[Mailman-Developers] Dates again
Stephen J. Turnbull
stephen at xemacs.org
Mon Nov 22 10:28:20 CET 2004
>>>>> "Steven" == Steven Kuck <scrib at afn.org> writes:
>> At 3:31 AM -0500 2004-11-20, Steven Kuck wrote:
>>> Since all of these messages are, in fact, being sent by my
>>> server I think it quite reasonable to change the "Date" to
>>> reflect the time that it was processed and changed by the
It's equally reasonable to disobey "no trespassing" signs if you plan
to avoid making a mess. In both cases, however, you're using a
facility that is somebody else's property, and in both cases you may
inadvertantly cause problems for the person who has the right to use it.
Steven> Can you point me to the definitive text for mail header
Steven> definitions? RFC 733 only defines the format - it doesn't
Steven> say they are any more inviolate than the "Subject" line.
That's true. In fact, they're equally property of the author(s) (in
the wording of RFC 822 and reaffirmed by 2822), and I detest mailing
lists that insist on munging the Subject header, too. Even this
one.<wink> If an MUA can't do the right thing based on the List-*
headers, I want no part of it, not even to share a mailing list with it.
Of course, unlike "no trespassing", there are no courts to enforce the
RFCs. But don't let that fool you. They are "legislation", the
product of the skull sweat of dozens of "representatives" far more
competent and hardworking than my Congresszombie, tested in practice,
introduced and maintained through the cooperation and hard work of
hundreds or thousands of programmers, distributors, and sys admins.
The social contract says "These headers are for the use of authors",
and you (as a list admin) want to hijack the "Date" header.
You say "well, it would be very convenient if I could use these
facilities for my own purposes". I can't blame you for _wishing_;
still, an honest man would create an appropriate header _for the use
of the mailing list manager_, and then lobby and write the software to
get it accepted and diffused through the community.
Specifically, you could use either an X-List-Sequence-Number header or
an X-List-Receipt-Date header for your purpose. Teach the archivers
about them (and the versions without "X-", since of course you'll
submit an RFC, right?) for the benefit of web readers.
Re your examples:
Steven> Why do I care if the message was stuck (unless it's my
Steven> server) or if the user's clock is wrong?
You don't. You've already proved that. But authors will care, if
they are not responsible for the stuckage, and they look like idiots
because their post was written without the benefit of a week's worth
of further information, but nonetheless appears with a current date.
Steven> As I said, I can guarantee messages from the future are
Steven> wrong. Disagree?
Depends on what you mean by "wrong". Your server may very well change
the interval between, or even the order of, two posts from the same
(ie, consistently translated) future. Whether this information is
interesting or not is another question; you are, however, destroying
it, and there's nothing in the RFCs that permits you to do that.
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Ask not how you can "do" free software business;
ask what your business can "do for" free software.
More information about the Mailman-Developers