Comments on: Yet another post (like everyone else) on Google Chrome OS

By: Ankur Banerjee

Ankur Banerjee — Sun, 12 Dec 2010 12:31:32 +0000

Cheers!

By: Pulkit Kaushik | TheDolt

Pulkit Kaushik | TheDolt — Sun, 12 Dec 2010 10:40:41 +0000

“Would you but a laptop on which you…”

*buy :)

By: Mrittunjoy

Mrittunjoy — Sun, 29 Nov 2009 13:14:50 +0000

True. Probably people get confused between SaaS and Software+Service. Feasibility is definitely a factor for which Google may have face the doodles!

By: Pritesh Gupta

Pritesh Gupta — Thu, 16 Jul 2009 12:08:09 +0000

Hmm.. Nice post, I also think that there OS may fail as its not possible to be connected to internet 24/7.

By: Ankur

Ankur — Sun, 12 Jul 2009 22:05:53 +0000

Frankly, I think that Google's announcement was partly influenced by Microsoft's announcement of Gazelle. At some level they felt threatened by it because Google (rightly) thinks of itself as the leader in the field of web applications right now and wanted to have something in place to match Gazelle - just for the heck of it. The idea's nice, it just won't work now. Web apps are terrific for collaborative work and there's certainly at least that niche they can corner. 5...maybe 10 years down the line Chrome OS would have been a viable idea with more reliable Internet connectivity. BTW, Google is working on a 3D standard for browsers but not in the way you're thinking. The official blog of the O3D project has demos and videos of developers working on the project. So are Mozilla and Opera.

By: Anonymouse

Anonymouse — Sun, 12 Jul 2009 18:31:35 +0000

I think that they have done something that is far ahead of its time. Distributed cloud operating systems is the best way forward as devices get tinier and tinier, while our expectations get bigger and bigger. So, having the processor do the dedicated graphics computation makes sense, and of course the odd 2 bits when lag kicks in. However, this presupposes as you said perfect connectivity, and I think that the day for that shall come but it is still a bit far away to go out and buy such a notebook.

Although there are ways and means to fool the user into thinking that they have perfect connectivity, for e.g. use a buffer and send the data as strings for stuff like games. Basically send in the computed vector co-ordinates, lighting, and rendering details as a small text file that an interpretor can reconstruct and run. So, as you jack off around in the game the text files go back and forth and the computation takes place on the server and the text file comes back where your computer is instructed to display it using its decoder.

Something like what used to happen in the early days of computing when they would move around these tapes they wanted to read them, but different computers used different formats. So, what one of these guys did was that he added a bunch of data at the starting of the tape consisting of pointers, and instructions to read the tape itself.

Basically, the tape itself carried its own runtime environment(analogous to it, anyway. I wouldn’t call it a codec as it did everything). Now what we can do is that we can download a r.e. at the start of the session and just send in text files with the appropriate commands, and function calls to do something. This would be a more efficient, better manner than downloading a whole image every time. You might say that a cache already does this, that we keep content on the disk, but that’s not what I want. I want the entire recipe to be downloaded to the user’s books so that seamless integration can work on even choppy net speeds.

I’ll bet you all I have that this is precisely what Google is going to do.

Oh and I won’t buy it. I am saving for OS X.