Draft specification for future X-Trace header

[Russ Allbery]

Forrest J Cavalier <mibsoft at epix.net> writes:

> There are understandable reasons to rewrite Path:, and perhaps X-Trace
> (although adding a new X-Trace makes sense to me.)

I'd much prefer to just reject if X-Trace is already present.  That's what
we already do and the various suck/rpost sorts of things already cope with
that by and large.

> What case can be made for rewriting Date and Message-ID?  If they are
> missing, they can be added.  If the Date or Message-ID are invalid, the
> post can be rejected.

> What do the working USEFOR, et al drafts say?

Precisely that.  That hasn't prevented some news server implementations
from not doing things that way, though.  :/

-- 
Russ Allbery (rra at stanford.edu)             <http://www.eyrie.org/~eagle/>