Chris,
CN> 1. making the timeouts rational.
CN> 2. correcting the authentication references in Submit now the relevant
CN> SMTP AUTH and STARTTLS RFCs are published and we have operational
experience.
Your original list was longer. But mostly none of that is relevant to
the concern I was expressing.
To repeat: You did not say what problems there were, that require these
changes. Some others have posted thoughts on the matter, but there is
still no coherent statement of why this is worth the effort.
CN> Now you seem to have some questions about the need for a submit protocol.
Apparently you missed:
DC> There are a number of things I like about the path you are going down,
Which is another way of saying that your assessment of my assessment is
dead wrong and unfounded.
CN> That question is answered in the already published submit RFC, and
CN> is a distraction from the technical discussion I have started.
And if I were to misread your statement here, the way you seem to have
misread mine, I would note that you are suggesting a) that the mere
existence of an RFC means that there is a real community need, no matter
how poor the demonstrated adoption history, and b) that we should all
just create technical changes, with no concern for why they are needed.
Now, I know you do not mean either of those things. Yet here we are,
with you stating that an effort to get a clear statement about problems
and needs is a "distraction".
Very puzzling.
Let me try to state this differently: If you want the Internet to adopt
a bunch of changes, you need to tell them why it is worth their time and
money.
d/
--
Dave Crocker <mailto:dcrocker(_at_)brandenburg(_dot_)com>
Brandenburg InternetWorking <http://www.brandenburg.com>
Sunnyvale, CA USA <tel:+1.408.246.8253>, <fax:+1.866.358.5301>