Android remains the only mobile OS still patterned after Microsoft's 2002 vision of the mobile value chain

Belfiore wasn’t shy about criticizing Google’s Android OS. Even though Google currently dominates the mobile OS market, its strategy of licensing the Android OS to manufacturers is similar to Microsoft’s previous approach with Windows Mobile: It’s open-ended, and there are few restrictions on how manufacturers can use or modify the OS.

As a result, Android is suffering from some of the same issues as Windows Mobile did: Android works better on some phones than others, manufacturers are shipping different versions of the OS on different phones, some Android phones are shipping with bloatware made by carriers, and some app developers complain that it’s difficult to make software because of the hardware and OS fragmentation.

via How Microsoft Hit CTRL+ALT+DEL on Windows Phone | Gadget Lab |

Android was initially an attempt to commoditize Windows Mobile with a symmetric but “free” alternative. The idea was to conform to OEMs and operator business models. This was directly copied from Microsoft’s early 2000’s playbook for entry into mobile.

Trouble is that Microsoft realized their purely modular, operator and OEM centric approach was failing to be competitive on the user experience and decided to move into a more integrated approach, closer to Apple’s integrated model.

But while Microsoft dropped their vision, Android was still patterned after their approach.

  • One can argue that Microsof'ts Window Mobile problem wasn't so much fragmentation but poor software.

    In any case, Google can always take control of the Android but re-visiting their own hardware option but this time sold via Carriers.

    • asymco

      I would bet on it. In another product cycle, the goal posts for what we consider a mobile device will have shifted yet again and the Android ecosystem as it is today will not be competitive

    • Joe_Winfield_IL

      Google could do this, but they would have to take a HUGE step backwards before moving forwards. They would lose the support of the current OHA partners, and would have to choose a device manufacturer to receive "favorite nation" status to manufacture white label phones. Google would have two choices – either go with one of their current partners or design in house and contract to a Foxconn-type builder.

      If they went the Foxconn route, Google would destroy all goodwill with existing OHA partners. This is exactly the same thing that Microsoft did by abandoning WinMo – a fitting end to the cycle of Android as emulation of WinMo. This action would not put Samsung, HTC, Motorola, et al. out of business; rather it would create adversaries where partnerships currently exists. These guys would flood the market with WP7 devices for lack of a better alternative.

      If they chose a current partner, such as Samsung, they would be orphaning all existing devices from competitors. They would also be constrained to the manufacturing capacity of a single OEM rather than the current array of OHA manufacturers. The OEM would also likely continue to build and support their own devices in parallel to Android, taking the economies of scale for themselves.

      Either way, this would confuse customers, contradict dozens of statements by management, and cut dramatically into market share. Google only really cares about market share today as a licensor with an advertising based revenue stream. The company would have to abandon its own DNA to achieve your vision, and Android would no longer be a strategic fit to complement its main search business.

      • Absolutely spot on comment, Joe. Google and their partners simply have conflicting interests. While Google tries to standardize to avoid/reduce fragmentation, the partners need to differentiate, which creates it.

        Without differentiation in product features they would compete mostly by price, which creates a market with razor-thin margins. We will see this soon with WP7, and eventually with Android 3.

        The big advantage of Apple is that they don't have to balance that out, and can consequently keep their margins at record level, while providing great and consistent user experience to their users.

      • dchu220

        I don't think Google has the support of it's OHA partners and vice versa. Google did very little publicly to show it's support for HTC when it got sued by Apple and I doubt the OEMs care much for Android except for the fact that its the best option they have right now. The moment sales slow, the OEMs will jump ship also.

        I wouldn't put any move past Google. They still haven't shown that they know how to manage a brand or that they have any loyalties to their partners.

      • Joe_Winfield_IL

        I'm sure the OHA partners were terrified of Google at first. As great as free software is, Google didn't have any financial incentive to devote resources to Android if it never caught on. However, everything changes when the numbers get big enough. Google now has $1 billion reasons per year to support their mobile platform, according to their latest quarterly filing. The market is growing explosively, and Android is at the forefront. Rate of growth will begin to slow, but actual unit numbers won't drop any time in the near future.

