Android Fragmentation

Android is fragmented and geting more so. This is a challenge for those that develop on it for sure and has been often cited as a big negative for the Android ecosystem. But it also a big plus.

I have a Kindle Fire on my bedstand. I use it primarily to read on in bed having moved to a Nexus 7 as my primary tablet device. The Kindle Fire uses Android as its OS and then puts a Kindle shell on top which makes it look and feel like something other than an Android. But almost every app that I have on my Nexus 7 is also on my Kindle Fire. The reality is that if you build for Android, you are also building for Kindle Fire.

When Amazon launches a phone, it would be my expectation that the experience will be a lot like Kindle Fire. Meaning it will be running Android with a Amazon designed shell on top.

And then there is Facebook. I have to believe that Facebook will build a phone in the same way. Start with Android and then put its own wrapper and apps on top. If that happens, I would imagine I would be able to run all my favorite Android apps on the Facebook phone.

So imagine a world in which three of the top four consumer tech companies have phones running Android. Does that sound like a fragmented world for Android? Yes. Does that sound like a recipe for having a massive number of Android devices out there to build to? Yes.

In my view, we are in a two OS world for mobile and I think we are going to stay there. I think Apple will own the high end with the best and most integrated experience. And I think Android and its many variants will own the rest of the market. I think everyone else is playing for crumbs in terms of market share and would be better off joining the Android variant parade.

What does this mean for developers? It means build for iOS and Android and ignore everything else. And I think it increasingly means you have to be on both iOS and Android as soon as you can. I have advocated for building for Android first and iOS second. I think that strategy will start making more and more sense for apps that aren't looking to be paid.

Fragmentation cuts both ways. It's bad and it's good. Long term, I think it is a big plus for Android.


Comments (Archived):

  1. Carl Rahn Griffith

    Fragmented also helps to aim for a demographic. This is a pretty fat long tail.

    1. fredwilson


      1. JamesHRH

        My last gig (2007) was for a company that now has 17M users ( on 3 continents) doing local search on a non-apple mobile phone platform. Ever heard of it ( rhetorical )?Being where the heat is ( geographically or platformatically – is that a word ?) is what matters.They declared bankruptcy last month.

  2. Laurent Boncenne

    With Microsoft finally waking up as a whole, and the upcoming Windows 8 sharing the same core as Windows Phone 8, and seeing how the two will interact, I highly doubt you’ll end up with a two OS ecosystem, but rather 3.And that’s actually a good thing.People these days seem so quick to dismiss the small (big) players with a pile of cash (billions) behind their small and young platform, it makes me cringe and laugh at the same time…As far as fragmentation goes, for Android, I see it as a pain point rather than a good thing, sadly it’s innevitable =/

    1. fredwilson

      I would love to see it happen but I just don’t see it happening

      1. Laurent Boncenne

        really? with the Surface and WP8 coming ? pretty sure you’ll have a decent player, don’t forget that Android took a bit of time to become a real threat to the iPhone in the beginning…I give it a year or two tops.

        1. fredwilson

          I hope you are right

          1. Laurent Boncenne

            It’s an interesting time to be watching those three giants fighting for market share, and from a consumer perspective, I really wish to see MS somewhat succeed at least, their strategy challenges both Apple and Google (android) because they’ll be in both categories (high end a la apple, and middle to low end).

        2. kidmercury

          leaning towards laurent in this beef. too early to rule msft out in my opinion. i think there is enough room for probably a few more operating systems. also, i think as the android federation grows, the most powerful ones will ultimately divorce themselves more from android, and thus will morph into their own full-fledged OS.

      2. kenberger

        I didn’t see windows phone becoming my phone of choice but it sure was in the mid-2000s. Gates didn’t see email becoming a big deal in ’94, but then changed his mind and implemented so quickly that Outlook dominated.Also (and I know absolutely nothing about this), how do we know that when and if there’s a FB phone it won’t be “anything but Google (Android)”? you might well have better info than i on this.

        1. fredwilson

          i don’t but i wish i did

          1. Wavelengths

            I love FB and I hate FB for very specific reasons. We need an alternative to FB and Google.Some people want something very much different from what they deliver.

      3. Todd Werelius

        I am a long time MS phan boy, but I don’t see anything from a developer perspective that is impressing me so I have to agree.Are developers really going to stop doing Android and iPhone and dilute their precious resources to develop for an OS that might provide a 1 digit uptick in revenues?MS used to get that user adoption of their OS was dependent on massive developer adoption of their OS. Sure the new device is pretty and slick, so what, without some really nice stuff developed by third parties it’s a paperweight.I never count MS out, but so far I have not seen any indication that they are all in on mobile or that they really even get it, I see the wide adoption of mobile as a ground shifting event similar to the adoption of the internet.There are a lot of things they could do to move this in the right direction but they aren’t doing them, not so far at least.

        1. Wavelengths

          Yes. Yes. See my other comments.



    2. John Revay

      I was going to comment 3 vs 2……The reason why the MS phone might gain some traction – would be Office integration…I worry as time goes on Office is becoming less of a factor.

      1. Laurent Boncenne

        I think the Office integration is certainly a good feature at launch, but I believe it’s the whole “one OS, many form factors” that will become the killer feature of the whole Microsoft empire, being able to develop easily for so many types of hardware is quite satisfying not to mention easy.I believe MS and their OSes to hold the middle ground in the one or two years to come, they’re finally waking up and I for one am glad to see them becoming relevant again.

        1. brian trautschold

          i agree – the surface actually seems really competitive, WP8 looks slick, the new office package & outlook have impressed – im currently all apple but for the first time in n years i am actually *considering* msft products – i think if they can reign in oems to build highly consistent & competitive devices (or self source)- they can more than challenge by entering strong via business channel with new outlook, office, yammer…

        2. PhilipSugar

          One OS many form factors is what killed them.

    3. Tom Labus

      I’m with you on this one.MSFT woke up maybe 2 years ago and seems to be going back to full life again.The new Outlook with a new toned down and simple look and feel. The new SlyDrive is now live this morning.I love the Lumia 900 and I realize it’s well behind in the race but my contrarian instincts are going crazy on this one.

      1. Techman

        Yes, so you love the new SkyDrive as well? I did an in depth look at it at my website if you are interested in viewing. http://techmansworld.blogsp… @wmoug:disqus @fredwilson:disqus

    4. William Mougayar

      Are there real examples of big co’s that were on BlackBerries and now switching en masse to Windows 8?

      1. baba12


      2. PhilipSugar

        Here’s the issue. Fewer and fewer companies are doing anything en masse when it comes to personal devices.Why? The whole point of the Blackberry was I can now do company work in what used to be personal time. It started off differently, only for those that traveled a ton, but now you can spin it any way you want, but that is the reality. Well if its personal time, then I also will do personal stuff. If I’m doing personal stuff its going to be on a device I like, not one you force on me.So force that shitty Windows phone on me, I’ll keep my Android, or I’ll keep my iPhone, and that shitty Windows phone stays at work or only goes on business trips. How does that work for the manager or customer that wants to get a hold of me NOW!

        1. William Mougayar

          I hear you. The mess will get messier before it gets prettier. The 2 big unknowns are still:BlackBerries- how will they migrate?Windows Phones – will they penetrate the enterprise market?If this was a 2-horse race, it would be a clearer race, but there are 2 other horses that are still on the tracks, even if they limping. Either they’ll join the pack or move off the tracks.

          1. PhilipSugar

            I am saying the consumer not the enterprise is going to decide. The enterprise market for Windows Phone’s or any other phone is dead. It is what the consumer wants.

          2. William Mougayar

            That’s a very astute observation if I paraphrase it: The consumer will decide the enterprise market for smart phones. That choice will not be imposed by IT or management anymore. The consumerization of the enterprise is alive and going to get even more interesting!

          3. PhilipSugar

            Which is why saying I want one standard is a fantasy. Look at the diversity of cars. When the consumer gets to decide diversity is what happens, when the enterprise decides everybody gets the same USPS Truck.So you can whine about it or live with it. Sure BigCo and the biggest BigCo, Government love standards, but the reality is when it comes to consumer, even if you have a standard, companies will bastardize it to try and get a leg up on the competition at the moment of purchase.

          4. FAKE GRIMLOCK


          5. Wavelengths

            Depends on the consumer and the enterprise. If the consumer has a CDL (Commercial Driver’s License) with Hazmat endorsement, he’s probably not the guy to be shopping for a high-end Android. The enterprise needs him to move that oil to market, at approx. $100/barrel. If his boss gave him the phone because it was the only one that worked in the 1000 square mile area of West Texas desert he was driving, then guess what he will use. Oh, and he’ll call his wife on the same phone. (Sometimes we have lady truck drivers, but I’m going with stereotypes.)

          6. William Mougayar

            Lol. Is that the business you’re in?

          7. Wavelengths

            Nope. But those trucks are all around me. Maybe 80% of the traffic around here is oil-related. Maybe 1% is windmill-related. BTW, I watched the Pecos town parade that opened the celebration of the Pecos rodeo weekend. Along with the horses, the floats of various athletic teams, the mariachi band, the souped-up, dressed-up classic cars, the town and sheriff’s police cars with sirens blaring, etc., etc., the parade included a big tanker truck. At first I thought it was just part of the usual traffic through town that mistakenly got caught in the parade, but no! It was part of the show!

          8. fredwilson

            certainly true in the US. i wonder about developing markets

          9. PhilipSugar

            Here’s what I’ve come to believe. Anytime, anyplace, anywhere you expect an employee to use a device outside the physical workplace, the ownership/choice of that device will eventually go the employee.In the office you control. If you expect anything out of the office the employee controls and the best you can do is to make sure all of the infrastructure in the data center is secure and you take every step you can to secure outside access.The dilemma is this: We’ve come to expect answers now. So because people don’t live at the office the only way you can achieve that is to have access outside the office.People don’t want/won’t carry shitty products mandated by a large IT department. It doesn’t matter where you are, IT will always manage to the lowest common denominator, not take risks, make vendors deal with long/expensive sales cycles, bludgeon in bloat by demanding little features that are their pet projects be included or you risk losing the sale, and require backward compatibility/long cycles. So in turn the products from a user perspective are shitty.This is ok, when you are talking about the products that run the company. Its why its called work. You deal with the inflexibility of the accounting system because you have to and if it screws up, its not an oops sorry my bad, it can be a $400mm mistake like Knight. Many startups don’t realize this when selling to the Enterprise.However, when its something outside of work the user becomes in control. Give me a crappy product or no outside access???? To bad so sad boss you don’t get the answer now. Boss says ok, screw that policy I need answer now! IT grip: broken.

          10. FAKE GRIMLOCK


          11. PhilipSugar

            This goes for software too. Dropbox is a good example.

          12. Wavelengths

            The enterprise has an opportunity to evaluate the environment and affect the employees’ choices.Here in the Permian Basin, AT&T and Verizon rule, but they both offer sketchy service. My T-Mobile is often better, but not always.If a major oil company has an well producing $2,000,000.00 per month, they will probably want the employee to have a cell phone that will allow them to stay in touch. If they have 20 or 50 or 100 of those wells, well, do the math.Technology is awesome, and I suffer from technolust, but here are some from-the ground observations.

          13. PhilipSugar

            I think we actually agree. (btw: my father is an oilman from East Texas for a large Oil Company)The Enterprise is saying you need to stay in touch, because of exactly what you say. I live in a very rural Eastern Shore Town on the Chesapeake. No AT&T service, none. If an Enterprise required a resident to use AT&T they would not be available. Hence the employee will choose.

          14. Wavelengths

            If the enterprise realizes that only one type of service will serve, then the enterprise will rule.I am in West Texas. What service will cover the tanker trucks, the oil rigs, the geologists, the landmen, and all the others who serve those folks? Or, if you like, the windmills and the people building and delivering them?It depends on the money behind the market. If the incentive is good, the enterprise or the individual will change.

          15. Wavelengths

            Your comment comes from a perspective of evaluating the present. It used to be that if you weren’t using WordPerfect, no one could communicate with you. Then Microsoft showed up with Word, in a bunch of iterations that were more or less usable. It took awhile before they became a standard platform for documents.I hated Blackberry from the start because I was independent and I hated the configuration for my way of thinking and the way my fingers worked, and no one was telling me what I should buy.I chose T-Mobile’s alternatives partly because they allowed me to hook into wi-fi to seamlessly receive my phone calls — a major deal when I was in an area that had no cell phone service at all, from any carrier, but I could still get my calls through T-Mobile with a wi-fi connection.

    5. ShanaC

      why cringe?

      1. Laurent Boncenne

        poor use of english =) sorry about that. I meant to say that it makes me annoyed that people forget so quickly how these business cases go over and over again in history and everybody jumps right into the easiest conclusion without considering how losers once became winners in a similar situation. MS, to me is in this kind of situation and while they may not succeed, they have a shot. Dismissing them and not even mentioning them as an enventual player in the field is wrong imo.I like to think they’re kind of pulling a pivot right in front of us and no one really sees it completely, forgetting that a pivot for a starship like that is no ordinary task. It’s quite impressive to me, and I’d just wish people noticed it more…

  3. William Mougayar

    But for the developers Android still a bad challenge because browsers & OS versions are all over the place. Add various screen sizes on top of that and you are a 3rd factor. Now try to think of all the various possible permutations: screen size X OS version X browser type/version. Fragmentation and an increased market footprint are good for Android, but does it have to be that way? Why can’t Google put a minimum of order to that mess? 

