Why Google Chrome is not a “Windows Killer”

Here we go again. The chimera that can sink a thousand ships, let alone web browsers masquerading as operating systems, is at our port one more time. The internets are abuzz with the introduction of Google Chrome and many already announced the (re)start of the OS wars:

Folks this is the operating system war in full action. One between Microsoft and Google. Google is coming out with their own browser called Chrome.


TechCrunch promptly declared it the “Windows Killer“:

Make no mistake. The cute comic book and the touchy-feely talk about user experience is little more than a coat of paint on top of a monumental hatred of Microsoft…Chrome is nothing less than a full on desktop operating system that will compete head on with Windows.

There’s no point revisiting the colossal misunderstanding that confused the web browser with an OS when none other than Netscape co-founder Marc Andressen said half a decade ago, “Windows is just a bag of drivers.”

A modern operating system geared for general purpose computing must provide out-of-the-box hosting of multiple apps, managing of physical and virtual resources, disk access and file system management, 2D/3D graphics rendering, computer hardware interface, device drivers, printing, networking, multi-format media playback, DRM, and a thousand other, often inter-related, chores. Can Google Chrome playback Windows media files? Can it hook up to 1,000 different scanners, webcams, joysticks and other input devices? Can it even print?

Now, Mr. Arrington like others might have in mind limited (and perhaps embedded) OSes that can only provide a very small subset of what we commonly expect from general-purpose operating systems like Windows or Mac OS X. Unfortunately we have seen that movie some years ago. Remember the web-surfing appliances of yesteryear like Sony eVilla or 3Com Audrey and Kerbango? How about Compaq iPaq or Netpliance I-opener?


What I find most ironic is the fact that what is supposed to be the “Windows Killer” only runs on Windows, thereby giving Windows users yet another reason to stick with that OS and perpetuate the common assumption that good things happen first and usually only on Windows. What a way to start dismantling the Windows machine!

33 thoughts on “Why Google Chrome is not a “Windows Killer”

  1. There are huge technical and non-technical barriers to replacing local running applications with the cloud.

    For one, how do you want to expose the majority of the hardware layer to 3rd party websites while maintaining security?

    For two, how do you expect to allow people to work disconnected and how do you deal with privacy/intellectual property rights issues?

    Why should I get a Linux based operating system with generalized drivers to run on my machine only using a fraction of its capabilities when I can run Windows and actually use the hardware I have?

    I like Linux, don’t get me wrong – as a secondary machine or secondary OS to use when I’m not doing things like gaming, working with hardware, video editing, etc.

    Unless Google pulls out a few miracles, “Chrome OS” will get lumped into that same category of toys that the i-Phone is in. Except, at least the i-Phone makes money off of sales and isn’t delusional enough to think it is seriously threatening Microsoft Windows.

  2. Pingback: Google Chrome: The Other Half of the Review « Idiosyncratic Protectional Field²

  3. Chrome is not an OS and it can’t kill Windows, but that’s not Google’s aim – it wants to make the OS irrelevant. If your apps are all online and you spend all your time in the browser, then any old OS will do (as long as it supports your browser of choice).

    Microsoft has no incentive to make the web as good or better a place to run apps than Windows. MS wants users running Windows, and that happens if users need it to run their apps.

    I see two possible outcomes: either Chrome will take over the browser market (not likely – MS is a veteran of browser wars) or they will compel MS to improve IE enough to keep up, which will have the effect of making the web a better place to run apps. Google wins either way. The difference between the Netscape battle and this one is that Netscape could only win if they won the browser war. Google can lose the browser battle and still win the war, by forcing MS to help them abandon their local Windows apps.

  4. Kontra –

    I’m getting the sense that there’s some confusion about where the OS ends and the Windows GUI begins. Today’s netbooks are shipping with Linux and many users are replacing that with Ubuntu.

    I’m struck by history — many of the arguments herein were the same arguments that a Windows GUI would never replace DOS. Why would someone use something so useless, when the DOS prompt was faster, more predictable, and more elegant?

