On Thu, 7 Jan 2010, ned+ietf-smtp(_at_)mrochek(_dot_)com wrote:
I agree with Cyrus. Even if the market has reached a workable compromise
for the time being by using well-known host names, it has done so at the
cost of flexibility which SRV records could provide. I think it would be
very useful to document the naming conventions that will allow current
clients to find the servers, but I don't believe that's a reason to *not*
document a more robust and formal SRV record protocol as well in the hope
that clients and configurations will move to that.
Exactly right IMO.
Thanks for all your arguments. I have been convinced. Perhaps I should be
less negative about the likelihood of deploying improvements...
I also fear that an effort to do something more elaborate with XML and web
servers and so on carries with it a very high liklihood of failure, and
even if
we manage to agree on something (probably at some distant point in the
future),
it will then fail to deploy.
Actually I'm less negative about this because it has clearer benefits.
My concern arises not from a lack of benefit - the benefits here are pretty
clear - but from the fact that when you go this far you're necessarily engaging
in a schema design exercise.
The past history of schema design work in the IETF has not been good. Partial
or even total failures are pretty common, and even when such efforts succeed
they have tended to take far more time than anyone expected them to.
For
example it allows an operator to override an MUA's built-in preferences.
The big question is whether we can get the various MUA vendors to
co-operate instead of continuing to do their own thing.
Yes, that's also an issue.
Ned