    1. fredwilson

      There are roughly four or five devices that make up the bulk of android. Build for them to start

      1. brian trautschold

        i tend to agree – but ‘conventional’ wisdom currently is build & build very fast, launch, measure, then scale out offering…thoughts on gaining acceptance/feedback on 1 platform then move to other with built demand

      2. William Mougayar

        We are in the final stretches of the html5 web mobile app that will work for both Androids and iPhones, but even within the top 5 devices, they are different versions of the browsers/OS/screen sizes, so we’re catching incompatibilities and small fixes as we go. Why did we have to wait til Ice Cream Sandwich to have standardized software updates? Google has been too lax on hardware requirements for Android phones that’s a fact and the reason why there is so much fragmentation.

        1. JamesHRH

          They don’t do standards. Not a closed, orthodoxy based outfit.

      3. Ivan Vecchiato

        That’s what we did with a sailing application, targeting it to the Samsung’s most important phones. But you are on the Android Market, and when a user reports an issue on a phone that is not totally supported, that is a user that you should not ignore, as Garyvee says.In fact, with another application (interactive digital menus and ordering system for restaurant) we decided not to be on the market and selling the application already installed on low cost tablets that we guarantee for. As you can see, a totaly different and less powerful business model.

        1. kidmercury

          “selling the application already installed on low cost tablets that we guarantee for.”that is a very powerful and very under-appreciated (for now) business opportunity. well done.

          1. Ivan Vecchiato


        2. William Mougayar

          What is the name/website of your app?

          1. Ivan Vecchiato

            Regatta Pro / Regatta basic is the sailing application.The website of the restaurant app is (sorry, italian only, at this moment).

      4. Dan Wick

        I agree with aiming at the top 5 android devices when developing.I think the fragmentation of Android’s OS version is an equal issue to the device fragmentation. The flagship Android devices, like the SIII and Nexus 7, run a recent enough version of Android to develop efficiently against.When Amazon forked Android for the Fire the picked a too old version of Android in Gingerbread. Hopefully the bump that up for the Fire2. Same with the Nook.The combination of hardware and Android version is really the key to being able to build a native or HTML5 app that provides a good enough experience. Just because a device can technically support a feature doesn’t mean it’s going to render it responsively enough.Also, with a full blown version of Chrome being the default browser on Android, that opens more opportunities for reliably developing mobile web apps. But again, you need a mobile device that has the hardware to reliably render the experience in a browser.

    2. Ivan Vecchiato

      I think you are right William, Google definitely needs to put a minimum of order. Fragmentation is an opportunity, but if things are going on this way, this fragmentation is going to be like developing for more platforms. Really, developing an application and revising it for three or more phones, according to how many important phones on the market you want to target, is not far from maintaining different main lines of the software project. I would not find it strange if a large software company had a development deparment subdivided into “samsung team”, “sony team” and so on. That would be totally inefficient, but not so impossible.

  4. Wells Baum

    This begs the question for a new app – do you soft launch on Android or build for the steady and predictable iPhone/iPad and worry about additional Androidscale later?

    1. fredwilson

      I vote launch on android and iOS at the same time

      1. adrianoconnor

        I agree with this, if you have the resources. We went for both at the same time because it’s becoming clear that Android is the most popular platform for ‘regular’ people. Internally though, between releases, the iPhone version of our app tends to be slightly ahead — that’s where we add features first. Partly thats because we get a week while Apple review the app to work on the Android version, and partly it’s because the developer tools are a little more refined for iOS, so it’s slightly more enjoyable developing iOS, especially if you just want to quickly prototype stuff.

  5. awaldstein

    Fragmentation of platforms may equal in this case, fragmentation of markets as well.I agree, build on both. Neither one will help you find your market though. Market discovery and customer find are pretty well separate from the platforms.Android and iOS are platforms not channels. Not to be confused.

    1. David Semeria

      Well said Arnold

    2. Wavelengths

      Some markets want the technology to be invisible and they are the ones who will buy the cheap phones/tablets, etc. I think a big market for those who really want to be insulated from the sizzling technology that so many of us love.

      1. awaldstein

        Don’t think of it that way. If fact, the opposite.Apple users see themselves. This is why Google has and will continue to fail in the TV market. They dont’ understand that the market is not at all interested in tech. It is interested in themselves as the users of it.This community as much as I cherish my membership in it is not the market.

  6. David Semeria

    I’m a big Android fan, but I must say the fragmentation is both frustrating and worrying.If the motto “divide and conquer” is correct, then Android’s future looks messy.

  7. Rohan

    For me, this misses a big chunk – the enterprise market.That’s why I’d be inclined to go with @lboncenne:disqus – between iOS, Android and Windows – I see Microsoft in the best place to eat into what was once a Blackberry dominated world. We’ll have to give it 5 years atleast though given corporations will take a while to change.If Windows phone can nail Outlook, Excel and PowerPoint integration, I can’t see why BigCo’s wouldn’t go for MS.

    1. Martin De Saulles

      I don’t see better integration with MS Office really helping Windows Phone very much. Working with Office on a phone is not very practical – not much better on a tablet either.

      1. JamesHRH

        MS Office so entrenched. Compatibality, even if it has no material benefits, has selling point benefit.

        1. Martin De Saulles

          Good point – nobody ever got fired for choosing Microsoft

          1. Rohan

            Exactly. I am not saying they will win because they are good. I am saying they will win because they are acceptable.Apple will always be a B2C company, at heart. Being the thick walled garden that it is won’t help with B2B sales as well.Android is a bit too fragmented for Big Co taste, I would think. Plus there is the massive element of security which might be a problem with such fragmentation.In short, MS has a much stronger shot than it is made out. And if it is enterprise, there’s likely to be money in that. Imagine being pre-installed on millions of company phones.

          2. PhilipSugar

            I used to agree with that statement but now I don’t.I was on record having said that Blackberry wouldn’t crash as fast as people thought. I was not only wrong, it crashed faster.I think we are at the stage where it comes to personal devices the end user chooses. I have said (and it is true at my company) we are going to get to the point where the employee is just reimbursed for the device they use to attach to the corporate network. PC, Laptop, Tablet, Phone.Like many tech companies we have all devices, and I will say this I use an iPhone, have no problems if you gave me an Android, but our Windows Phone sucks, and not only that it is becoming an orphan at three months old.Now when it comes to the applications that run the corporation. Users will have much less choice.

          3. mydigitalself

            I heard this referred to the other day by an internal IT administrator as BYOD (bring your own device) and it’s fast becoming the norm.

          4. PhilipSugar

            The reason is simple. You are asking people to basically be available during what used to be personal time. If its personal time, you want it to be a personal device with personal stuff on it. You can try and “force” a company device on people but the reality is they can always pull the trump card and say nope I couldn’t get back to you because it was the weekend and I don’t carry that device then.

          5. Cam MacRae

            + 1I’m glad I scrolled before typing a carbon copy reply.

          6. Shawn Cohen

            I wonder if we’ll be saying the same thing about Apple a decade from now, as new execs start steering the ship. Apple, like MS, has all its eggs in one basket.

          7. PhilipSugar

            Oh, I’m not going to argue about what will happen next year, much less next decade. I’m just pointing out that when it comes to what device is in the users hand, even at big corporations the control has gone to the user and that means the strategy of entrenchment at the user level is completely gone. For proof, I give you the diversity of devices I see Federal Government employees using.For the big systems those devices are connecting to……that is something I can see in decade terms.

          8. Shawn Cohen

            who said anything about arguing? 🙂

          9. ShanaC

            how many devices have you seen the feds use. and which feds?

          10. PhilipSugar

            I was at a very large military funeral at Arlington for a Fraternity Brother on Thursday. Brass from different services, grunts performing the service, employees of huge military contractors, other gov’t employees from I don’t want to know where. Afterwords watching people catch up on work at the Fort Meyer Officer Club I saw as many different devices as you would see at a Tech Crunch conference, other than the dress and haircuts no different.

          11. ShanaC

            Hmm, interesting, if only because device security is a big deal to these people…

          12. Erik Neu

            Even reimbursement is becoming a thing of the past. Most employees, and all IT employees, are going to be carrying smartphones. That is part of the point, as others have made–it’s the employee’s device.

          13. PhilipSugar

            It is not reimbursement it is an allowance. I give you X dollars per month towards your phone, don’t give a shit what you use. AT&T, Verizon, Android, iPhone don’t care.

          14. Wavelengths

            Depends on where someone is in the hierarchy. See my other comments. Employees are not lawyers, doctors, or other business owners. Those are the ones whose hours are worth big bucks.

          15. ShanaC

            I still don’t think that – there is a specific kind of chokehold that happens when SMBs deal with BigCo. And that chokehold involves a lot of software issues that microsoft owns

          16. PhilipSugar

            ??? I am talking about what device BigCo’s users use to access BigCo Network.

          17. ShanaC

            I know, but some of bigCo users are not the same as their employees. Think of your doctor’s relationship with your insurance provider when it comes to billing. BYOD doesn’t work nearly as well in this sort of contract situation.

          18. Wavelengths

            My neurosurgeon has my MRIs on his I-pad, which he can access even when the hospital internet is down.My neurologist keeps his charts and sends his prescriptions from his laptop PC.If you look at what these guys charge per hour, you will start to get an idea of how likely they are to want to change platforms, yet how ready they are to change platforms when they can become more efficient and have better access to the info they need.

          19. fredwilson

            spot on

          20. Rohan

            I don’t know Phil. I agree with you that employees are deciding which phone they want increasingly.But for the really big companies (big banks etc), I think we are overestimating this behavior..

          21. ShanaC

            Especially when you are talking SMBs. Think your doctor – your doctor has to have software that works with bigcos (insurance companies), you, and his little office. Microsoft has a chokehold on that dynamic. Breaking it is critical for success in mobile

          22. PhilipSugar

            The web killed that, and mobile is killing the fact you need a desktop. Exactly how does MS have a chokehold?

          23. ShanaC

            When your scans are sent to a hospital, or insurance provider, usually the only software supported is Microsoft. Same thing with bills. Largely it is due to the software developed being microsoft only and very buggy.

          24. PhilipSugar

            Hmmm, I don’t know. I watch my Nurse Practioner wife fill out secure web forms for insurance companies all day and from all sorts of different devices.

          25. ShanaC

            they said that about IBM once upon a time. While still true, it isn’t the same anymore for IBM

        2. Wavelengths

          Back to the “Duh!” factor. Where anyone can jump in and use it.

    2. Abdallah Al-Hakim

      Blackberry might still have a chance to consider license their enterprise service unit. They were recent reports that IBM expressed interest. Also, the BB10 is rolling out in early 2013 (if it is not delayed again) and would be interesting to see if it the big shaker that RIM claims it to be. I can’t help but cheer for them since I am Canadian but I am not too optimistic

    3. fredwilson

      here in the US we are past the time when employers dicate what device you use.

      1. Techman

        If they do need you to use a device you don’t have, they will most likely provide the device to you.

      2. Rohan

        I don’t know about that, Fred.. Are all the employees of the big banks close to where you live running on Android phones?

        1. fredwilson

          they use what they want because any device they would choose to use can access the company email system

  8. jason wright

    I’ve always had the feeling that Google’s motive for Android was to throw a whopping great big spanner in the Apple works. Chronologically this may not stand up to scrutiny, but it still feels that way to me.I’m waiting for the iphone 5 and then I’ll decide if I want a simple or more complex experience.

  9. jdelvat

    Luckily there’s also another player coming in (or out)?Blackberry / QNX is still around in the professional environment, however probably not for long …

    1. baba12

      The professional environment you talk about is basically a fiefdom of people wanting to hold on to their turf. It is rarely based on technology. If you are a business there is no real difference to the user if a application runs on one or another platform or whether you use a Dell or a Compaq machine etc. These so called standards and rules of engagement are not determined from the business need but more from a need to either fluff up resume or to protect the IT functions.QNX just got sold and I don’t think it will be for the masses maybe some unique NSA need will let them be run as a boutique player. RIM and Blackberry are stuck in time and at this point in time they are most likely to just shutter down in a couple of years.As Mr.Wilson points out the Android fragmented platform will become a solid platform, but I will maintain that in time this debate or decision about which platform to build on will be gone, HTML5,6,..n will become the defacto standard and will make many fiefdoms crumble.But then again humans are known to find ways to create barriers.

  10. John Frankel

    What if Apple moves to lower price points based on its ability to have lower costs of manufacturing? i.e. Dominate not just the high end, but also attack the low end. The android fragmentation is huge (… and that leads to increased expense for developers to build for different screen sizes, resolutions, with flash/without flash etc. Apple offers a platform that is simpler to build for and one that seems to attract customers that pay for apps. I suspect that build for iOS first to prove your business model is still the default decision for most mobile apps, and then decide when Android makes sense.

    1. William Mougayar

      it gave me a headache looking at that table (btw- the right parentheses is tacked on to the url and it breaks unless you delete it)

    2. fredwilson

      samsung is making more android phones than apple is making iphones

  11. chrisdorr

    I am curious why you would recommend a developer build first for Android when Apple as you say “has the best and most integrated experience”? Shouldn’t developers think first about the user experience and create apps in environments that create the best user experience? Isn’t everything else secondary to that?

