One Year Later, Google’s Vision of Android Apps on Chrome Has Collapsed

Posted on May 5, 2017 by Paul Thurrott in Android, Mobile with 122 Comments

One Year Later, Google's Vision of Android Apps on Chrome Has Collapsed

One year ago, Google announced that it was bringing Android apps to Chromebooks in a move aimed directly at Windows. But that vision has never turned into reality. And this week, we found out that it is delaying this hybrid solution yet again.

What the hell is going on here?

Granted, this is a slow-motion train wreck: Google originally promised that Android apps would be broadly available on Chromebooks by the end of 2016. And media reports throughout last year were perhaps overly positive about Google’s expected impact on Windows PCs and Macs. This was supposed to be game-changing.

To be clear, if Google can pull this off, Android apps on Chrome will indeed be disruptive, as I opened pondered a year ago in Can Google and Apple Pull the Plug on the PC Market? A June 2016 video described the wonders of this solution.

And then things got silent. As I wrote in January, in Still Waiting for the Chromebook Revolution that Never Came (Premium), Google’s late 2016 promise was smoke, and only a very slim selection of devices ever got Android app support, and then only in pre-release form.

The back-to-back releases of the Samsung Chromebook Plus and Chromebook Pro in early 2017 were supposed to turn things around. These Surface-like hybrid PCs offer touch screens and pens, and can be used like a traditional laptop or like a tablet.

There’s just one problem: Samsung has only released the Chromebook Plus so far, and this expensive device hasn’t moved the needle at all. The Pro, which features an ARM processor instead of an Intel processor, was delayed from March to April. And then to May.

And now we’re told that this new version of the device will ship sometime “this spring.” And according to many reports, the reason for the delay is, yep, you guessed it, that Google actually cannot figure out how to combine Android apps and Chrome OS. This attack that Google announced a year ago is effectively vaporware.

I still believe that this kind of solution can be hugely damaging to Microsoft … but that is predicated on Google actually making it work. And so far, they have not.

So breathe easy, Microsoft fans. Once again, the company has been saved. Not by its own technical excellence, but because of the feeble ineptitude of its competitors. Classic.

 

Tagged with ,

Join the discussion!

BECOME A THURROTT MEMBER:

Don't have a login but want to join the conversation? Become a Thurrott Premium or Basic User to participate

Register
Comments (123)

123 responses to “One Year Later, Google’s Vision of Android Apps on Chrome Has Collapsed”

  1. Darmok N Jalad

    This is standard operating procedure for Google. They start many projects and either never finish them, or they get them to a good place and then replace them with something else. Take their multiple attempts at a messaging service, the death/decline of things like Reader and G+, or like the latest move of Google Earth to Chrome. It feels like their software and services department is run like Hogwarts.

  2. Jorge Garcia

    Fact of the matter is, if Susie Q. College can open up her laptop at Starbucks, start typing out her essay, make her powerpoint presentation, research her subject matter on the web ANNND still have an instance of Facebook open, an instance of Snapchat open, an instance of Instagram open, an instance of WhatsApp open, and even one of many SMS-forwarding services as well....ALL open in the "desktop" background (in separate phone-shaped windows, probably) then that that would appeal greatly to her. That's how Millenials roll in 2017. Currently Windows can't replicate that scenario at all, only Android and iOS can, but neither has the right desktop implementation YET. ChromeOS won't cut it yet, maybe never will. Samsung, among others is working very HARD and intelligently on this problem.

    • skane2600

      In reply to Jorge Garcia:

      Apparently Millennials "roll" by using a UI that's just in your imagination. One could argue that a student who has Facebook, Snapchat, Instagram, and WhatsApp open is going to be too distracted to complete that essay.

      • Jorge Garcia

        In reply to skane2600: The only reason they don't currently is because there is no acceptable desktop solution that checks off every box. They now just keep the phone nearby and switch clumsily back and forth, but I know that if their social media apps could run adequately in the background, it would be the elegant solution they want.


        • siko

          In reply to Jorge Garcia:

          Have you tried windows 10 and some of the apps you describe? even a low end pc would run facebook, messenger, telegram, several web pages/browser, powerpoint (from store) and heck word and sway all at the same time no sweat in any way you'd like to arrange your windows.... should give it a go!

          • Jorge Garcia

            In reply to siko: Yes I have two windows 10 desktops, two Windows 10 laptops, and my "main" rig which is windows 8.1 + Stardock Start8, and will be until MS pries it from my cold hands. I also have tons of Android machines and my girlfriend ONLY uses Apple...because, you know, Apple. So yes, I've tried it all and literally nobody except businesses and nerds like us give a crap about Windows anymore. The people using it in 2017 are doing so because they HAVE to.


    • SvenJ

      In reply to Jorge Garcia: "Currently Windows can't replicate that scenario at all, only Android and iOS can"
      How so? iOS is still essentially limited to one app on the screen at one time. That isn't a desktop with instances in little windows. Android on a tablet doesn't do this either. There generally are options for split screen, but two side by side, not windows. Samsung has shown this DEX thing, but that is on an attached screen, not on the device, phone/tablet, itself.
      Windows has the greatest potential for making this happen. I can already resize UWP apps into windows on the desktop that look remarkably like their phone counterparts. The shortage is the apps themselves, not the ability to display them as you suggested. Android of course has the apps, but at this point displaying them on a Chromebook seems elusive.


      • MikeGalos

        In reply to SvenJ:

        "I can already resize UWP apps into windows on the desktop that look remarkably like their phone counterparts."

