John Levine wrote:
If there were a son-of-DNSBL that published more complex information,
e.g., if we figure out reputation well enough to understand what a
generally useful reputation record containing more than one bit would
contain, I'd be less opposed to a new RRTYPE since both producers and
consumers would need new code anyway to support it. But inventing a
new record that is just like an A record only different is silly.
Yeah, I think I said that in that argument ;-)
Furthermore, if we were to do something more featureful than DNSBLs,
it'd probably not be DNS anyway, so a new RRTYPE is moot.
Re your other point, if it seems likely that the BCP-ish doc can come
out at the same time as the spec, I would of course adjust the reference
to it from the current vague language to be more specific.
What's preventing the BCP-ish document coming out at the same time as
the spec?
I've been handed a few more typographical nitpicks, still nothing
substantive.
_______________________________________________
Asrg mailing list
Asrg(_at_)irtf(_dot_)org
https://www.irtf.org/mailman/listinfo/asrg