[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

DEPLOY: DNS Record Types

On Aug 24, 2004, at 1:25 AM, Mark Lentczner wrote:

On Aug 23, 2004, at 5:09 PM, Jim Lyon wrote:

I believe that the latest Protocol draft (draft-ietf-marid-protocol-03)
contains changes that don't correctly reflect the consensus of the group
concerning DNS record types.

The changes you cited were all made in response to review by the people of DNS-EXT. In particular, Olafur Gudmundsson, co-chair of DNS-EXT provided much of the wording you see verbatim. He and I discussed the changes and in particular the very points you bring up. He said he'd like to see this version of the wording go to last call, and that he'd be willing to be involved in discussions.

My comment about the current wording is that either senders or receivers who do not implement the "MAY" clauses are acting against their own best interests. This is misleading at best. As a sender, I am going to publish all record types that receivers might look up as soon as I can. As a receiver who is interested in discarding as much forgery as possible as quickly as possible, I'm going to look up whatever records I can look up as quickly as I can.

As a practical matter, I'll be publishing and checking TXT records very shortly, and the SPF2 record type at some point in the future. As the spec is written, a new reader has to study carefully and work this out for themselves. This is not helpful to the email community and it will slow down adoption.

I am strongly in favor of Sender ID, and would like to see it deployed as quickly as possible.