----- Original Message -----
From: "Chris Haynes" <chris(_at_)harvington(_dot_)org(_dot_)uk>
<big snip>
I don't like the idea of an open-ended invitation to use the record for
as-yet
unidentified purposes, unless the publisher _really_ intends that..
I hereby modify my proposal to the use of a modifier "op", as you suggest,
but
it must give specific authority for the use of the record with each
specific
protocol in a comma-separated list, e.g.
"op=pra,smm"
Perfect solution imho :-)
So now the PRA test may only be applied if the "pra" identifier is among
the
values in the "op" list.
The permitted values for the "op" modifier list contents are to be
recorded in
experimental I-Ds.
Note that the use of SFP records in SPF tests is _always_ permitted.
If any record publisher wants to be really helpful to whatever experiments
may
arise in the future, we can define a pseudo-protocol "any", so that
publishing
"op=any" means that the publisher is inviting unrestricted use of her SPF
records (and is prepared for the consequences).
Yes - I'd agree with this idea too.
This whole idea is an elegant way through to a political minefield. It
means we do *not* have to mention PRA or MS in the spf spec, and it also
means
that we allow for future developments - including, possibly, some of our own
;-)
Slainte,
JohnP.
johnp(_at_)idimo(_dot_)com
ICQ 313355492