        They look remarkably like their phone counterparts because they ARE their phone counterparts. Same app. Same code. Just that the phone doesn't have a big enough screen to support the larger screen size UI in the app so it uses the small window UI just like it does when sized down on a desktop/laptop/tablet.

      • Jorge Garcia

        In reply to SvenJ: I did state that iOS has the potential, in the sense that all the ingredients (apps) are available TODAY, but that the implementation isn't there yet. There is no doubt in my mind that Apple is working on Windowed iOS, and it will be a revelation to Millennials. Android will need to respond to it, as not everyone will want to/be able to buy into the Apple world. Windows has zero chance of success in the consumer/young people world moving forward. The Apps will simply never materialize,and when they do, they will be inferior and/or too late.


        • skane2600

          In reply to Jorge Garcia:

          When did Millennials vote you in as their spokesman?

          • Jorge Garcia

            In reply to skane2600: I'm just trying to push a point based on what I see every day. I have several young siblings, many nephews, many friends, many peers and I still spend a lot of time at USC, which is my Alma Mater from 18 years ago. It is clear to me that everything they do that is computer related is somewhat of a work-around, and the reason for that is that NO ONE as of yet makes a laptop INTERFACE capable of addressing their needs, or that they would enjoy using. (You need both to move forward). I think HP and others are missing out on a very important demographic that will spend cash freely if you can sell them on a convenient experience.


  3. Mark from CO

    Paul:

    We'll have to reconsider the perceived strategic and execution alacrity we thought Google possessed.  Android Apps on Chromebooks are indeed a strategic threat to the life of Microsoft's PC business.  Surely Google has made the investment in this effort, the prize too large to ignore. 

    Maybe... Perhaps... Against all odds, Microsoft may really have an opportunity to fix itself in this market.  But they better not be complacent, but continue to move aggressive to meet needs and to innovate.

    Mark from CO

  4. Ted O'Hayer

    Fuschia happened.


    Its an OS running their own kernel (avoiding Linux's unstable driver API/ABI) and is designed to run progressive web apps written with the flutter toolkit.


    This solves the Android update cycle problem but ensures new applications are easily written, since Flutter apps will also work on Android and iOS. And since they're progressive apps, they'll scale to a notebook form factor.

    • Jorge Garcia

      In reply to Ted O'Hayer: Hopefully they can make this work...but a lot of the World's developer's have already put a lot of effort into making and polishing Android Apps, so there will have to be some kind of "bridge" that is totally transparent to the end-user...otherwise, it's DOA like Windows Mobile and their chicken-and-egg problem.


  5. Lateef Alabi-Oki

    How long did it take Android to become the most dominant computing platform?


    How long did it take Google Chrome to become the most dominant web browser?


    How long did it take Gmail to become the most dominant email service?


    How long did it take Google Maps to become the most dominant mapping service?


    How long did it take YouTube to become the most dominant online media platform?


    How long did it take Chromebooks to become the most dominant devices in the education sector in the US?


    These were all products tech bloggers predicted will "collapse".


    I wonder if 5 years from now I'll look back at this post and chuckle.

  6. BoItmanLives

    Let me get this straight: Paul gushes over MS shipping a crippled Windows 10 SKU that's artificially locked into the worst aspect of it (the Store), but Google taking more time to do something as herculean as combining mobile and desktop is a "slow motion trainwreck". Got it. Maybe MS should have taken more time.

    I suspect this article is going to be hilarious to visit in retrospect. One word: Fuchsia.

    • SvenJ

      In reply to BoItmanLives: "as herculean as combining mobile and desktop"
      You realize MS has already done that right? The reality that Win 10 mobile hasn't caught on does not change the fact that Win 10 is as single an OS, which runs across multiple form factors, as there has ever been. You can write a single app which runs on phones, tablets and desktops. You have to take the different platforms into consideration when you write it, but you can do it once. You only need to resize a (well written) desktop UWP app down to a phone sized window to realize it is the same app.
      Google only needs to ensure that Android apps run in a little window on Chrome OS. Maybe the issue is that Chrome OS is a browser, not really a full OS.
      • skane2600

        In reply to SvenJ:

        Back in the day, you could run C programs that ran on a variety of different OS's but it didn't make those OS's all the same. There was a time when MS Word source code was shared between Windows and the Mac even though the Mac was using a different OS and a Motorola 68000 and Windows an 8088. So common code is nothing new and doesn't create "universality" of applications. Windows 10 may make it a bit easier, but the platform-specific tasks are still there. Resizing doesn't solve the platform-independence problem in the general case. Sometimes you have to change the "objects" in the UI. What usually happens with UWP apps is that they accommodate the mobile environment to the detriment of the desktop (e.g. Wasted space, overly large objects, etc).

        • MikeGalos

          In reply to skane2600:

          And that proves you have no clue about how UWP works and how you develop for it.

          Windows really is one OS from IoT through superservers and from POS systems to Tablets to Phones to laptops to desktop to AR/VR devices to wall displays supporting a vast array of I/O models including pen and touch and keyboard and mouse and data stream. All of these run the same code on the same libraries and those with a UI use the same UI code with hints to readjust when sizing crosses a threshold you specify.

          Comparing that to having a language that's standardized is a massive demonstration of ignorance and being decades behind the state of the art.

          • PeteB

            In reply to MikeGalos:

            Only an MS employee or deluded zealot still thinks UWP is viable, let alone "state of the art". Comedy.