        The HTC suit is another matter. Google isn't going to indemnify its OEMs. Ever. The IP on Android is sketchy, but even if it were air tight, Google would be foolish to give blanket support to open source code that is given away for free. Google's liability would grow exponentially with very little opportunity to grow revenue correspondingly.

        Again, Google's only business is advertising. Everything else they do is in service of their ad business. Maps – great way to serve local ads. Youtube – great way to grow in display. Search – the foundation of the company. GoogleTV – the first step toward mainstream media buying. Chrome – a browser with Google Search built right in. Android – a giant funnel to all Google services, and the best hope for market share in the developing, non PC owning, world. They don't need to have strong brand management skills to see that upsetting the OEMs would set the business back. This shared interest creates a symbiotic relationship with their partners that naturally wouldn't exist. There will always be friction, but until a better choice emerges, there is plenty of loyalty among OHA members.

      • dchu220

        The OHA partners weren't scared of Google. They were scared of Apple.

        The HTC suit is a big deal even if Apple loses. Yes, it's the smart thing to do by not protecting your manufacturer's, but not standing by your code is just as problematic. It decreases the leverage that Google has to push things through in the future. Business is still done through relationships. You won't read about that in a 10-K filing.

        A perfect example is how content providers are blocking GoogleTV and I don't see them changing their mind anytime soon. The economic models are completely different. Content providers leverage their best shows and channels by bundling access with their average shows. You want ESPN? Well, you have to order ESPN3 as well. The same goes for buying ads. Until Google can show the networks that can make as much money with their model, the content providers are not going to join.

        I would disagree with you about brand management. The iPhone would never have happened if Steve Jobs couldn't convince AT&T give them control over the content on the phone. The fact is that many industries are not controlled by the consumer, but by entrenched industry power brokers. You've got to have massive selling skills to get these guys to change their ways.

      • Joe_Winfield_IL

        Why does it have to be binary? Of course they were scared of Apple, but that doesn't exclude a fear of being abandoned by Google. My kid is afraid of both zombies and ghosts.

        I don't think we are really taking opposite sides of an argument. I agree that Apple much more closely manages its software, hardware, distribution, and brand. I agree that it is hugely advantageous. I commend almost all of Apple's business practices, and I recently doubled down on my investment in AAPL.

        However, I stand by my previous comments. Google has a massive disincentive to abandon their OHA partners. This is an important differentiating factor between Android and WinMo. MS had already collected their licensing fees on WinMo before killing the platform, so it didn't matter much if the users were left with unsupported devices. Android derives its revenue from continued use of Android products. The cash flow on Android is much slower, and as a result Google is required to do everything possible to enhance the UX over the long haul.

        I still see an inevitable friction between Google and the OEMs. Google has to support the OEMs, but the opposite is not true. Oddly, I think the handset makers have all the leverage. They can walk from Android to another platform, or they can modify Android to the detriment of Google's unification efforts. Google needs volume to succeed, but the OHA partners don't need Android in the long run. They are mercenaries that allow the consumers and carriers to dictate the terms, and if the terms change to Android's detriment, so be it.

      • FalKirk

        "Google now has $1 billion reasons per year to support their mobile platform, according to their latest quarterly filing…"

        That billion dollars you're talking about were projected numbers based on one quarter's sales. Very speculative. Let's not count those dollars until they come in.

      • Joe_Winfield_IL

        You're correct, it is speculative. But I tend to think they will get there will today's growth trend. Ultimately though, it doesn't matter if it is a real number. As long as Google THINKS the revenue opportunity is real, they have incentive to support Android.

  • FYI. For some reason this last blog post only shows in Firefox on my PC but not on Chrome??

    • Joe_Winfield_IL

      I've got it on Chrome no problem. But I've noticed that sometimes Chrome is a bit buggy on my PC and a reset does the trick. I had the same problem with today and Facebook last week.

