From: Russ Allbery (rra@stanford.edu)
Date: Thu Sep 12 2002 - 09:54:56 CDT
Ralph Babel <rbabel@babylon.pfm-mainz.de> writes:
> 2. Same as #1, except that for news-only headers,
> in particular those that are critical to articles
> in transit ("Path:", "Message-ID:", "Newsgroups:"), we
> may well devise our own mail-compatible 7-bit encodings
> for efficiency or to avoid problems with current server
> implementations, or if relaxing the traditional syntax
> would adversely affect the distribution of articles
> within the current server infrastructure.
> 3. Same as #1, but no restrictions on news-only headers,
> e.g. if someone includes an 8-bit news header in mail,
> it's their own fault. After all, there's
> application/news-transmission.
I'm somewhere between these two, with a bit of a change. In general, I
think that we should simply use the mail standards as is, but there are
some cases where the mail specification simply will not work for Usenet
(such as allowing spaces in the Message-ID), and in those cases I think we
should outlaw the problematic constructions and then try to get them
changed in the next version of the mail standard, whenever that happens,
so that we can re-establish compatibility. Unfortunately, there are some
things that mail allows that just badly break news in a way that can't
easily be fixed.
Those changes should be as minimal as possible, however.
I think it should be possible to do 7-bit encodings of every header for
mail compatibility, and it may make sense to require this for postings to
moderated groups, or at least have the server check for characters in
ranges mangled by mail.
However, use of raw 8-bit characters in national hierarchies is already
widespread and I'm hesitant to just outlaw it completely. I do think that
we should discourage any use of untagged 8-bit content in the headers,
probably at the SHOULD NOT level, but then try to figure out what to do
about it in the other areas of news since it will still keep happening.
-- Russ Allbery (rra@stanford.edu) <http://www.eyrie.org/~eagle/>