    It took Microsoft three versions before Windows 3.0 came out — and they made a lot of mis-steps along the way, such as using Cntl-P for paste instead of Ctrl-V.

    I’d argue that all of you power users are not the audience. Surprisingly, there are a number of computer users out there that can’t afford to buy the Windows OS; whose government is subsidizing Internet access as a way to improve their Third World status, and who don’t need multi-channel audio nearly as much as they need a cheap word processor and email.

  5. Pablo, the computer vs. user “view” distinction is a contorted one.

    Think of it this way. On Mac OS X, there are literally dozens of desktop apps that are directly integrated with the Internet. How do they do it? They all share WebKit as a browser. Where do they get that capability for free? The OS. How about accelerated graphics rendering within the web browser? The OS. How about multi-channel audio? The OS. How about zoomable text rendering? You get the idea. If you think browser or Internet app vendors will create their own versions of these myriad resources absolutely necessary to experience the web, that’s madness.

    The browser is just an app layer on top of the OS. That’s all. You may increasingly access traditional OS functionalities through the browser, but that doesn’t obviate the OS at all.

  6. I agree with lolec. I think you guys like Kontra are comparing a browser with an OS in terms of functionality, by the “computer view”. Google are thinking in to replace the OS by a browser in an “user view”.

    And I think Google guys are right. You can replace the OS by the browser in the “user point of view”. Of course, the computer still needs the OS, but not the user.

    I can say that looking to my own “work process”. Two years ago, my main tools used to be: a notebook with MS Office, including Word, Excel, PowerPoint and Outlook.

    Nowaday, replacing those softwares by GMail, Google Calendar and Google Docs I can use any computer to work. Doesn’t matter the OS. I can use a laptop with Windows XP, a desktop with Vista, or a Mac or even a Linux with much less impact than just two years ago.

    I’m using a notebook because I travel a lot, and I use the travel time in airplanes, taxi cabs etc to write a text or to reply an e-mail. But even to reply e-mails I prefer to use my iPhone than using my notebook.

    So, if we think in more 4 or 5 years (think in Android, modern cellular phones like iPhone being popular, high speed internet connection almost everywhere (using cell towers or WiMax etc), I need to totally disagree with you, Kontra.

    I think a good, fast e stable browser can replace an OS, not by the computer point of view, but by the user point of view. And if I could choose one monopoly, computers or user, I choose users, no doubt. As Google.

    Best regards

  7. Good article. Finally someone not as arrogant as Arrington. My point why Chrome (or any web browser for that matter) won’t ever kill windows is because the world is not like America. Not everyone all over the world is connected to the internet. Not everyone has a fast internet connection. Here in Asia alone, only 2 or 3 countries have blazing fast internet connections (japan, korea, china?). Windows will still exist no matter what Arrington and other web as os supporter say. And if they think that it will only kill windows, think again. Windows has a huge share in the OS market, what’s stopping Chrome from killing Linux and OSX, which are minority in terms of OS usage? These guys are just fanboys and losers like Michael Arrington.

  8. Kontra: That isn’t exactly new territory, there already are applications on Mac OS X that do that, such as Mailplane (for GMail) and Fluid (for Facebook, Campfire, Pandora, etc).

    Indeed, it isn’t a new technical territory. But Google want to control that territory, I think. That excludes the idea of just joining some existing project.

  9. David: “I suspect that the answer is that Google wants to control the out-of-browser UI.”

    That isn’t exactly new territory, there already are applications on Mac OS X that do that, such as Mailplane (for GMail) and Fluid (for Facebook, Campfire, Pandora, etc).

  10. Kontra: Google still needs to demonstrate why it was necessary to release their own browser rather than help Mozilla, Apple or anybody else where Google thought they needed help.

    I suspect that the answer is that Google wants to control the out-of-browser UI. Specifically, they want to be able to make their web apps feel more like desktop apps. That’s slightly different from wanting to take over (or replace) an OS. That’s why Chrome has such a bare-bones out-of-browser UI: It’s supposed to just mimic the “window frame”.