    • Fritzlan

      Comes up fine on my iPad.

      • Marcos El Malo

        lol. I'm sure this is not the reality, but I couldn't help but imagine you feeling smug as you typed that. 😀

  • timnash

    Since Microsoft was always going to push its browser and search on Windows Mobile, Android was a sensible defensive strategy for Google to stop being locked out of the mobile advertising market. It therefore made sense to adopt Microsoft's business model and, given the number of ex-Microsofties now in Google, it is unlikely to change.

  • yuri

    I've read some of your Android related articles and I think that you are consistently missing one aspect of the competition between the platforms – the point of view of the niche application developer.

    Very soon all smartphones will offer the same hardware and software basics. I'll leave the hardware aside. On the OS level we'll have more or less the same features. There will be some differences in UX and though some will be passionate for their UX preference, most users won't care one way or another as long as they feel comfortable with the UI. The mass market applications like Skype will be available for all platforms, so no differentiation there too. The battle will be decided by the long tail of niche applications, because many users will need one or two of those and their needs will vary – just like on the desktop. Whichever platform catches that long tail will be the dominant one.

    For small software developers or companies whose business is not focused on mobile it will be prohibitively expensive to support all platforms. They'll have to pick one or two platforms that give them the best bang for their buck. And the situation there is pretty clear – Java wins hands down and will probably be followed by Qt in one or two years. Java is currently the main development platform for Android, BlackBerry, it is a first class citizen on Symbian, it is also preinstalled on all Samsung OS devices and others. The fragmentation problem between Android and the rest of the J2ME supporting devices is big, but it is significantly smaller than having to maintain a complete rewrite for iOS only. The talk of inter-Android fragmentation is complete nonsense on the global fragmentation scale.
    Qt will be the main development platform for Symbian and MeeGo. An iPhone port is on its way and will probably become quite potent in one or two years. I think it will be possible to create a Bada OS port too. Windows Phone 7 is out of the picture for Qt, because of their no-native limitation.
    I am not considering the Web platform, because it is too young and for medium to large applications it is a no-go.

    The bloatware problem is a minor one – all Google has to do is put a clause in the compatibility requirements that all applications should be removable. Then somebody will write an anti-bloatware application and anyone who is disturbed by those would be able to remove them easily. I don't think that the majority will care though.

    • DB2010

      "all Google has to do is put a clause in the compatibility requirements that all applications should be removable."

      Then the MFR. can just switch to a more compliant OS. Which, might be more profitable. In the long run Android is at the mercy of the MFR and carriers.

      • kevin

        Even f I write an app for Qt, it will not perform the same on a low-end smartphone and a high-end smartphone. What do I optimize it for? If I optimize for the low-end smartphone, a competitor could write a similar but more capable app to take advantage of the hardware on the high-end smartphone and not only take my business but also my more affluent customers (who shelled out for the high-end smartphone). If I optimize for the high-end smartphone, then a bunch of features might be unusable or poorly usable on the slower, low-end smartphone (that might also be missing some sensors). What's your solution as to how I keep both user experiences top-notch? Or do you really don't get that user experience is what counts?

      • kevin

        My reply was not to DB2010 but to yuri. Sorry about that…

    • I'm not sure I understand.

      You say a small developer has limited resources, and therefore should pick one platform to target. You then go on to argue that a small developer should target as many Java-using platforms as possible.

      You seem to think that there is no additional effort involved to do so. Do you think all mobile platforms that use Java are equally easy to program for/port to? Java or not, you're not going to straight-port anything between Android, Blackberry, and Symbian.

      And bang-for-the-buck speaking, consider the iOS App Store: I don't think anyone argues that it is the top distribution channel in terms of sales and availability to consumers. It's probably the best single platform anyone could pick to develop for, exposure-wise.

      Do you believe that every Java-using platform, combined, can outpace the size and scope of the iOS distribution channel? Even if you do, is the advantage worth the task of still having to maintain ports for widely different Java environments to do so?