    1. Irv Remedios

      It depends on the problem being solved. Apple limits developer access to certain things (i.e. core telephony features). It’s hard to innovate in those areas on iOS.

      1. chrisdorr

        Irv, That is true, but in the end the iOS experience is superior for the user. Now on Android the experience is all over the map as it changes for every version of Android. How can people who constantly worry about user experience (as tech VCs and developers should) let Android get a pass on this? Surely Google has to correct this in order to compete with Apple in this very important area. I find it fascinating that very sophisticated tech people don’t hammer at this more often. Instead they give Android the benefit of the doubt.

    2. fredwilson

      i think developers should reach the widest audience

      1. chrisdorr

        Yes, I see that point, but is a heavily fragmented audience which has a less than optimum user experience in fact the “widest audience”. The numbers in the abstract might be there–but the users as real engaged users are not. My experience dealing with companies/developers that have applications in both environments is that user engagement is much higher on the iOS platform as are the opportunities for making money. So the higher numbers for Android really hide what is a much more pressing reality for developers–higher numbers of handsets, lower engagement, less money. So one has to pick carefully which version of Android one is to develop for as a result. The reach is not as wide as people think. Whereas with iOS it is very straightforward. The numbers of handsets are much closer to the real reach. Many people just naturally like Android better as an abstraction (it is more open) and don’t like Apple because it is so closed. But then they don’t have to make the day to day decisions about where to deploy their development resources to really reach people in mobile applications.

        1. fredwilson

          our portfolio companies don’t see that effect

      2. Wavelengths

        Yes. Even Homer Simpson should be able to run the app.

  12. Just asking

    Fragmentation perhaps in one perspective. But consolidation of the market place in a duopoly appears to be the landscape. Bring in MSFT and an Oligopoly in the communication structure is once again firmly in place. . Does this pose leaving little room and high barriers of entry for others both in launching new platforms and which apps may eventually be “allowed” to populate the platform? Wu’s worry rising again?

  13. Max Yoder

    From my perspective, native apps are a necessary Band-Aid. But, in the long term, it just doesn’t make sense for HTML5 (or 6 or 7) to NOT become the de-facto development medium. Right now, it’s too slow, but it won’t be that way forever.To use Nick Carr’s analogy, there was a time when electricity was a decentralized thing. If your business wanted electricity, it hired engineers to build and maintain the necessary infrastructure. This was the model that Edison pushed on the world because it was the model that made his investments the most money.Over time, the inefficiencies became clear. Thousands of electrical engineers were solving the same problems day in and day out. So we developed the centralized utility, like we have today, which allows you and me to pay a company to employ engineers to satisfy our electrical needs.What I’m saying is this: you, me, and a million other people shouldn’t all have to manually download a 3MB update every time an app wants to evolve. One person should update the app, and when I launch it, that update should already be there. Maybe native apps will evolve to curb this inefficiency, but I just can’t envision a future where mobile doesn’t go in the same direction that desktop applications did, and computing did, and electricity did.Editor’s Note: Sorry if there are typos or clarity issues. I wrote this while running to a meeting.


      “If your business wanted electricity, it hired engineers…”My grandparents had to pay to have electric wires run to their store way back when.

    2. Greg Athas

      I think the fundamental problem with HTML5 right is now is one of discovery. People are very comfortable getting apps from a store, seeing ratings, sorting by popularity, and launching from icons. I don’t think however that HTML5 apps should mimic the native app model, it’s already staring to get too cumbersome to manage. We need something that combines the comfort level of a curated store with the elastic nature of the web.

      1. mydigitalself

        If the ‘app’ that you download from the App Store is just effectively a browser wrapper and receives notifications then the rest of app is HTML5, remotely updateable, cross-platform. This is the model we are currently developing towards right now so it’s fine for both distribution and fragmentation.

        1. Wavelengths

          So you are saying “seamless for the end user.”Many, many smart and powerful people don’t have time to learn the new technology. If you make it bombproof, and “duh-proof”, you have a huge market.I remember reading in some viral email that if Bill Gates bent down to pick up a penny it would cost him $16,000 (or so) in terms of his attention. We should remember that the target market may sell tires, cars, oil drilling rigs, and information you could never conceive of. But in their context, their time is more valuable than we might imagine.

      2. Max Yoder

        I agree that discovery is a problem, but I disagree that it’s the fundamental problem. I think the fundamental problem is that, as of today, native apps can run laps around HTML5 apps.

      3. fredwilson


    3. falicon

      Of course now larger companies are starting to move back towards generating at least some of their own power via solar (and then selling it back to the grid)…which just goes to prove even more almost all this stuff is cyclical (and that Roy Williams actually might know what he’s talking about too -> http://www.mondaymorningmem… ) 🙂

      1. Max Yoder

        Interesting perspective. I’ll check out the link. Thanks!

    4. matthughes

      When/how do you think HTML5 apps will improve?

      1. Max Yoder

        It’s less about HTML5 improving and more about technology improving. Faster, smaller processors and faster, more ubiquitous high-speed internet will help a lot.In the short term, HTML5 could use better offline caching capabilities, geolocation authorization options, and, as others have mentioned, distribution and monetization channels.

        1. matthughes

          Good insight.Do you know of any slick HTML5 apps that are ahead of the curve?I’m with you – I would like to see HTML5 become the de-facto standard.

          1. Techman

            And as HTML5 becomes more popular and widely used, the better layout engines such as WebKit and Gecko will get.

          2. Max Yoder

            Sun. It’s simple, it’s far from revolutionary, but it’s nice:

          3. matthughes

            That’s pretty slick.You’re right, super simple.Would love to see apps trend towards HTML5…

          4. Abdallah Al-Hakim

            This is not my domain of expertise but my understanding was that LinkedIn iPad app was HTML5

          5. matthughes

            Good find.Interesting points their developer makes about responsive design for apps v. web sites.

          6. Abdallah Al-Hakim

            they do make some good points and reaffirms the notion that simpler is better. Just make it functional first before obsessing with design and look.

    5. Todd Werelius

      HTML/5/CSS etc. is hardly universal in it’s application or implementations on the desktop and I don’t see any reason why that would change on a mobile device in some ways it’s tougher because of the hardware support requirements for certain types of app’s.I don’t see low drag continuous deployment as that big of an issue, there are other ways to address total app replacement issues outside of the web model and Google is actually already starting to implement this and it’s fairly developer transparent. Continuous deployment will become more of a political (security) issue than a technical one I think.You would also still have to deal with the actual feature support differences on each device, sure you can hide some of this behind mediated API layers etc. but I suspect that will also become an area of fragmentation and frustration (well it already has on x-platform toolsets)I think HTML5-x will end up getting better and therefore broaden it’s appeal for certain classes of app’s in regards to x-platform but I would not bet that it will follow the same path as the desktop, or at the same pace.

      1. Max Yoder

        You could be right. Great input.

    6. hypermark

      I think the hard truth on this one is that HTML5 is just not ready yet.Believe me, there is nothing that Facebook would love more than to avoid native client apps and build everything in HTML5, or for that matter, simply provide a native wrapper on HTML5 (which is what they are doing now). Instead, they are rebooting that strategy and embracing more native development modes.Plus, I would note that we were talking about the Web obviating the need for native client software back in 1994, and I even remember having to defend why native software “mattered” back in 2003.If anything, the promise of ‘write once, run anywhere’ is always just around the bend, a choice between “good enough” lowest common denominator and the best, most native, most integrated experience.I’d argue that today, more than ever, people want the best, most native, most integrated experiences, but as the Android vs. iOS debate underscores it’s not an ALL or NONE.

      1. Max Yoder

        I agree that HTML5 is not ready and, as of today, people want native apps.If I were to bet on a future, I’d bet toward my beliefs above. I wouldn’t bet the farm, mind you, because I love my cows too much to do that.

      2. fredwilson

        i agree

    7. ShanaC

      Can battery life support HTML only though. Those radios, they are power intensive. Other than that, I think you are right!

      1. Max Yoder

        Good question. I don’t know.

        1. ShanaC

          I’m of the opinion that until the battery problem is solved/browser/antennas fix their power usage, talking about pure HTML apps may never come to be. If you kill the phone using your apps, what is the point of using HTML?

    8. fredwilson

      i agree with you max but i think we are a ways away from that day

    9. JohnDoey

      That is not possible.The iPod could replace the CD because they both hold the same content:- CD audio is PCM 16-bit 44.1kHz stereo encoded as a Red Book CD-DA track- iPod audio is PCM 16-bit 44.1 kHz stereo encoded as AIFF, Apple Lossless, MP4, or MP3… it’s the same audio. So you can swap CD player and iPod, but the user still hears the same exact sound. Even a trained ear cannot tell the difference between a CD or lossless rip of a CD. You gain mobility, you lose nothing. You can get rid of the CD.HTML cannot replace C/C++ because they do not both contain the same content. HTML is a tiny subset of C. Tim Berners-Lee used an OS written in C, an IDE written in C, and then wrote a C app called WorldWideWeb which then evolved essentially into today’s WebKit. Your browser and OS are written in C. I can always make a C app that does way more than your HTML app and that will always be the case. C becomes C++ and Objective-C, not HTML.How is Avid going to publish their iPad video editor app as HTML? Maybe their 2011 C app could be rewritten in 2025 HTML, but it will look shabby next to even the 2015 C version.Also, HTML tracks behind C in standardization and gaining new features. I put an audio file in a Web page in 1994 and it worked but there were issues. Today, you put an audio file in a Web page and it works, but there are issues.What has happened is that HTML got full of itself and thought that it had a monopoly on the Internet” Then, Apple stole 1-click Internet installs from the Web (BTW, that is the proper use of “great artists stealing”) and added that to C apps that each are as powerful as the Web browser itself, and then with App Store reviews made the C apps safer than the Web, strike 2. Now, add 1-click iTunes purchasing overlaid on the install click — and please, no more talk of the Web replacing everything.(What has the Web done to strike back? Google built FlashPlayer into Chrome, Mozilla blocks ISO audio video playback.)Another thing to remember is that the everyday average Joe and Jane now do non-linear digital video editing and high-res photography just as a part of their day. Even users with the most basic computing demands require Mac-like software today. All that stuff is centered around OS X, not the Web. (For example: Apple is the top vendor of both consumer and pro video editors, even though their apps only run on their own devices, and MPEG4 is a standardized QuickTime movie.) If you give a kid a computer right now, it is more imoortant that they can shoot and edit a 5 minute documentary or news report than it is for them to work with Microsoft Office. HTML will be used to present the final movie on YouTube.

      1. Max Yoder

        Hi John, I think you make a strong argument, but it appears we’re coming at this from two different angles. I’m not suggesting that back-end languages are going to be supplanted by HTML; I’m suggesting that HTML will provide the user-layer, while the back-end computing happens in the cloud.


    Is it just me or is this more “50 First Dates”?This reminds me of Peter Norton’s replacement windows shell. So we’ll get a bunch of bastardized phones then a company will come along and create a generic phone for all shells then the shells will all disappear because a cross-platform nightmare will ensue.

  15. LekanB

    This is a good convo we have going here… I’d say my biggest issue with the Android fragmentation is the amount of time it takes to get users to the newest os update.Apple is able to get to 50%+ adoption in weeks. Google still lets carriers/mfg’s too much control over when os updates get pushed to users. That’s the greater contributor to the fragmentation challenge.

    1. fredwilson


  16. Jonathan

    We should clarify that Apps you purchased for your Kindle Fire are also available on your Nexus 7 — but you have to re-purchase them. People like myself who have bought many apps from the Google Play store feel strong hesitation to buy an Android device that requires a different storefront (Amazon, Facebook, or otherwise).

    1. fredwilson

      i don’t think i had to repurchase. maybe all the apps i use are free

  17. RichardF

    Google building the Nexus 7 was a great move. The price point is perfect. My 80 year old mother has one and loves it. So she has a tablet with Jellybean and my Samsung 10.1 has only just been upgraded to ICS. Samsung will have to ensure that they are up to date and providing the latest Android experience or they will lose market share. I don’t use any of the shit that they preinstall on their phones or tablets.Android first, completely agree

  18. William Mougayar

    Let’s call a mess a mess. Fragmentation is too pretty a name for it.As a user, this mess is making me appreciate my iPhone even more.As a developer, I’m shaking my head. Android has had a pretty messy start. Google: Fix it.

    1. Abdallah Al-Hakim

      I agree with the mess part!! I have an andorid at the moment and the fragmentation a.k.a mess one major reason I would switch to apple.

    2. baba12

      Google has a simple clean page for search, but everything else it has it”s hands in are a mess, why because it serves to keep people employed, create the mess then use people to fix the mess or attempt to fix it. Will google become a Microsoft, time will tell…

    3. ceonyc

      Yeah, there’s absolutely no reason for this–in the same way that parents shouldn’t be cooking three meals at once to satisfy the food quirks of three year olds. When I was growing up, you ate what you were told to eat. Google owns Android and *can* make Amazon and Facebook confirm to a standard that makes it worth developing on. There’s no reason why they need to fork the whole system. They can also allow wrappers to fall within a set of predictable constraints that you can develop for.

      1. Cam MacRae

        And to this day I live in terror of smoked cod!There’s no real reason why Amazon or Facebook can’t have their own launcher, and leave the rest of the system as they found it, albeit “hidden” from normals if they really must.

