o, the google OS. it was only a matter of time. kottke points out that a web platform/OS effectively already exists, since users can already run most of their lives online. this is true, but misses the point.
the point is this: the main (only? probably not) promise of the web as OS is client thinness -- if everything happens there, very little needs to happen here. we should have ultra light, incredibly cheap clients that consist essentially of a small graphics-optimised processor, a display, and input device/s. but, of course, as it stands, in order to be able to use the web as an operating system, users still almost always require a heavy operating system. (i discount smartphones as promising but ultimately incapable of replacing full-featured systems as they are currently configured. android is nice, but it doesn't cut the mustard. if chrome OS, as promised*, can jump you straight into the browser while still providing some of the client-side lifting necessary for a low-latency user experience, it will surmount this almost-last hurdle to the world of thin clients. (after that, we'll just need immersion wireless connectivity, but 4G is theoretically on the way.)
* why did they have to go and announce vapourware? what happened to the company i used to work for anyway? i expected better of sundar.