[Mailman-Developers] archive

Barry A. Warsaw barry@digicool.com
Thu, 28 Jun 2001 12:27:03 -0400

>>>>> "YN" == Yanuar Nugroho <y.nugroho@student.umist.ac.uk> writes:

    YN> How is the principle of archiving in Mailman? Is it related to
    YN> storing and retrieving data when the archives are accessed
    YN> thru the website?  Can mailman take a particular message in
    YN> the archive and stick it to new message to be distributed?

Mailman is designed so that either its internal Pipermail archiver, or
an external archiver of the site's choice, can be used.  Pipermail is
pretty braindead simple: it simply creates the indices and html-ifies
the message.  It has the advantage that it's written in Python and
easily bundled for the one-stop-shopping crowd.  It has several
disadvantages though, and many serious Mailman sites use a separate
3rd party archiver instead.

IOW, it would be the backend archiver's responsibility to implement
things like thru-the-web replies.  Mailman supports direct injection
of messages from any external source (and MM2.1 will be even better in
this regard).  I'd love to see something like this added so its On The
List, but who knows when there will be time to devote to it?