      Your mention of Qt is also odd. You present it as another way to target the widest number of platforms, but Qt will only be officially deployed on one ecosystem (Maemo/Symbian) and hasn't even been rolled out yet. And again, it will not be available on Windows Phone or Blackberry. Even giving Android, considering that Qt uses C++ and not Java, that contradicts your argument for the use of Java-ubiquitous platforms anyway.

      You and I reach opposite conclusions. I don't see Java as a zero-friction mechanism to deploy across as many platforms as possible. I'm not sure anyone ever used it as such, despite how it was advertised for such a purpose.

      • yuri

        You are still thinking in mass market terms. The basic idea of any app store is to target the mass market. Many vertical (niche) applications will be distributed directly through a web site, or will be installed by company's tech department. The iPhone and Windows Phone are out just on that requirement.

        I never said there is no cost to porting from Android to BlackBerry, but this cost is much smaller than a complete rewrite for iOS.

        Generally speaking any platform choice depends largely on what is the nature of the application and what is the target audience. My current focus is on applications that are not mass market and generally don't care about any App Store. Let's look at several such cases and the reasoning behind the platform choice.
        Private business applications that need to connect to company's server applications will favor Java based platforms and Windows Phone, because the company's dev team is probably using one of the two for the server applications.
        Field oriented applications, say for the geodetic or forestry market, will favor platforms that have a good choice of rugged devices. If you count WinMo 6 as dead, we are left with Android and J2ME devices. Speaking from experience, what happens here is that many people are just fine using their regular phones, if they can, instead of buying a rugged phone, and those who don't have a compatible device, have no choice but to buy one, rugged or not. The end result is that the platform wins, not necessarily the rugged phone manufacturers.
        I have a more hypothetical, but not so niche example. If you take a fitness center in Europe that already has a web site where its members can set up a training program and update their daily progress. The web site is also viewable on mobiles so that members can work with it while they are training. The next thing they want to do is to provide a mobile application, which will allow their members to track outdoor activities. Extending their mobile web site is not an option, because of potential coverage problems and lack of maturity of the web platform. They take a poll to see what devices their members use and after removing all devices that don't support GPS, they get say 38% Symbian, 15% iPhone, 15% Android, 10% Samsung touch screens, 10% BlackBerry, 8% other J2ME devices, and 7% WinMo, Palm, etc. They can address 70% of the market with one application at 1.5X cost or 15% of the market with X cost. (I am excluding the long tail of J2ME devices). They are not interested in any App Store because if you type 'fitness' you get like 100 hits, and they anyway want to target that application at their members and web site users. They also prefer that the application can be installed through the web browser…
        So they build a Java based application and they promise their iPhone users that next year they'll have an iPhone version. Just that next year half of the iPhone owners have moved to Android and then an iPhone app won't pay itself anymore. So they decide to invest instead in a Qt app to improve the UX of the largest group and hopefully be able to address iOS eventually.

        As the scenario illustrates, by supporting 4 major Java platforms (Symbian, Android, BlackBerry, and Samsung) you can reach a very wide audience. The porting cost between those is not very high. Also, I don't see Qt as an option if you want to target the widest possible audience, only as a second-best choice after Java, though a distant one at present.

      • So you believe that "the porting cost between those [Symbian, Android, Blackberry, Samsung] is not very high." That's fine. I just wanted to ask the question.

        Your examples are a small-business custom built app, and a local app for a local business. You are correct; I incorrectly assumed you were talking about the mass market.

        But you must admit…there's no business in that. If you're not actually out to sell your software to as many people as possible, then it doesn't matter what platform you pick at all as long as it satisfies your customers.

        As to your hypothetical fitness studio, I pity them if they really decide to make four or five different Java ports of their hypothetical app. Especially since now there's no actual financial incentive to do any of that work, since it *is* a very niche market.

        You also aren't clear on whether these apps are being written from scratch or not. If they are, then Java and Objective-C are equally accessible; of course if you have a pre-written Java app, you'd probably want to continue in Java.

