ietf-mxcomp
[Top] [All Lists]

Re: consensus call on MUST/SHOULD language for TXT records

2004-09-03 16:14:26

mazieres(_at_)gmail(_dot_)com:

On Fri, 3 Sep 2004 17:33:48 -0400, Andrew Newton <andy(_at_)hxr(_dot_)us> 
wrote:

Despite lack of agreement on the outcome, there is agreement on the
stated positions.  In the words of Olafur:
1) Specify everyone to be compliant at all times.
2) Specify clearly full complaint state and tolerate non-compliant
state during phase-in.

I would favor position 2.  I think the recommended document should
describe where we want to be in the long run, whether or not we can
get there quickly.  While I strongly suspect Sender ID will undergo
future revisions anyway, why intentionally adopt a standard that will
have to be superseded?

I second this. The document should bless the usage of TXT records
during the initial deployment phase. It does not need to define
the time schedule for the txt -> spf transition.

        Wietse