Robert:
 
You said:
 
> Seriously though, I once used the VMS SEND command to PUT a file that
> contained an 8-bit character that I had forgotten to strip via BITNET.
[...]
> The character was stored in the filelist but I don't
> remember whether it was garbled by any of the steps.  I'm sure that
> somebody remembers how this works.
 
It works OK (i.e. the files are stored with the 8-bit characters intact)
with BITNET SEND for GET and PUT.  However, fewer and fewer LISTSERV
administrators and editors have direct access to BITNET.=20
 
> Of course, as
> soon as you are using the higher order bit, you also have to worry about
> character set standards.  Should you use LATIN-I, DOS, Mac?  Who will be
> using the data?  What software will be used to access it?  These are
> rhetorical questions ;-)
 
These questions have real answers.  The MIME standard is quite complete
and incorporates at least 146 different character set specifications,
along with another 286 aliases for these sets.  For mail programs like
Pine and Eudora which have the ability to specify which character set you
are using, this information is included in the three lines of MIME data
that I am suggesting be available to all LISTSERV lists subscribers. At
this point, I think that rather than creating a new MIMEHDR setting, it
would be best to just add these lines to all HDR settings, including
SHORTHDR.=20
 
The three lines are as simple as this:
 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=3Diso-8859-1
Content-Transfer-Encoding: QUOTED-PRINTABLE
 
This tells that the message uses the ISO-8859-1 character set and that
characters above Hex 80 are coded like =3DE9. If your mail program is
capable, when you receive a message like this, it will inform you that it
may not display these characters properly. On the other hand, if you have
your character set properly and you have font support for it, you can send
and receive E-Mail in Hebrew, Cyrillic, Arabic, and ahost of other
non-Latin-based languages.  The person who started this thread was asking
about Albanian dialog and, while I am unsure if they need non-Latin
characters, I am reasonably sure there is an appropriate character set for
them which will allow them to read and write their real characters without
resorting to kluges for =E4 like 'a' or 'a"'.=20
 
Finally, someone mentioned that the default for LISTSERV was FULLHDR. As
somone in another thread pointed out, there is no "out of the box"=20
configuration of LISTSERV and I find that this list (LSTOWN-L) is
apparently set to default to SHORTHDR (which has some interesting
implications about what is considered necessary and/or efficient) and so I
did not get the MIME information in your message, hence your signature
line looked like this to me:=20
 
> Bob  (qui pr=3DE9f=3DE8re tout de m=3DEAme =3DE9crire avec ses accents)
 
For people with FULLHDR, this would have appeared as:
 
  Bob  (qui pr=E9f=E8re tout de m=EAme =E9crire avec ses accents)
 
For people without FULLHDR, *both* the above lines will contain some too
many weird codes. I had to paste together the above line by looking up the
characters in ISO 8859-1 for E9, E8, and EA.=20
 
So... in closing, while it may offend the sensibilities of Eric and others
to include three extra lines in a SHORTHDR, I find it to be a virtual
necessity for those who wish to dialog in languages other than English
that use the appropriate character sets and who also happen to be using
systems that they cannot control the administration of.=20
 
Cheers
 
Mark Hunnibell                  Email: [log in to unmask]
KIDLINK Gopher/WWW Coordinator  http://www.connix.com/~markh/index.html