      • yuri

        Who says that the online services of a fitness center have to be cheap and won't be profitable. One key lesson from Apple's iPhone success is that the right product at the right price will sell well.

        Then you should also take into consideration, that the fitness center will most likely hire a software shop that already has expertise in mobile GPS applications and some part of the functionality will come at lower cost. The custom logic like communication with the existing web site or integration with the training program rules will have to be custom built. In the case of Java-supporting devices all of this can be implemented once and reused across all of them. Depending on the requirements they may either use a custom-drawn UI (LWUIT) or go for the platform specific UI. In the first case the UI logic will also be quite portable between the Java devices, in the second the cost will be higher, but it is still easier than maintaining two completely different code bases.

        And no, Java and Objective-C are not equally accessible – having to do memory management adds development overhead. The exact overhead will vary from developer to developer, but software shops are not exactly full of star software developers.

      • If they're going to hire out to a software shop, again, it depends on what platform they want to target.

        If they're really going to pay the money to have the app developed, they'll have to find some way to make it financially feasible. If you're saying they'll choose a Java developer because 1) their customers use more Java phones than iPhones, 2) there are more *good* Java developers than *good* Obj-C developers, and 3) the developer house will be reasonably priced, then yes.

        It all depends on what platform they target for their customers, what budget they have to work for, and what compromises they have to make. What platform they choose must make sense based on what their customers actually use, not how accessible Java is.

      • TomCF

        iOS explicitly has a way for corporations to roll out software that is not tied to iTunes. I expect Windows Mobile to do the same.

        Small businesses wanting a custom application will (if they're smart) simply ask their customers what kind of phone they're using, and get an app built for their customer base.

        If they want something multi-platform, they'd be better off with a mobile optimized web app, since that'll cover all smartphones.

        Using Java when that doesn't support one of the top 2 or 3 mobile platforms doesn't make any sense.

      • Ted_T

        The thing is, user facing businesses that have smartphone apps, tend to have iPhone apps and only iPhone (and iPad) apps — be they fitness centers or in this case museums:
        Why? To quote Al Capone: "That's where the money is."
        And bonus: their developers don't have to worry about cross platform compatibility.

        Good luck with your project, but you might want to consider who the target audience is. Beyond shear numbers, the iPhone offers customers who are the most likely to install third party apps and the best demographic in terms of disposable income (see Al Capone, above).

    • Joe_Winfield_IL

      If I'm reading this correctly, you are proposing that in one or two years the application publishing business will be commoditized via Java and/or Qt. There are a number of unforeseeable factors that can impact this vision. Apple only recently decided to allow ported applications, and there is nothing to prevent iOS, Symbian, Android, or anyone else from rebuilding this type of wall. Also, you were very quick to dismiss the web itself as a universal platform. We all know that HTML5 isn't yet fully baked, but a smartphone without a browser isn't a smartphone. At a certain point though, this will likely become a truly universal standard.

      In the interim, I would think that revenue is the single driving force in a niche developer's decision about which platform(s) to support. This site tends to underweight Android because there are mountains of data that suggest Android to be a smaller revenue opportunity than iOS, regardless of market share. For all of the negative press thrown at Apple's approval process, to date Apple is the only smartphone provider (either OEM, carrier, or platform owner) to have successfully created a consistent, trustworthy storefront with a single payee for all transactions. The products are all publicly and transparently rated, there is a good discovery system, and there is no malware or pirated software. As a result, the lion's share of iOS users are willing to store and use credit card accounts on Apple's servers. RIM, Windows, Symbian, etc. have and will continue to try to replicate this success, but none have hit critical mass to get developers pouring in. While Android has hit this mass with both users and variety of apps, they have not engendered the goodwill and comfort required for massive volume of purchases. This is not an easy problem for Google to remedy, as a techincal solution will not necessarily affect the psychology of customers.

      • yuri

        You are talking about small developers targeting the mass market. This may be the rage now, but very soon the mass market will be dominated by several big companies and the majority of small developers won't be able to survive there regardless of which platform they target.

