RE: [ga] DNSO ICANN board member

[Dave Crocker <dhc2@dcrocker.net>]

At 01:24 PM 9/3/00 +0200, Roberto Gaetano wrote:
> >>How would you call a root that does not use the ICANN root as baseline?
> >
> >I'd call it a "Private Root" since it does not reflect the publicly

The term alternate root has been used to cover any DNS hierarchies that is 
administered independently of the IANA name root.

This is not affected by the choice of some independent administrations to 
incorporate other name administrative assignments.  In other words, the 
fact that some alternative roots incorporate the IANA/ICANN assigned names 
is entirely irrelevant.


> >viewable internet name space (anything less than the ICANN baseline is a
> >private name space). It would not be an "Alternative Root" (alt.root) to
> >the "IANA/ICANN root" in the publicaly-viewable sense.

This bit of nomenclature subtlety might be helpful to the logic of those 
running an alternative root service, but is irrelevant to the public 
discussion, now and in the past.


>The main thing I am worried about is that two alt.roots have different
>name servers for the same TLD.

The ability to have the name space assignments be different is the key.


> >We all saw what John Postel and Paul Vixie (at John Gilmore's prompting)
> >did a couple of years ago in splitting the root servers up into US
> >Gov-controlled and non-USG-controlled groups by changing the
> >non-USG-controlled root servers to pull their root zone from an IANA
>server instead of a.root. What they actually did was attempt to create an
>alt.root out of the non-USG-controlled root servers. Had no-one noticed the
>change of root authority then the 7 CORE TLDs would have been added to half


I don't know how Simon came by these bits of fantasy, but they are 
factually wrong in every respect.

John was testing the ability to switch to a new root server.  At the time, 
there was some concern that NSI might turn rogue and he was exploring the 
ability to remedy the problems that would cause.  In addition, the idea for 
a "stealth master", that is finally being deployed now, was developed back 
then.

The view that it was a surreptitious resulted from people who had never 
before been involved deciding, after the fact, that they should have 
been.  Those who were regularly part of DNS root operations, and some of us 
who had not been, were well aware of the plan.  (I did not know the planned 
timing, but John described the effort to me some weeks beforehand.)

Addition of the IAHC/POC-specified new gTLDs -- CORE did not create the 
list -- was an entirely public sequence and the test of a new root server 
administration had nothing at all to do with adding new names.

The fact that the DNS root had always been under the direction of IANA and 
that it was IANA performing the test makes the claim that it was an 
"alternative" root particularly silly.

Since IANA had always made the decisions about top-level domains, and since 
IANA supported addition of the new IAHC/POC gTLDs, the additions would 
merely have been continued enhancement of the IANA DNS.  There would have 
been nothing "alternative" or "independent" or "rogue" about it.

Lastly is the fact that the test "master" root got all of its data from the 
NSI 'a' root.  There was no independent name administration.

d/

--
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