From: Charles Lindsey (chl@clw.cs.man.ac.uk)
Date: Mon Jan 31 2000 - 04:29:43 CST
In <ylsnzikp28.fsf@windlord.stanford.edu> Russ Allbery <rra@stanford.edu> writes:
>G James Berigan <usefor@war-of-the-worlds.org> writes:
>> They aren't equivalent. DRUMS requires at least one FWS whereas ours
>> requires either one FWS or one comment without FWS.
>I see that now that Pete has also realized the same thing. *wry grin* So
>MESSFOR has gone back to what we currently have, which means that my
>original statement is correct; in order to ensure at least one whitespace
>character between references, we need msg-id *(FWS [CFWS] msg-id).
No, we are OK as we stand.
Currently, we say that all comments are a MUST recognize but SHOULD NOT
generate (except in the few places they are already customarily used).
So posting agents that adhere to that SHOULD NOT will always generate
genuine whitespace between the references, which present software will
understand. (And present software would not accept comments there anyway,
whether or not WSP was present as well).
Future software MUST be written to accept either WSP, or a comment (or
both) between references (and if it happens also to accept neither, then I
don't care). So when "new" software is prevalent and reading agents start to
relax that "SHOULD NOT" (which they really should not be doing until the
next news standard), everything will then be found to work.
Actually, I doubt posting agents are going to want to generate comments in
funny places. The real reason why we are moving to allowing comments (as a
MUST accept) is so that valid email messages which happen to find their
way into the news system will (as the "new" implementations become
prevalent) have less and less reason to barf.
-- Charles H. Lindsey ---------At Home, doing my own thing------------------------ Email: chl@clw.cs.man.ac.uk Web: http://www.cs.man.ac.uk/~chl Voice/Fax: +44 161 437 4506 Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K. PGP: 2C15F1A9 Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5