Hi.
In order to make the posting deadline and, with luck, move this
forward, draft-klensin-rfc2821bis-08 is now in the posting queue
(I'm awaiting the confirmation message).
Please see the change log for a summary of what has been done.
Note that the problematic "such response" sentence has been
removed per comments yesterday and today.
After a discussion with Tony, we've tentatively decided to take
a hard line toward post-Last-Call suggestions for small
improvements. The "Last" in "Last Call" needs to mean
something; if we keep iterating every time someone has a new
idea about an issue that might reasonably be addressed we will
never get this long-overdue project finished. My own position
at this point is agnostic about the merits of more text for RSET
-- had the issue come up several months ago, I would have wanted
to participate in the discussion and possibly include some text.
But, at this point, absent evidence that the spec is actually
broken, I think it is more important to get finished.
It is not an accident that HT is still prohibited in addresses.
The change is easily made in an RFC Editor note should we decide
to do so and Tony will address that issue. In the meantime,
please don't re-post the same comments on the assumption that
you are being ignored. My one observation on the subject is
that both 821 and 2821 permitted _any_ ASCII character if they
were escaped. Neither HT nor SP were ever specifically
permitted; they were just part of "any". We needed to permit SP
because there is a known (although old) use case for it. As
far as HT is concerned, I recommend avoiding the discussion of
whether it is a nice white space character (since there are no
white-space matching rules in 2821bis) and instead focusing on
the question of whether it is actually helpful in email
addresses.
Tony will generate a new Last Call to cover the changes.
john