At 18.23 -0400 01-05-22, Keith Moore wrote:
no disagreement with any of the above. I was just saying that the
generalization "lists are users" has often led to incorrect conclusions,
even if a list is closer to an MUA than to an MTA.
The problem is that IETF has never made any complete standard
for mailing list behaviour. Some details are mentioned in
some places:
"-request" convention in RFC 2142
"List-" headers in RFC 2369 and RFC 2919
Loop control in RFC 2821
MAIL FROM in RFC 1123
I wrote, two years ago, a document summarizing some commonly
agreed ideas about how mailing lists should work. I have
just written a new version of it and submitted it to IETF
drafts. It has not been announced yet, but you can download
it from http://dsv.su.se/jpalme/ietf/mailing-list-behaviour.txt
This document contains things which cannot go into a standard,
so it could become an informational RFC or a BCP RFC.
I wrote this document mainly as part of an effort to persuade
the developers of FirstClass to modify FirstClass to behave
more in accordance with standards and generally agreed
ideas how things should work. My experience is that
groupware products, like FirstClass, for which e-mail is
not the primary goal of the products, tend more often than
pure e-mail software to do things the wrong way.
In particular, FirstClass converts incoming mail to its
own internal format, and then when the message is sent
out to mail, converts it back to mail format again.
This results in a lot of "header munging", for example
the outgoing message has a FirstClass internal Message-ID
instead of the Message-ID of the incoming message.
The proper way to do this, which we have done in our
own system, is to store the original incoming e-mail.
Even if you show your own users the converted version,
you use the stored, original incoming e-mail as a
basis when sending it out again, and modify this only
in accepted ways (adding "Received" and "List-" headers
when it is sent out to e-mail members of a meeting
in the groupware system, which properly should act
as mailing lists, when seen from Internet mail.)
--
Jacob Palme <jpalme(_at_)dsv(_dot_)su(_dot_)se> (Stockholm University and KTH)
for more info see URL: http://www.dsv.su.se/jpalme/