[ga] Re: DNS/ICANN understanding

[Jefsey Morfin <jefsey@wanadoo.fr>]

Dear Dave,

1. Candidate filtering
I thank you for this. This should be rule nr°1 in all the
Internet so called "governance".
At 06:11 23/08/00, you wrote:
> The ICANN Board is responsible for global
> policy of DNS (and IP) administration.  Having experience with a
> particular registry is, of course, helpful.  However it tells us
> nothing about the person's understanding of large-scale operations,
> design or policy issues. They might have great insight or they might
> not.  We can't tell.
> An auto mechanic might be quite good at auto
> repair, but would you automatically assume that they are good at
> designing a new automobile, or even able to suggest engineering changes
> to existing cars?  And just so there is no confusion:  the
> answer to both parts is
> no.)


2. DNS nature
Now, let see your key point which is also well stated.
> > > e.  
> > > ICANN should not require new TLD registries to sign contractual
> > > provisions or make other assurances that they will not participate in
> > > alternative root systems.  I am not aware of any evidence that such
> > > alternatives present problems for the Internet, and ICANN should avoid
> > > practices that will be seen as monopolistic or
> > > anticompetitive.
> Most of
> the statements pertain to fuzzier aspects of legitimate social (and
> legal) policy and, therefore, can reasonably be debated.  This one,
> however, falls victim to the efforts at trying to move technical
> realities and constraints into that fuzzier area.  That is, it tries
> to pretend that technical issues are not technical issues.  We had
> all better be very careful that ICANN does not start trying to make rules
> that violate technical realities, such as the nature of the DNS
> root.

User demand 
.

First I will get rid of the non technical issue: the technical solutions are there to address the user's demand Mr. Love wants to represent. I am sure you would not accept if I pretended you position reflects your technical inability to match the user's demand!

DNS Technicalities and NASDAQ

Warning: I quote two RFCs here to support my text. They are of 0 technical degree of complexity. Anyone can understand them. Anyway, the quotes I make are enough to fully understand my point.

You obviously refer to RFC 2826 which is the state of the art on this question. Being a technician you should read it in a technical way, not in the biased way of an old stubborn lawyer :-).

1) a technician you will probably accept the words "old stubborn" ...

when I show you the misvision of "God" vs the Yokohama acceptance of the need of new gTLDs. You will read Jon Postel's RFC 1591 (this is the vision funding the TLDs). The text says:

'In the Domain Name System (DNS) naming of computers there is a hierarchy of names. The root of system is unnamed. There are a set of what are called "top-level domain names" (TLDs). These are the generic TLDs (EDU, COM, NET, ORG, GOV, MIL, and INT), and the two letter country codes from ISO-3166. It is extremely unlikely that any other TLDs will be created.'

As you know, the ICANN sells away this Jon Postel's vision $50.000 a study. Real world is not always technical.

2) Now you are not anymore biased ... :-)

and have acknowledge there may be some different technical visions, we may come back to the ICANN "funding RFC 2826". This RFC is the central reference used by ICANN, NSI, and al. to justify their anticompetitive conduct.

Any layman can read the document summary and understand. So let do it.

-  this RFC is only a comment. Its Title is:

"IAB Technical Comment on the Unique DNS Root".

   (IAB stands for Internet Architecture Board)

-  this RFC is for information. It starts with the following note:

"Status of this Memo

This memo provides information for the Internet community. It does not specify an Internet standard of any kind. Distribution of this memo is unlimited".

I underline: "it does not specify an Internet standard of any kind". An entire economy and millions of dollars in Domain Name value are built on the two quoted RFC... When NASDAQ understands it ...

-  this RFC then supports a seemingly obvious yet contradictory concept:

"To remain a global network, the Internet requires the existence of a globally unique public name space. The DNS name space is a hierarchical name space derived from a single, globally unique root."

...

"This does not preclude private networks from operating their own private name spaces, but if they wish to make use of names uniquely defined for the global Internet, they have to fetch that information from the global DNS naming hierarchy, and in particular from the coordinated root servers of the global DNS naming hierarchy."

In lay words this means that:

- to get all the information you must have an access to all the information.

- but it adds: this does not prevent you to access it together with your own information.

This means you will have your information (the addresses of your private network) plus the information of the Internet. This is contradictory since your private network information is not available to others which therefore have not the more extended information you have (this contradiction is adressed and solved in my last part "globally" consistently with the RFC).