        1. ShanaC

          there is something wrong with you for disliking smoked fish. I’ll eat it for you!

          1. Cam MacRae

            Oh I like smoked fish, especially smoked Tasmanian salmon. But smoked cod gives me chills.

      2. ShanaC

        What of anti-trust?

        1. JohnDoey

          What of it? Who is making 100% of the sales in what market?

      3. fredwilson

        i don’t think google can tell facebook and amazon what to do

        1. Bla1ze

          They can’t because what Amazon and presumably, Facebook, for the sake of the conversation are doing is exactly what Google built Android to do and that’s forking it and making it their own.

          1. JohnDoey

            Which is why we should be speaking in terms of making Apple apps or Amazon apps or Google apps or Samsung apps. Which open source projects a particular vendor used is developer information, not for users. Users do not care, and further do not want to care.

        2. Techman

          Well in some cases they could, like when Facebook and Amazon want to submit their apps to the play store. And these days normal Android users don’t know how to side load apps. And some get scared by turning side loading on for security reasons.

      4. CJ

        Goes against the whole Open Source thing.

        1. JohnDoey

          Collaborating on a functional, practical operating system project goes against the whole open source thing? Maybe that explains a lot.Yet Apple’s open source projects work great. Apple WebKit is the most widely-used mobile open source project. Used by Google, Amazon, Facebook, Adobe, RIM, HTC, Samsung, Sony, and many others.So I don’t think “open source” is an excuse for what is wrong with Android.

      5. Techman

        But that would go against open source, one of the main reasons why Google made Android in the first place.

        1. ceonyc

          Let’s be clear: Google bought Android… and they don’t exactly participate the way other open source caretakers do:From Wikipedia:*”The rest of Android is developed in private, with source code released publicly when a new version is released. Typically Google collaborates with a hardware manufacturer to produce a flagship device (part of the Google Nexus <http:”” wiki=”” google_nexus=””> series) featuring the new version of Android, then makes the source code available after that device has been released.[131]<http:”” wiki=”” android_%28operating_system%29#cite_note-130=””> **In early 2011, Google chose to temporarily withhold the Android source code to the tablet-only Honeycomb release, the reason, according to Andy Rubin <http:”” wiki=”” andy_rubin=””> in an official Android blog post, was because Honeycomb was rushed for production of the Motorola Xoom<http:”” wiki=”” motorola_xoom=””> ,[132]<http:”” wiki=”” android_%28operating_system%29#cite_note-131=””>and they did not want third parties creating a “really bad user experience”by attempting to put onto smartphones a version of Android intended for tablets.[133]<http:”” wiki=”” android_%28operating_system%29#cite_note-132=””>The source code was once again made available in November 2011 with the release of Android 4.0.[134]<http:”” wiki=”” android_%28operating_system%29#cite_note-40sourcerel-133=””> “*They develop with the manufacturers, and THEN release it…. so they’ve got more influence then you give them credit for.

      6. JohnDoey

        You are pretending that Google, Amazon, and FaceBook are divisions of the same company rather than viscious competitors.

    4. Max Yoder

      “Fragmentation is too pretty a name for it.”LOL

    5. awaldstein

      So…and you knew I would ask this…I’m listening to you and others and no one talks about their market.Is your market fragmented across devices? That is your first core market? No one goes after everyone at once.

      1. Mark Essel

        “Is your market fragmented across devices?” that’s a serious barrier to entry. The influence and attractiveness of a single experience is very attractive (aka apple). Even there you have to deal with versioning, and support for old iOS/devices. I was surprised to see so many apps just not work on Michelle’s 3GS

        1. awaldstein

          Yup…this platform fragmentation makes broad based consumer plays even more interesting.

        2. CJ

          Exactly my point, the fragmentation is present in Appleland as well, it’s just hidden behind an OS that looks the same as it did in OS rev. 1. You don’t see it until you bump up against it, but it’s there nonetheless.

          1. Mark Essel

            It’s there but less so.Even the web has it, look at html5 vs html, webgl vs 2d web, and consider all the folks that have no script and adblocker installed. Add to that all the various views on web data (mobile, desktop, tablet, etc.)

          2. JohnDoey

            A summer shower and a hurricane are both rain. That does not mean it is the same thing.To support every phone Apple ever shipped, you need only one graphic layout. And you get all the iPod touch and run 2x on iPad for free. With one layout — and all your users see the exact same thing. If you break down and do a second layout, you can support every iPad at full-res also. And all your users see it pixel-perfect. That cannot be done on Android. The artist time alone goes up 10x at least. Then you have to deal with dark screens on some devices, PenTile screens, screens that lack various common colors.Android fans keep saying this is easy. Developers and producers are telling you it is not. Why do you think giant Apple, 35 years making PC’s, only made so few phones with so little variation in screens and API? Because otherwise, an iOS-style app platform is not possible. They brought most of the Mac over to mobile in 5 years, including key apps. Google’s tiny Android division cannot outdo that across thousands of devices.

      2. William Mougayar

        I am thinking both as a user and as a developer who is going to market. As a developer, it’s a no brainer I need to reach both markets – Androids and iPhones user, but was debating the messiness of each and hurdles/work you have to go through to make that happen. As a user, I often think about my switching costs or inconveniences. The only reason I carry 2 devices now is so I can test the Android against my App, and that’s a different Android than the one my developers have, so we’re seeing different user experiences.

        1. awaldstein

          I think most users just buy what they can afford. If you can afford Apple you buy it, if not you don’t.The upside of having a retail store to bring your device into is just so huge for many that they simply think Apple. IMO.

          1. JohnDoey

            Android is not necessarily cheaper, though, either.

      3. Elia Freedman

        I assume Fred is referring to consumer markets, which are spread across platforms. Business customers are a different issue, but even then I am seeing more “bring your own device” strategies, which means a mix of iPhone and Android.

        1. awaldstein

          Agree….The enterprise is one thing (not my world) but but half the people in the country work for companies under 100 people and they are definitely ‘bring your own’.

        2. CJ

          Yeah – we’re now a BYOD business in what was before almost exclusively blackberry. Truth be told, we’re probably 70% ActiveSync now, Blackberry has fallen a long way.

    6. Mark Essel

      Damn right William.

      1. William Mougayar


    7. RichardF

      I’m an Android user and I just don’t see the mess that you are talking about but maybe that’s because I’m on Samsung devices.The only drawback to being on Android for me is that sometimes Android receives an app after the IOS version having said that I’m noticing that more and more frequently that there is either little or no time difference in release. Pinterest releasing their mobile apps today is a good example of that.Google are playing catch up but they are catching up quickly.

      1. fredwilson

        samdroids are the best

      2. Druce

        Do you root it? AFAIK Verizon Fascinate (original Galaxy S) stopped at Gingerbread and isn’t going to ICS or Jelly Bean. Being stuck on an old carrier-crippled OS isn’t my definition of catching up quickly.Next iPhone is crucial. I used to think it was a repeat of the PC wars, Microsoft was the new IBM, Google was the new Microsoft, and Apple was, well, the new Apple, innovative, closed, overpriced, loved by early adopters, loathed by OEMs and probably doomed when the market went mainstream. But the early PC was an IT buy, and the mobile device is a consumer buy, and Apple’s brand and retail distribution/service are paramount. And they have the apps and the ecosystem, and iOS is now their DNA, whereas one feels it’s a defensive afterthought to Google.Used to think Android would eventually win out but today would rather play Apple’s hand. The success of iPhone 4S which is old tech, smaller screen, no LTE etc., a generation behind even Galaxy S2 convinced me. If Apple introduces a truly competitive phone, smaller tablet, TV etc. they will clean up.The bull case is they extend the music franchise to all of digital media, including over the top video and e-books, controlled by or consumed on tablets, and become the Microsoft of mobile/cloud,disintermediating Google on its profitable products, first maps but ifthey can do that they could do same for mobile search, and take onAmazon on electronic media distribution.edit… would also add iPhone 4S now $50 on Sprint, so Apple’s margins might go down but if they keep old phones around and Sprint as a low-end carrier Android’s price edge is blunted. And Apple still dominates crushingly in tablets. but Apple needs to get back on top of the product cycle with iPhone 5.

        1. JamesHRH

          That is, absolutely, the play for Cupertino.The trick will be making the ‘media hub’ strategy move from HW (iMac) to iCloud. That part is iffy – they will rock the HW.

        2. RichardF

          I don’t root them, so far I’ve not really found a need to. I used to have an iPhone 3G that whilst in theory it was possible to upgrade to the latest IOS after a while it just became pointless.I like the nexus 7 and it will be interesting to see when Apple produce a 7″ tablet. I say when because if they don’t they will be allowing Google to make an inroad into that market. The product and price point is very attractive.I agree the iPhone 5 will need to impress, I have plenty of friends who are waiting to upgrade and if it’s not significantly better I can see them jumping to Android.

      3. JohnDoey

        You are talking about applets like Pinterest or Facebook. Things you can run on the Web, or now you can run as an iOS or Android app. The action in iOS apps for a couple of years now is apps from the Mac, Windows, DOS, console games. Things like Keynote, Avid, GarageBand, OmniFocus — PC class apps, not Web apps. Also, hardware accessories like USB microphones and so on. Google has not even begun to support these types of apps. That is part of why there are almost no PC-size Android devices — no PC apps.Windows RT will be the second ARM-based platform to support PC apps, so over the next few years we will see even more apps going from Mac, PC, game consoles over to iOS and Windows RT/8.Basically, you can replace a Web browser with Android. An iOS user can replace a Mac or PC or game console. I just got rid of my second Mac because the 2 apps I used it for are on my iPad and iPhone now. Same exact apps, working with the same documents.So it is not close unless you are only counting applets — lightweight apps. You are saying that Android throws the javelin almost as far as iOS right now, but you ignored that iOS is a decathlete. iOS is doing everything that Android does plus everything the iPod and Mac do as well. Hundreds of thousands of hardware accessories, cases, creative apps, video editors from the 2 leading makers. It is not close. Only the javelin event is close.

    8. Elia Freedman

      I have a hard time seeing how Google can fix it. All of these guys want to use Android and present their own front end, to have their own differentiation. That will slow up deployment of new OS versions. Furthermore, we are in an experimentation phase for device sizes, so we can all kinds of screen differences. That will continue for a while until this shakes itself out.

      1. CJ

        There is one way that can cut it down a bit. Google needs to create a theming engine inside Android that lets you skin it, mod it, etc but keeps that code separate from the updatable ‘core’ code. That would satisfy the companies I think, but the carriers still demand too much control over the product itself, Verizon being the most notorious.

    9. mcbeese

      Another word for ‘mess’ is ‘opportunity’. When I was a big company I guy, I used to think of messes as ‘problems’. Now, whenever I see a big market that looks like a mess, I find I automatically start scoping out the opportunities.

      1. fredwilson

        i just upvoted this comment

      2. WhosWho23

        i found an amazing network to make your professionel connections wider, alot of Vip’s sharing and doing biz together .. u shud check it

      3. JohnDoey

        That is a great attitude. Good luck feeding your family while working on Android.

        1. CJ

          Good luck getting discovered in an increasingly crowded app store.

    10. ShanaC

      welcome to open source. I don’t think they need to fix it, what I do think is that they need stronger controls and a clearer sense of how to monetize outside of ads on phones.There is no red hat model for Android, in other words. If there were, I think a good deal of the fragmentation problem would be solved.

    11. CJ

      This fragmentation thing is over-argued and overstated. It’s like saying that because there is a PS1, PS2, and PS3 that there is fragmentation in the video game space. That’s bollocks. But for the sake of argument, Apple has the same issue, though it’s glossed over because each iteration of iOS largely looks the same as the one previous!The first iPhone is stuck at 3.1.3 with no MMS support, the 3G is stuck at 4.2.1. The 3GS and above run iOS5 and iOS6 but not all features will be supported. (Siri anyone?) And remember, Apple is the only company that manufactures iPhones, There are dozens upon dozens of companies making android devices. So why does Android get criticized where Apple doesn’t? In the end, there are two very easy ways to always have the latest version of Android on your phone; buy a Nexus device (which are already the cream of the crop), or hack the hell out of it, which is both easy and rewarding.

      1. fredwilson


      2. William Mougayar

        Ok. 4 dead ends for iOS vs. dozens of dead-ends for android phones. see that Wikipedia link that John Frankel shared. Hacking & rooting an Android are easy? For whom? If you are a developer, yes. Most users aren’t technically inclined.

        1. CJ

          Not really, 90% of Android devices are running only 3 versions of Android: Gingerbread, Froyo, and Ice Cream Sandwich Most apps are written to work perfectly fine on either of those OS versions.Maybe I was overstating the ease of rooting a bit, but I think most of the issue lies in gaining the knowledge rather than executing it as the steps themselves are relatively simple and often outlined to be almost foolproof.

      3. Abdallah Al-Hakim

        the numerous devices available for android users are very desirable for some because they favour hardware over software. However, for some such as myself who prefer software – the situation becomes messy because of the fragmentation. I have had an android phone HTC desire for 2 years and have been frustrated at the lack of software updates. Also, I agree with William Mougyar that hacking is not a trivial matter for non-developers. I would agree with your advice that buying the Nexus device is probably better choice for those wanting to get an Android with latest OS

        1. Wavelengths

          Remember that a “rocket scientist” may be very smart about rockets, and even stars, but have no clue about new phones. Smart people are not necessarily smart about computers or communication devices and protocols.

