Sender header (was Re: ietf-nntp BCP ...)

[Chris Newman]

On Wed, 18 Dec 1996, Nat Ballou wrote:
> I believe all we are talking about here is describing how the
> sender: line works today for a current practices doc.  Granted,
> there are cases where even today, it is not of much use.
> 
> However, I believe everyone agrees that the Sender: line (as defined 
> above) does NOT belong in any RFC since not every authentication 
> provider is able to derive an email name for an authenticated user.

Resolving the problems with the Sender header has been addressed in
the DRUMS working group.  There was rough concensus at the San Jose
meeting that the "Sender" header refers to a deliverable mailbox.  So
constructing it from NNTP authentication info may be impossible.

I'm working on an independant header proposal to address the trace
information issue.  It should be coming out as an internet draft in the
next couple days, probably with my suggested filename:
"draft-newman-msgheader-originfo-00.txt".