Scott Kitterman wrote:
I think it was also suggested (and the reason the WG chose not
to pursue it) that it was outside the scope of the WG charter.
It's hard to define what "sends no mail" actually means. With
SPF you can cover "neither as HELO nor in MAIL FROM". With PRA
you can say "not as PRA". A draft trying something better could
still miss say Reply-To (no originator header field at all).
At the end we might arrive at "has no MX and either no SMTP at
its address, or no address at all" as definition. And if folks
really want this they didn't show up on the SMTP list to fight
for a "mandatory MX at least for IPv6" in 2821bis to support
Doug's proposal in this direction here. So far 2821bis didn't
adopt it, and we're in Last Call.
None of those problems apply to a 'sends no mail' SPF record.
+1, a "v=spf1 -all" is simple. However it doesn't cover a PRA
in my 4408 universe, no matter what 4406 said, or what the IESG
said about 4406. Adding another SPF record "spf2.0/pra -all"
is no rocket science, but maybe Bill wants "really nowhere, even
in non-PRA header fields", and then he IMO needs "neither MX nor
IP" to determine it, not SSP or DKIM.
I'm not sure how many ways we need to say that.
Receivers not otherwise directly interested in SPF or PRA would
miss it, but they could use a service aggregating this source
(among others), e.g. in the form of a "no mail" black list.
In other words I also don't see how adding SSP as another source
could help Bill, unless he intends to work on SIQ as aggregator.
Frank
_______________________________________________
NOTE WELL: This list operates according to
http://mipassoc.org/dkim/ietf-list-rules.html