Cyrus, for us at least, the issue is mostly moot because we have little
interest in either of these extensions. (And I speak as someone who has
implemented the vast majority of Sieve extensions so far.)
In the case of convert, we've been providing on-the-fly conversion capabilities
in our software ever since the early 90s. We've gone through multiple
iterations of this idea, including system-level facilities, user-level
facilities, and even schemes that allowed for hierarchical control of
what's done.
To be blunt, essentally none of it has proved to be worth the time spent coding
it. Yes, there has been occasional usage, but the biggest use by far has been
to do stuff the facilities weren't really intended for, like fixing up various
sorts of brokenness - something that could have been accomodated better by a
less general and more targetted mechanism. About the only conversion that has
ever seen significant use is charset conversion, and that's invariably best
done as a system-level thing since users rarely understand how to configure it.
(And at this point it's all built in anyhow; no need for Sieve to control it.)
The problem appears to be fundamental - people simply do not know how to codify
their conversion needs in advance.
I will also point out that complex per-user conversions at the MTA level have
some serious performance implications in a high performance environment such as
ours. I have always regarded this extension as mostly intended for MUA-side
actions where such costs are essentially irrelevant.
As for SIP notifications, it's not that we don't use notifications - we use
them extensively - but other sorts seem to have us covered.
Now, if a specific customer application arose for SIP notifications, the fact
that we are aware of patents in the aware would immediately result in the
corpordate legallie-beagallies getting pulled in. I honestly haven't a clue
what the response from them would be.
Ned
_______________________________________________
sieve mailing list
sieve(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/sieve