[wg-c] RE: Give me a plan.

["Roeland M.J. Meyer" <rmeyer@mhsc.com>]

> Behalf Of Einar Stefferud: Sunday, July 30, 2000 2:36 AM
>
> Hi Roeland --

Hello Stef,

> "Roeland M.J. Meyer" wrote:
> >
> <---SNIP->
> >
> > Yes, this is all true. But, my paper never got into voting
> > issues. The assumption is that meta processes would be
emplaced
> > that would protect the minority until it came of age. The
second
> > and third year would definitely see some changes. If the
process
> > were to be frozen for the first two years then the slght
> > numerical advantage NSI had should be effectively
neutralized.
> > Note also, the paper purposely stayed far away from political
> > forms. It is strictly an operational submission, a how-to.
>
> I think it is also time to consider that it should be possible
for TLD
> holders to be acceptable even though they might not populate it
for
> public use.

Robin has already stated disagreement with this concept, in
private mail. However, I think that was a reflexive disagreement,
on limited principles, and not fully thought out. I didn't chose
to argue it at the time. Due, in large part, to lack of time.

> Their cannot be anything wrong with someone signing up for a
TLD and
> making it totally private, but also reserve the right to make
> it public
> at some future time, which means they must have the right to
not make
> the SLDs public according to any schedule.

Indeed, a charter for a TLD could (and in my case, does) include
restrictions on who could access the resolver and what they
resolve. Currently, VPN registrations are done manually. FTN
registrations are automated from a nodelist, but resolution is
restricted to defined systems. The former for security, the
latter to avoid protocol interaction problems (no internal
Internet technology). Also for the latter, as FTN gateway
technology is developed it may be opened up to the public.

> I note here that there is now no need for anyone to populate
their SLD
> with public names below their SLD;-)...

Also a case with MHSC.NET right now. I have entire sub-domains
(3LDs) that are not visible to the general Internet. Also, my
name servers do not allow zone transfers, except to selected
hosts, in my ACLs. Also note that my off-site secondaries
(blacktop) do not resolve everything that NS[1-3].MHSC.NET do (I
love BIND8 <grin>).

> I can see a very large demand for TLD holders to not want to
> display in
> public all their SLDs, though we might require that they at
> least occupy
> their TLD space with at least one public visible site, even if
it only
> says "Keep OUT --- Private Property".

We, as a group, have never gone down to this level of detail
because of problems/arguments in the meta-layer. But, I have long
maintained that some sort of naming convention needs to be
emplaced. Just what that convention may be is almost irrelevent,
as long as it exists and is consistent. It is the root-registry's
job to create and maintain that convention, IMHO. In my WG-C
submission, I allude to this with the requirement of a minimal
auditable service level. That SLA should spell out the nameing
convention for the minimum SLA that qualifies a TLD.

> I say his in reply to Roeland because I think he might be in
this box,
> so he just might agree with me on this.  And, be able to
> further explain the point;-)...

Yes, I am in this box. I hope that I have explained adequately. I
am actually a little impatient, like Richard, and want to get on
down to operational issues. My WG-C submission as an attempt to
get past the meta-layer bickering and on down to the how-to,
which was the original point to ICANN/DNSO/WG-C, IMHO. That some
chose to rehash five-year-old issues instead was something that I
could not avoid <sigh>, nor could Jonathan. Given the
circumstances, I think Jonathan did an adequate chair job. But, I
don't think that WG-C, as a group, did the job it was chartered
for, unless you take my paper into account (for the record:
http://www.dnso.net/mhsc-tld.htm). My paper was not a group
effort, it was an MHSC effort. IMHO, many on the WG-C chose to
bicker at the meta-layer because there is more wiggle-room there.
At the operational how-to level, it either works (makes sense) or
it doesn't. Also, many didn't want to profess their ignorance of
actual operational practices (had no clue of what made sense).
Many of the WG-C wouldn't know a root-level login if it bit them
on the nose. They're not even sudo-ers.

NOTE: I hope no one minds that I CC WG-C on this.