[ga] root vs roots - the false but usefull confusion

[Jefsey Morfin <jefsey@wanadoo.fr>]

Dear Name Critic,
I think your post summarizes the level of confusion reached in people's
mind with Start's paper. This comes from the deliberately ill chosen
"alternate root"  wording and from the naming of the ga-roots ML. There is
only one root (it should be ga-root. There are several sources but a single
root without a single point of failer].

It is not the root of the iCANN: the iCANN has deliberately chosen to have
its mini-root (authoritative on its TLDs] being only a part of the full and
unique root.


On 10:05 30/05/01, NameCritic said:
>A Standards Competition between Roots was my understanding. Maybe I'm just
>reading too many lists and I'm mistaken. A likely analogy for the
>competition between roots would be the standards competition between VHS and
>Beta awhile back. Which one was better? Beta. Which one won consumer
>support? VHS.

There is not such a thing.

There is a capacity for millions of TLDs (cf. BoD response to the GAC in MDR).
undere the common free DNS technology.

There are people thinking that TLDs are to be used by the users to the best
common advantage of the users and there are people thinking TLDs are to be
used by the users to their own best business advantage.

The first ones think that stability comes from everyone using an open system.
The other think that stability is the stability of their income through a 
controlled
system (hence M$ 200 by VeriSign planned on the DNS).

This translates into those thinking the user is authoritative for his own 
machine
and those thinking the merchant has to be authoritative over his clients 
machine.

The rest is pure window dressing and personal pride of technicians not
wanting to acknowledge they have been manipulated by merchants. There
is not such a thing as alternative root vs authoritative root. Please note 
that
the iCANN contract talks about Authoritative Root-Server System, not 
authoritative
root, and describes this system as the galaxiy described the named.ca file 
(they
never dare to speak of their mini-root in a legal document, so nothing 
prevents
them to install the common full root if the FoC asks it).

There a single root including all the TLDs, including the iCANN TLDs. The
iCANN mini-root is part of it. If you want I can copy you the real root as 
I am
one of those trying to maintain it.

There is only two problems: collision and exhaustiveness. Some people have
started TLDs in parallel or attempting competition (iCANN, NEW.net), some
do not care too much (Name.Space) and some do not document (China).

The TLD community is quite adamant at not accepting colliders. And is
obviously upset on technical grounds by the bis.biz collision (now with
added bad faith). Should someone come with this attitude, none of the
RSCs (Root Service Centers would accept him).

The collision problem is treated quite simply: on the basis of the first
date in operation (as the US belong to the UK due to the first landing
date), of he good faith and of the reality of the service. No one decides
who is right: only the RSC decide to accept a collider or not. If the collider
is unhappy he may go to another root. Market will decide. RSC try now
to talk together and would be more than happy to see the iCANN joining.

Why would someone chose a root rather than an other one? For
a few reasons:

- it is kept updated (new TLDs show up and zone servers IP addresses
   are to be maintained: I publish every day a report on the discrepancies
   I found and broadcast it. I use the other roots as inputs too. So this
   stabilizes a lit and becomes very precise and  professional). Just
   remember that this responsibility is carried at the iCANN by an open
   position and no report are published. It happens (from former Directors
   reports) that when they come in the morning the root has crashed...

   Obviously all the iCANN changes (gTDs and ccTLDs are integrated)
   in the common full root.

- they may filter some TLDs (their service is to the users not to a
   a twisted legal obligation) that their user do not want (there is a
   religious oriented root filtering adult TLDs)

- they may integrated DNS+ services the user favor. Actually we will
   see the ISP becoming RSP (root service providers) with a menu for
   the users. This will certainly develop in commercial roots with special
   bargain to access TV shows, discounts, etc....

So the root is obviously unique but common, not hi-jacked by a non
profit no member VeriSign supper contracted USG delegated corporation.
It is open and belongs to everyone. An example is English:  there is
only one meaning for each word, but everyone may [try to] speak English
without having to their thinking in the Oxford Dictionary. Is some one
get the idea for a new word: he may be praised for that, or he may TM
it, or he make a lot of money of it. Same in TLD. There some free,
there are some costly.

This shows you that most what is discussed on ga-root has no real
meaning except when coming down to reality and getting histerycs
out. The question is only: when and how the DN replace the TMs with
their - non yet defined rules. Many people are really unhappy with that.
Starting with large US corporations due to the US law. But we are on
the move and this is not 94% of the people's "cup of tea". (Vint Cerf
testified that non iCANN TLDs are not his "cup of tea").

So there may be commercial competition (I suppose it would be
more complementarity). But with a common basis. So instead of
talking of alternative roots (for most of those you know: ORSC,
Pacific, etc...) it is better to speak of coopetitive roots and to
count the iCANN among them with its mini-root conception.

Jefsey




--
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