          • skane2600

            In reply to MikeGalos:

            If it's truly one OS, then every single application that runs on one device can run on all others. This is simply not the case. There are different device families that enable features that are unique to that family. If MS wants to define a "single OS" as an OS that has common features on different devices, that's fine, but as a practical matter these differences matter. Even MS's nomenclature argues against the "One Windows" branding. What need is there to call Windows 10 builds that run on a phone Windows 10 "mobile". Why, for example, is the PDF enhancements not included in the mobile version yet if it's "one windows"?


            UWP advocates like to talk about resizing, but as I've said before, that is insufficient in the general case.


            I see in your last sentence that you've resorted to insults. In that case, I'm dropping out of this discussion.

    • MikeGalos

      In reply to BoItmanLives:

      Microsoft has a demonstrated history of actually producing operating systems and locking out a feature is relatively easy.

      Google has never written more than a shell on top of an existing operating system, is an advertising brokerage house and not a system software shop and is attempting, with no track record, a task you call "herculean" with no history of ever succeeding on that level of software development.

      Yeah, anybody thinking Google's plans should be treated as anything more than vaporware issued by their PR department is not being a journalist but a cheering section repeating unqualified press releases. Sorry if that's the bar you want the industry press to follow.

  7. rds2587

    "Not by its own technical excellence, but because of the feeble ineptitude of its competitors."

    Such a strange statement. Firstly, Google is anything but feeble and inept, it is one of, if not the best and most innovative tech company humans have ever produced.

    And secondly the fact that Microsoft has been able to do device agnostic apps is pretty technically excellent. Sure the strategy and communication has been all over the place, but from a technical point of view, the fact that apps with the same codebase can target such a diverse set of devices is no small feat.

  8. Joseph Savage

    Collapsed? It hasn't even gotten started yet. A lot technology we use on a daily basis had 1.0 or 2.0 versions that were terrible.


    I never used Android until after 2.0. I did fire up Windows 3.0 but didn't really use it until Windows 95. It's too early to use such dramatic words as collapsed.

  9. Adzprazolam

    Disappointing to know that for the majority of chromebooks the android app dream never came to fruition....they work perfectly on my chromebook pixel (which I love) but then again, Im fully aware that device shouldn't factor in to any measure of success regarding chromebooks more broadly.

  10. Jules Wombat

    It will be Android Apps on Android OS [laptops/ Desktop] that will [in a few years] be killing off Microsofts ambitions in Education and at low end business, just as Android now dominates Mobile, Auto, IoT etc.

    ChromeOS has passed "Vive La Android"

    • Jorge Garcia

      In reply to Jules Wombat: This is exactly right. Android is a very flawed OS, but just like VHS, which was crap compared to the competing formats, Android has floated to the top to become king due to openness and its cost..."free". People who think that another OS is going to come along any time soon and make developers jump ship and re-write all their rapidly-maturing software clearly did not pay attention to the Windows UWP trainwreck. If mighty MS could not make a "third" ecosystem happen, then who will? BezosOS? MuskOS? ChromeOS was a mistake. Had Google, say in 2012, started making a varaint of Android that ran well on a desktop/laptop, we would have a very credible competitor to Windows right now, as developers would have tweaked a lot of their software to fit the larger screen. Instead, we must wait and watch this happen over the next few years.


  11. burog25c

    I honestly have NO idea why Google didn't just tweak Android and use it for Chrome. Seriously, they would already have every app you can think of. Just skin Android like lots of phone makers do.

  12. Mike Cramer

    I'm reluctant to count out Google, but after 33 years of listening to "Microsoft is circling the drain," I'm highly skeptical.


    More than that, who really wants Google to take over the computer industry? They provide "free" bare-bones software. They make all of their money from advertising, fuelled by tracking users. That will always be their priority. Microsoft makes their money from software and services.

    • SvenJ

      In reply to Mike Cramer: When discussing Google's intent I always like to reference the "Your Content in Our Services" section of their Terms of Service (https://www.google.com/policies/terms/)
      "When you upload, submit, store, send or receive content to or through our Services, you give Google (and those we work with) a worldwide license to use, host, store, reproduce, modify, create derivative works (such as those resulting from translations, adaptations or other changes we make so that your content works better with our Services), communicate, publish, publicly perform, publicly display and distribute such content."
      Yes you still own your work, but you give Google rights to do ANYTHING with it. Think about that in terms of Google Docs and Photos. This beyond advertising.


      • nbplopes

        In reply to SvenJ:


        Yes. That bugs me a lot. That is why I only use Gmail, Search and Google Maps. Even that I'm thinking about moving out.

        I bought my first Android device some months ago. It was relatively cheap in compairison with other phones I had. Before that I had iPhones and Windows Phones. I'm planing to go back to the iPhone ... heck I'm planning to go back to Apple ecosystem all the way again. Let's just say I was not very pleased with using either Google or MS as daily drivers. The bummer is really the price of Apple's devices. Operationally they are the most robust and durable across the board in my experience, but sometimes we just want something cheaper ....

        If Apple decided to split the water in the $1000's, meaning iPads to sub 1000's computing and OS X for over 1000's, along with providing official support to the track pad I would be quite happy and probably would go for several home runs in this decade and the next not just with iPhones ...

