Windows 8 Tablet Fragmentation and the App Dilemma

by Ben Bajarin   |   September 21st, 2012

I was having a discussion with an iPad software developer recently and we were discussing the iPad mini. Interestingly he was still a skeptic about the iPad mini and I thought his reason was interesting. He noted that the apps they develop, and primarily the user interface, are specifically designed for the current iPad screen size. He said that everything is placed where it is for a reason.

His skepticism about the iPad mini was based on his conclusion that if a 7-8 inch iPad was to come out his current app would not work. His point was that 2 or so inches may not seem like a big difference but for many apps that have menu’s and touch based navigation interfaces, 2 inches is a lot of screen real estate to loose. Basically he had concluded that for many applications developers would target each tablet screen size independently.

His points got me thinking. First of all I agree with him. If we have learned anything about Apple’s developers is that they are willing to take the time to make sure their app experience is ideal no matter what the screen size. The iPhone 5’s larger screen and app developers already starting to take advantage and optimize their apps for the new 4” screen. Interestingly Apple, during this transition, is faced with having apps with two different looks and feel in their app store for both 3.5” and 4” iPhones. So as developers look to tweak their UI for the iPhone 5 which app UI will we see in the app store? Apple is solving this elegantly but only showing consumers the new app UI for 4” apps only if they have an iPhone 5. That way consumers who don’t have the iPhone 5 will still see app preview screenshots of the 3.5” UI.

Now with all of that in mind let’s turn our thoughts toward Windows 8. In all the above examples I mentioned we were talking about screen size differences ranging from .5 to 2 inches of difference. And within that extremely small range we should expect to see developers uniquely tweak their app experience and UI. In essence they are not simply shrinking or expanding their apps to work on smaller or larger screens, they are in essence creating new app experiences for those screen sizes. Windows 8 touch based hardware will be so fragmented in screen size that we will see touch based Windows 8 hardware ranging from 10” all the way up to 27.” If developers feel the need to optimize their software for a screen that is anywhere from a half-inch and even a 2” difference, what will they do when they have 4, 5, or 6 different screen sizes to target in the Windows 8 touch hardware ecosystem? And more importantly will they feel that their energy and resources will be worth the investment and hard work?

Microsoft needs developers to be writing touch based applications but my concern with the touch based hardware fragmentation is that it will may cause them to target only specific screen sizes and not others. This would mean that the touch based software experience will be better on some Windows 8 hardware but not others. I can tell you right now that an application that is built for 10” Windows 8 hardware is not going to be a pleasant experience on a 27” all-in-one running Windows 8 with a touch screen.

Some categories, like games for example, may work fine within this fragmentation. However, it would seem logical that even developers of many of the popular games may want to make tweaks for larger vs. smaller screens that may run their apps.

The bottom line is that I expect developers who are looking to sell software to the masses to want their software to be the ideal experience on any screen size. To do that they will inevitably need to write software and create user interfaces that are specifically made for certain screen sizes. This is where I feel developers may feel the need to pause and truly evaluate the effort they put into Windows 8 touch based software.

You can make the point that the screen size fragmentation I mention has existed for decades in the Windows ecosystem. This is true but I fundamentally believe that when it comes to mouse and keyboard software and UI this fragmentation is not an issue. Because of the unique way touch based software UIs are made, I believe fragmentation becomes an issue when it comes to touch computing in a way it never was with mouse and keyboard computing.

I believe touch computing is the future, and so does Microsoft with the emphasis they are putting on touch. Microsoft’s challenge over the next twelve months is to convince developers who also believe in touch based computing that their platform is the one worth investing in.

Ben Bajarin

Ben Bajarin is a Principal Analyst at Creative Strategies, Inc - An industry analysis, market intelligence and research firm located in Silicon Valley. His primary focus is consumer technology and market trend research. He is a husband, father, gadget enthusiast, trend spotter, early adopter and hobby farmer. Full Bio
  • FalKirk

    Ben, fantastic insights. I was so inspired by your article, that I wrote one of my own entitled “With Apps, Size Matters”.

