I've written a lot on the efforts in converging web and desktop applications (Adobe Air, Mozila Prism, Silverlight, JavaFX, Google Gears etc.). Now Google ups the ante with the introduction of its new webkit based browser - dubbed Chrome (You can see a comics explanation of its main concepts/features).

Let's do a quick recap before taking more about Chrome. Basically we see all the major players trying to blur the lines between desktop applications and web applications (a.k.a. Rich Internet Applications) some players are on the offensive (Adobe, Google) and some on the defensive (Microsoft, Sun) but the direction is identical. The web oriented companies understood that RIAs are becoming more real application and not "just" web pages. They also understand they need presence on the desktop for easier accessibility and better acceptance by users as a "serious" applications. Furthermore, the fact the applications become more serious and more mission critical, along with the fact that they can (and are) be used on-the-go where disconnects occur, the need for occasionally connecteness becomes more apparent. This is where smart-clients have a lead and technologies like Gears are trying to catch-up.

Now,in my opinion, Google makes a bold move to change the rules and re-define the playground - if webapss need to run on the desktop, let's make the browser the new desktop.
What makes me say that? because it is focused on application (see the comics),because the browser runs each tab in its own process, because it has a process monitor, because it is a link on the google home page...


From the chrome "OS" point of view we can look at javascript,HTML etc. as the IL (bytecode in java speak) on which the application run. This makes cross-compilers like GWT and the good side of MS Volta (vs. the bad side) the next abstraction layer. I expect these will be more significant in the future

Anyway, you can see for yourself if you download it now from www.google.com/chrome