    • obarthelemy

      In reply to Mike Cramer:

      I think there's a huge difference between Pro and Consumer. Consumers want whatever is cheaper, and don't value their data. There are some exceptions, but niche; and product quality of course plays a role, I'm assuming all things are equal on that front, which they pretty much are between Android and iOS. Also, consumers want socially valuable and ego-reinforcing (ie, Apple).

    • PeteB

      In reply to Mike Cramer:

      If you think Microsoft isn't tracking users - to the tune of about 2000 data points in Windows 10 - then I'm afraid I have some bad news..

      • BluetoothFairy1

        In reply to PeteB: Pete, you need to read GOOG and MSFT privacy agreements before making this blatant implication. Unlike GOOG, your data will not be sold to third parties with the intent to make profit. MSFT drains your data to make their products better, with GOOG they suck your data because you are the product. See the difference?


      • James Wilson

        In reply to PeteB:

        Those data points are very different. Google takes info about you. Microsoft takes info about the phone. How many apps opened, which apps etc. Google reads your mail, logs your position, who you contact, how often, what you talk about etc. At least Google are open in that they say we collect any and all info.


        Far more invasive.

  13. illuminated

    Google wanted to have their own version of windows phone.

  14. nbplopes

    I agree. It seams that Google failed in bringing Androids Apps to Chrome in a sound, robust and convincing way. I

    I don't think Google in this case thought, "Oh let's attack MS by bringing Android apps to Chrome". I just think that Google was doing its own thing and it did not work as expected. But maybe they thought in these terms, who knows. If they did, I hope they learned their lesson. The lingo used its just to ramp up the drums for the Windows audience :)

  15. Chris_Kez

    Google I/O is coming up. I have a feeling we'll hear more then.

  16. Nicholas Kathrein

    Paul. You got a line of the article wrong.


    "The Pro, which features an ARM processor instead of an Intel processor, was delayed from March to April. And then to May."


    The Pro is an intel process and the plus is ARM. Part of the delay is most likely that the Pro is Intel m3 processor and getting apps to work reliably with Android apps has been the issue. Everything running Arm works pretty well. I'd wager we'll hear a lot about this @ Google IO which takes place this month. In fact that could have been brought up. Saying Google IO is around the corner and if not addressed there then that would be telling.


    My 2 cents

  17. obarthelemy

    I'm really wondering what Google's long-term plans are:

    1- are they adding more Desktop nous to Android ( Ă  la DeX), or just the bare minimum hoping users will switch to ChromeOS in Desktop/Laptop/Docked mode as much as possible, and just use Android/Desktop apps as a temporary band-aid ?

    I'd much prefer a DeX-like solution, something less drastic even . Android is this close to being Just Fine on the desktop: it supports mouse, webcam, external storage, network, DACs... (basically all standard USB and BT peripherals) I don't miss floating Windows that much, what hurts is the absence of mouse zooming (maybe touchpad/touchmouse support) and consistent keyboard shortcuts.

    2- Are they keeping Linux as the underlying OS or are they switching to their pinksomething OS ?

    3- Are they planning on hitting the security and/or updates reset button (like MS did a few years back) anytime soon ?


    I can understand Android on ChromeOS being deprioritized if any of that is going on. But they should tell us.

    • James Wilson

      In reply to obarthelemy:


      Google's long term plans are fairly easy to understand - what can they do to get more data points from their users. If the underlying OS of Android is preventing them getting enough data points on users, they will change. If not, they won't.


      ChromeOS was good in schools as it prepared kids to use Google when they grow up. Now they know kids will get Android phones, they don't need to focus on the school side any more as they can get the data on the kids now and stay inside the law (Parents creating fake Google accounts etc)


      Google is morphing from an advertising company into a data capture / analysis company. Once it has ingested the data, it can be used for advertising (currently, number one revenue) but this data can also be sold to other companies / used in various analysis - both good and bad. To be fair to Google, it is morally neutral - but it's clients - that's another matter.


      So - Google's long term strategy is 'What can we do to get more data from people'. Once you have the raw data, captured and stored in Google's giant data centre - you can worry about how you analyse it later.


      Capture -> Analyse -> Report

      • obarthelemy

        In reply to James Wilson:

        Nice rant; but irrelevant for 2 reasons:

        1- MS and Apple are doing the same, as are FB and Amazon. And now US telcos. And everybody else.

        2- Google et al. living on our data is not the point. To get more data, Google has to get more users and more usage. Controlling the OS/Platform/Ecosystem is a key advantage; what's their playbook to achieve even more users and usage after getting 85% of phones and 66% of tablets ?

  18. MikeGalos

    It's pretty easy to create a press release announcing vaporware that will be "a threat to Microsoft". Any Silicon Valley PR firm (or ad agency) can do that and many, many have. What's actually hard is creating an actual product that even works.

    Look at the OS market. Right now Windows is the only OS that's been commercially successful that was created in the last quarter century. All the rest are shells and tweaks on top of an old minicomputer OS designed almost half a century ago.

    Apple tried to create a new OS multiple times and failed.

    IBM tried to create a new OS multiple times and failed.

    IBM and Apple tried to get together and jointly create a new OS and failed.

    Google didn't even bother to try.

    Press releases are easy. System Software is hard. That's worth remembering the next time somebody issues a press release and it gets lots of breathless prose.

    • obarthelemy

      In reply to MikeGalos:

      Wasn't Windows created on top of MS-DOS which itself was an 8086 CP/M clone ?

      • fishnet37222

        In reply to obarthelemy:

        Windows 1.0 up until Millennium was created on top of MS-DOS. Windows NT was a completely new OS that did not depend on MS-DOS.

