Draft:  draft-ietf-sieve-imapflags-04.txt
Reviewer: Gray, Eric [Eric.Gray@marconi.com]
Review Date: Wednesday 5/3/2006 12:56 PM CST
IETF LC Date: 5/4/2006

Summary: Almost ready - questions/nits

Questions/Comments:
==================

In section 2.1, you say that spaces are acceptable (MUST be always allowed)
in flag names.  You also say that a string containing a space-separated list
of flags (I assume flags in this context is the same as flag names) is the
same as "a string list consisting of the flags."

I am not sure what this means, but it seems to say that we MUST allow spaces
as valid both as part of a flag name and as a separator between flag names.

If this is the case, does this rely on a carefully chosen set of well-known
flag names to avoid confusion?  Or does this rely on the fact that (system)
flag names must always begin with a backward slash ('\')?  If the latter, is

a space always required as a list delimiter?
___

In section 4, in the second example, it is not clear why 

'hasflag :contains "MyVar" ["label", "forward"]' 

would evaluate true, given that "MyVar" has value "NonJunk Junk gnus-forward
$Forwarded NotJunk JunkRecorded $Junk $NotJunk". 

Does "forward" 'match' with "gnus-forward" for some reason?

Also - are all of the values, in the section 4 examples, valid flag-names?

NITs:
====

Section 3, third paragraph, last sentence: "MUST cause a runtime
error" as opposed to "MUST cause runtime error"...

Section 6, first paragraph, last line: "side effect" as opposed 
to "side affect"...