wildmats (was: Re: ietf-nntp Draft summary of IETF 48 meeting)

[Andrew Gierth]

>>>>> "Russ" == Russ Allbery <rra at stanford.edu> writes:

 [Russ]
 >>> I'm in favor of generalizing comma support; it's supported in enough
 >>> different places and not in others that trying to maintain that
 >>> situation doesn't strike me as a good idea.

 [Clive]
 >> I would be happy with that. Are we allowed to do it (I remain confused
 >> as to what changes we are and aren't allowed to make) ?

 Russ> It's *somewhat* borderline, but I'm comfortable with making
 Russ> this change because some currently deployed NNTP commands
 Russ> already support commas, so there's existing practice and
 Russ> existing implementation experience.

Nonsense. Commands like LIST ACTIVE currently take single wildmat
patterns, and existing implementations will treat commas in those
wildmats literally; therefore this is an incompatible change.

-- 
Andrew.