    I’m pretty sure Apple would not want to relinquish any sort of UI control from Safari, and Mozilla/Firefox may not either (despite getting quasi-all their funding through Google).

  11. First: if you like innovation, you’ve got to love the chaos that surrounds it.

    Why doesn’t George Stephanson (George WHO?) have a monopoly on railway stock? Why didn’t those Victorian Englishmen help Stephenson improve the locomotive but start their own enterprises to work out their own silly, crazy and crappy (and a few sound) ideas around the concept railway?
    Well, nobody knows… The only thing is this appears the environment where real innovation thrives, an environment where a lot of freedom exists and a lot of ideas can be tried out.
    Today’s webbrowsers seam to be in a comparable phase as Victorian railways. Time isn’t ready to come with a unified solution. I was amazed by Googles move, I am amazed even more by how much room for improvement there still appears to be. We get more improvement, more innovation, at the expense of users having to choose between Firefox, Opera, Safari and Internet Explorer. At the expense of users having to spend time and effort in determining what’s really goor and what’s not.

    Second: my impression is the current innovation is an addition not a replacement to existing systems.

    The core OS running on most systems will be some flavour of open source Unix (called Linux, Darwin, Android or anything else). The only company that is trying to keep up and running a proprietary core OS is experiencing hard days because it simply isn’t worth the money. Core OS routines are well understood, extensively tried out and freely available. The money lies in services with added value. So what companies that really matter (like Apple and Google) do is try and create the preconditions to deliver those services to their clients. Google wants to deliver (sponsored) search results, Apple wants to deliver iTunes and iLife content. Apparently it is worthwhile to do some open source development for this goal. Amazing yet fascinating.

  12. Neil: “Why do they ‘need’ to demonstrate this?”

    Because when you’re battling a 80% browser-share gorilla, you need all the help you can get from your potential partners, like Apple, Mozilla, Opera, Adobe, etc. You seem to have a rather simplistic view of what it takes to build a platform where others can play as well.

    Making five-year predictions is wholly problematic, I don’t know anyone who knows what will happen in 15 years, that’s an eternity in technology. So I couldn’t tell you what might happen by then. I’m fairly certain though that OSes aren’t disappearing anytime soon.

    As I said above, OSes may indeed get slimmer and trimmer. Mac OS X, after all, fits into an iPhone. But that’s different from narrow-purpose appliances, which have been soundly rejected. Ironically, the iPhone, the quintessential mobile net device, gets its mojo from having a robust version of a desktop OS.

  13. “Google still needs to demonstrate why it was necessary to release their own browser”

    Why do they “need” to demonstrate this? You “wnat” them to demonstrate it, but they don’t “need” to do diddly-squat. When they can just buy whatever they want, they don’t “have” to demonstrate anything to anyone, other than that they can keep up with the revenue expectations of the investors.

    All my employees use Google Apps. I didn’t need to spend a brass razoo on software, including an OS, as Linux is free. Three years ago I would have had to spend tens of thousands of dollars, and the vast majority of that would have been on MS software. This does represent a killer app to MS, however, they haven’t gotten as far as they have wothout the smarts to compete, and they will. And that will make things better for us all. Didn’t tabbed browsing in IE get there directly as a result of Mozilla implementing it?

    Don’t kid yourself kontra, most people don’t give a toss about the OS so long as they can get their stuff done – Chrome appears to help them vastly in this endeavour, and it’s free… And yes, we are probably ‘clouding’ the definition of OS with overlying apps, but the blurring doesn’t matter, it’s irrelevant. Google justed sucked millions of dollars of revenue away from MS by releasing Chrome. In fifteen years we’ll probably be cursing the demonic monopolists at Google instead of today’s demonization of MS, but things will move ahead.