        • skane2600

          In reply to fishnet37222:

          Again, it depends on definitions. The fact that NT didn't depend on DOS didn't mean it was "completely" new. If it were, it would have failed due to an inability to run legacy programs.

          • MikeGalos

            In reply to skane2600:

            Nope. It actually had a Windows personality module loaded if Win32 applications were called but equally loaded the POSIX personality for those apps and an OS|2 personality model for OS|2 applications.


          • SvenJ

            In reply to skane2600: Of course it depends on definitions. When you get right down to it there is nothing that is completely new. The ability to run legacy applications doesn't require that the old code is somehow buried in there. Look at the Linux Subsystem for Windows. It's not like they are running Linux code in there, They are producing a 'Black Box' into which you stick Linux directions, gets 'translated', executed, and returns Linux responses. What you put in and what you get out isn't new, but how it happens sure is. Same with legacy applications on NT.


            • skane2600

              In reply to SvenJ:

              It's a matter of judging how much "newness" qualifies for "creation". In any case, Linux is no less "new" with respect to AT&T unix than Windows NT is with respect to earlier versions of Windows. I don't say that because I'm Linux fan, because I'm really not, but the claim that "Windows is the only OS that's been commercially successful that was created in the last quarter century" is simply false. If one includes Windows NT, one must also include, Linux and possibly iOS, and Android as well.

              • MikeGalos

                In reply to skane2600:

                Even if we pretend that Linux is anything but a knock-off clone of Unix, you should realize that iOS is a licensed Unix. (as are all the operating systems Apple sells). Apple tried repeatedly to write their own modern OS and failed even when partnered with IBM and then had to buy NeXT to get an object-based shell to run on top of the OS they licensed.


                System software is hard and, as the discussion points out, Google has no track record of being able to do it.

                • skane2600

                  In reply to MikeGalos:

                  Linux's userland has a lot in common with Unix's just as Windows NT's userland has a lot in common with prior versions of Windows. In both cases the kernels aren't the same. I don't buy this distinction you are trying to make.


                  I can't find any reference to OSX or ios being licensed from AT&T, but I assume you have read about it somewhere and can provide a link.

                • MikeGalos

                  In reply to skane2600:

                  Feel free to look at any history of BSD which is the basis for NeXTSTEP which is the core of iOS and macOS. BSD is licensed from AT&T through a sublicense to University of California.

        • MikeGalos

          In reply to fishnet37222:

          Actually, that's a misconception. Windows 95 didn't load MS-DOS unless for backward compatibility it had to load MS-DOS drivers (device= line in the CONFIG.SYS) or had commands to process in the AUTOEXEC.BAT file that thus required loading an instance of MS-DOS to load COMMAND.COM to process the batch file.


          Since lots of people needed those the Windows system VM had a copy of MS-DOS loaded but it was not needed for the OS itself unlike with Windows 1.x-3.x.

        • obarthelemy

          In reply to fishnet37222:

          So, Windows too succeeded and got established as a shell on top of an old stale OS.

      • MikeGalos

        In reply to obarthelemy:

        Not really. See my reply to Skane2600.

        • obarthelemy

          In reply to MikeGalos:

          You said "Windows is the only OS that's been commercially successful that was created in the last quarter century. All the rest are shells and tweaks on top of an old minicomputer OS designed almost half a century ago.".

          How does "shells and tweaks on top of an old minicomputer OS designed almost half a century ago" not apply to the original, run-over-DOS Windows ? And Windows was created, and successful, first as its DOS-based iteration.

          And since it does apply, or did apply for Windows first couple of decades, why are you unwilling to give as much time to, say, Android, to swap out its internals just like Windows did when it switched to the NT kernel ?


          I'm not grokking your logic. Nor your point, frankly.

          • MikeGalos

            In reply to obarthelemy:

            Windows NT and consumer Windows since XP do not use the same architecture at all in relation to Windows 1.x, 2.x, 3.x or 9x. It's basic design is totally different from microkernel on up and is no more related to them than an Atari 2600 game emulator app running on an Apple iPhone makes the iPhone an Atari.

            The point is very simple. System software is a very, very hard job and of ALL the tech companies around now only Microsoft has actually written that level of software successfully.

            When ANYBODY else makes system software level project claims beyond adding yet another shell on top of *ix they're outside their area of proven expertise and those claims should be taken not with a grain of salt but with a full firkin.

            • obarthelemy

              In reply to MikeGalos:

              Again, how is MS establishing the Windows GUI and APIs by first tacking them on top of DOS any different from Google doing that with the Android runtime on top of Linux ? MS then switched DOS out, do you *know* Google isn't working on doing exactly the same ?

              • Waethorn

                In reply to obarthelemy:

                Linux isn't an operating system - it's a kernel. NT is a kernel. Windows NT is Windows built onto the NT kernel. The difference is that Microsoft won't license the NT kernel for developers to build an OS with their own bootloader, shell or libraries separate from a complete Windows installation, aside from niche usage in embedded markets, of course.

                • obarthelemy

                  In reply to Waethorn:

                  How is that relevant to the discussion ?

                • skane2600

                  In reply to Waethorn:

                  The funny thing is if you point out a vulnerability in a Linux distro, some fans will say "No, it doesn't count because Linux is a just a kernel" but if you were to say "Linux doesn't have a word processor" than suddenly Linux is not just a kernel to them.


                  The distinction between a kernel and the rest of an OS is useful from an architectural point of few, but as a practical matter, it's the totality of an OS that matters.

