Re: LYNX-DEV Re: reorganizing the lynx-archive

[Al Gilman]

  From: Andrew Kuchling <address@hidden>
  
  > QUESTION: "TRN" turns each thread, somehow, into a branching
  > tree.  On what basis does it decide on starting a new
  > branch (sub-thread).  Is it just the subject line?
  
        I'm writing a Hypermail-like tool (Pipermail), and have had to
  implement threading.  Most mailers and newsreaders include an
  In-Reply-To: line in the header of a message, giving the message ID of
  the message being replied to.  If that isn't present, one can next
  look at the References: line, which lists the preceding articles in
  the thread.  If neither "In-Reply-To" or "References" are there, the
  archiver just has to guess and look for an earlier article with the
  same subject.
  
        I don't believe structuring our e-mails is any solution; the
  time would be better spend actually maintaing a FAQ, improving the
  archive, or improving the tools used to search the archive.
  
Andrew:

How near is Pipermail to ready for pilot use?  Could an installed
copy be turned on to be either 1) the primary archive if it's
stable enough 2) the site where we experiment with smart linking
if it ain't??

Everybody:

My workflow model has been that one has topical editors.  Each
editor reads the full stream and has post privileges to topical
archives.  The trick is to get cross-referencing connected across
the "topical archives."  One way to do that is via email aliases
whose incoming streams are marked up with metadata that the
archiver understands, and then combined into one stream fed into
the archive.  You can get the editors to follow structured rules
like using authorized keywords reasonably consistently in a way
you can't get the general subscribership to.  But you want the
editor delay in the archiving phase, not prior to broadcast of the
original raw post.

Al Gilman

;
; To UNSUBSCRIBE:  Send a mail message to address@hidden
;                  with "unsubscribe lynx-dev" (without the
;                  quotation marks) on a line by itself.
;