  14. Ramon: “the idea of a general purpose computer will become more and more absurd.”

    Anything could happen. But this has also been promoted a generation ago during the Netscape era, but didn’t turn out that way. You may be underestimating what an OS does for users, and they have become accustomed to all the functionality. Reducing that will be a problem. To the extent that you try to duplicate what an OS does in a “browser” you’re merely shifting stuff horizontally. Now surely OSes can become trimmer and leaner, but I just don’t see their disappearance any time soon.

  15. All of which can be provided freely by a thin Linux install that requires vastly less hardware to run than Windows. With a browser like Chrome, more importantly the v8 dynamic language VM behind it which smokes all other JavaScript VM’s and can handle modern multi-core processors, many if not most users will be easily be led away from Windows and into an era of smaller, cheaper, internet aware appliances. If all the apps a user needs are web apps, then a browser like Crome is the only windowing system an appliance needs. The underlying OS need not ever be visible to the user. As computers get smaller and smaller and embedded into more and more devices, the idea of a general purpose computer will become more and more absurd.

  16. BCumbers: “Is VMWare a kernel, operating system or application?”

    Does it have to run on another OS?

    “All of which could be done with net-based apps and the cloud”

    Only if those net-based apps run on an OS. They still need local “hosting of multiple apps, managing of physical and virtual resources, disk access and file system management, 2D/3D graphics rendering, computer hardware interface, device drivers, printing, networking, multi-format media playback, DRM, and a thousand other, often inter-related, chores,” as I mentioned above. Otherwise, back to limited appliances, a category that flopped miserably.

  17. Ziad: “Now let’s see how well they integrate their other products, and what that does to these markets.”

    Google still needs to demonstrate why it was necessary to release their own browser rather than help Mozilla, Apple or anybody else where Google thought they needed help. For example, one doesn’t need to release a new browser to speed up JavaScript. Do the new engine and release it under open source. Ditto, with multi-process, etc. Do they need to position Chrome as technology that blends into other browsers eventually, general purpose infrastructure or specific app delivery vehicle? Is it a browser for static sites or an RIA runtime? Does the introduction of Chrome make Apple, for instance, more or less likely to adopt Gears? You can’t expect collaborators to play ball with you if you directly compete with them.

  18. Is VMWare a kernel, operating system or application? I think the point is that the traditional view of an OS for executing applications is evolving and concepts like this over time may not kill off Windows but may kill off its dominance.

    The current desktop OS metaphor is based on the “prehistoric” assumption that the desktop must be used for general purpose computing, file management or the hosting of anything. Tell me if you know any non-techie or sophisticated user of a computer that does anything beyond email, surfing the net, document writing or managing their photos or music?? All of which could be done with net-based apps and the cloud; “smart” hardware can eventually do everything else.

    Evolution does not happen overnight…. :)

  19. Good argument, Kontra. I see Google’s strategy with Chrome as being primarily defensive, with a platform for many offensive thrusts.

    Defensive because IE8’s privacy features are expected to interfere with Google’s ability to deliver targeted ads, and to properly record click-throughs. Although Google controls the default search in Firefox and Safari, losing the revenue from IE in Windows would be devastating. How could Google argue that Microsoft’s interference with Google’s use of IE for ad revenue is monopolistic, when Google just introduced its own browser to extend its search monopoly. In a field where IE was already losing ground to Firefox and Safari, both of whom default to Google Search.

    Platform for offensive thrusts: Chrome will be a walled garden in which Google can properly integrate many of its existing products that are not getting enough traction, such as:

    1. Apps – Google would not be able to get from Firefox’s open community the integration with Google Apps it could have with Chrome. And forget getting Apple to do that in the browser given Mobile Me and iWork. Maybe Chrome is more of an Office Killer than Windows Killer, attacking Microsoft’s dominant position there with a superior product.

    2. IM – Google Talk has not done much against MSN, Yahoo and AOL but with an integrated browser client and common web-synch contact directory and calendar, it would not only gain ground in the IM space, but also deal another blow to Outlook.

    3. E-commerce – Google Checkout could also be better integrated and secured than Paypal.

    All the open-source code made Google’s entry into the browser market easy. Now let’s see how well they integrate their other products, and what that does to these markets.