    • skane2600

      In reply to MikeGalos:

      "Right now Windows is the only OS that's been commercially successful that was created in the last quarter century. "


      It depends on what criteria you use. Windows was created over 30 years ago. Yes, there have been updates and architectural changes, but Windows 10's relationship to original Windows is at least as strong as Linux's or OSX's relationship to Unix.


      But I agree with you that System Software is hard which is why I'm not taking the claims of Windows on ARM at face value. When it's finally released, we'll see.

      • MikeGalos

        In reply to skane2600:

        Windows operating systems since Windows XP on the consumer side and Windows NT on the commercial side is based on the NT architecture that Dave Cutler and his team wrote in the late 1980s. It's core is OS independent and has a Windows personality that is the "public" face of Windows. It's no more based on 16-bit Windows than it is on it's other original personalities of OS|2 or POSIX.

        • skane2600

          In reply to MikeGalos:

          Yes, I know the history quite well, but your position is that the kernel makes the OS which is an opinion that isn't universally shared. Besides, Linux has been very successful outside of the desktop and its kernel is completely independent of any kernel running in an "old minicomputer OS designed almost half a century ago." So I think you're bending your definitions to suit your point.

          • MikeGalos

            In reply to skane2600:

            As for the hoary old "Linux is not Unix", sure. technically they're not identical in the way that an IBM PC clone in 1982 wasn't an IBM PC. Or as the song lyric from an episode of Futurama parodying The Wizard of Oz put it, "We resemble but are legally distinct from The Lollipop Guild..."

            • skane2600

              In reply to MikeGalos:

              Come on. Even the worst of the early PC clones had more in common with an IBM PC than Linux does with Unix. MSDOS and PC DOS were nearly identical. It was mainly the BIOS that was different which wasn't technically part of the OS.

          • MikeGalos

            In reply to skane2600:

            You could argue that the basic architectural design and not just the design of the kernel is the OS but that wouldn't change anything I said.

            Now if you're arguing that a change in shell is the OS, you'd be hard pressed to find any takers for that definition.

            In any case, I'd suggest reading the Helen Custer authored first version of Inside Windows NT to get the basic concepts and architecture of modern versions of Windows.



  19. skane2600

    Windows on ARM might turn out to be similar. Certainly it should have been far easier to port the Android environment from one ARM device to another, then it would be to port a much larger OS functionality from Intel to ARM. This is true regardless of what form the "porting" takes.


    Having said all that, I still think that running Android apps on Chromebooks might be considered a disadvantage to the market segment that has embraced it - education. The fundamental philosophy of a Chromebook is that the web is all you need. Adding Android apps undermines that philosophy.

    • MikeGalos

      In reply to skane2600:

      Actually, Windows is relatively easy to port since it was designed with portability in mind. Remember that when Dave Cutler's team wrote what became modern Windows they intentionally wrote it for obscure RISC chips first and then ported to Intel so that they wouldn't inadvertently use any Intel specific ideas. It's already, over the years, been ported to RISC, CISC and VLIW architecture chips from half a dozen different vendors.

      • skane2600

        In reply to MikeGalos:

        Of course modern Windows is a lot bigger and more complex than the original NT and even the original version on Alpha wasn't 100% compatible. Besides, only a tiny, tiny percentage of people ever ran those other versions so the compatibility across thousands of applications was never tested. Presumably the number of users of Windows on ARM will be orders of magnitude greater.

        • MikeGalos

          In reply to skane2600:

          Something has to actually ship before it's anything but ludicrous to tout its massive installed base.

          As for Windows, it's been developed or released on

          Intel i760 and i860

          Intel x86

          Intel/AMD x64

          DEC Alpha (32-bit commercial, 64-bit unreleased)

          MIPS R3000 and R4000

          PowerPC (IBM and Motorola PReP versions shipped, Apple version worked on unreleased test hardware that Apple decided to kill.)

          ARM

          Intel Itanium

          Fairchild/Intergraph Clipper

          SUN SPARC

          • skane2600

            In reply to MikeGalos:

            Windows on ARM is obviously intended to appeal to a market segment larger than any of the non-intel CPUs. If it doesn't achieve that, it will be considered a failure. So I was giving it the value of the doubt. You're right it could be a flop in which case my "orders of magnitude" comment would obvious be wrong.

  20. dhallman

    This will be a welcome delay for Microsoft's Windows S vision. Now the race is on to see if developers will a) enhance their Android Apps for Chromebooks (assuming Google ever gives them a workable platform) or B) 'bridge' their Windows 'apps' to the store, a system that seems to be ready for prime time, but has less of an engaged userbase.


    Google should have 'bet the farm' on getting their app store into PC form by the 2016 holiday season. It is amazing Microsoft is getting a 2nd chance. Or is that 3rd? ...4th?

  21. VancouverNinja

    It is too late Paul. The window of opportunity has closed and MS is launching Windows 10 S. Google's joy ride hits reality in Education as a result. Consumers will be buying sub $200 Windows 10 S devices that a Chrome book cannot best. MS has worked hard to shore up their core user base on the PC. Whether the PC market shrinks due to changing device formats or not MS has done what needed to be done. I think the two interesting questions are 1. Can Google stay on top with the Mobile Phone business over the next decade and 2. Does Apple have any ground breaking technology or new devices in the near future? The last shoe to drop is in the mobile space - Does AR/VR - Virtual Assistants - Bots = a new form factor that does leave smart phones behind? That's MSs vision but can they pull it off? This is an exciting time to watch all of the coming innovations unfold as well as the competitive strategies behind them.

