Re: [dnsext] RRTYPE request: template for ZS record

[Ted Hardie <hardie@qualcomm.com>]

At 12:44 PM -0800 11/21/08, Jim Reid wrote:
>On Nov 21, 2008, at 19:46, Ted Hardie wrote:
>
>> I don't see how the proposed ZS RRTYPE solves this problem.  It seems
>> like it has exactly the same problem as TXT does:  it will be 
>> completely
>> unstructured data.  It also seems likely that different zones might
>> diverge on how they present zone status information.
>
>Well, I suppose the I-D could say something like "don't put more than 
>one ZS RR at any node/zone". An unstructured (human readable) text 
>string is all that's needed here: anything fancier is overkill IMO. A 
>TXT record can't easily provide that info because these are already 
>used for too many ad-hoc purposes.

Imagine for a moment we're dealing with four zones.  One is
subdomain.delegateddomain.example; one is .example;
one is in-addr.arpa; and one is nic.edu.cctld.  Do you expect
zone status from each one of them to share any common
vocabulary at all?  Is there any common set of tokens or
understandings about what semantics this carries?  In nic.edu.cctld,
if the zone maintainer put "Gleep" and in subdomain.delegateddommain.example
the zone maintainer put "Gleep" do you expect them to mean the same
thing?

I think I would understand this much better if ZS had a well-described
set of potential statuses, even if these are not standardized. 
"Bob's looking at it" and similarly ad-hoc descriptions seem really unlikely
to be helpful outside of the tiny set of people who know who Bob is.
But I can't tell if that sort of thing is in the universe of what you consider
reasonable things to put here.
				regards,
					Ted

--
to unsubscribe send a message to namedroppers-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/namedroppers/>