Re: [ga] ccTLD dialogue with Joe Sims (wasRe: your comments

["todd glassey" <todd.glassey@worldnet.att.net>]

Michael - this ccTLD nightmare may actually be an outgrowth of the way in
which DNS was architected, and to a large extent may be based in that DNS
services are not like telephone numbering. Also there is the issue that the
DNS of today (BIND 9.x) is really still only designed for a single root
system through a single registrar so global scaling and the need to offer
billions of name possibilities were never considered. Especially when some
of these names would need to be "identical" to others. Likewise the BIND
protocol is only a delivery mechanism and is not capable of managing the DNS
infrastructure it supports further complicating the process.

With that said let me propose that should have happened is that the
country-code (what now is the ccTLD URL anchoring nomenclature) might have
been made the prefix on the URL as opposed to a part of the URL itself. This
would have allowed an essentially  infinite number of possibilities for the
dot com, net and org domains since each  ccTLD code would have its own set
of dot whatever's. Then to round this off, all we would need some globally
"equated" names as well for folks like Cocacola or other global players.

So something like "http://[us]www.nasa.gov" would have the [us] country code
in the URL and would have been NASA globally, but to keep things simple the
already existing "http://www.nasa.gov" would have been the same from within
our TLD.

Such a mindset would also allow for us to do "[us1], [us2], and [us..X] zone
names such that the underlying marks (the textual representation of the
URL's) could be identical. It is no more confusing than the use of ccTLD's
now to achieve the required physical expansion of namespace.

My feeling is that this model would have required simple additions on the
processing of DNS to make real and would have saved a ton of pain in the TLD
creation since each TLD would only need to support its own root servers to
step in here. This means that registrars or Countries could set up their own
TLD's which is what true privatization should enable.

I think that the adoption of such a naming model convention would have made
life much easier for all concerned.


Todd



----- Original Message -----
From: "Michael Froomkin - U.Miami School of Law" <froomkin@law.miami.edu>
To: "Elisabeth Porteneuve" <Elisabeth.Porteneuve@cetp.ipsl.fr>
Cc: <barrister@chambers.gen.nz>; <ga@dnso.org>; <jsims@JonesDay.com>;
<apisan@servidor.unam.mx>
Sent: Monday, June 03, 2002 5:28 AM
Subject: Re: [ga] ccTLD dialogue with Joe Sims (wasRe: your comments


> Actually, I don't think this will help as much as it should, since the .us
> contract written by DoC takes the Sims line: .us must do what ICANN tells
> it.
>
> On Mon, 3 Jun 2002, Elisabeth Porteneuve wrote:
>
> >
> >
> > Peter Dengate Trush wrote to Joe Sims:
> > > But step one is for you and ICANN to accept that the cctlds do not
operate
> > > under a licence from ICANN. They operate under a duty to their LIC.
> >
> > I think that now, since the .us ccTLD has been put back to life,
> > we will see a better understanding of ccTLD position from the US
> > colleagues. There is a clear difference in .us and .com managements,
> > the former being US matter and duty to their LIC and USG
> > (actually the USG did not sign any contract with ICANN for its
> > own space, which was analysed as a good example to follow by
> > more ccTLDs), the later is an ICANN experiment in devising
> > extraterritorial rules for global TLD, and their reinforcement.
> >
> > My personal interpretation is that for many of our US collegues
> > there was a confusion between .com (extraterritorial) and .us (USA),
> > and a kind of perception that US did accept other countries
> > to influence the .com matters, while they do not get "reciprocity"
> > for other ccTLD. Now with .us up and running well, they have a
> > clear, American example of the difference with gTLD.
> >
> > Elisabeth Porteneuve
> >
> > --

SNIP

--
This message was passed to you via the ga@dnso.org list.
Send mail to majordomo@dnso.org to unsubscribe
("unsubscribe ga" in the body of the message).
Archives at http://www.dnso.org/archives.html