        The niche market is the one that by definition does not want to distribute through an App Store either because your application will be berried among 100 others that look the same, either because your application is not for public use.

      • Tim F.

        Your distinction between mass market and niche are absurd. A vertical doesn't discover a small app perfectly tailored for their industry through ESP. There is no reason to avoid an app store because it will be buried — one among hundreds of thousands is still better than one among hundreds of millions of web sites. I think you eliminate iOS and WM7 because of the App Store/iTunes management (unsure what you are really gettting at), but I've been in several large cap companies distributing specialized, vertical apps within their own infrastructure FOR iPHONE. There is central IT administration. You keep saying complet iOS port, which tells me you are already creating Java mobile apps. If so, just say that — I've been making niche java mobile apps and I'm going to keep doing so. But your own tendencies are not a proof that after losing the mobile market, Java is going to somehow win it back again. Not happening. It had its chance in mobile in 2000.

      • yuri

        Initially I was including small apps as potential niche apps, but actually I don't think it is possible. A niche market application has to cost more to make profit, say more than $100 per year. A small application will have hard time convincing clients to shell such money. So you either have to beef-up your application or target the mass market.

        I am not saying you *can't* distribute through an App Store, just that in many niche cases you don't have any reason to, and you wouldn't want to.
        In my particular case the Internet is not going to be the discovery mechanism, and once we've established contact with prospective clients, we would want them to go to our web site, instead of pushing them to an App Store where they may see competitors too.

        Java may have lost its total dominance, but the need for low-cost cross-platform development is still there and it will be a factor in the platform wars. I would have loved for WebKit to be that platform, since it practically has the whole mobile market, but it is not there yet and I am not sure it will ever be. Microsoft was killing Web technologies before because they were afraid to lose their Windows dominance, now Apple would stop any new technology that would come to a Flash or Java level of productivity, because they want to build their own app ecosystem… From where I stand, however, their's or Microsoft's mobile proposition does not look good. Apple has neither a low-cost device, neither a rugged one. One may argue that mine are very specific requirements. The point that I am trying to convey is that a big part of the market will have some specific requirements covered by some niche application. Niche applications would often need a specific hardware solution, or specific distribution needs, and Apple is not covering that space at all. Android on the other hand is everywhere. The niche market has interest in its success and generally the success of open ecosystems that will allow reduction of development cost through a unified development platform, be it Java, Qt, or Flash.

      • Simon

        I don't agree with your conclusions. Apart from details such as your gym example, where you say that the internet/gym web page would be the natural discovery mechanism whereas I think the app store/market would be simpler (by inputting the name of the gym), the whole thing about ease of use for developing for different OSes is flawed:

        You claim that by targeting Java you would get Symbian, Android, Blackberry and Bada, hence covering a majority of the phones with little work. While true, the cost of supporting all of them is not negligible since they all have completely different GUI kits, thus necessitating you to write separate GUI code for all 4. Furthermore, on Symbian you would not get the native look (which requires Qt), so your app would likely look & work suboptimal.

        But if you still think this is a reasonable way to define what to target, I present to you C++. This enables you to target Symbian (Qt is C++), iOS (C++ works fine here as well), Android (using the native SDK). You would of course have to code the GUI separately here as well (using Obj-C for iOS), but that is apparently OK to you.

        So no, Java is in no way the natural choice (unless you are already familiar with it and don't want to change). And as I said, I really think app stores is far simpler way to both distribute and find applications.

        As for being buried and thus only big-name devs remaining in the store, I think you are wrong here as well. I have a small, geeky app in the AppStore which sells appr. 5 copies / day à $2 -> annual income about $2000. Not getting rich, but a fun hobby, and the niche is so small I cannot see a big developer entering. So I'm way above your $100 threshold, and no signs of sales dropping off (the app has been in the store since opening day).

      • yuri

        No argument against Qt – I'll be happy to see it targeting all main platforms. I am not sure Qt will be able to support Android, since all main platform functions above the kernel are implemented in Java. Maybe it will be possible to do an NDK-to-Java bridge. BlackBerry is also a problem – their OS is also implemented in Java AFAIK, and they don't have anything similar to NDK.

