Hi John, I understand and respect your knowledge/opinion (as I recall reading something you wrote on the history of gTLD and ccTLD) on the difficulties of setting up another type of script hierarchy for the <idn>.<idn>.
However, if one day it is indeed brought to the table, I most certainly would like your vote of confidence (assuming that I have not missed other technical difficulties that you see with it.) Thanks Ben ----- Original Message ----- From: "John C Klensin" <[EMAIL PROTECTED]> To: "ben" <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]> Sent: Friday, October 26, 2001 9:56 AM Subject: Re: [idn] An ignorant question about TC<-> SC > --On Friday, 26 October, 2001 10:20 -0400 ben <[EMAIL PROTECTED]> > wrote: > > > Hi John, > > > >> > >> For those who want the history, at least as I remember it, > >> read on... > >> > > > > Thanks for the history lesson and I now have a much better > > understanding of all the efforts that have gone into this. So > > I have the same question for you as I do for Martin. Why > > can't we simply label an IDN with the what language script it > > is in. In my opinion, it is much more important for an IDN > > user to know "what script the IDN is in" and far less > > important to know "if the IDN will take them to a company or > > goverment or military or organization website". > > I'm not sure I understand your question, but I think you are > asking a basic question about the organization of the DNS. > I.e., if we decide languages are really important, why don't we > reorganize things into a language-based hierarchy, rather than a > country-and-function-based one? > > The only _technical_ answers are that it would be hard to get > there from here, and that, given the administrative nature of > the DNS delegation structure, finding a single, agreed-upon, > authority for each language would be extremely hard. There are > also a number of administrative obstacles, not the least of > which is that many people would not agree with your opinion. > > I hate to keep saying this, but your question takes me back to > what I think is the basic problem here. The DNS was designed to > provide identifiers for computer resources, using an > administrative structure that was fairly convenient for those > computer resources. If we need to find names, or understand > strings, along other dimensions, we are either going to need to > remove the problem from the DNS or we are going to be faced with > tradeoffs among a range of undesirable solutions. > > john > >