  20. I think when people claim that Chrome is a Trojan Horse for a new OS they really mean a potential replacement for the traditional desktop. Google already has its own internal OS – custom built Linux that runs 65% of all searches online – you don’t see them ever open source that. It’s not what you would want to install on a desktop or phone. I suspect Chrome is Google’s counter to MobileMe. Despite its startup flaws, MobileMe shows a level of Desktop, Cloud/Web, and Mobile device integration unseen before. When it gets better and if Apple add in more features related to iWorks, I personally would drop my Google Docs and do it all on MobileMe – at least I own the content. I think Google’ is learning from their Android team how far Apple is along with WebKit to to make it the best mobile browser, and despite having its own mobile OS already in Android, Google realize it needs a combo platform of Android+Chrome to compete with iPhone+Safari, lest it gets lump with Nokia+Opera? or WinMo+IE?. The interesting story here is that Google seems to be saying what it needs now can be better served by WebKit than with Gecko/Firefox mobile version.

  21. Not to disrespect, but those who claim a browser kills an OS probably never did any serious study in computer science, which is why they don’t understand the convenience to have a dedicated layer for OS and one for applications. If a browser is to replace an OS, it has to take care of all the drivers, scheduling, resource allocation, security and so on. It quickly gets out of hand if tries to do all these, and we would be talking about a browser the size of hundred megabytes at least. The sad truth is most of these journalists don’t understand computer science yet they keep on spreading such nonsense just to attract eyeballs.

  22. If you think of the mobile phone as an example of an appliance, then yes, we are going to appliances.

    There are many comments here that are quite valid but are based on a particular computing paradigm. But one comment stuck out:

    “So if you’re Google, why not help Mozilla and Apple to become better and let us all move to the cloud faster?”

    So…what do you think Google has been doing all along? Perhaps you’ve heard of or used Google Browser Sync…great product, not a lot of traction, and consequently hard to sustain innovation. There are a LOT of sacred cows in today’s browser architecture…by creating its own browser, and making it open source, Google is able to lead by example and thereby “helps” Mozilla and Apple to become better.

    I can’t begin to try to debate the merits of a netbook with an audience that has sophisticated needs and can’t see how they would ever use such a device. Ultimately, the markets will speak far louder than any blog post as to whether your assertions continue to hold. I once shared your perspectives on appliances but in my conversations with laptop manufacturers, 2008 appears to be an inflection year.

    Let’s revisit this in three months.

  23. I agree, but I wouls like to address some of these comments. First, a web browser is an app and very far from an OS!!! Second, for Google to build another OS would be reinventing a wheel to say the least. Third, you cannot have a web-based OS, it makes no sense from a performance perspective. That would be like walking to work to get your car, then pushing it home so you can drive it to work. It’s absolutely silly. Finally, NOT EVERY COMPUTER IS OR CAN BE CONNECTED TO THE INTERNET!!!!!!!!

  24. Brian: “I believe it is only a matter of time before Google attempts to make Google Chrome the default operating system for a new generation of sub-$500 notebook computers that will never be touched by Microsoft Windows but instead will run Gmail, Google Docs and so much more.”

    So you’re saying we’re going back to ‘appliances’? As I pointed above with examples, all single- or limited-purpose appliances have been unmitigated flops, without exception. Yes, we have the beginnings of a ‘cloud’ now, but not too much has really changed since then. The first time that limited netbook is unable to do what’s commonplace to any other computer user, from playing back media to printing, complaints will pour in.

    “Interestingly, Google Chrome imitated Apple in choosing WebKit…”

    Just wanted to point out that before Apple picked it as a layout engine, KHTML was virtually unknown in the general public. It was Apple that forked it into WebKit and made it tangible in the marketplace as Safari.

  25. “Soon you will get near to OS performance in web based products.”

    How can you? Seriously? How can you even access the web without a computer? How does that computer run without an OS?

    Statements like that are just completely meaningless.

