At 08:58 04/02/05 +0000, John Berthels wrote:
> Anyway, take a look, and feel free to fill in any gaps you see.
Since the scope seems to encompass some retrieval aspects (the
"downloading of N megabytes" performance requirement) do we also want to
have a stab at the "mailbox problem" as well as the "transport problem"?
Or is that biting off way too much (as if we weren't already).
Various kinds of software does various kinds of things to email
headers not only in the transport phase, but also in the 'mailbox'
phase. One of the reasons in my opinion for the current header mess
is that such issues haven't been thought through. So for the moment,
I would assume that everything is connected, and that we may want
to split things up later.
Please feel free to rule this out of scope (and sorry if this takes this
off at a tangent), but some requirements in that area might be:
- Users want to be able to keep their mail organised into folders
Very good point, but I would remove the word 'folder' from all the
following items to make them more independent.
Regards, Martin.
- Users want access to their mail folders from anywhere on the internet
- Users don't want anyone else to access their mail folders
- Users want to be able to search their mail folders
- Users want to be able to automatically file mail into incoming folders
- Users want to share some of their mail folders with other users
[which looked at from the right angle makes mail transport redundant...]
regards,
jb