At Mon, 11 Apr 2016 15:54:05 +0200, Shane Kerr wrote: > At 2016-04-08 11:28:12 -0300 Ray Bellis <r...@bellis.me.uk> wrote: > > > May I please remind the WG of draft-bellis-dnsext-multi-qtypes-01 > > I note that your idea was about 3 years ago. When it was mentioned, > Alfred Hönes noted his ideas about his presented 3 years before that. > My guess is that we could probably go back and every 3 or 4 years find > a similar proposal. :)
Going back at least to the mid '90s, yes. Don't recall whether this came up in the '80s. :) As I recall, the thing that stopped this every time was lack of consensus on pesky details such as "to which QNAME does the RCODE apply when this fails" and "to which QNAME does the AA bit apply?" It's possible that DNSSEC-aware stub resolvers would provide some leverage here, since fields like RCODE and the AA bit are somewhat redundant if one can just check the freaking signatures. _______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
