[gnso-rds-pdp-wg] Some reg'n data I think necessary (was Re: GNSO Next-Gen RDS PDP Working Group teleconference)

[Andrew Sullivan]

On Tue, Mar 22, 2016 at 09:14:49AM -0400, James Galvin wrote:
> 
> While it is true that a good portion of the work that is conducted by the
> ICANN community is around domain names that are “on the Internet”, the fact
> is there is no requirement that a domain name be operational in any way.

Aha.  I think we're disagreeing about a term, not in substance.

When I say that these are domain names "on the Internet", that
includes the cases where you don't turn them on.  They're still only
domain names with respect to the Internet, as opposed to something
else (like, for instance, trademarks or baby names, which get
registered elsewhere).

To put this another way, the act of registering a name in a domain
name registry creates an association between the name and the
registrant with respect to the Internet, but _not_ necessarily with
respect to other systems.  By registering anvilwalrusden.com, even if
I do not add any name servers to that registration, I get the right to
use (or not use, as I see fit) that name on the Internet.  It does not
mean that my (non-existent) baby is named "anvilwalrusden.com", or
that my future automotive and bait shop business sells anvils and
walrusses, or anything like that.

And this is quite proper, because ICANN's mission is about things on
the Internet and not about other stuff.  It can pursue activities
outside the Internet context only in so far as those activities are in
support of the core mission.  This WG shouldn't investigate outside
that boundary because if we do, ICANN won't be able to implement the
recommendations: they'd be outside the ICANN mission.

> It is a fair question for this group to ask whether that is the degenerative
> case or that is the baseline case.

Well, I think it's obviously a second case as a matter of history, but
I think it doesn't matter for our purposes, and would probably just be
a distraction.

> The operational data you describe is not necessary for domain names that are
> not operational.  The operation of a domain name could become one of many
> use cases to be considered separately.

Every domain name, by definition, contains the potential to become
operational on the Internet.  For instance, the data that I listed
included things like name servers; obviously, if the name does not
actually have authoritative servers then they can't be provided.  That
lack of name servers is also operational information, however.  If the
RDS for a name has no authoritative name servers and yet I can
retrieve an answer from the DNS, then either there is an illegitimate
spoof going on or else the name servers in question are lame.

> Today there is a tendency to require the operational data, even for names
> that are not operational.

Yes, but today we also have the problem that any class of data that is
published for anyone must automatically be published for everyone
because the protocol is stupid.  We have available to us a smarter
protocol, and so the question of what data is to be collected (in
absolute terms) can be unhooked from the question of what data is to
be available and under what circumstances.  We must continue to keep
that distinction in mind.

Best regards,

A

-- 
Andrew Sullivan
ajs at anvilwalrusden.com