    If you just mean that some desktop software will be replaced with either pure web-based applications, or hybrid applications, then yes, I agree with you. But, you still need a computer, and that computer still needs an OS. And that OS will have to be multi-purpose.

    We applications and integration is great, lots of possibilities. But, given what computers are used for, you will never replace local processing, and local services.

  26. Google can see a gazillion people not making full use of their Internet experience every time your Firefox browser breaks.

    Google can see by Apple’s 25x mobile browser share that a superior user experience can break all sorts of expectations of what a browser is supposed to “do”.

    Google has already invested into Mozilla and Apple’s platform. There is no need to reinvent a wheel that is designed for horse-drawn carriages. Google has taken all of these observations and created a browser designed for the AJAX-driven world that Google itself helped create.

    Note that Google Chrome and Safari are both built atop WebKit, and its investment in Chrome helps WebKit and by extension, Safari, move forward towards the cloud.

    For the foreseeable future, there will always be people like Kontra that expect certain things from their computing environment.

    Robust sales of netbooks suggest that there is a small yet growing audience of people whose needs are less demanding. I believe Apple will launch its own sub-$600 netbook late September and by this time next year Google will lobby hard to have Dell and other netbook manufacturers preinstall Chrome on their netbooks.


  27. lolec: “Nothing stops Mozilla, Apple or even Microsoft from taking Chrome and re-brand it as Firefox or Safari.”

    So if you’re Google, why not help Mozilla and Apple to become better and let us all move to the cloud faster? Why introduce a browser of your own that might pose various different kinds of challenges to Safari and Firefox?

  28. Nothing stops Mozilla, Apple or even Microsoft from taking Chrome and re-brand it as Firefox or Safari. Actually that makes it windows killer. Even if no one use Chrome, i bet all browsers will take Chrome approach to the web, making the web stronger, think about it, Google is already the world leader in web based products, that’s what it is, so, making a browser to run web applications better and faster, and making it open source just will get more people using web based products instead of OS based. And if you are going to use web products, you will use Google products.

  29. Where’s the Justice Dept when you need them? GOOG gets a free pass until it’s too late, the horse is already out of the barn. Same thing happened in antitrust against IBM in the 80’s, the legal action was way past the monopolistic moment. Ditto MS in the 90’s, punished way after the monopoly crime. And now GOOG. Ooh but let’s give them a free pass, because after all, they said “don’t be evil…” Wake up people, GOOG is the new Borg, resistance is futile…

  30. lolec: “google needs a native product in every computer to compete with windows.”


    What is it, for example, that Safari won’t do for Google that Chrome can? Faster JavaScript, multi-process, Gears, etc? Most of those are supposed to be open source and available to WebKit folks, if needed. What prevents Google from helping Apple or Mozilla to add some of the things that Chrome has that Safari or Firefox might lack at the moment? Why does Google need Chrome then if not for potentially proprietary stuff that hooks up tightly to their services? And if that is the case, why should people jump on the Chrome bandwagon?

    “they can do everything online with Chrome”

    That remains to be seen. Until and unless Google provides the basic functionalities of an OS, as everyone expects it, it’s, as I said, a chimera.

    BTW, I have absolutely nothing against Chrome, just the notion that it is/can be an OS.

  31. Well i think you are not looking at the big picture here.
    Shure, Chrome is a windows only product for now, and its similar to other browsers, so, why isn´t firefox a windows killer ? You will not find the answer in the present or near future, but with netbooks growing in popularity and the web becoming faster and faster, google needs a native product in every computer to compete with windows. Shure, the browser still needs an OS to run in, but it will not mater if no one uses OS based software. Soon you will get near to OS performance in web based products.

    maps, calendar, mail, rss, documents, photoshop, im, social, music, video.

    Almost all the stuff people do in their desktops can be done online. so yea, its a windows killer when people will not mind having a ultra basic version of linux or whatever because they can do everything online with Chrome. Windows will throw a web based bomb soon. but this is a tiny stem in the end of windows as we know it.

Comments are closed.