Except it turns out that obsolete stuff like local storage and direct hardware access and being able to run offline is actually sort of useful, which is why everybody is falling over themselves trying to jam all of it into HTML5.
I assume that Chrome OS will completely sync with Chrome on the desktop. This means that if you're working on your web-based apps on your computer, you'll be able to pick up your Chrome tablet and seamlessly continue the work and then switch back again.
In other words: what do Chrome OS's limitations add to the experience? It's interesting from a technical standpoint to build a fully-functioning system with Chrome OS's limitations, but how does it make a better product?
If that is true, then Chrome will certainly be violating the spirit if not the letter of software - "free software" that only allows you to do things on server controlled by someone else...
It is _exactly_ what I've wanted since netbooks came into existence, and I've put a lot of hours into setting up stripper Arch setups that do this.
Is it better than anything else on paper? No, but on paper isn't the only way to measure things (to the chagrin of us engineers). Sometimes less is more.
If touch tablets ends up upending the netbook category, then Chrome OS won't really have a home. It doesn't have the capability of taking advantage of advanced hardware, and netbook functionality might be fully accounted for by Android tablets. Chrome OS was an interesting bet in the direction of computing (smaller, inexpensive notebooks as the norm for Internet consumption and light office work), but things might not pan out the way Chrome's founders had thought.