- the RFC summary ends with another confused yet apparently pretty obvious point  :

"[the need of a globally unique root] This is a technical constraint inherent in the design of the DNS. Therefore it is not technically feasible for there to be more than one root in the public DNS. That one root must be supported by a set of coordinated root servers administered by a unique naming authority."

"Put simply, deploying multiple public DNS roots would raise a very strong possibility that users of different ISPs who click on the same link on a web page could end up at different destinations, against the will of the web page designers."

This intends to mean:

- 'if you want to have the information of a database you must use that database': this is OK.

But it also includes two non discussed suppositions:

- people must obey the author, removing the Internet navigation freedom rights.

 

I may want to subscribe to a [family] DNS root which removed all the adult and gambling sites, or to a [business] DNS root which also removed all the family related sites, replacing by little flowers, or NASDAQ reports. These DNS roots are value added DNS roots.

Value added DNS services is a coming industry, with broad ambitions. The alternative root pionneering time will probably be over very soon as alternative evoluates towards super (cf. infra).

- global DNS = public DNS.

.

The entire issue is here: from the occurrence of the two words in two different paragraphs  about two complementary issues in the same document inattentive people assumed they were an unique concept because of the ill worded "globally unique public" of the introduction (othewise it would be like explaining that 2=1 because 1+1=1).

To be true this confusion comes from a lack of technical and semantic consideration by the authors who focused on the inner rather than on the external technicalities of the DNS using inconsiderately the adverb "globally" which is meaningless if you consider it in details and should probably be replaced by "finally" or "eventually".

Warning: the value added DNS industry will disseminate a better understanding of the DNS system by the public and a deeper analysis of its technical, legal, commercial possibilities. This will certainly lead to a revaluation of the value of several NASDAQ companies. It is of  particular interest to consider yesterday NSI announcement of foreign language domain name endorsement: it creates in the public understanding scores of new local gTLDs. This is pure value added DNS industry on a large scale (millions of domain names ahead).

The global/public issue being a key issue I will discuss it in a separated part.

GLOBAL vs PUBLIC DNS

RFC 1591 can only be understood if you realize it talks about TWO different data bases.

Prior to discuss this point I want to tell I happened to head, back in 1984/1986, Tymnet "public" network  "Extended Services". Being a French national, I got such difficulties to make the word "global" heard and  ny concept about it understood that I was nicknamed "the Global One" :-).  

So, I am quite used to people meeting difficulties at these very simple, yet very powerful, new concepts.

The equation is not "public = global", but "public + extended = global" services.

I tried for years to define the extended services (it was my job definition!). I would say they are everything  a technician may build that the user may request and the network operator may not consider as his job.

One side: an offer defined by technicians (you David). An other side: needs represented by people like James L. Love. 

I modelized the concept, enlarging the ISO model and I tried to analyse all this. IMHO a basic element to consider is the very nature of a network which allows both the operator and the user to control their use of it according to their own views, duties and interests. 
.

- Dave says "I owe to the public stable operations allowing the author to make available pages to his readers".

.

this is a public top quality service which can only be achieved by a public unique root.

.

I am not short minded so I help people extending my common offer

- James says "I am the user and I have the right to a global solution" including:

 

the default public service offered by the network

.

and to any addition by Value Added/Extended services I want to use even not perfect.

Now, if you go into the details of the RFC 1591 you will see that my reading is perfectly technically correct if you accept that the public service is part of the global network, not the otherway around. This translates into a scheme where the user or the ISP has its own virtual intranet (I name it externet), and  runs its own super root (not an alternative root) and defaults to the public root when a TLD is not matched. This way:

• the user system has correctly an unique root (his super root) and the routing is exactly what he wants, preserving his Internet navigation freedom rights, at his own risks the public service  is a top quality reference well behaved and managed by the Public Root Systems under the ICANN/DNSO control on the 13 existing machines: the users may default to it or value added clones may copy and filter it according to their own specs.

• BTW this shows you the futility of the Domain Name/TradeMark issue. I would be very very interested if someone could give me a legal definition of what is a Domain Name ???  I asked the WIPO in a very polite and documented way, to know if I could use the URDP for the two gTLDs I propose which introduce new addessing semantics. Eventhough I am a prospect for them I got no response for weeks now.

I hope this helps. Anyone interested in discussing this in a pragmatic/business way is welcome.

Jefsey Morfin