>>>>> "MH" == Manar Hussain <email@example.com> writes:
MH> Is that us ?
I believe so.
MH> My last mail was that we'd discussed things here and were very
MH> interested in making some contributions - in particular seeing if the
MH> new "archive format" for mj and our "message db" format could converge
MH> as a starting point (our current system is somewhat different from what
MH> Jason is proposing).
In what way is what you're doing different? The only real requirement that
I have is that the backend file be in the standard mbox format and that it
be a fast operation to extract a message from the archive. The index can
be anything, but I can't see how you could get away with much less than I
proposed and I don't think that the actual format of the index is terribly
important as long as the relevant information is there and there's a way to
associate a line number in the archive with the message number it contains.
The really big part is the web presentation stuff. I don't see it as being
closely coupled to the backend (or at least it shouldn't be) as long as the
backend can provide individual messages and enough quickly accessible info
to build the indices in a reasonable time.
MH> I left it at a "please let us know if there is interest" with an
MH> overview of what we are doing and access to code once we'd done a bit
MH> more documentation following should that be the case - but I didn't
MH> hear back from anyone so we've just continued on our merry way
MH> developing it for our own use.
Sorry, I've been away and busy even when I'm here. I'm going to start work
on this today because I need the basic archive functionality (add a
message, extract a message by number) in order to make digests work.
Why don't you key me/us in on your plan and how it differs from what I've