At 10:02 PM -0600 9/16/00, Alexey Melnikov wrote:
So you like the idea, but you don't like the wording (more precisely the
explanation).
Yes, I like very much being able to set IMAP flags with Sieve.
I just think the draft should specify what happens when the new
commands are used, and leave the underlying implementation alone. I
realize you were introducing the global variable concept to help make
your point, but I personally think it's better so simply say that
this command adds a flag, this command takes away an added flag, etc.
If the message is kept, the flags (if any) are also kept. If the
message is discarded, the flags have no effect.
Also, I think the descriptions of 'mark' and 'unmark' are unclear. They
should specify exactly what is to happen, and not have so many MAY do this
or that statements.
Barry originally said that draft is too tied to IMAP. mark/unmark
allows to specify
importance of the message without enforcing the way how it is implemented.
I think it makes it harder to interoperate, especially since this
affects the email client. Perhaps what we need instead is to give up
using IMAP flags for this, and use Annotations instead. Sounds like
the message/priority is needed here.
I like the idea of setting flags independently of 'fileinto' or
'keep'. I'd really much rather not have 'fileinto'/'keep' be a way of
setting flags.
As I've heard voices from both sides (from people who want to use
separate commands
or to use tagged arguments) I can't make judgment which way is preferred.
I will have both in next draft and hopefully I will hear more comments.
I think the draft should make it clear that this is an open issue: to
have separate commands, bundle with keep/fileinto, or both.