ietf-nntp <0> and message IDs

[Matthias Andree]

Russ Allbery <rra at stanford.edu> writes:

> I think we have consensus on eliminating the <0> wart and requiring that
> all articles have message IDs (possibly assigned by the server if the
> incoming message didn't have one).
>
> If anyone disagrees that that's the consensus, please speak up now.
> Otherwise, Clive, please go ahead and make those changes, and then we can
> move on to making sure that the wording on requiring a message ID is okay.

(Ceterum censeo "<0>" esse delendum).

On the topic of how the Message-ID is to be obtained, and on the server
providing one, the Usefor drafts make a distinction between "Injecting"
and "Relaying" agents. Only the former are allowed to add the Message-ID
header (and turn a "Proto-Article" (which may lack Message-ID) into a
article that has all mandatory headers)) -- so POST might be able to add
the Message-ID, but not IHAVE. Opinions?

-- 
Matthias Andree