On Mon, 06 Sep 2004 01:55:50 -0500, wayne <wayne(_at_)midwestcs(_dot_)com>
wrote:
In <F10C6AA6-FDF0-11D8-B6CF-000A95B3BA44(_at_)hxr(_dot_)us> Andrew Newton
<andy(_at_)hxr(_dot_)us> writes:
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.
...
So, if we are planning on supporting SenderID, it appears that we MUST
allow everyone to always be conformant at all times. This means that
we MUST NOT require implementations to query SPF RRs.
The standard can be written in such a way that senders MUST publish
SPF records and MAY publish TXT records, while receivers MUST check
SPF records and MAY check TXT records.
Only the receiving MTA is covered by the patent license. Compliant
implementations will have to check SPF records. If the recursive DNS
server an implementation talks to is broken, that probably doesn't
mean the MTA implementation is non-conformant.
But it would be kind of amusing if Microsoft sued everyone with a
broken DNS server. (I know they are always looking for new ways to
get people to upgrade...)
David