    • http://www.facebook.com/michael.noonan.108 Michael Noonan

      It’s so refreshing to read comments where people disagree with each other without feeling the need to hurl insults!

      • steve_wildstrom

        Thanks. We work hard at trying to keep the discussions civil.

  • Geoff
    • FalKirk

      When you design your app to run on everything, by definition you’re optimizing it for nothing.

    • benbajarin

      Hi Geoff, unfortunately nothing new there. EVERY platform has such best practices and tools. Yet app developers on Android and I expect even Windows now will still try to create only one app that works across all screens. From my several dozen interviews with iPad app developers there is a conclusion that this is simply not an ideal experience for customers.

      Also, I am absolutely convinced that not only trying to make one app package that works for touch across all screen sizes is going to be difficult so will it also be to write one that spans both touch and mouse and keyboard.

      SO again my fundamental point is that how the app performs and the ultimate experience with that app is going to vary within form factors. This is why you see iOS developers making custom versions of their app for different screen sizes.

    • steve_wildstrom

      You’ll notice that MSDN article does not really come to grips with the question of designing for the huge variation in screen sizes that Windows 8 will have to work on.

  • steve_wildstrom

    I’ve now seen, and in many cases tried out, Windows 8 running on touchscreen systems (tablets, laptops, and all-in-ones, plus one freestanding monitor) ranging from 10″ to 23″. The Metro UI works well on the smaller sizes, but degrades badly as the screen gets bigger (in a sense, that’s the opposite of all previous versions of Windows, which always seemed happier the more real estate you had.) The real problem is the full-page Metro apps, which are just silly on a big screen.

    I haven’t seen a dual-screen Win 8 setup yet, but would be interested in hearing from anyone who has tried it.

    • Mhe She

      Steve, Good point, plus I don’t understand why Microsoft is hell bent on putting touch for Big screen desktops. It is completely non intuitive to use tocuh on vertical big-screen desktops. They should realize their folly soon and focus on touch screens from 7 inches to 11 inches for tablets.

  • http://www.appleoutsider.de/ AppleOutsider.de – Sebastian P

    “I can tell you right now that an application that is built for 10” Windows 8 hardware is not going to be a pleasant experience on a 27” all-in-one running Windows 8 with a touch screen.”

    Based on what?

    I’m serious. You make this assertion based on what data?

    The sheer fact alone that nobody on earth would be able to wield a 27″ screen with one hand to enter information with the other results in the usage situation being a complete different one, namely propped up on a table. As a result, you’d be using the 27″ machine with BOTH hands. Of course it would be DIFFERENT but how on earth can you make the assertion right now that it would be NOT PLEASANT?

    I am not trolling here. I didn’t buy the iPad 3 because it is too heavy for me to use it more than 30 minutes without getting too tired to prop it up with my left hand while using it with my right. I am waiting for the next one and hope it will be lighter (i.e. the iPad Mini or a lighter iPad 4).

    Usage experience is a combination of everything. Weight. Screen size. User position. UI design.

    What Windows 8 needs is great designers. Simply stating that the usage experience would be bad just because the screens differ in size is as if you were saying that Windows and OS X suffer from different screen size. I doubt that you have ever programmed something in your life, for you would know about UI elements that are used as filler to adjust for variable width and height.

    Fixed width makes it EASIER to design, that’s a given. It also makes it easier to remember usage patterns for the user so they can pick up another iPhone 4 and instantly know where everything is. But the sheer fact that apple now offers devices with two different aspect ratios and with that two different resolutions in the iPhone 4 and will most likely soon offer a shrinked version of the iPad should’ve told you that different sizes soon won’t matter anymore. Sure, Apple is successful because they iterate in small doses, but saying that the next version of an OS that is installed on 90+% of all Personal Computers, of which more than 100 million were sold every year for the last decade (bringing the installed abse to what? roughtly five to eight times the installed base of the iOS environment) would suffer from widely different screen sizes is just… well. Let me put it like this:

    Have you read a Gizmodo article about Apple within the last three years?

    It reads something like that.

