Hi Russ,
[I-D author requested Cc]
At 19:35 24-09-2008, Russ Allbery wrote:
Well, I find that statement unobjectionable but essentially meaningless,
in that I don't think the document says anything substantively different
including that statement than without it. But if it makes others feel
more comfortable, I don't object to including it.
If you don't find that the addition changes anything, there is no
point in including it. The point was that the author address should
be fixed if it's not according to the specifications for that
medium. I did not elaborate as the medium is not restricted to mail.
That provision doesn't document what happens in practice, nor would it be
possible to limit the header additions. (For example, it's essentially
impossible to send something as an e-mail message without adding Received
headers.) I prefer to give implementors a better idea of what to expect,
and in practice arbitrary headers will get added by the transit through
the mail system, including all sorts of X-* headers, trace headers, and
random detritus from spam filters.
Okay.
We could easily remove that specific header field name from the example
and instead just say:
The news-to-mail gateway adds an X-* header field to all messages it
generates. The mail-to-news gateway discards any incoming messages
containing this header field.
Would that be an improvement?
Yes, that's better.
Regards,
-sm
_______________________________________________
Ietf@xxxxxxxx
https://www.ietf.org/mailman/listinfo/ietf