As someone who has cited I-Ds in published papers (if you're curious,
see http://www.research.att.com/~smb/papers/badesp.ps -- I even
identified one of the I-Ds as an "obsolete Internet draft), I have to
support the notion of keeping them around. In my case, I wanted to
point out the inherent problems in a particular approach to IPsec. I
also needed to point out the evolution of the ideas over time. I know
of no better way to do that than to refer to successive versions of the
same document. In fact, I had to hunt through the net to find various
versions of the documents in question...
--Steve Bellovin