Re: [dnsext] [dane] TLSA == RRtype 52

> On Mon, Apr 16, 2012 at 03:24:22PM -0400, Olafur Gudmundsson wrote:
>> But the application in this case referenced a particular version of an
>> Internet draft:

We applied in good faith, assuming that our application was for the protocol, not a single draft. If the DNSEXT WG chairs felt that we should wait until the draft could not have any wire changes, the DNSEXT WG chairs should have told us that at the time of the application.

The registration at IANA does not name the draft; it lists the name for the DANE WG chair who applied for code point. A sensible interpretation of that is "the person who is named gets to decide what the code point means", not "go ask that person which draft he applied for". If given a choice between what one co-chair and IANA thinks, versus the other co-chair, I hope most people pick the former.

This would not be an issue if a DNSEXT co-chair (Olafur) had not exhorted people to implement using the assigned code point even though the spec is not stable. In that message, he (wisely) did not say "implement only the -18 draft, and we might screw you over later if the spec changes". Most developers care about implementing a stable spec. The spec will be stable when the IESG sends it to the RFC Editor, hopefully in a few weeks. If a developer wants, they can use the code point now; nothing stops them other than the desire to ship stable code.

--Paul Hoffman

_______________________________________________
dnsext mailing list
dnsext@ietf.org
https://www.ietf.org/mailman/listinfo/dnsext