On Sat, 2003-09-27 at 09:59, Kevin McCann wrote:
Try it for yourself. Maybe your needs are different. But if I want an archiver that will process huge mbox file from other systems - without crapping out, will decode encoded binary attachments, have flexible, customizable output, etc., I can't think of anything else I would rather use. If MHonarc is so bad, why do so many people go the extra lengths currently required in order to use it instead of pipermail?
As I say, it works for me. And very well. Maybe worth a second look? If you try it again let me know what you find.
Pipermail definitely has some distinct advantages, so I'm glad we include it. It has some problems too, most notably its memory consumption, and I think inefficient on-disk storage. The most often cited problem is that it cannot handle regenerating entire very-large archives because of memory usage.
I really really want to use something like message-ids to generate message file names. I want to be able to generate links to archived messages in the footers, but I think the best way to do that is to agree on a reproducible, independent algorithm for calculating them. Another approach would be to put even the public archives behind a cgi and have that implement a mapping between message-id derived links and the sequential file names (although that won't fix the regen problem).
It would also be nice if we could more easily customize its look and feel. Anything more and we're getting into arch-ng and zest territories.
-Barry