Bruce Lilly <blilly(_at_)erols(_dot_)com> writes:
Since it is unchanged from RFC 822, that's pretty hard to pin on 2822.
If the 1036 restrictions need to remain in effect until software authors
have had a chance to make revisions, fine.
RFC 822's References specificaction is useless junk. RFC 2822 is the
first mail standard with a useful specification of References, taken
almost entirely from Usenet (and from mail clients copying existing
practice on Usenet).
I think the issue here is comments, yes? If so, I think news software
should ideally learn to cope, although I have no idea why one would
intentionally put comments into the References header other than just
because one can.
--
Russ Allbery (rra(_at_)stanford(_dot_)edu)
<http://www.eyrie.org/~eagle/>