    • MutualCore

      In reply to VancouverNinja:

      IMHO Google's biggest strategic mistake was not launching physical retail stores years ago. For all the flack Steve Ballmer gets, he started opening them in 2009 and now there are like 200 across the world? Sure they are a money loser at the moment, but they are just one super hit product away from being prime retail space.

  22. dcdevito

    Google simply can't push products out the door. And for the ones they do, they can't support them for very long. This is what happens when you build a culture internally as chaotic as theirs. Sure, you can do a lot of "throw it at the wall and see what sticks" approach, but any company can do that. The fact that the tech media has always seen them as a darling has prevented the public from really going after them. I like their services a lot, who doesn't? And they sure are smart people, they just can't push, support or build products. They're a services company plain and simply. They can't master anything else.


    With that said, the success of Chromebooks will never rely on Android apps anyway. Maybe a longer term plan would, but they haven't shown their hand in this regard. Who knows what's really going on? I/O is in a couple weeks, maybe Andromeda is indeed real and on its way. You never know with Google.

  23. ProgrammerAl

    I've never heard much about there being a lot of in-fighting politics in Google like Microsoft (used to?) have. But I'm wondering if that's the case here. Since you're such a Google lover now Paul (sarcasm I swear), have you heard anything about that possibly being the case? It would explain things if the people in the Chrome OS group wanted to take over certain things that Android has, or vice versa. Or if that new OS (Magenta?) is taking away a lot of internal resources for this.

  24. Demileto

    Anyone who had the slightest understanding of the complexity of building operating systems and application models knew Android apps in Chrome OS was never going to be as trivial as Google claimed to be. What Microsoft did with OneCore and UWP, unifying them across such a wide range of devices with vastly different needs is technologically unprecedent, there was no way Google was ever going to succeed with a half-assed shortcut in a considerably shorter time than MS's years long converging of its many Windows kernel variants into OneCore.

    • Narg

      In reply to Demileto:

      Paul does seem a bit over-dramatic on something that is far more complicated that most understand. Computers in general are still quite new in the whole scheme of humans and the planet. I would not expect anything other than what works today. Promises of even the best companies often miss their mark in one way or another.

  25. mjw149

    Well, they never finished a tablet version of Android, so, yep, this checks out.

  26. Jorge Garcia

    SuperBook by Sentio, Remix OS, Phoenix OS, DeX OS by Samsung, and even that venerable old Motorola Atrix phone-laptop combo (6 years ahead of its time), are/were all trying to accomplish the same thing....DO WHAT GOOGLE JUST BOTCHED. ChromeOS was a nice experiment, but they should have simply done a desktop-friendly fork of Android instead, or alongside. Now, other people all over the world are racing rapidly toward this inevitable future. Microsoft is probably very glad that Google botched their OS Game, as Windows would be even farther down the black hole than it is now. IMO, companies like HP and Dell are very dumb to release ALL of their laptops in 2017 with either Windows or ChromeOS installed. Nobody who grew up after the 486 era WANTS Windows anymore, and nobody really understands what ChromeOS is, and they are usually disappointed when they find out all that it CAN'T do. Android currently does 99% of what 95% of people could want, so HP, Dell Etc. would be wise to be rapidly making their own forked versions of desktop-friendly Android, if they want to sell laptops to young people going forward, that is.

    • skane2600

      In reply to Jorge Garcia:

      You took your percentages out of your.. well, head. I expect that HP, Dell etc would prefer to cater to the much, much larger market for Windows PCs than to go to all the trouble of forking Android for desktop systems with the hope that demand will eventually appear. Android is a decent OS for what it was designed for - smartphones (if you ignore fragmentation), but a desktop OS has different needs and one of them is running industry standard applications.

      • Jorge Garcia

        In reply to skane2600: Ok, I stand corrected. 95% of non-business types. Of course you are right that because of legacy/business software, Windows will "need to" remain relevant for a good long while...but explain to me why there are still tables and tables of beautiful laptops at Best Buy, all running that overkill, super easy to destroy, OS known as Windows? Even when Best Buy is packed to the rafters with shoppers, NO ONE ever approaches these tables. The reason is Windows itself, and that's coming from a 25+ year DOS/Windows enthusiast who reads Thurrott.
        • skane2600

          In reply to Jorge Garcia:

          So do the workers at Best Buy ever approach you and ask why you show up every day and hang out during all the business hours? Or do you fly around the country observing all the Best Buys? Because if you don't, how would you know that people aren't buying Windows PCs there? Best Buy is a business. If nobody is buying PCs they wouldn't still be in the the store.

        • Simard57

          In reply to Jorge Garcia:

          why do you think an Android desktop is more appealing. Granted my toe dip was when the Transformer Prime came out but it fell way short to replace my laptop.

  27. Tony Barrett

    Maybe Google jumped the gun on the whole Android apps on Chromebooks thing, but there are Chromebooks out there now that will run Android apps, and Google recently published a much bigger list of devices that will support it. Hey, unlike MS who release half-baked software to the world and expect everyone to be their beta testers, maybe Google want to get it right. Who knows, but I think this article is very negative, especially in a week when Thurrott has been gushing over this not-as-good-as-it-sounds Surface Laptop that isn't revolutionary, or ground-breaking or a technology breakthrough. It's just a laptop with a Surface badge, and an expensive one at that for that you're getting.