          1. JohnDoey

            It is almost impossible for gadget nerds to understand that. They literally think a doctor should be hacking his phone and then use it in a clinical setting, or that a musician should expect to work with device drivers.

      4. fltron

        The problem with this logic is it attempts to parallel similar issues without properly weighing them. Few people are affected by the original iPhone not having MMS (few, really means none). You can (I suppose) debate that Siri should have been supported on older hardware, though knowing that the iPhone 4S had hardware specific for Siri and knowing Apple’s higher standards, this is probably a poor argument.Fragmentation is a far bigger issue on Android. Friends are stuck at 2.3 and get paranoid about which hardware they buy because they’re not sure it’ll be supported over the next 6 months. With Apple, you might not get all the new features of the next iOS release, but your hardware is likely to get some new features and certainly the new version of iOS. Apple has been getting better at supporting older hardware, not worse.I don’t get why there’s a need to defend Android’s very real fragmentation by attacking Apple on limiting feature-sets on old hardware because Apple feels that older hardware can’t support those features. The latter choice, much like with your PS example, is normal. That’s life. Older hardware can’t support newer software. In Android, the issue is older hardware gets lost/forgotten from the support stream even though it can support the newer release of Android.I’m not all anti-Android fragmentation. I think we’re seeing companies like Samsung take the lead in the platform with fewer products, and companies like Amazon going their own path to address customer concerns. There’s plenty of opportunity in Android. But let’s not fool ourselves by comparing Apple and Android on this. There’s no comparison when it comes to support. And it’s not because Apple is this fantastic company that supports older hardware, they simply have the business model more favorable for that.

        1. CJ

          Siri will run on an iPhone4. It’s been proven. The original iPhone still doesn’t support MMS a glaring oversight for a $600 device that affected everyone who owned it, quite inexcusable but another argument altogether.Your argument seems to me that Apple can fragment because everyone with an iPhone buys a new one every year and even if they don’t and the new iOS doesn’t support EVERY feature, it supports some. That seems to me to be the very definition of fragmentation, so what’s the difference?

          1. fltron

            I don’t really want to get into a Siri debate because I’m sure you’ve heard all this before and if you’re unconvinced then not much else I can say… however yes, Siri runs on an iPhone 4. Saying that, this doesn’t take into account that Apple may have wanted to stagger Siri’s roll-out due to concerns on server load (reasonable) and that they were unhappy with the performance of Siri on the iPhone 4 (also reasonable).It’s hilarious when someone is like ‘Hey, some hackers made this worked, and it worked just fine’ ignoring a billion dollar company that ran millions of tests to see how the feature functions. Do you really really think that Apple puts so little thought into things that they tried it out and said ‘hey it works’ and left it at that? They obviously test everything. They obviously (proven) included hardware features so that Siri would function better.This argument is so lost on your side that you need to ignore logic to make it. It’s not impossible that Apple didn’t include Siri so that the iPhone 4S would get more sales, it’s just really very unlikely.My argument is that Apple can fragment because hardware is a natural fragmentation point. Newer hardware is more capable and thus can offer greater flexibility. Most iPhone customers get a phone every 2 years, and on that cycle, they’ll always have the latest release of iOS. They may not have all the features, but they’ll always have many of them. They will also have any security, performance and bug fixes.The guy buying the HTC phone with stock Android 2.3 is not going to get any of that on the Android platform.

          2. CJ

            Maybe but SIRI was THE killer app for the 4S, without it the 4S was just a spring refresh. I know it’s hard for most Apple fans to consider money when thinking about the company but I assure you they iterate only because it makes them more money. I also assure you that the rate at which they trickle out features and obselete new features on older hardware is carefully calculated to maximize profit as well. Sure, the user experience is considered too, but they didn’t get 100 billion in the bank by dumb luck.Secondly natural fragmentation point or not, it is a fragmentation point that forces you onto the hardware upgrade treadmill if you want to keep current, same as Android. In the end 90% of Android devices run either Froyo, GB, or ICS that’s equivalent to Apple. The real difference, the one that Apple fans like to gloss over, is that each version of Android varies significantly while iOS looks and feels like the day it was born.

          3. fltron

            You continue to argue the 30% argument showing bias. Yes, Siri is a (probably poor) selling point of the iPhone 4S. Apple is advertising it plenty so we know this is where Apple is leaning. However, iPhone users upgrade every two years. Chances are if you’re going from a 3GS you’ll get the iPhone 4S. A tiny number of users would go from a 4 to a 4S. Your logic would work if not for: 1) special hardware in the 4S and 2) high standard iPhone has on this and 3) Server load is already an issue on Siri (especially originally) with just 4S users using it. Can you imagine the 4 also had it?Sure, sure, I’m sure part of the reason Apple made this decision is to make more money, but you’re not arguing the most obvious of reasons which is technical, serverload, etc..If you can’t see issues with Android fragmentation, nothing I say will change your mind.Your last little parting shot about how Android has evolved and iOS looks the same.. that’s because Android looked like utter crap at first and iPhone looked good from the start.Oh, one more point.. Android fragmentation issue is partially because of hardware, yes. Of course. The problem is, it is also because carriers and manufacturers forget about the phones they make. You get double-wammied – You have to buy a phone with good hardware (exists in the Apple world also) -and- you have to buy a popular phone that’ll be incentive for carriers to upgrade. The latter isn’t even a guarantee.The valid point that you’re making that is often understated in these debates (in Android’s favor) is the version of Android people running is less of an issue. This is completely valid. However, arguing that iPhone and Android fragmentation is the same is a bogus argument lacking clarity of numbers and proportions.

          4. CJ

            We’ll just have to agree to disagree.

          5. fltron

            Hey Malcolm,You obviously don’t need to respond to this but if I might bug you for additional info on your opinion, because I’m having trouble understanding your disagreement.Are you denying either of these two facts?1) iPhone 4S has specific hardware to -improve- Siri functionality (not saying it is not possible with older hardware, but I am stressing the improved part)2) Siri servers have had issues (especially early on, but still today) with downtown. That downtime could be due to either a) server load, b) bugs, c) upgrades to the beta software.I feel like if we can agree that the facts above are true, than I’m having issues with your conclusion. It is a very well established fact that Apple has high standards of functionality for feature products (yes yes, they’ve released crap, but with regards to stuff they consider feature) and it is also well established that Apple almost never releases what they would term ‘beta’ software.I think you have a stronger argument against the iPad 1 getting iOS 6, because the first iPad’s processor is the same (similar?) to the iPhone 3GS. There are certainly differences between the iPad and iPhone 3GS (such as resolution and graphic requirements) but overall this seems more of a grey area to me.Whether or not Siri is technically possible on the iPhone 4 is unquestionable. Yes, Siri is doable on the iPhone 4. However, if you consider both the server issues Apple has had with Siri, Apple’s continued demonstration that they will err on the side of customer experience (not wanting to sacrifice it), and the hardware specific to the iPhone 4S to improve Siri behavior… I’m having trouble seeing how a rational person would disagree with this point.Anyway, no need to respond, but you do have me curious.

      5. JohnDoey

        You have it totally backwards. 3 PlayStations from the same vendor over 15 years and 5 phones from one vendor over 5 years is absolutely nothing like Android fragmentation.When the first iPhone was left behind at iOS 3.1.3 (it had MMS by then) the model was over 3 years old, had not been sold for 2 years, and represented less than 1% of the total user base. It was also US/AT&T only. Those users were fine for at least another year also.The reason Android gets criticized over this is that Android is broken in this way. Apple does not get criticized for this because iOS is not broken in this way. > there re dozens and dozens of companies making AndroidBut the consumer only buys one phone every 2 years and they use it 10 hours a day. Excuses for why Android is broken don’t help the consumer.> buy a Nexus> hackNexus is a niche product, very few are available to phone buyers. Hacking is even more niche. So you are explaining why you and Sergey like Android, but not explaining why a consumer should suffer through it.

    12. awaldstein

      They are not going to fix anything soon IMO.

    13. fredwilson

      as a user the “mess” as you call it makes me love my android more. sandboxes are messy but fun to play in. sterile jail cells are not.

      1. William Mougayar

        I’m torn between being a user and a developer when issuing an opinion on Android vs. iOS. It’s like a 2X2 matrix that has 4 answers. Where you sit determines where you stand.

        1. panterosa,

          @wmoug:disqus I think much as you do WIlliam, as a user and as a developer. I have an iPhone and an iPad, and I’m doing an app for iPad. I can’t wrap my head around the app for the android market, and won’t, until I have many requests, or more revenue.

          1. William Mougayar

            I can’t wait to see your new iPad app! Great.

          2. panterosa,

            @wmoug:disqus Actually, there will be two apps, one botanical and one zoological. Gaming to learn by, FTW.

          3. ShanaC

            That sounds fun! I wish I could test, but I am on android

          4. panterosa,

            You can test the new board game, which is seeking a licensing deal.

          5. ShanaC

            are you actively playtesting?

          6. panterosa,

            ALWAYS!!!!! Every age group.

          7. ShanaC

            maybe I can get a group of friends together to playtest 🙂

      2. Crispo

        So you say … But not fun to play in when you need to get work done. The overwhelming majority of consumers prefer predictability and consistency. The whole point of what we do is to serve the needs of the consumer. Jail isn’t a pleasant place to be and you don’t go there by choice. Apple doesn’t force consumers to buy their product – it’s a weak and tired analogy that misses the big picture. Don’t confuse what WE (creators and hackers) need – a sandbox to play in and what our CLIENTS (consumers, the end user) wants or needs – a consistent, pleasant experience.

        1. fredwilson

          not all clients want clean

      3. JohnDoey

        That is a sort of Computer Nerd Machismo that is completely irrelevant to 99% of humanity. Most people are working way, way higher in the stack. Most people find iMovie and GarageBand to be much more fun than computers, and those apps are only on iOS because it is not a mess. The graphics and audio video subsystems from the Mac are there. Most of my iOS time is spent recording multichannel audio and MIDI. The wireless MIDI on Apple platforms is fun, and productive. Controlling a million dollar recording studio from an iPad is fun. Android just doesn’t have anything like all that. Not because a 3rd party did not build an app, but because the OS itself is lacking the infrastructure” Somebody has to build dozens of operating system components before Android can be fun for audio, video, photo editing or music recording, or a real 3D game platform. The things that most users want to do, not just the computer nerds.I’m not sure why computer nerds can’t understand — there are like 1000 kinds of nerds, not just computer nerds, and WE ALL NEED COMPUTERS. Music nerds, art nerds, photo nerds. Imagine a15 year old kid with little money and all he wants to do is record his own music. He can plug an Apogee MiC into an iPod touch running GarageBand and it just works. No computer stuff. No drivers, no setup, no glitches, no measuring megabytes, no folders, no file storage, and NO CRASHES, and PRO QUALITY. That is not a toy setup or hobbyist setup. He has more tracks and instruments than The Beatles ever had. There is just nothing like that outside of Apple platforms. The infrastructure is not there to support it. It only works because thousands of people at Apple did not just ship a mess and tell users to learn Java and look for opportunities. GarageBand literally goes back to 1986, when it was called Notator. This stuff is real computing, it is not going to appear on Google’s 2005 -era Java phone OS any time soon. Computer nerds who sell Android to music nerds are basically destroying their computing lives. You are giving your musician friend a chemistry set because you are into chemistry. Total fail.Google+ — also a mess, also popular with computer nerds. Not an alternative to Facebook for most people.

        1. CJ

          Actually those apps are not on Android because Apple wrote them. As far as ease of use, yes Apple wins there. They give you one bowl of dogfood and tell you to eat or starve. Really easy to use.Google leads you to the Dogfood aisle and tells you to take your pick. If you can’t find something suitable still, they’ll take you to the plant and let you create your own recipe. Still not liking it? Fine, let’s go out and gather the raw ingredients. You might call it a mess or fragmentation, I call it choice.

          1. Donna Brewington White

            Well said, Malcolm. I’m not a developer or even much of a techie (although maybe a latent geek), but android seems to open up the playing field. I want to support that with something as simple as the smartphone I choose to use.

  19. baba12

    Why build for an OS, why not build using HTML5 and javascript, then you are OS agnostic and you should be able to run on top of any device that has a open standards based browser.There should be very few applications that need to have access to specific OS functions that forces them to have to develop for a specific OS. As usual I see enterprise IT defining some rules ( cockblocking) to make vendors have to conform to their stupid rules to get their business. Maybe Bring Your Own Device (BYOD) will become the norm at the enterprise level and the work of enterprise IT will become less about fiefdoms and more about service.Having seen the past wars of developing for a particular stack UNIX, Windows, OS 400 and the underlying stack that basically allowed certain players to stay entrenched is not the way I see or wish to see technology adoption. TOo much energy is expended with this model.As a developer one wishes to keep users from having to define what they use as a service to be based on a platform.Having the last post about open standards, I would hope that Mr.Wilson (Fred) would proselytize the open HTML5 and talk about how one can build really rich applications/services that run on top of a standards based browser.Will Microsoft finally throw their towel?Fragmentation is good, watch out Dell’s of the world what will you do when the BYOD concept becomes the standard for the enterprise.Oh blimey!

