[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [atlarge-discuss] Question to the candidates on IDN and Domain Names



At 02:30 p.m. 16/05/2003, J-F C. (Jefsey)  Morfin wrote:
There are questions I would like to ask to the candidates. This is about the Internationalized Domain Names.

It seems to think that ".fr" is only associated with French scripting - but France also accept Berber and dialectal Arab as large languages of different scripting and probably more than 50 languages as "langues de France". Tamil is a major language without a country.
Good that .fr has the resources to do this. Many other ccTLD's will not want to do it for lack of demand.


Another issue is the disrespect of international conventions and commercial reciprocity that ccTLDs and ICANN seem to head for. This translates into the fact that today I can register http://cardin.cn or http://dior.jp. This means that Chinese and Japanese names should be supported under ".fr" and this to be mandatory for every language. This is not seemingly what ICANN press gTLDs and ccTLD to do.
Reciprocity is a worthwhile concept, but to introduce it now would mean more likely a restriction on IDN registrations (registrants NOT being able to register IDN's, because there "home" ccTLD does not have the resources to provide for full reciprocal Chinese, Korean, Cyrillic, Arabic, etc IDNs)

I would therefore not be in favour of a reciprocity rule at this moment. Instead I would advocate for each ccTLD registry to be an "open" registry, not restricted to nationals and residents.

Another issue of importance is that the IETF has defined how IDN will be supported in making two big mistakes, in following ICANN and Verisign choices. The first one is that the IDNs will be "ML.ASCII" (ie Muti-Lingual.ASCII, the DN in Chinese and the TLD in ASCII so Verisign may sell ".com"). This is a violation of the cultural right of many countries and cultures.
Ah Jefsey, you know full well that allowing "DN.non ascii" would put an end to ICANN's controlled issuing of new TLD's.

The second mistake was to copy the solution of Verisign to transform ML names into ASCII names built as two letters+"--"+ a transcoding. IETF chose to transcode Unicode names into ascii sequences via a process named "punnycode," after a namepreparation equivalent to the transformation of upper-case in lowercases etc. IANA (ie. ICANN) chose the two letters as "xn". So a "Chinese scripting.com" will print as "xn--cbsds.com" and will be accepted by the DNS.

However, "xn" has been radomly chosed among a list of bigrams decided acceptable by Louis Touton. It only happens it was not acceptable because he ignored that:

- "xn" means "Chrsitians" as "Xms" means Christmas. This will be odd for many users to discover their "hidden" face is christianized.
- "xn" is also the official European standard to mean "danger" http://www.eurolinc.org/xn.htm
- "XN Inc." is he leading insurrance company for Expatriated or "internationlized workers"
- "xn" is understood by many people as an abreviation for "xenophobic" or at least "xeno" meaning foreigner - hardly a way to name what is supposed to make homeon the net non-English speakers.

When an equimpent cannot print in Chinese, Arabic, etc... it will print "xn--abcd.com". This permits to register "babel names". Names chosen for their hidden ascii value, but legally registered and possibly protected, in using the internationalized scripting. This means that one can plan to legally register an Ethopian language IDN which will print worldwide as "xn-coca-cola.com."
In the "real world" I fail to see that this would be a real problem, unless the address would be used for a site that would support an active policy of name confusion.
Not likely, with the xn--prefix.

We known many limitations on the Internet due to the WIPO. Is it necessary to make such mistakes which will necessarily lead them to an over reaction? As to adopt non multilateral and non reciprocal attitudes which will lead the WTO to interfere?
So if I understand you well, you are not really concerned with the lack of reciprocity, just by a possible WTO interference?

At a time the ITU has said that multinational domain names are a key issue for them (consensually voted by the ITU GA in Marrakech).
Well , yes, it is a way to get a foot in the door.
Whether the intentions otherwise are something different from simple providing regulatory space for a market remains to be seen.
The forced common protocols of English and ascii create the common language of the internet, much as Air Traffic Control worldwide has adopted English as its lingua franca.

The IDN's could promote a Babelization of the Internet, allowing for islands of non-communication. WHOIS with IDN's is still not fully solved.

The Domain Name extensions now used may not fully satisfy every registrant, but they have the advantage of being universally resolvable.


There is also a fear from the confusion of scripting characters in different languages (like when I register "lbm.com" to type it as "IBM.com"°. This currently lead to hyper-complex schemes which would arbitrarily impeach the registration of billions of names. Created by the lack of simple real life analysis by IETF this solution would try to patch 5000 languages to make believe the bug was a new feature for them.

This will have an impact on the "first come first serve" rule because a requested name may be refused today and accepted tomorrow due to politico/technical changes (this is what ICANN initially proposed).



2. My questions are:

1. are you aware of this major problem?
I am aware of many problems pertaining to the introduction of IDN's. Challenges to answered or overcome.

2. do you support a test by ".us" to register the Spanish, Chinese, Japanese, French ,etc names of the US citizens for free (to replace their fulty current sprlling) and to shows the rest of the world how IDNs work?
Any TLD is free to run such tests.

3. do you support the demand to ICANN to replace "xn--" by "x--n" (there is no greap appeal in registering "http://x--ncoca-cola.com..
I have no problem with it.

4. do you support that users want and should get "ML.ML" domain names, ie in their language also in the TLD.
Mmm. A loaded question. Anarchy looms here. A fully unregulated marketplace, where IDN.ML registrants can be held hostage by robber barons who can do with "their" registry whatever they want is not something I would support.


5. do you support that this organization should have a WG-IDN and a doctrine on the matter,
When we get our house in order first, yes.

6. do you support that this WG-IDN should cooperate with the language organizations and be a member of the ccTLD WG-IDN; and also of the ICANN WG-IDN lead by Katho San.
And with any other organization that deals with specific IDN issues.


7. do you support that every TLD should support every language by international reciprocity. Also that this support, contracts etc should be proposed in using the same words, all over the world for the same language, with the same meaning
No. (see above)

8. do you support that the conflict resolution procedure should be removed from the WIPO and ICAN and given to panels of users only able to understand what the users will figure out form a name and decide of the conflicts and of their best resolution
NO. Unless the formation of such "panels of users" and their operational procedures, fees, etc. are precisely defined. I would never support the emergence of hundreds of (captured) "kangaroo courts".

9. more globally do you accept that a domain name is a service to the users to identify, name and access the site of the registrants before being an advantage to the registrants.
I do not understand what you mean by "before being an advantage to the registrants".


As such do you agre that the users are entitled to see the domain names to stay with the registrants on a life long basis (and on a final basis for archiving services).
Only for IDN names or for all DN's as a general principle? Do you include the "life" of a company or a non-profit foundation?

I do support that DN's are property, can be inherited, pledged as collateral, transferred, assigned and sold and that they should be given full property protection rights.

10. as such and to permit the permanency of the DN, do you support that DN must be free; and that the Registry must make its money as a trusted service to registrants (RFC 1591, ICP-1),
If that is the case, then the DN cannot be totally "free".

the first of these services being a profesional optional WhoIs?
The WhoIs must be professional and optional at the registrar level.

I have answered your questions, because I they ask me about something I have an interest in, but I frankly do not think that this Panel has the mandate to set up WG's and evolve icannatlarge 'doctrine" on IDN's.

The Panel should first and foremost concern itself with setting up the structure of this organization and launching legitimate elections for its officers.
"Doctrine" about issues like IDN's should be developed via Polling Questions to the membership directly.


-joop-



---------------------------------------------------------------------
To unsubscribe, e-mail: atlarge-discuss-unsubscribe@lists.fitug.de
For additional commands, e-mail: atlarge-discuss-help@lists.fitug.de