On Thu, 2004-11-11 at 12:49, jpinkerton wrote:
It's an extremely high-risk strategy, as we know, and not one I would have
recommended, but it's where we are so let's deal with it.
In the good case, where we don't get induated with help requests from people
with existing records who are having problems created by PRA, then we can
relax and say that Meng made a good call. The problem is going to be moving
forward from that point. Do we assume that PRA will die, and continue with
the v=spf1 records as they have been, or do we tell everyone to add "
spf2.0/pra ?all " in retrospect? What happens then when PRA fails and MS
come up with a new idea - "son of PRA" = SOP and we have to add a record "
spf2.0/SOP ?all " to avoid it's machinations?
The "what about future expansion" argument _further_ supports the
position that requiring opt-out in order to not participate is a bad
idea.
This strategy looks to me to be very short sighted, which is why I was
promoting the idea of the optional modifier to (dis)allow *any* protocols
which might be threatening to mangle v=spf1.
I vote for the optional modifier to only allow, not disallow. A
modifier that disallows moves the opt-out from another TXT record into
the modifier -- still an opt-out.
Andy