    1. William Mougayar

      Baba- good points, but even at the HTML5 level, you need to test for compatibilities across the various Android platforms (see my other comments here). A mere difference of browser version can break something. We have had experience with that.As for BYOD, it does come with its management nightmares / overheads. You need another platform to manage that, and BYOD is a problem for the Enterprise, not for the user or the developer.

      1. baba12

        Agreed nightmares for management, but they don’t have to be, that is exactly my point, nightmares are created and sustained for a purpose for someone to manage. If you follow certain standards and adhere to those standards then nightmares shall go away. In all my years in technology I have not seen a situation where a given process that was technology enabled had an issue with technology, it was more of an issue with the management of that technology and we humans tend to add barriers/layers to keep things complex as a way to sustain our own lives. As for the difference in browser versions etc that is again a human made problem not one of technology. I have been advising a small fortune 500 company on getting away from the traditional IT business unit that supports the company to just having a very robust network based on standards. They are in the process of not purchasing any more computers for their 7000 desktop users. We will see how this strategy works out but so far the CFO has been happy with the pilot program and the users have all been very happy.By the way when are you coming to Brooklyn, I believe we still have to have a drink.

        1. William Mougayar

          Ah yes…..I’ve got 2 people to see there now 🙂 End of Sept hopefully 🙂 Thanks for the reminder.

    2. ShanaC

      why do you think IT cockblocks?

      1. baba12

        That’s the general nature in most enterprises. If anyone in IT comes up with a better solution i.e. removes friction or reduces complexity etc they are considered a threat to others in the IT department and thus they will find they get blocked out. IT is one of a few business units that tends to have a relatively low attrition rate, if you are in sales, you under perform chances are you get laid off, in IT under performance is generally rewarded and you get to keep your job. Generally when there is a desire by business units to use a product/service, instead of supporting it IT , will state that said product/service opens up security holes etc and will vehemently try to deny use of that product/service by the business unit. If IT departments instead were prepared to say we have issues but we shall find a solution to make it work in the environment that would be a progressive ideal. Most IT departments are reactive and thus are always behind in deploying solutions. IT departments in Type B & C companies are more like that than Type A companies i.e. say wall street banks or some small media agency. Type B would be a Proctor & Gamble and Type C would be your Federal/State/Local Governments.Cockblocking in Type B and C companies is prevalent and strongly desired.

  20. Josh Rutstein

    Given the early investment in FB by Microsoft, and the shared enemy in Google, I wonder if the FB phone would be launched on a MSFT platform. There must be a lot of lobbying going on there

  21. kenberger

    I agree so much that I’m adopting the motto: Embrace Fragmentation.When I attended the Tmobile G1 launch a few years ago where larry and sergei skated in to a venue near the queensboro bridge, they proclaimed that Android would usher in dozens if not hundreds of models from an endless array of manufacturers and carriers, bringing hi-end and low-end phones. The clear benefits are WIDESPREAD INNOVATION that you’ll never get from just 1 company, and MORE FREQUENT CHOICES (the clear downside is developer headaches). Critics were skeptical, but it has totally happened.The Galaxy S3 I have here, and the Nexus 7, are things of grace and beauty. And there will be ever more devices soon that will blow us away*. I wouldn’t count on Apple taking the entire part of the hi-end mobile market for long.*See bgurley’s article “The freight train that is Android”.

    1. fredwilson

      i am getting an S3 any day now. can’t wait. i hope it is running jelly bean

      1. kenberger

        It won’t have Jelly Bean yet (at least not this week or next), *especially* if you got it via a special Google PR contact, and you’ll notice JB’s absence compared to ICS. To load JB today, the best bet is the Cyanogenmod 10 build, which is pretty good but still buggy (I can help if needed).You also won’t be blown away by the S3 compared to your GNex phone, other than a noticeably longer-living battery and slightly warmer-looking screen.

      2. Srikanth Yada

        Why pick an S3, if you are going to put Jelly Bean anyways? Why not a Galaxy Nexus or HTC One (industrial design, not plasticky looking, etc).

        1. fredwilson

          i have a galaxy nexus.i just like to play around with different devicesit’s so easy to switch from device to device on android with all my services in the cloud

      3. kidmercury

        samsung has stolen the coolness factor from apple — a very impressive accomplishment. i will be rocking the big ass galaxy note in a few weeks. can’t wait!

      4. Vitomir Jevremovic

        It rocks. iPhone 5 will have to do so much more to win the fight.. But not with S3.. With S4 and S5. Also have in mind Samsung dominance in LCD business. S series could become native second-screen controlers. Future is fun

  22. kidmercury

    fragmentation is where it’s at and it leads itself to integrated ecosystems designed for a specific niche that promote a certain set of values. or to put it more simply, just do your best at copying amazon.

  23. Ryan Frew

    If this were 20 years ago…iOS = AppleAndroid = MicrosoftMicrosoft = Linux, or somethingIt’s almost the exact same discussion

    1. ErikSchwartz


      1. PhilipSugar


      2. Techman

        It would actually be nice to see Microsoft adopt Linux, since it seems that Microsoft does not love Linux, but they love Mac OS, a form of UNIX.

        1. Wavelengths

          The copycat not looking at the origins?(Sorry, I seem to be on a tear tonight.)

    2. Elia Freedman

      On the surface it may look that way, but in reality it is nothing like that at all. I encourage you to look more closely at what is happening in these markets.

      1. Ryan Frew

        I’m certainly guilty of oversimplifying. Although I wouldn’t stretch to say that “it’s not like that at all”.

        1. Elia Freedman

          I would and it is a long list why. MS had 90%+ market share. It had no competitor. No one chose whether to develop for Mac or Windows, there was only one choice. The only people developing for Mac were a very small group of enthusiasts.iOS is the dominant platform for developers who wish to charge for their apps today. In the US, smartphone market share is ~50-50 and tilts heavily in iOS’ favor when considering tablets and phones. Even worldwide, with tablets, the numbers are much better than the 60-20 market share split. Unlike the PC market, developers risk a lot by ignoring iOS even for a free product. In fact, I personally can not come up with an Android-only success story off the top of my head.

          1. Ryan Frew

            Market share is a great point. It was probably a miss on my end saying “twenty years ago”. Hell, I’m 21. Maybe it would be more accurate to say the conversation was more similar as recently as 7 years ago.

      2. Ryan Frew

        I took an excerpt from Fred’s post to convey what I mean. Although this is certainly just on the surface, like you said,”So imagine a world in which three of the top four consumer tech companies (Gateway, HP, IBM, take your pick) have computers running Microsoft. Does that sound like a fragmented world for Microsoft? Yes. Does that sound like a recipe for having a massive number of Microsoft devices out there to build to? Yes.In my view, we are in a two OS world and I think we are going to stay there. I think Apple will own the high end with the best and most integrated experience. And I think Microsoft and its many variants will own the rest of the market. I think everyone else is playing for crumbs in terms of market share and would be better off joining the Microsoft variant parade.”

        1. Wavelengths

          This looks to me like an opportunity for someone to create a “third world” environment, and I mean that in the sense that the 3rd world could leapfrog the two existing dominators.Even Apple is stale. And Microsoft has always seemed like a huge mish-mash on top of DOS, to me. (Although I may be terribly naive.)So, now that we know a lot about how people interact with computers on many different platforms under many different operating systems, how about someone comes up with something truly new and integrated and built for the semi-literate majority? Oh, which would instantly pass the litmus test with “enterprise.”

        2. Vitomir Jevremovic

          High end is also Samsung with S3

    3. ShanaC

      except android and ios are built on linux….

    4. Wavelengths

      Microsoft always seemed to me to be a “me too.”

  24. Abdallah Al-Hakim

    One of the biggest issue with Android fragmentation is that not all devices receive most up-to-date OS version. This is very frustrating which is why I recommend Android users to at least stick with Google Nexus phones. Also, Apps don’t work as well on all devices and I am constantly receiving updates to apps because they have been further tweaked to work better on other devices.

    1. fredwilson


  25. LaMarEstaba

    iOS is for a consistent, standard level of performance. Android is for those who want to customize their phones. Android has a much better share of the market outside of the US because people need to have phones for different purposes. When I studied business at Peking University, my professors told me about the shanzai cell phone industry. It’s worth taking a look if you want to understand how emerging markets or growing economies use the existing technology:

    1. fredwilson

      thanks. i will take a look.

  26. Dave W Baldwin

    You need to develp for both…keeping in mind it will be three.The AskZiggy thing in Windows will be placed in Android and looking ahead two years, the evolution of Android into a less fragmented will take place. Windows will be developing to their own vehicle, but look to being in multiple vehicles.Something you left out @fredwilson:disqus is the probability of Google producing their own phones now that they have Motorola. If they don’t, a competition will grow between them and Micro to be on multiple phones.This plays well into the evolving into ‘control’ devices joining the user’s toys. So we should thank the hard work of Nuance.

  27. markslater

    great post. Bang on.Android is tricky – we have comfortably settled in on IOS and are 7 or 8 updates in. Except for the horrendous metrics that the app store provides ( i mean horrendous – i feel like i am looking at lotus 1-2-3) – IOS is a very stable and simple dev environment.But we are taking our time on Android.the world is indeed changing. its a 2 platform race – that’s it thats all – sorry to all you BB lovers – game is over.There are massive tectonic shifts beyond just mobile platforms – how companies are being run – its amazing to watch…..i picked up my wife’s ipad this morning and it started barking at me with meeting notifications – i said to her – you have 9 meetings today. her response – “i know – death by meeting” – i said – why don’t you try and get everyone to stand up in the room… will go quicker…..(she works at P & G)

    1. kidmercury

      you took the time to diss BB but not MSFT? i’m starting root for MSFT in the mobile wars because they are such an underdog here. they have xbox, surface, evil troll patents on everything, legacy customers, and an enterprise foothold. they ain’t done yet! msft ftw!

      1. markslater

        i didnt know that they actually built mobile software…..

  28. markslater

    i am thinking of creating an ODE TO TWITTER conference room at our office. When you enter, you have 160 characters to get your point across…..what do you all think?

    1. fredwilson

      i love it

  29. rohun

    Fred, I would add two additional considerations.1) Google’s “Android” and Kindle’s “Android” operate in different distribution ecosystems. The Google Play store (and native google apps) exist only on Android devices “with Google”, which the Kindle Fire is not.For developers to distribute on Kindle, they must sign a separate legal agreement with Amazon for distribution, which has different economic considerations than that with Google.2) One of the biggest issues with Android fragmentation is tied to adoption of new releases of the Android OS.As of Today (see:…, only something like 16% of Andriod devices in market are running Ice Cream Sandwich. This is the version of Android that was released over a YEAR ago! Additionally, only a small fraction of the devices in the market are going to receive the update! With average phone ownership in the range of 18-24 months, you would think that Google would work with OEMs to at least support the newest OS during that period. This is a major issue.Contrast that with iOS where even the latest, yet to be released operating system iOS 6.0 will run on every single iPhone 3GS. These are phones initially released in June 2009 (now over 3 years old).That means developers for iOS can take advantage of most of the latest and greatest SDKs and APIs (hardware specific ones like front camera would be off the list of course) and know absolutely that there is a huge % of devices in their adressable market.Hopefully the recently announced google PDK will help OEMs come to market sooner with devices launching closer to the release of the new OS versions, but I’m not sure Google can force OEMs like Samsung who are making a ton of money on devices to keep from making their devices prematurely obsolete by witholding OS updates. It’s a real problem.

    1. fredwilson

      excellent comment

  30. Cam MacRae

    I love my Fire.I love it even more having replaced the Amazon launcher with Go Launcher HD.But I won’t be getting another one because dead branches are a bitch. Instead I’ll be going straight to the source and getting a Nexus.

    1. fredwilson

      you will be happy when you do that

      1. Cam MacRae

        I’ll take that as a recommendation. Google annoys me as they’ve priced the Nexus 7 $52.40 higher if I buy it locally, because, if you can believe anything published by Denton’s rags, of the relative income difference. Bah! That makes the 16GB $319 delivered.

    2. kidmercury

      amazon has many tricks up their sleeve — kindle fire 2 probably comes out Q4 this year. they need to up their game for sure but i think they could win the hearts of developers. i think amazon is best positioned to make it easy for developers to make money, and is most inclined to cut developers a deal they can’t resist. if they can pull this off… ftw!

      1. Cam MacRae

        I spend $100 a month with Amazon, so I’m fairly committed to their ecosystem, but the Fire should have been (can be!) so much more. I’ll gladly use both app stores on my Nexus.

  31. Dan Epstein

    I’m a big Apple fan, but I want Android to do well, because a little competition is best for everyone. My concern with Android is who’s making money on the platform. Is it the developers? The hardware companies? The carriers? Google? I want there to be (at least) two choices for mobile. If Android’s going to be a success, it has to make money for some, if not all, of the aforementioned. And I’m not sure it’s doing that.

    1. fredwilson


      1. Dan Epstein

        Good point. Horace Deidu (asymco) also has a nice series on “Android Economics”. Here’s the first post:

  32. Dudu Mimran

    The problem with Android from a developer standpoint is the inconsistency of user interface and experience across different devices and since most mobile apps are mostly about experience then anyway a serious work needs to be done. The variety of Android versions across different devices where Google made sure there are major changes over there makes it also a big issue. The main obstacle as I see it is even understanding on which devices your app works and on which not and why, there are som many out there with different distribution level in different markets which makes testing very costly and cumbersome. This situation makes the plus much smaller. I see Android becoming more and more as “Linux for mobile” where it is the core mobile kernel with different shells run on it. If you are developing a low level utility then it is great, otherwise…

  33. matthughes

    Isn’t it possible that eventually there will be a fallout between Amazon & Google (Android)?As well as Facebook & Google?Apple & Google were cozy and now they kick each other in the shins every chance they get.Right or wrong, it just seems like Amazon & Facebook will eventually want complete control of their respective mobile platforms.

