Here are the things I still believe are unresolved.
1. Request: Implementations are required to decode header charsets.
(Chris and Harald have probably resolved this, but I need to figure
out wording.)
2. Request: Reject should be optional.
(I just want to know that this is consensus.)
3. Request: forward adds Resent-* headers.
(I just want to know that this is consensus.)
4. Request: Short-circuit evaluation should be either MAY or MUST, and
must be discussed in case it matters in the future.
(I have heard contradictory opinions on this. Someone (I forget
who) wanted it removed since it doesn't yet matter. Chris (and
probably others) want it in there because it will probably
eventually matter. I propose that it has to be discussed, is a MAY
for implementations, and a MUST if extensions offer tests with side
effects.)
I have a list of things I expect to change in -05 at
<http://www.contrib.andrew.cmu.edu/~tjs/sieve-changes.html>, which is
the list I posted last week with all the changes I know about.
Thanks!
Tim
--
Tim Showalter <tjs+(_at_)andrew(_dot_)cmu(_dot_)edu>