Google says the software architecture will basically be the current Chrome browser running inside “a new windowing system on top of a Linux kernel.” So in other words, it basically is the web as an OS.
Or, in a bit less sensational words, they are not developing an OS but a window manager for an existing OS. They might also be dropping X11 as they go, but the article doesn't give that away.
This might actually work for netbooks that are mostly used as a fullscreen browser. But I don't see this making any significant inroads on bigger laptops or desktops in the foreseeable future. The technology behind "the web" just isn't up to the task. Wrong tool for the job.
The problems begin on such a basic level that I find it ridiculous someone with the engineering muscle of google is even considering this.
How basic? Well, we don't even have a Socket on this soon-to-be-an-OS platform. That basic. Admittedly it should be relatively simple to add and hopefully the one outcome of this project will be that Websocket finally gains traction (across all browsers).
But it doesn't stop there. The HTML/JS combo is lacking severely in many areas relevant to RIAs. People still have to reinvent basic widgets like comboboxes, datepickers, auto-completion, richtext input, sliders etc. in javascript. CSS is still a nightmare even for simple layouts. Will google fix these problems and break compatibility to other browsers? Don't think so.
So here's what I think they'll deliver: A mangled linux distro with chrome running fullscreen, likely even on top of X11. It'll have a fancy start-page that wants to log you into the google universe first. There will be some flashy widgets and a fancy javascript framework for the windowing - but overall it will just be a fullscreen browser with all its problems.
I hope they surprise me and do better. Because if this is their plan then I think it's time to sell some of those GOOG shares. Google never needed to hype their products upfront thus far. In that light this is a worrying announcement.
Some are claiming that rather than hurt Windows, this is going to take the market share Linux "might have had". But I really don't think that Linux users are going to want a system that gives up more of their privacy to Google, and restricts the power of your own computer to what you can do with Google.
I hate to use the expression "killer app", but if there ever was one, it would be a computer that "just turns on". This has been Microsoft's Achilles heel forever. I've always said that if there was a laptop that turns on as quick as a Palm pilot, it would steal the market. Now we'll see...
The BIOS is one area that is long overdue for some modernization. If the folks over at coreboot.org can do it from the outside, imagine what would happen if mobo manufacturers opened up their specs to the community and embraced what came out of it.
Yes, it certainly is overdue for modernization. Speaking as someone who has had the misfortune of interfacing with it, nothing would make me happier if it went away.
One of the other stories claimed they said it was because (last Thursday) the NYT and Ars Technica reported it. That's what Google said, but of course they didn't have to confirm the stories (and I think they haven't on previous occasions unless it suited them).
So your question remains: why does it suit them now? (Guessing), perhaps part of their evolution into a huge corp?
Wild speculation: perhaps they are trying to get out in front of upcoming announcements on actual products to show they are working in this space too? The crunch pad is due out soon and Apple is also believed to have a product in this space as well. If they put this out there now, any publicity for similar products on blogs and news articles will also likely have a blurb that "Google has also announced they will enter this competitive space..."
Im pretty sure google base can be written off as a failure - not total since its still in operation. But it was supposed to be a repository for product information that google could throw into search results like video and news.
What Google is doing is not recreating a new kind of OS, they’re creating the best way to not need one at all.
The OS should seem like it isn't there at all. It should stay out of your way as much as possible, but be there to do what you want as quickly as possible, then get out of the way again. The OS should be like a storybook butler. Discrete, unobtrusive, quietly useful.
A nuclear bomb? Come on, they didn't even release a direct competitor to Microsoft's OS. Yes, I'm sure it will be streamlined and efficient and very well designed for web apps and things programmed for it, but it still doesn't replace a Windows desktop machine because of its inherently different architecture.
No, you're right, it doesn't. But if you take all of the pieces that Google have been assembling over the past few years - Chrome Browser & OS, Docs & Spreadsheets (ie. Apps), Gmail, Android, NaCl, etc... these all add up to a very slick platform that ticks a lot of boxes.
For a very large number of people, a cheap PC/netbook/tablet thing with a browser that can do web, email, light word processing & spreadsheets, boots in 2-3 seconds and costs $150 would be ideal. I'd buy one.
Chrome OS + Chrome Browser + Google Apps puts Google in a position to deliver this, using super light & cheap ARM netbooks, taking them to a place which Microsoft will find it very hard to follow. This puts them in a position to take away a pretty large proportion of MS's home users, who are currently running a ~2-3yr old PC with XP on it for these tasks.
The current state of the Chrome browser on Linux would make this "web OS" the equivalent of Windows 3.11. In it's current state, there is really no way to compare Chrome and Chromium and claim they represent the same kind of polished product. Chromium had so much issues with stability and performance, that I had to give up on it. I took a short detour by Opera, before Firefox 3.5 became usable, and I was back where I started.
If I were going to make a web OS, even for netbooks, I would rather have it use Firefox 3.5 as performance is much more predictable and reliable, not to mention it can be extended with addons, something I would consider a minimum for any platform.
One word: games. Consoles have really stepped it up, but a lot of people still use Windows at least partially because of hardware accelerated 3D gaming. I don't see that being Google's netbook OS focus, or many people porting such things to web apps, obviously. Unless they have some secret alliance with OnLive ;-)
Google has the O3D API, browser-based access to 3D graphics, presumably hardware-accelerated for machines that have it. http://code.google.com/apis/o3d/.
DirectX has done almost as much, if not more, to preserve the Windows moat as MS Office has (moat in the sense that Warren Buffet says it). Many people buy Windows (or steal it) just to play games, just as many businesses stick with Windows because MS Office is the resume-standard office suite and VB apps are so easy to hammer out.
Really it'd be nice if either a) game developers switched to OpenGL, or b) the open source community came up with something better than DirectX from the developer perspective, creating a natural impetus to switch.
(Yes, I am being entirely selfish here, as I'd like to stop dual-booting for my non-Wine games.)
just 2 days a go i predicted that google will come out with its own OS - though android - but never expected that in just couple of days my gut feeling will come true ...
since google chrome runs on microsoft windows, I think it would only be fair if microsoft internet explorer would be allowed to run on google chrome os (under the condition that ms would port it, which seems rather unlikely)
Or, in a bit less sensational words, they are not developing an OS but a window manager for an existing OS. They might also be dropping X11 as they go, but the article doesn't give that away.
This might actually work for netbooks that are mostly used as a fullscreen browser. But I don't see this making any significant inroads on bigger laptops or desktops in the foreseeable future. The technology behind "the web" just isn't up to the task. Wrong tool for the job.
The problems begin on such a basic level that I find it ridiculous someone with the engineering muscle of google is even considering this.
How basic? Well, we don't even have a Socket on this soon-to-be-an-OS platform. That basic. Admittedly it should be relatively simple to add and hopefully the one outcome of this project will be that Websocket finally gains traction (across all browsers).
But it doesn't stop there. The HTML/JS combo is lacking severely in many areas relevant to RIAs. People still have to reinvent basic widgets like comboboxes, datepickers, auto-completion, richtext input, sliders etc. in javascript. CSS is still a nightmare even for simple layouts. Will google fix these problems and break compatibility to other browsers? Don't think so.
So here's what I think they'll deliver: A mangled linux distro with chrome running fullscreen, likely even on top of X11. It'll have a fancy start-page that wants to log you into the google universe first. There will be some flashy widgets and a fancy javascript framework for the windowing - but overall it will just be a fullscreen browser with all its problems.
I hope they surprise me and do better. Because if this is their plan then I think it's time to sell some of those GOOG shares. Google never needed to hype their products upfront thus far. In that light this is a worrying announcement.