[Mailman-Developers] in-reply-to problem

[Brad Knowles]

At 3:22 PM +0900 2004-10-04, Stephen J. Turnbull wrote:

>>>>>>  "Brad" == Brad Knowles <brad at stop.mail-abuse.org> writes:
>
>      Brad> 	You could spend a trillion dollars on this problem and
>      Brad> not find a solution.  The entire computer industry has been
>      Brad> working on AI since the early 1960s, and they haven't solve
>      Brad> it yet, and don't appear to be much closer today than they
>      Brad> were back then.
>
>  That's a bit pessimistic, don't you think?  What can be done without
>  I-R-T or References doesn't require AI.  Plenty of MUAs solve the
>  problem of threading messages with no I-R-T header satisfactorily (ie,
>  messages that are in the same thread almost always appear within a few
>  lines of each other in the summary).

	I don't think I'm being pessimistic here.  I think I'm being 
realistic.  There are no deterministic ways to figure out precisely 
which message is being replied to and which other messages are being 
referenced, unless you can guarantee that you can perfectly parse all 
possible quoting methods, and you can guarantee that you will always 
have enough information available in the quote to guarantee that you 
can determine which message is which.

	Unfortunately, in the real world, you can't guarantee any of those things.

>  On the other hand, an archiver (such as Pipermail, distributed with
>  Mailman, or MHonArc, a popular third-party archiver) does (typically)
>  have that information, at least in "rebuild-the-whole-shebang" mode,
>  and I would say that you should look at archivers, not at Mailman
>  proper.  Also look in the mailman FAQ for how to use a 3rd-party
>  archiver with Mailman; you can probably find one that does a better
>  job than Pipermail does.  MHonArc would be a good bet.

	In terms of re-creating threading data that was destroyed by an 
the MUA?  No, I don't think that you're going to find much here.

>  http://www.jwz.org/doc/threading.html describes a threading algorithm
>  that is fairly robust to this problem in my limited experience with it

	Right, and as complex as this algorithm is, it still requires 
"In-Reply-To:" and "References:" headers in order to be able to do 
it's job.

>  That's as far as I'm willing to go though.  Really, you should get
>  your users to switch to an MUA that can handle headers that were
>  standardized in 1975 or so.  And there are plenty of MUAs that can do
>  something sane about pseudo-threading messages without I-R-T.

	Now there's a statement that I can agree with.

-- 
Brad Knowles, <brad at stop.mail-abuse.org>

"Those who would give up essential Liberty, to purchase a little
temporary Safety, deserve neither Liberty nor Safety."

     -- Benjamin Franklin (1706-1790), reply of the Pennsylvania
     Assembly to the Governor, November 11, 1755

   SAGE member since 1995.  See <http://www.sage.org/> for more info.