    • benbajarin

      My point was that an application built for 10″ would not be pleasant on 27″. I am not saying ZERO apps will be pleasant just that I know for a fact from interviewing developers of touch based computing applications that the philosophy is in line with what I just outlined here. I have also seen and spent time with many of the touch based all in ones and seen some of the software running for touch that was made for smaller screens. SO call it hands on experience.

      So for an app to be pleasant on each screen size it will need to be custom made for each screen size. Thus causing the fragmentation issue. And right now here are the screen sizes we will most likely see on touch based Win 8 hardware over the next 12 months.

      10.1, 11.1, 13.3, 14.1, 17, 21, and 27. I see this as a real issue when it comes to how developer of touch based computing applications think about the best experience possible in this touch computing based future we are heading in.

      If this is the first article of ours you have read, I strongly encourage you, (if you have the time and or the desire) to go through our archives and look at the analysis we have put forth regarding tablets at large and touch computing as a philosophy.

      And no I have not read anything at Gizmodo for quite some time as I rarely frequent tech news sites.

      Here is the link to our tablet tags and our Windows 8 tags in case you have interest.

      http://techpinions.com/tag/tablets-2

      http://techpinions.com/tag/windows-8

      • http://www.appleoutsider.de/ AppleOutsider.de – Sebastian P

        I see your point.

        What I tried to get across is that you are coming from the wrong direction. You think about the “touch experience” in the terms Apple defined, e.g. there’s this one product that needs to have programs optimized for this one screen size and you try to apply it to Apple’s newly defined product category of post-PC devices.

        But this is Windows. Windows was _always_ able to be used on different screen sizes and while I agree that the user experience will be _worse_ than the one on the iOS devices we all know and love, it will be just the same as it was in previous versions of Windows.

        IMHO it will be the same as the difference between Mac OS X and Windows. Windows 8 will definitely “suck more” compared to OS X, but I am really disagreeing that because in the past you had to use devices with programs that were basically made for phones (e.g. Android tablets that made you use blown up versions of the smartphone versions of programs) it is impossible for programmers to make programs that DO adjust just fine to larger screen sizes.

        In short: Windows (Microsoft) has a ****-load of programmers behind it. Just like they manage to offer programs that are usable on different screen sizes they will be able to make programs for different screen sizes on Surface devices.

        I kind of feel that you are “tainted” by Android device manufacturers offering tablets with Android 2.3, a phone OS.

        Furthermore I think we _need_ this approach, e.g. a company offering an OS that is not dependant on a handful of fixed resolution displays for the whole OS to be successful.

        Because I think the fixed-screen-size-optimized-programs market is pretty much taken by Apple, this way or the other. Your article was basically about how Microsoft can only succeed if they mimic Apple.

        And I doubt that that is the case.

        I think it’s more likely that the problem with a program not working on a 21″ device is the size of the device in the first place. I doubt that a touch device should be of that size at all. So maybe we kind of agree on that point – maybe Microsoft should’ve “restricted” the OEMs to a maximum size of screen but then again we all know that there would be some manufacturers who put Windows on anything anyway because the Hardware is standardized.

        I think it’s a false dichotomy to say that if an OEM that produces a device that is unusable Microsoft would fail. They don’t really care on what you put their OS as long as you pay for the license. And customers will always flock to those devices which ARE usable.

        Microsoft ist basically taking the natural selection approach whereas Apple is designing the whole experience (narrowly avoided commenting on religion there :-)

        I think both approaches can and HAVE worked in the past and just because Apple was successful with their approach IMHO doesn’t mean that it’s the ONLY way to success.

        • steve_wildstrom

          See my comment below for some experience credentials. I, and I am sure Ben too, am trying very hard to judge Windows 8 on its own terms, not in comparison to anything else. For one thing, since Apple does not offer touch on Macs, there’s nothing in particular to compare (although I did see a 22″ device running Android yesterday. It was a bit peculiar.)

          My feeling is the Windows 8 UI is impressive on tablets; much less so on bigger displays. The few apps we have seen so far are awful on big displays, but while they look good on tablets, the generally offer limited functionality.

