Having not (yet?) gotten permission to make it a working-group
document, I've updated vacation-time with the individual-submission
name, as noted earlier:
OK, I'll buy that. Good. So for the next version:
1. I will change ":time" to ":seconds".
2. I will change "vacation-time" to "vacation-seconds".
3. I will add text saying that "vacation-seconds" implies "vacation",
and that a script that includes "vacation-time" in the "require" list
MAY omit "vacation" from the list.
The new version is here:
http://www.ietf.org/id/draft-george-sieve-vacation-time-01.txt
Overall this is looking good. I only see a couple of issues with it, both
having to do with compliance language.
The first paragraph of section 2 says, "Only one of ":days" and ":seconds" is
allowed -- both MAY NOT be specified (it is a compile-time error to do so)."
"MAY NOT" is not legitimate compliance language, nor does it make any sense as
such. Obviously a script can specify both parameters; but doing so is supposed
to generate an error. To the extent there's a compliance issue here, it's what
the Sieve processor has to do.
It would also be inapprorpiate to require this to be handled as an error at
compile time - use of ihave precludes that, for one thing.
I therefore suggest changing this to:
Only one of ":days" and ":seconds" is allowed -- use of both parameters in
the same vacation action MUST result in a Sieve processing error.
Next, the capitalized "MAY" in the sentence
Note that "vacation-seconds" implies "vacation", and a script with
"vacation-seconds" in a "require" list MAY omit "vacation" from that list.
is also inappropriate, since it isn't talking about what a conforming Sieve
implementation can do, but rather about the allowed semantics of Sieve scripts.
It should simply be changed to lower case.
Other than these two issues, IMO this looks good to go.
Ned
_______________________________________________
sieve mailing list
sieve(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/sieve