John.Maloney at disney.com is widely believed to have written:

> Thus, I think a better fix for the slow startup time for large mail
> databases would be to switch to a binary format for the .index file
> that can be read much faster than the current text-based format.
> As I recall, parsing the large integers used to represent timestamps
> and message ID's is very slow.
Don't forget to look into using image segments. As an example, we
changed to Imsegs for loading >2.5Mbform & string resources and went
from ~25sec to ~2secs load time. If each index were saved in a suitable
segment file it might make it very practical.

I'm a moderate email packrat but not completely obsessive; I have some
120Mb of email for the last fifteen years. I'd certainly need a system
that could cope with that.

One other facility I'd like would be something I can only call
name-echoing. I get mail from a variety of sources and dump them all
into one mailbox. If I reply to a mail sent to tim at sumeru, I want the
reply to be _from_ tim at sumeru; and if the mail were sent to
hunkaburninlove at eros.net replies should appear to be from there. Amongst
other things this would be helpful at work where people play a variety
of roles; email to 'vmboss' might get passed to fred but when he replies
it is helpful if the reply comes from vmboss.

tim
-- 
Tim Rowledge, tim at sumeru.stanford.edu, http://sumeru.stanford.edu/tim
Useful random insult:- So dumb, he faxes face up.