        As this conversation evolves, I am starting to see small apps as non-deciders in the platform wars. Its great that you are making money with your application. There is also the advertisement option, which I didn't consider and as a monetization option may very well pay better. The thing about small applications is that they are easier to replicate across all the platforms. If you don't port your application, somebody else will. The net result is that anything found in the App Stores will not decide which platform wins. The software outside of the app stores will be a decisive factor, though.

        The application I am working on is not going to see an iPhone version in the foreseeable future, because it is complex enough to make a port prohibitively expensive. And our clients are ready to base both their device and carrier choices based on what we require. Going with Java we can yet address 70% of the smartphones out there and this is perfectly fine for us. Maybe in two years we'll have to revise that, but Android (and BlackBerry) will already have their native versions regardless of what we decide then. Also, due to their superb Java support Nokia have been the main benefactors of our platform choices so far.

      • berult

        Your train of thoughts reflects a Business Plan more than unbounded logic. It is circular, hence impenetrable and easily defendable. And it is flawed in that it is based purely on what you wished and intended Apple were not.

        You can build promiscuous niche partnerships all you want with Java and Flash, special needs induced by specious feeds "templates" yours and Google's greed. You're both ideologues of market differentiation based on opaque substantiation. Make it as geek-like complex as necessary for your aims to be digested as comfortably ethereal, and you get a fail-safe, unApple Business Plan.

        So you wish it to be…

      • kevin

        Obviously, you can choose whatever audience you want to focus on. But up above, you made a big assumption: "Just that next year half of the iPhone owners have moved to Android and then an iPhone app won't pay itself anymore."

        That is just so unlikely to happen when iPhone owners have 90%+ satisfaction rates (with a majority of that being very satisfied). Surveys have shown iPhone owners are more engaged with their phones and their apps than any other phone. Plus worldwide demand for iPhone, four months into its model year, still hasn't abated.

        Simply then, you makeplans based on that assumption at your own peril.

      • OpenMind

        You already made decision and investment according to your assumptions. Now you want to justify your decision and investment with your wishful market forecast, even though hard data point otherwise. Good luck to you and your employer.

      • yuri

        I am sorry guys, but my thinking is not wishful. Here are some hard facts that have little to do with my platform choices:

        From my new (uninfluenced) clients there are 0% with iPhones. Most of them have Nokias – from the most expensive smartphones to the low end models.

        And some charts from an iPhone-centric site:

        The way I read those charts is like that: Java-supporting platforms have more than 70% of the market, iOS is flat, Android is skyrocketing.

        My logic focuses on my business needs and my choices are right for those needs and they are far from based on wishful thinking. What you fail to see is also very simple – Apple is one company with one device, it simply cannot shine everywhere. But it behaves like it can – high walled garden, arrogant attitude towards cross-platform development. It does so at the peril of its own market share. It doesn't mean Apple will stop being profitable, but most likely less so if they don't change something.

      • berult

        Here is what reality postulates on success and failure in mobile computing: if you can make it with consumers, you can make it with consumers. Cherry picking needs not apply. The trickling down effect from Apple enthusiasts to mainstream consumers to small businesses to large corporations has to work both the clock and the referee; "quality" usage time is of the essence here.

        Argumentative story-telling based on washed out clichés such as  

        "…What you fail to see is also very simple – Apple is one company with one device, it simply cannot shine everywhere. But it behaves like it can – high walled garden, arrogant attitude towards cross-platform development. It does so at the peril of its own market share. It doesn't mean Apple will stop being profitable, but most likely less so if they don't change something.",

        reads as if it had been pulled out of a wishing well in Mordor, for the story line is not being borne out by " off the charts " consumer and mindshare actual and projected metrics. You may feel ios takes you for a ride, but why wouldn't it, as it owns both the horse and the sleigh ?  

