Re: WGLC results: EDNS0

[Thomas Narten <narten@raleigh.ibm.com>]

> I think the group is "tired" because it is required to make a decission
> that is irrelevant to the working of the draft. After all, I think that
> placing the burden on the working groups that propose new items to be
> managed by IANA complicates matters: Why should IANA need to keep track
> which policy to use when allocating a new code, simply because different
> WGs happened to reach different decissions when other resolutions would
> have worked equally well?

The reason the WGs need to be involved in setting the IANA assignment
policy is that only those folks who understand the implications of
assigning new numbers are in a position to make "good" decisions. If
the IANA just started handing out (say) new RR code points to any
random user, wouldn't it be advisable to have someone with some DNS
clue actually review the request to make sure it wasn't totally bogus?

> Anyhow, we will have to decide the one or other way in order to get the
> draft advanced. But I strongly suggest to urge IESG and/or IANA to publish
> an RFC that will lay down this political (not technical!) decission for all
> items to be assigned by IANA.

Please take a look at RFC 2434 (BCP 26). Writing an IANA allocation
policy does not have to be very hard to do. If you don't want to think
very much what the policy should be, yet be sure that numbers aren't
assigned badly, simply use the "IETF Consensus" or "Standards Track"
allocation policy defined in 2434. Those policies tell the IANA not to
assign any numbers unless there is an RFC. Since all RFCs are reviewed
by the IESG before they become RFCs, all DNS-related assignments will
then likely be sent to the DNSING WG for review before any assignment
is actually made.

Thomas