    1. ShanaC

      I’d be surprised if there wasn’t more fallout. I’m waiting for the patent wars to get hotter to see what that fallout will look like

    2. kidmercury

      absolutely. everyone is just piggybacking until they have the muscle to roll their own. but, they may never have sufficient muscle to do that.

  34. Elia Freedman

    We are developing an Android version of our primary products right now. It is a paid app and have minimal expectations for it. I will say that it is amazing how much customization is required for every screen size. In the end, we are going to have a phone version, a 7″ tablet version and a 10″ tablet version. (We are highly screen dependent.) there is also significant differences and have been told we will likely need graphics in as many as four different sizesGiven that these problems of fragmentation have been around a long time. Once upon a time it was Palm OS and Win Mo. Then it was Win Mo, BlackBerry and Symbian. Symbian had many of the same issues as Android from a fragmentation perspective. And RIM released minute differences to every carrier for every phone model, too.In general I agree with Fred. Especially for consumer facing free apps, iOS and android are becoming requirements.

    1. William Mougayar

      I’m glad someone else is echoing my concerns about screen size being also a fragmentation factor.

      1. leigh


  35. Todd Werelius

    First let me say that different home screens from different mfg devices is not a very serious form of fragmentation in my opinion. Most users only use a single device for quite some time, so they don’t really notice or care about this.The general usage patterns are all the same anyway and from a UX standpoint that’s all that really matters. While it might take a few seconds to figure out where a mfg stuffed the settings option when you get a new device, or orient yourself when a major pattern change appears it’s generally not really an issue, iOS has also added new patterns I might add.I have a much harder time moving between Android and iOS devices. My wife prefers to carry an iPhone but likes the Android UX pattern and misses her “back” button pattern from Android.Now as far as actual fragmentation it’s much worse for developers than the average user thinks. There are multiple types of fragmentation that Android developers have to deal with and all of them add up to a lot of development friction, as well as bad PR if not handled properly.Mfg Implementations. Any of the hardware API’s are problematic to use without deep device certification since Mfg’s all seem to introduce glitches of one sort or the other. Sometimes this means it doesn’t work correctly, other times it provides the wrong results, sometimes it’s poor performance etc.For instanceThe camera for should work the same across all devices using the same API level, but of course it doesn’t which means you either have to code for glitches, not all of which are known, or use a lowest common denominator approach.On most carrier/device combinations you can ask for the most accurate GPS provider and it works fine, on others it hands back the wrong answer you get lower resolution results, same for power management features etc.Since most developers don’t have access to all devices this issue manifests itself asangry users posting that the app sucks since they have no idea it’s a device specific problem and quite honestly if they did they would and should not care, in the end it’s always the developers responsibility anyway.Getting this resolved is a major headache unless your pockets are deep enough to buy a lot of devices, or pay a service that allows you access to the same.Google should either provide a very low cost “Virtual Lab” for developers, or demand that mfg’s do so. They should also do a better job with certification by actually updating their acceptance testing to reflect known gaps that allow these kind of things to happen. The could also demand corrective action for devices that have issues in order to continue to use the market.OS API. If you’re willing to put up with 9x% device capture then it’s not a terrible problem since you can basically just write off the older versions, if you can’t then it adds quite a bit of overhead if you want your app to upscale with the newest features.Google does provide a support library that makes some newer features available to older devices but they could do a much better job with this.There are well known add on’s that do a better job than Google does and to my thinking that is just ridiculous. Adoption of new OS upgrades should be better and more timely, but considering that each mfg has to do this on their own it’s not surprising they are not.Screen Sizes. I think Google has done a fairly decent job with coming up with ways to manage different screen sizes ( much better recently ) though it’s a bit of a pain to implement in all circumstances.For instance the native Map library doesn’t play nicely when using the support library and fragments which makes it difficult to provide universal supportThere are work around’s but implementing them is not a good use of one’s time and Google could easily do better with this by hiring the guys that fix their problems for them or at least making a document that aggregates the problems and solutions.In general this adds time to implement correctly though at least there are “classes” of screen size that are easy to test on the emulator, but it still requires multiple levels of resources (graphics) to support properly which adds cost as well as bloat.Device Capability. Android has a fairly good way of detecting and handling this both on the platform and in the market, so it’s more of a nuisance than a real problem but it also adds time to defensively develop for this and make sure that the device the user has can run your feature set properly.Market FragmentationThis is a newer problem, with Google having it’s market, Amazon having it’s market, and if Facebook takes the plunge same there. This has advantages and disadvantages but it also adds to the general deployment overhead, analytics etc.When you add all of this up development on the iOS platform (which does have issues of its own) seems like nirvana in comparison.</rant>Once you learn where the gaping pits of fragmentation are it’s not that hard to avoid them, well most of them, it’s just fatiguing and makes each deployment cycle longer and more costly than it sometimes needs to be.

    1. falicon

      Good stuff.It’s def. daunting as a dev. right now…without really knowing it, everyone that cries for you to build out a version of your stuff across both platforms (and the web) is basically saying “You must go build your service/app 15 or so times, right now, quickly”…really really tough to do quick and with quality…

    2. fredwilson

      this is a great comment

  36. Carl Rahn Griffith

    Reminds me of the heady days of Unix. Now, that was a smorgasbord of choice!

  37. Andy

    It might not happen for 5 more years, but I believe you will see the majority of apps to be in the cloud eventually – when this happens, WHICH mobile OS you have will become less relevant – and as a developer, you will be developing for the cloud, not IOS or Android …. Shorter term, you should recognize that IOS is also fragmenting….even more so if/when they release a smaller tablet, and a TV device

    1. leigh

      From your lips to Gds ears my friend. The day we only have to test for ONE system and stop the madness will be a happy day.

  38. Anthony Ortenzi

    As entrepreneurs, isn’t having a problem to solve the starting point for building a business? If we assume the solution isn’t “give up, all hail Apple”, then there’s a huge opportunity *because* of the fragmentation.

  39. chudson

    I am CEO of a company that builds for Android and I think it’s actually more complicated that simply saying “fragmentation” – there are actually at least 4 major variants of Android that you need to focus on when thinking about how to design your application and UI / UX. To me, these four variants are more like sub-markets than points on a continuous spectrum that “fragmentation” suggests. The four key markets, in my view, are the following:1. Amazon Kindle Fire – Similar to Android minus some of the Google-specific functions (Google IAP, native Gmail client, Google auth, etc).2. High-end Android – Designing for phones like the S3, Galaxy Nexus, HTC One and other high-end devices that have fast processors, large screens and lots of memory.3. Low-end Android – There are tons of Android devices out there that are spec-light in terms of screen size, screen resolution, and processor. Designing apps (especially games) that run on those devices can be really difficult.4. Tablet Android – The 7 and 10 inch tablets are great, but they too are a different beast. In the same way that some folks are focusing just on optimizing for the iPad, developing apps for tablets requires some special considerations if you want the app to run on a tablet. For example, the Nook, like the Kindle Fire, runs it’s own version of Android and has its own store.Just wanted to add a bit more perspective to the conversation about what fragmentation looks like from a developer standpoint.

    1. fredwilson

      great comment. thanks for sharing that.

    2. ShanaC

      this looks like a headache to deal with.

      1. chudson

        Simple way to make life easier is to not stress out about supporting the most basic set of Android phones.

        1. ShanaC

          Thanks for the development advice!

    3. kidmercury

      damn this is a good comment

      1. chudson

        Thanks, glad it was helpful!

    4. Abdallah Al-Hakim

      great points and hence are the reasons for being wary of getting another android device. As a developer, do you treat all 4 subgroups equally or do you focus on particular ones?

      1. chudson

        We focus mostly on tablets and high-end Android phones – it’s a headache to deal with the low-end ones.

        1. Dave Lee

          Agreed that the low end phones are difficult to deal with… but there are so many it’s hard to ignore.

    5. Matt A. Myers

      Any thoughts on using responsive frameworks and then packaging them? How do they compare to native from your experience?

      1. chudson

        No direct experience there. I will say that we’ve tried some cross-platform tools and native wins in terms of user experience every time.

        1. Dave Lee

          Totally agree here. We’ve decided to go native across all the mobile platforms because of performance. But I think the cross-platform tools can make sense for some developers who have different priorities.

        2. Matt A. Myers


    6. Dave Lee

      Actually, I would add one more – Android China. This is a huge market and it’s inaccessible through Google Play. Meaning as a developer, I need to take my localized build and manually create accounts in various Android China stores to get them uploaded. Totally a pain.This is in addition to the other 4 markets chudson mentioned.Vs. developing for Apple – we just need to focus on iPhone, and iPad if we want that market. It will be interesting if the 7″ iPad Mini adds more complication. Also, the new iPhone with it’s 16:9 aspect ratio is another interesting move that hopefully won’t make things more complicated for developers.

      1. chudson

        Dave, I agree – I have about 50 different markets in China that email me monthly claiming to be the largest source of Android in China. I could also add Korea to the list, as the SK Telecom store and several of the other carrier store offerings are as large if not larger than Google Play in terms of ability to move installs.I’ll be curious to see how diverse the iOS ecosystem can get before developers begin to howl about fragmentation.

        1. Dave Lee

          chudson – yeah, we also have a difficult time keeping up with the popular China android markets. We had a to hire a Chinese person to research and apply to the markets for us. Your point about Korea is interesting as well.Regarding iOS, Apple seems to come up with easy/creative ways for developers to deal with retina/non-retina, iPhone/iPad, etc. But it will be interesting with new devices and screen sizes. Not just the new iPhone, but what if there’s a 13″ or 15″ iPad that comes out?

    7. Techman

      Developing for iOS could sound easier, but at this point I wouldn’t focus on them so much because Apple is moving to a point where everything is going to be locked down and you have to do whatever they want you do to just to get your apps on the Mac App Store, not including having to pay for a developer account every year.

      1. JohnDoey

        Developing for iOs doesn’t just sound easier’ it is easier.All vendors are locking own more stuff every day because users want that.Dealing with Apple’s one app store is child’s play compared to dealing with dozens to hundreds on Android.You have no idea what Apple is doing in the future, but Apple is on a more consistant path than any Android vendor.

  40. David Kuriakose

    Why would Facebook build an Android platform when they have their partnership with Microsoft? iOS / Android … they will all migrate to the web just as desktop applications have migrated to the web.

    1. fredwilson

      i don’t have any inside info that says they are building on android. but if i were them i would do that

  41. hypermark

    Dumb question, but what are the “breakout” examples of non google apps that launched first on Android? I can think of many on the iOS side of equation.

    1. Irv Remedios

      Lookout Mobile is one. Most apps take the iOS route since it’s an easier development path (no fragmentation issue). There’s also a heavy press bias towards covering iOS apps.

      1. hypermark

        Thanks a bundle. That’s an interesting example, inasmuch as there is basically no security software market on iOS.Are there good consumer-y examples?My bias on this one is that the tilt for developers towards iOS is two fold. One, far more developers are iPhone (and iPad) users than Android, and developers tend to develop first for the platforms that they actually use.Two, creative types see their greatest value-add higher up the stack, and when you have to get into the bowels of worrying about OS variants, device type variants, API variants and disparate channel and distribution considerations, that changes the creative equation.

      2. ShanaC

        Why does the press lean so heavily on IOS? I get the feeling that I should be looking more closely at potential breakouts on android (pintrest style)

    2. fredwilson

      very few, if any

      1. hypermark

        I think that there is a moral of the story there. Can a great platform emerge on sloppy seconds alone? One could argue that that was the story about Microsoft during PC era, but the distinctions are pretty material.

  42. Dale Allyn

    The blanket statement of “build for Android first…” must presuppose that one is building a completely market-agnostic product. If one (reliably) determines his/her market to be weighted toward iOS, then of course one must adjust to cater to users if not launching both OS versions simultaneously.In the past, iOS was an obvious first choice, but as you are suggesting, Fred, times have changed. Know your market and build for it. That may mean deliver a great web-app (mobile optimized) first that accommodates both well, then launch the mobile apps on each platform. In a perfect world, launch for both. If resource constrained, launch for the largest opportunity first, but don’t delay in delivering the second.

    1. fredwilson


  43. Eric Falcao

    It’s very disheartening. I am very excited about Android in general but only as it comes packaged from Google. I love my Galaxy Nexus. It’s a fantastic phone and has really turned me on to Android as a platform. Too bad most other incarnations I’ve touched are sub-par (Kindle, Sense, etc).

    1. fredwilson

      as native android gets better, the carrier variants may die off. but it seems we are moving to another form of variant

  44. gregorylent

    fragmentation doesn’t matter to young consumers from shanghai to sydney …the large form-factor of the htc and samsung galaxy products has swept the coolness factor right out from under the iphone’s feet.

