Full Disclosure: I recently switched back to IE from Firefox after the latest browser update (1.0.2, I believe) caused slowdowns and instability. Is that the source of the problem? I don't know, nor do I care. I'm enjoying using IE again, even without tabbed browsing, plus it automatically logs into GameDev.net, something I could never get Firefox to do no matter how many times I deleted and recreated my cookies.
I don't even miss password manager that much.
Which brings me to browser obsolesence (and this started out as an addendum to my enthused gushing over Spotlight and Mac OS X Tiger). I've never understood why everyone makes a big deal over Firefox gaining ground on IE. So what? The browser remains important for now, but will become increasingly less so - as a standalone component. As web services proliferate, and not only the .NET variety but also XML-RPC and SOAP, local applications will be able to transparently query remote data stores for the information that is pertinent to the individual user. iTunes Apple Music Store (aka iTAMS - go figure) is an online application, but can not be accessed without iTunes (well, it can, but it requires some creative hacks) or via the browser. It is integrated into the local iTunes application that resides on the user's hard drive, transparently querying the online data base and conducting transactions.
In that light, in the larger scheme of things, so what if Firefox overtakes Internet Explorer? Where the real work should be progressing is in developing and popularizing the Gecko layout and rendering engine as an efficient, accessible, cross-platform way of embedding a standards-compliant web presentation component in your application.
Firefox is a very good browser, but as digital convergence occurs - and it will occur, whether you personally like it or not - and web content becomes just another form of retrievable content, the appeal of the standalone browser will wane.
Cheap shot: I believe open source works as a store of common knowledge, as a democratization mechanism for technology that has become infrastructure. At the point where open source provides a truly viable alternative to a proprietary solution across all critical usage contexts, the technology is basically a freebie and proprietary developers should move on to something else. The web browser is infrastructure, though the emphasis on Firefox over Gecko is hurting the perception, even among developers, of the web browser component as distinct and useful.
Frankly, if not for a few glaring deficiencies in Firefox (say what you will, but being able to seamlessly navigate my local machine from the web browser and navigate the web from my filesystem explorer is extremely useful; and the lack of COM interop hurts Windows intranet applications), I would say Microsoft ought to stop developing IE and just adopt Firefox as their browser of choice. Of course, that would hurt the public perception of Microsoft (because the press enjoys building IE vs Firefox up as some sort of meaningful "conflict" - bitches), so it's not going to happen.
Plus, you know IE7 is going to be badass. Microsoft has taken far too much heat, like they did with IE3, like they did with VC6 as it aged, not to come back with a killer product. Should be fun to see what they do with it.