer is always right” and while that is usually a good business maxim, we all know the customer is also frequently wrong, and sometimes it isn’t worth doing what it takes to bring certain customers on board. I’m sure even our esteemed host would agree that doing the things it would take to not alienate say RMS as a developer is not always worth the cost. Clearly there are times when for the sake of expediency, practicality, or simply politics it becomes necessary or the lesser of evils to alienate a fraction of developers.

  95. > A prisoner who escapes prison will find themselves hunted down

    Strawman. The prisoner may feel he was wrongly prosecuted by a corrupt DA, kangeroo court, victimless crime, etc. The relevancy is that if the prisoner has sufficient control, he can restore justice. An example of sufficient control would be a bankrupt society unable to enforce its bankrupt and corrupt system.

    I do not think much of jails. I would prefer to let armed victims exact punishment without fear of retribution by the state. Any way, I am apparently censored by ESR and unable to comment on his pessimistic anarchism blog. I was going to provide a link to the math of socialism, wealth, etc at my forum/blog. I guess he (or his automated filters) doesn’t like competition. Any way, it is okay, this blog is intellectually fun, but also a way to waste time that could be used for programming.

    > This sounds an awful lot like “the customer is always right” and while…

    I understand your point and I must admit I am not privy to all the issues that Apple likely sees. They probably figure they can’t beat Windows Phone on integration any way.

    They just won’t get me to invest in their APIs if they don’t demonstrate the things I need to see which show the APIs will propagate eventually to open source and live a long time. I am done with programming to niche markets. Windows Phone 7 and/or Android will destroy iPhone on diverse integration.

  96. btw, i wrote a long reply because it was censored. Any one who steals my time, gets my to eat my dust.

    ESR says: Nothing has been censored. Some apparently duplicative messages landed in the spam queue and I deleted them; I tried to let all the content through exactly once.

  97. > A prisoner who escapes prison will find themselves hunted down

    The prisoner may feel he was wrongly prosecuted by a corrupt DA, kangeroo court, victimless crime, etc. The relevancy is that if the prisoner has sufficient control, he can restore justice. An example of sufficient control would be a bankrupt society unable to enforce its bankrupt and corrupt system.

    Note, social commentary was deleted to try to get this through this blogs filter…

    > This sounds an awful lot like “the customer is always right” and while…

    I understand your point and I must admit I am not privy to all the issues that Apple likely sees. They probably figure they can’t beat Windows Phone on integration any way.

    They just won’t get me to invest in their APIs if they don’t demonstrate the things I need to see which show the APIs will propagate eventually to open source and live a long time. I am done with programming to niche markets. Windows Phone 7 and/or Android will out compete iPhone on diverse integration.

  98. >Strawman. The prisoner may feel he was wrongly prosecuted by a corrupt DA, kangeroo court, victimless crime, etc.
    >The relevancy is that if the prisoner has sufficient control, he can restore justice. An example of sufficient control would
    >be a bankrupt society unable to enforce its bankrupt and corrupt system.

    Either I’ve had too little sleep recently, or your talking about something completely different from me. You have complete control over your iPhone hardware because no one has the legal authority (nor the time or inclination) to force you to not do it. A prisoner does not have complete control over his situation because the state has the legal authority and the guns to force him to do what they want. This has nothing as near as I can tell to how the prisoner feels or how much corruption there is in the system, the point is when it comes to iPhones and you, you have complete control, Apple just doesn’t make it convenient.

  99. Making something difficult enough, is the same as slavery.

    You know, this kind of overwrought hyoe really doesn’t support your position. If you don’t like the product, don’t buy it. If you don’t like the terms of a contract, then don’t agree to it. It doesn’t mean that those who make a different choice than you do are slaves.

  100. Comscore just published data for July, showing a healthy growth for Android, but Apple is still leading:
    http://www.comscore.com/Press_Events/Press_Releases/2010/9/comScore_Reports_July_2010_U.S._Mobile_Subscriber_Market_Share
    These are US figures.
    Figures for Europe are vastly different, as Nokia still dominates the market this side of the pond, and there are still three iPhone users for each Android user:
    http://www.comscore.com/Press_Events/Press_Releases/2010/9/European_Smartphone_Market_Grows_41_Percent_in_Past_Year

Leave a Reply to Mike Swanson Cancel reply

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