      • kevin

        Since none of your clients have iPhones, and most have Nokias, you're likely justified in the choice you make for now. Clearly your clientele today is in a different portion of the market than iPhone. Fine.

        So there really was no need for you to defend your choice by making assertions about iPhone. But having made those assertions about Apple, you should also ask yourself what data it would take to change your mind. People said most of what you said almost last year, and yet Apple has increased smartphone unit market share year-over-year for every single quarter since 2Q08 (its first comparable yoy quarter). That's 10 straight quarters. NO OTHER HANDSET MAKER CAN SAY THAT. How many more quarters of this pattern need to occur before you question your assumptions or ideologies?

      • Marcos El Malo

        Furthermore, there is nothing stopping an iOS developer from using the gym (U.S. for fitness studio) as a discovery mechanism. All the gym needs to do is provide a link on it's webpage to the app's location on the iTunes store. There are hundreds of businesses already doing this. I'm not sure why Yuri isn't aware of this.

        There are some downsides to relying on the iTunes store, but discoverability isn't one of them.

    • Steko

      "Whichever platform catches that long tail will be the dominant one. "

      Your whole argument hinges on this questionable premise. It's far more likely that no carrier will be dominant short of some crushing IP court victory that outright excludes competitors.

      • asymco

        I still don't see evidence that long tail is not addressed by iOS (or Android). The number of applications built on iOS is the largest number of software products that has ever been built on any platform. Developers also have the option of building web apps, an option which has largely been ignored but is there just for those "niches" that don't require native APIs.

    • dchu220

      Isn't the app store, with 300,000 apps the very definition of the long tail?

  • Iphoned


    “I would bet on it. In another product cycle, the goal posts for what we consider a mobile device will have shifted yet again and the Android ecosystem as it is today will not be competitive”

    What kind of shift
    are you referring to that Google can’t evolve android to meet?

    • !iphoned


      Do you not know how to click on the Reply button so that the conversation is contained rather than scattered all over the page?

      • kevin

        Currently, if you reply from an iPhone, there is no reply button, so you can't keep the conversation contained. Don't know if that's what causing iphoned to do that, though.

        Horace, I think there is a way to put in the reply button — see what does for elmer-dewitt's column.

      • Kevin, it's not a button, but a link, and it works fine on iOS…

      • Iphoned

        No I don't.

    • asymco

      If you look back at the history of smartphones (or "mobile computing" in general), over the last 15 years there has been a remarkable amount of innovation and turnover in leadership. Everything has changed, from the operating systems, the business architecture, the network technologies, the form factors, the input methods, the incumbents and the business models. I suspect changes will not have stopped after the announcement of Android. The way Android is architected today does not allow for evolution in all the dimensions that change can happen.

      • dchu220

        It reminds me of this post that showed the profits of integrated vs open handset companies.

        The idea that the cellphone of today is going to look anything like the cellphones of 3-5 years from now is the same type of thinking that is killing Nokia. The fact that there is so much room to innovate in this space is what makes it so interesting.

  • Iain

    I think what people are forgetting is that Google doesn't charge for its OS! So unlike MS, I think if problems with the Android platform occur they can start again with a new version backwards compatible with the things Google wants it to be compatible with, and a more restricted license. The hardware operators can then carry the can and pick up development of their bastardized version.

    • dchu220

      So if I am an OEM who saw my research and development dollars go down the tank because Google decided that they didn't want to support Android anymore, why would I want to support anything else?

      If I can't clearly differentiate myself from my competitors, how am I going to make money? With such low margins, I would be putting my whole company at risk during a recession.

      Wish life was so simple, but that's why leadership and strategy are so important. Because you can't always start over and your reputation matters.

  • Pingback: Korta klipp – 09 November 2010()

  • But with Microsoft approach they can never compete with apple on price,
    the UI is interesting but I'm doubtful if it's enough to draw eager clients.

    As a result they'll have to invest ton of money just to shove their leg through the door,
    by than then the market will be shifted again.

    It's might be that Android strategy is misguided,
    but trying to compete with apple on desirability alone,
    sounds to me even more misguided.