Hey, there isn't a commonly accepted scheme for citing papers or books. Well,
not a single one, anyway.
It's an interesting question though, because of the potentially transient
nature of the resource (not even 'potentially' for news). Electronic journals
are going to have to make sure that a single canonical version of each paper
exists, and that it always stays in the same place. It should not be altered
or corrected after publication. Although electronic means provide an excellent
medium for adding later notes and referring to later papers, there should be
a page that refers to the paper and which also links to later notes. But I
digress.
I would use something like the following. (Disclaimer: the below resources
do not exist!).
For e-mail
[1] D.E. Knuth, private communication
(or "private e-mail") as with paper letters at present.
For HTML documents
[2] D.E. Knuth, "The Electronic TeXbook," http://tex.org/book/
but I would be wary of citing them unless they were guaranteed not to change.
For news
[3] D.E. Knuth, Usenet article.
No point in saying more as no-one can go and look it up, unless there happens
to be a permanent archive of that group. Again, I would be reluctant to cite
it at all.
Just my tuppence (3.2 US cents) worth.
-- Stephen R. E. Turner Stochastic Networks Group, Statistical Laboratory, University of Cambridge e-mail: sret1@cam.ac.uk WWW: http://www.statslab.cam.ac.uk/~sret1/home.html "I always keep one big file in case I run out of space." A colleague of mine