Re: You can't win!

In <[email protected]> John Stanley <[email protected]> writes:

>>>There is a third possibility, which I have seen in action: ignores
>>>application/news-transmission header altogether and posts entire
>>>encapsulated message as the body of a new article with locally-generated
>>>headers.

>3. The HUMAN reading and approving the email was presented with an
>    encapsulated news message as the "body" of the submission, which
>    differs from unencapsulated submissions in that it contains the headers
>    for the news message, including any "references" headers that MUST be
>    maintained in followups. That should be a red-flag to anyone actually
>    looking at the submission.

Ah! So it was a human, who should have known better, and not some broken
script :-( .

>Yes, it knows the intended newsgroup because it was mailed to the 
>moderators submission address. The newsgroups header in email should be 
>ignored because it has no well-defined meaning. This points out exactly 
>why encapsulation should be the only defined means of forwarding messages 
>to a moderator from a news agent. That USEPRO promotes ambiguous behaviour 
>is a bug, not a feature.


Yes, we are all agreed that encapsulation is a _much_better_ way to
submit articles to a moderator.

But, unfortunately, the ghastly method of just putting a Newgroups (etc)
in the email is so widely used currently, and moreover a large number of
moderators are totally unaware of the 'proper' way to do it, that this WG
decided, some way back, that is was better to retain that ghastly method
rather than suddenly to break what was actually working reasonably well.

My message, which started this thread, was just to point out that adding a
Content-Type: message/news-transmission could actually break some
currently-working encapsulation-safe moderators' scripts, contrary to what
we had earlier assumed, and contrary to what USEPRO currently implies.

AND to ask whether anybody wanted to tweak our current wording during
AUTH48, which is our last chance to do so. But I hear no such suggestions
so far from this list.

-- 
Charles H. Lindsey ---------At Home, doing my own thing------------------------
Tel: +44 161 436 6131            Web: http://www.cs.man.ac.uk/~chl
Email: [email protected]      Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K.
PGP: 2C15F1A9      Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5