On Oct 04 2004, Bruce Lilly wrote:
If the mailing list doesn't keep an archive of itself, then there's
little hope of making it intelligent at all.
There's a big difference between keeping an archive (write once)
and grubbing through an entire archive when each new message is
submitted to the list.
I'm not suggesting rescanning the entire archive each time a new message
is submitted (accepted), an up to date index is all that's needed.
But it does require careful programming to ensure scalability, I
completely agree, and if someone tries this it won't be done with
ten lines of perl.
Also, it doesn't suffer from the problem of recognizing variations in
Re:, Re[2] etc. to identify replies (which is what is problematic now).
Presence of "Re:" does not identify a reply.
Subject: Re: the note between Do and Mi
Subject: Re: Element number 75
etc. do not indicate replies. The only reliable indication
of a response is the presence of an (optional) In-Reply-To
field (if the original had an optional Message-ID field).
I suspect I got confused. What is your position on threading in
mailing lists? Are you suggesting that:
1) In-Reply-To is the only reliable existing means of discovering
relations between messages?
2) In-Reply-To is not honoured by MUAs with sufficient frequency in practice?
3) Therefore, threading is not worth considering for mailing lists?
and separately, for example:
4) embedding a thread ID token into subject lines of list propagated messages
cannot mitigate the shortcomings of 1-3?
That's a security issue, which exists independently of the
issue of accomodating list member preferences. What's the worst
that can happen now on mailing lists, and would it get significantly
worse with an "intelligent" list server?
Most lists do not depend on accurately identifying an author
at submission time. You seem to be proposing that some automatic
action is taken by a server based on unreliable information.
That's generally a bad idea. While digitally signing messages
could help, use of signed messages on mailing lists is quite
rare.
How do lists handle posting privileges now? For example, what does
ietf-822(_at_)imc(_dot_)org say do about accurately identifying me when I post
to
the list? I don't sign my messages digitally.
You're right. If such a person is not a member, then his/her preferences
can't be known.
They can be known if and only if he as a message author indicates
where he wants responses to be sent, in his message (via Reply-To).
What about his receipt preferences for messages sent by anybody else
on the list?
Ok, it's a good question, but I think that's bordering on recursively
reimplementing the mailing list. My personal inclination is that a
response to a response is not a response to the original poster, so
needn't concern him or her.
Hypothetical situation:
A asks a question
B gives a wrong answer, which is copied to A
C, D, E, and F respond that B's answer was incorrect, and giving
the correct answer -- none of which is seen by A.
I don't see the problem with that. When did the matter of correctness
of an answer enter the debate? I see the point of what you're saying,
but I don't see that it is a list server's job to inspect the message
content for accuracy.
--
Laird Breyer.