...two things, really. Both are important to Microsoft, and both are, frighteningly within their grasp.
The first, and perhaps most crucial, is .NET's role as a massive, Microsoft-specific "innovation," which they can pull out and show off at their appeals of the DOJ antitrust ruling. You can't get much more "cutting-edge" than distributed, XML-based, full-blown application and desktop OS funstionality. It has all the right buzzwords, and is such vapor at this point that no one can tell them that it's crap.
Secondly, .NET really probably will turn into some product or products within the next year to eighteen months. The open source rallying cry of "many helpers, many eyes" applies in most ways to Microsoft, too, since they have more programmers to throw at most problems than God (or even most OSS software).
I'm not implying that it will be a good product, or that it will be anything like the grand picture they've painted so far, but what they're talking about is basically just an upgrade of their scripting tools to support a few more languages, and XML (possibly SOAP) protocols for distributed COM automation.
So, what do we do about it? If we can see through the marketspeak running rampant about this whole thing, it looks far less rediculous, and it doesn't seem like it would be that hard to pull the rug out from under them. How, you may ask? By beating them to the punch. We need SOAP, or a similar, open, easy to work with messaging protocol for remote object transfer and control. We need libraries for that protocol in Java, Perl, Python, C/C++, etc., and we need it soon.
Let's stop spending hours and hours tweaking Enlightenment themes and ripping MP3's, and see if we can actually pre-empt Microsoft on something. Any takers?