          It’s all going to be up to the app developers, both Microsoft and third-party. Will they continue to develop Desktop apps or will there be a massive swing to Metro? Is there a way to make Metro apps much richer on large displays while remaining usable on small ones? We won;t know the answers to this until Oct. 26 and later, although I will say that if Microsoft wants to make a positive first impression, it had better have a flock of good Metro apps ready on launch day. Otherwise, no one will buy Windows RT tablets.

          • http://www.appleoutsider.de/ AppleOutsider.de – Sebastian P

            I agree completely. I ordered a Lumia in March and sent it back because it didn’t offer a way to track your data usage (not built in and impossible for 3rd party apps because there are no background tasks allowed like on Android) and because there simply was not a single usable client that syncs with Google Reader and offers offline usage. There are at least 10 on the store (née market) but they are all garbage.

            The only reason I currently use Android is that there are _enough_ usable Apps. But I will most definitely get an iPad Mini because I miss the games. And that’s where RT is going to win it big if you ask me.

            Theres a several billion $ market for Games on Windows and it has _always_ been the reason why I didn’t switch to a Mac (I’m in my 30s). That might change in the near future because of Steam.

            All there needs to be is Games and a usable version of MS Office and RT will be a success*

            *of MS proportions. Nobody should expect it to perform like iOS. Ballmer hasn’t been fired for a decade with revenue and market cap essentially having remained the same so from a standpoint of an analyst there’s simply no way at all for MS to make RT a hit because Apple is so enormously more successful. But just like back in the day I rooted for Apple I think it’s ok to root for MS now :-)

          • unhinged

            My first thought on a method to solve this is to have a system where the developer can specify ideal, minimum and maximum display size/resolution for the app. Windows 8′s “full screen app” mode could be modified to respect the maximum values and display the app in a separate window if required.

  • unhinged

    On reading this article, I found myself wondering why no emphasis has been given to the target market for Windows 8. Is there a specific segment of the overall market that will buy Win8 no matter what?

    My intuitive answer to that is the business world. They may delay for a while, but business has historically bought new computers on a fairly regular basis and eventually adopts the latest Windows version because they plan their training out a few years ahead and the training services are preparing customers for Windows 8 (or, for small businesses, they outsource support and the support company makes the move to the new version to ensure continued revenue).

    I would hazard a guess that the bulk of software written in the world today is for business use. And for a business, the variance in equipment is usually a lot smaller than for the consumer market. Big companies buy in bulk to get discounts, and that means standardisation.

    So if Windows 8 is going to be largely used by businesses, and there’s less variance between hardware in that scenario, the arguments about sub-optimal user experience don’t carry as much weight.

    If, against all my expectations, Win8 succeeds with consumers, then, yes, fragmentation is very likely to be a big issue. But then we run into the second of my objections: historically, Windows software in general has not been developed with the idea of a fantastic user experience as the overriding goal. Windows developers (and users), in general, favour the pragmatic rather than the artistic.

    If we accept all that, then I don’t see how the fragmentation issue is going to matter all that much in practical terms. Developers aren’t going to drive themselves to include an improved UX when their experience tells them that customers won’t pay for the extra effort (those users who _will_ pay have already bought into the iOS ecosystem). Customers who want the optimised experience are probably already iOS users and won’t buy Win8. Customers who don’t care too much about the UX or who are buying bespoke solutions won’t be bothered by a generic interface.

    Does this mean that Win8 is worth developers investing in? Well, the sheer size of the market pretty much guarantees that business as usual will continue for existing Windows devs. I suspect that we’ll see more dev shops keeping their Windows efforts as is, using the regular revenue to operate, and experimenting with iOS to see if there are extra profits to be made.

    • Mhe She

      Developers for consumer apps should target the 7 to 11 inch size.

  • Mark Handowich

    Reasently i found a great app called easy budget. Charts in this app helps me to notice where i spend the most. You can try it yourself.

    http://apps.microsoft.com/webpdp/en-us/app/easy-budget/1216179c-7b0d-4cbe-af71-a96e27ba535d