These proclamations invariably miss the bigger picture while focusing on a narrow aspect of an issue. The death of the desktop is hailed because:

Installing and uninstalling and maintaining anything on the desktop (be it Windows or Mac or Linux) is hard, and more and more people won't bother. Why? Because there's a better alternative. Another way of saying this is, the browser is the new desktop.

Oh, so the browser makes all the problems of interoperability, specialized applications, and maintenance go away? Not exactly. From that same post the author describes how the he "solved" the problem of a slow computer:

The problem turned out to be too many plug-ins in the browser. She had a Upromise plug-in, a Google toolbar plug-in, a Real media plug-in, and a bunch of other plug-ins I didn't even recognize. I turned it all off, restarted the browser, and poof, "the computer" was several times faster. Cue fanfare.

Turning applications off will make performance issues go away but that is hardly a solution worth praising. Instead, this is an excellent example demonstrating that the problem, and therefore the solution, is not tied to a particular platform, be it Windows, OS X, or a browser, but to the complexity of what we are trying to do in IT. The desktop isn't dead, the browser isn't dead, and with Google Gadgets and a like they will increasingly work better together.

Now, how about a moratorium on "X is Dead" nonsense so we can spend more time on the real challenges.