>I've seen that. Microsoft has an opposite opinion, as you would expect. And, not being able to see their code, one can only run external tests to see if such problems would occur. If Gibson can create exploits, and he is not the brightest bulb in the box, you can rest assured other, more talented people are exploiting them to the max. I have often wondered if there was a relationship to the appearance of these bugs and the intrusion into Microsoft's intranet last summer. It would have taken about 6-10 months to scan code for weaknesses and test them out.

I think you're confusing holes in IIS with raw sockets support. Raw sockets support could be useful if you're writing your own low-layer protocols, but is most often used in syn/ack attacks to sent malformed packets. _ALL_ other OSs do this, and Windows can do it now too, it just doesn't do it natively. Whether or not raw sockets support in Windows is a giid or bad thing is _exactly_ an opinion, and it depends entirely on your philosophy. Sort of like gun control. What kills people, guns or people?