    1. ShanaC

      what about old customers. They have more money,and very different needs. Young customers have more time in a way to hack around

      1. Wavelengths

        Young customers may not even have a job that pays for the phone. Their parents are hoping …. Seriously consider what the older customers need.The old rule used to be that a company CEO could be right 51% of the time and still make money. So the company I worked for booted the inventor/CEO, and brought in a headhunter’s choice who had no clue about the market.The company tanked. I have no idea where the CEO is, but the technology exists today in a way that could be monetized in an instant. But there is no one to talk to.The floodwaters of new technology left that company behind. No one was looking at the macro vision.The old customers might pay serious money to get back in the game!

  45. Matthew

    The biggest problem with fragmentation is not knowing the answer to this question:”Will [insert app name here] run on my Android device?”

  46. Joe Beninato

    Fred, I think there is no denying that many mobile app developers avoid Android and go iPhone first because of the fragmentation issue. Instagram is probably the best example of this. We’ve done native iPhone and mobile web tested on iPhone and Android, and I can tell you that Android is one or two orders of magnitude more complicated to test for. For startups with limited resources, it is hard to invest the resources necessary to build and test great products that work across all Android devices. The good news from my perspective is that I believe Android handset makers will realize that having 10 SKUs doesn’t make sense, will concentrate in less SKUs that are more successful, and that will help developers from a testing standpoint.

  47. Max Yoder

    These aren’t the droids you’re looking for.



  49. vgnyc

    If crowd sourcing & mass user engagement are key and all data is valuable irrespective of origination then I want Android at the altar and IOS as a bridesmaid. Mess I see as corollary to reach.

  50. Wavelengths

    You have a Kindle Fire on your bedstand and I have a paperback book in my purse. Two days ago, as I finished the last 3 pages of the John Land novel, I was ready to hand it off to the woman next to me who was also waiting in line, with 10 people ahead of us. I have an Android Samsung with T-mobile, an I-book, a Sony VIAO, and I use my son’s I-Pad, and I love how my neurologist uses the I-Pad to spin those MRIs around.If I’m backpacking, with no cell phone service (as I have in Borneo, Thailand and peninsular Malaysia), I might want to leave a paperback behind. But, as a reader with easy access to wi-fi or cell-phone connection, I’m delighted to be able to download. If I plan to send a book I loved to my daughter, along with the 2nd and 3rd books in the series, I might mail them, media rate, and she can trade them in at Half-Price Books in Austin, where the many Luddites can access the same literature.Each to his/her own. Which is what we shouldn’t forget when we are looking at markets.

  51. Josh

    Fragmentation, it exists in the transportation market. Just because there are a dozen different car manufacturers doesn’t mean that horse and buggies are a better choice. All of them take gasoline and get you where your going. Fragmentation means choice. As long as all of them run android apps, and a developer can target android, is the consumer really affected by ‘all that choice err fragmentation”?

    1. fredwilson

      nice comment!

  52. MicroSourcing

    Android would do better to streamline their apps, but if these different platforms have their own business interests, the fragmentation issue may have to drag on longer.

  53. JohnDoey

    I suggest you talk to developers and content producers about this, because what are airy, philosophical, future visions for you are cold, practical facts for us.You make the common mistake of assuming “if it can be done on iOS, it can be done on Android.” That is fundamentally incorrect. Both may be planets, but iOS is like Earth, and Android is like Mars. You have to design a custom billion-dollar sky crane to land on Mars. On Earth, 10,000 flights daily. You buy a ticket. You are asking FedEx to ship to Mars because it is also a planet. The infrastructure is simply NOT THERE. Your answer is: it will be one day. Maybe, but I still cannot ship your package to Mars for $15.Excluding the Web, over 90% of the world’s client apps are in a C language. iOS, Mac, Windows, Linux, PlayStation, Wii, and Xbox all run C language apps. Android runs Java. A well-known media player was ported from Linux to iOS in weeks by someone who was not the original author, but it took the original author 2 years to rewrite it in Java for Android. Ironically, there was zero demand for it on iOS where playing media is second nature, and lots of demand on Android, but the limitations of Android made it hard to fulfill that demand.You are on the 14th floor speculating that your building has a basement that goes one mile deep, just like the Apple building next door. But your building has a standard basement. Nothing special was done there. Developers cannot build you a bomb shelter or hydroponic farm on level -200 because you don’t have a level -200. You have some pipes and a furnace like everyone else. The fact that Google has alternatively been telling you they are building such a basement or that you don’t need it is not the fault of developers and can’t be fixed by outside developers.And to your main point, myriad devices are not good for app development. There are millions of generic devices, but still, most consumer apps originated on iPad, iPhone, iPod, Mac, Wii, Xbox, and PlayStation. Word and Excel are Mac apps from 1985. Microsoft Office is a 1989 Mac app. Photoshop is a 1992 Mac app, it did not run on Windows until version 4.1 or so. WorldWideWeb is a NeXT app that did not come to Windows until 3 years later, and didn’t really work until 5 years later in 1995. Windows was originally just a way for Microsoft to bring their Mac apps to DOS — if Apple had not stood still from 1985–1997, then Microsoft may not have had time to do it — to herd all the DOS PC cats together and get a platform that could run Mac apps from the previous decade — and even so, Windows was a massive technical failure much worse than The Big Dig. XP was the most popular version, and it shipped with all users as root, created the bot net, and then after 3 years of improvement inside Microsoft, it had to be killed and they started over, making Vista from their server OS. So you are expecting Android to be a place where new apps spring from because there are a lot of devices, but that is the opposite of how it has always worked.

  54. felixc

    While I think it could get messy, I think back on the overall user experience with early “smartphones” and even a fragmented Android experience will still push the market forward for consumers.Personally, I’m pretty committed to the Apple experience since my time is limited and I want a consistent experience. For example, the AppleTv is a great accessory for the iPhone and iPad.

  55. John Rorick

    Forget all of this Android Apple banter – @fredwilson:disqus how you diggin’ the Nexus 7? A few of my very apple centric colleagues are gushing about it. Especially at that price point. One of the surprises being the usefulness of the size.

    1. fredwilson

      i love it

  56. george

    So to paraphrase, the problem is thesolution! Well, I’ve never seen a fragmented system scale successfully. PerhapsAndroid will be the exception, but we’ll have to see how this plays out. Froman experience and profitability POV, there’s no debate, simplicity and smoothintegration is winning and scaling. I see a much different picture – Android iseating up its own developer ecosystem while iOS is blossoming.

  57. Vitomir Jevremovic

    Use multiplatform tools like Unity. Multires UI and UX are a much bigger problem then the device or OS behind it. That is fragmentation too. A mess too.

  58. Scott Barnett

    I’m late to the game on replying to this post, but I feel pretty strongly about this and am very interested in this group’s answer…. I’ve had an Acer Android tablet for well over a year, and I bought a Kindle Fire for my wife 3 months ago. I’ve noticed a few things about the Fire:(1) It’s uniquely and nicely integrated with all things Amazon. Reading books/watching movies from the Amazon store is a seamless and simple experience(2) It’s not integrated with the Google Play store in any way. I can’t get my Google Apps (Gmail/Calendar/Contacts/etc) on the Fire, although I’ve read there’s ways to “backdoor” them, I haven’t tried that yet(3) I can’t access certain aspects of my Amazon Account (Amazon Prime Free Videos, Amazon Lending Library) on my Acer tablet – they are *only* available on the Kindle device itself.I find this highly annoying and evidence of the “bad” surrounding fragmentation. I don’t want to have to spend time thinking about what features Amazon offers *only* exist on *their* Android devices. That is counterproductive. I love the Kindle integration with Amazon, and it would be my only Android device if I could get all the other Android apps I run on my phone and tablet – but I can’t. That seems shortsighted and frankly ridiculous in this “open” age. Am I missing something?

  59. amh15

    I somewhat disagree with your first sentence. The Android platform itself is getting less fragmented on the whole. I calculate this objectively here: http://fragmentationwatch.i… . I believe you would be correct in saying that screen resolutions have got more diverse, but even the paragon of unity that is iOS is about to have five different screen resolutions if my mental math is correct, perhaps more if we get an iPad mini. It would be interesting to see how many popular Android screen resolutions there are. Google’s own data is not very helpful on this subject. I would guess that today, 800×480 and 1280×720 would be the most popular, with a handful of less popular lower resolutions in there from the budget end of the market. You might well find that 95% of the installed base is one of five resolutions.

  60. John Revay

    W/ Samsung/Apple ruling – wondering if they hedged their bets.I know these platforms take time to develop…and I don’t know if ruling extends to non driod devices…..just wondering if now we will have a three horse race.Samsung Announces Windows 8 Phone, PC and Tablet”Samsung Electronics Co., Ltd., announced a new line of Windows 8-based devices under the ATIV brand Wednesday, including the first publicly-revealed Windows Phone 8 smartphone”

  61. Steve

    This seems accurate on the face of it. But thanks to hardware fragmentation (different devices made up of parts with different capabilities) and software fragmentation (different versions capable of supporting different features), Android is a really crappy OS to build a media publishing business on.In particular, most Androids suck at video. Android 2.x and 3.x devices can’t support adaptive bitrate streaming – basically, network performance detection to enable efficient content delivery – that is otherwise standard best practice to deliver video to phones and tablets on the move. If I have to sit still to watch video on WiFi or over a hardline, I’ll watch it on my laptop or my holy-bejeezus-that’s-really-big plasma flatscreen THANK YOU VERY MUCH. The only version of Android that really can support mobile video playback reliably is 4.x – and that eliminates Kindle Fire (2.x) entirely. Kindle Fire HD (Android 4.0 customized and crippled) and Nexus 7 (Android 4.1, AKA Ice Cream Sandwich) offer a Herculean leap ahead of everything else because they do support adaptive bitrate delivery. What’s even better is that Ice Cream Sandwich lets you use the same kind of adaptive bitrate delivery required on iOS mobile devices. There is bad news. What sucks about Ice Cream Sandwiches is each device generally requires it’s own copy/version/rendition- whatever you want to call the file that actually needs to get to me in order to play on my Android device’s highly, highly variable screen size (aka, form factor). What also sucks is that Android 4.1 currently accounts for something like 2% of the installed base for Android devices. Which in conclusion, makes Android as a whole a truly crappy 2nd-best option as a platform for a mobile video business with significant market ambitions.

  62. Dave McLaughlin

    But it’s often hard to predict how those frameworks will be limiting vis-a-vis one’s specific application, especially in a build-test-iterate approach where you don’t fully know what you’ll need the app to do 6 months from now or whatever. I do agree w/ Fred’s premise that you need to be on both (especially if you’re building a communication utility). But I’m increasingly believing in building native apps, with iOS preceding Android since it allows us to design and build a more optimal experience in the “cleaner” OS, working through the choices and prioritizations in terms of our UX, and simplifying that difficult process by removing the dimension of device fragmentation for the moment. “Creativity loves constraints” is a quote that comes to mind here. Then trying to replicate that UX on Android. I find the fragmentation paralyzing if trying to think in the opposite order.Curious what people think of this logic?

  63. Elia Freedman

    I have come to the conclusion that relying on a third-party intermediate layer is a huge mistake. You are making a bet that their stuff will stay compatible when often times those frameworks are abandoned or can’t keep up wit the latest innovations of the platform.

  64. ShanaC

    That’s ideal, not reality. Better question is how to get to reality.

  65. Wavelengths

    I believe fragmentation creates an opportunity for a different way of doing things. Chaos = Opportunity.

  66. Todd Werelius

    In general I Agree.I tried this out of the gate and quickly determined that they were not ever going to be up to date quickly enough or provide a rich enough feature set for what we do.I think for some types of low integration app’s though it’s OK, things that are closer to web’ish type implementations that don’t couple strongly to the platform.

  67. Wavelengths

    Too many moving parts.

  68. ShanaC

    reality is something you create

  69. Wavelengths

    The reality in most of this conversation is Android vs. Apple. A good topic. But the chaos, as described by this conversation, may present an opportunity for the right people to exploit/explore/monetize, etc.(P.S. Send me some smoked fish.)

  70. Wavelengths

    Yes, and why get stuck thinking that these platforms will express what we need in even two years?I like Fred’s logic, but I’m suggesting we keep opening the box.

  71. ceonyc

    To what?

  72. JohnDoey

    I think the most important thing is that whatever users see on the Web, they won’t pay for that. Your app has to be much better than a Web app. If you build a Web app so you can PhoneGap it onto 5 platforms, you will be outsold by 1 native iOS app that thrilled Web-weary users.Xcode has CoreAnimation, CoreAudio, SceneKit, Storyboard, and other features that enable the developer to go far beyond the very limited things the Web can do. The original Web browser and server were made with Xcode. Putting those tools down for yet another kludgey Web app is a very serious limitation on what your app will ultimately be able to do.Having said that, though, using PhoneGap for v1 and ramping up Xcode for v2 was very good for a couple of apps I worked on. It enabled us to get audio and graphics in front of the user right away and then we essentially redesigned it to be native. Made it much smaller, faster, cleaner, better-looking, easier to use, more worth paying for.

  73. JohnDoey

    How much innovating are you going to do with Web or Flash tools? Not much, and the user has likely already seen that for free in their browser. If you want to innovate in mobile apps, you are running Xcode and working native.

  74. ShanaC

    we could just go out to brooklyn and eat the smoked fish together. I’m looking for a herring buddy!

  75. ShanaC

    Philosophically, I wonder about that.

  76. Wavelengths

    Loves me some herring! (Not many options out here in West Texas!)