  28. Abbaabba2000

    To be honest, nobody is interested in toy apps on a mouse/keyboard pc configuration. Even if it is UWP, Android or iOS.


    In this case Apple is right with their device specific approach

  29. Watney

    I own a Chromebook Pixel and have messed with Android Apps for several months. The Pixel is one of the few devices that support Android Apps in the Stable Channel.

    I do not understand why everyone is so jacked about Android Apps on a Chromebook. I've tried many apps, and not one app has stuck. Sure it was fun at first, but think with me: Why would you want to use the Android Gmail App, when you can open up a Browser Tab? Facebook?

    I see four possible use cases, no make that five.

    a. Games. Android apps will open up Chromebooks to gaming.

    b. Microsoft Office. Microsoft does not allow Office Apps to be installed on Chromebooks. It's disabled in the store. I've read Microsoft is "working on it." One day you'll be able to use Office Apps on a Chromebook that will function similarly to apps running on an iPad.

    c. Skype. Some folks may want to be able to use Skype, so that could be a win for the short term. Eventually, Skype will run natively in Chrome.

    d. EMR Pen Input. The Samsung Chromebook Plus uses Wacom's EMR technology, so Android Apps will enable PDF annotation and digital hand written notes.

    e. CrossOver - full Windows programs. CodeWeavers has released a beta version of CrossOver in the App Store. The CrossOver emulator enables you to run full Windows programs on a Chromebook. I successfully ran Word 2013, but it was super buggy and not ready for prime time.

    To sum-up, I'm unsure how "disruptive," to use Mr. Thurrott's vocabulary, Android Apps will be on Chromebooks.


  30. scribz

    Interesting. Thanks for posting this. Google doesn't exactly have the best track record for follow through.

  31. mebby

    Of course you can say that Google, Apple, others have been saved (thrive) due to the feeble ineptitude of Microsoft...

    • dhallman

      In reply to mebby: Sure. You can also say that Apple has dropped the ball as of late with less than impressive designs and innovations. As has Google in this article and with the poor quality of prior Android versions + OEM forking. They have all made mistakes.


  32. Waethorn

    I've been using Android apps on my Chromebook Flip ever since they announced support for it last year, so I don't know what you're going on about.


    Try writing a custom kernel and drivers that works with a hypervisor for putting applications in a container on a new pieces of hardware in 6 months time and see how well you do.

  33. Bats

    I said this once and I'll say it 1 million times, if I have to. Everything Paul Thurrott writes and predicts almost always goes wrong. When i say "almost always" I am talking over 90%.

    GOOGLE CAN TAKE ALL THE TIME THEY NEED.

    The fact of the matter is this: Android Apps can be used on ChromeOS. Paul knows that. He already reported it. People can get Android Apps now, if they want. If Paul care to do any research (like reading forums, google blogs, etc..), it's not how Google envision Android Apps to be.

    All in all, it doesn't change the fact that PC and Surface sales have dropped. All this "...BUT GOOGLE PROMISED!!!" is all in Paul's head, which I believe is finding an excuse to write something bad about the company that he hates.

    You know what's ironic? Remember years ago, when Paul was launching his personal Scroogled campaign? It was during the time of that commercial featuring the Pawn Bros or Pawn Shop (I forgot the name of the show). Isn't it ironic that with Paul using Calendar, Gmail, Chrome, Search, and now Android that he (himself) is purposely and willingly allowing himself to be Scroogled? LOL.

    WIth this said, you can't trust anything Paul writes about. The definition of "Collapse" is something or someone that fallen down. Android on Chrome clearly hasn't fallen.

    I think my percentage just increase to 91%. (LOL)

    When it does happen, that number moves to 92%


    • rth314

      In reply to Bats:

      So you're saying that Paul likes Google enough to use Google Calendar, Gmail, Chrome, Google Search, and Android, but not enough to write a fair article? That doesn't make much sense.

    • Daekar

      In reply to Bats:

      Wow, that's some serious bitterness there.


      Based on my reading, the Android experience on ChromeOS is not up to par. The native stuff might work fine, but the Android apps don't run as well, and the UI/display problems are manifold.

      Yeah, they can take as long as they want, but the longer they take the more of a chance Microsoft has to transition from Win32 to UWP or web apps. And lest we forget, Chromebooks aren't exactly flying off the shelves anyway - I have seen exactly one (1) in the wild, and I have been paying attention. Say what you want about schools using them, when they get to the workplace they're going to have to learn to use PCs.

  34. UbelhorJ


    I disagree. I think Google is really close.


    I bought one of the Samsung Chromebook Plus laptops pictured above. It still has flaws, but I love it. Google is just a few tweaks away from making this thing truly wonderful. The Android app store really fills in the gaps where ChromeOS fails. I wouldn't recommend it to normal users just yet, but Google is really close. The ChromeOS and Android apps need to be more aware of each other, and Android needs access to the SD card (so stupid this doesn't work), but it's 90% there.


    The biggest problem I have is that Microsoft is doing to ChromeOS what Google did to Windows Phone. They have all of the Office apps marked incompatible so you can't install them from the store. They even removed OneNote after they announced Windows 10 S. I had to remove it and re-install it from APKMirror to make it stop killing itself.


    This little laptop can do just fine for what most people use their computer for. Windows 10 S and their treatment of the Android app store shows that Microsoft clearly see ChromeOS with Android as the threat that it is.

Leave a Reply