On Mon, 2004-11-01 at 11:39, Richard Bang wrote:
I don't have to point out where it breaks, you already said that the changes
are MOSTLY BACKWARD COMPATIBLE.
I vote that if they are not COMPLETELY BACKWARD COMPATABLE then a new
version is required.
Please remember that the "let's not do anything" choice in itself
doesn't maintain backwards compatibility.
1. MS is currently trying to retroactively cause spf1 records to be
interpreted in a way that is not backward compatible with the
currently published records.
2. The current experimental draft is not compatible with the de-facto
standard.
3. My proposal only introduces backwards compatibility issues in the
most broad sense: It adds the ability to publish records that would
currently be ignored by current parsers, though recommends against
doing that for scopes that include mailfrom.
It does nothing to alter the meaning of the million published
records.
Not only are the compatibility problems my proposal introduces
practically nonexistent, they also have the potential to reduce the
compatibility problems caused by (1) above.
Please describe any actual, practical problems my proposal could cause,
and how they're worse than what already exists with MS's proposed
interpretation of spf1 records in pra scope.
--
Mark Shewmaker
mark(_at_)primefactor(_dot_)com