Changing the values of domain names and the need for mapping

Vint Cerf vint at google.com
Fri Feb 20 04:38:01 CET 2009


Paul,
plainly we disagree about the need to re-charter. I believe where we  
are with IDNA2008 lies within the charter.

Esszet and Final Sigma are not mapping procedures. That they may  
require accommodation for previously mapped registrations is  
understood but I believe, manageable based on what has been discussed  
in the WG mailing lists and in meetings.

The primary focus of the charter was to develop a solution that is  
independent of versions of any specific Unicode version and it would  
appear that this is where the current drafts take us. The context  
rules are limited in scope as they now stand.

It seems to me that we should be working towards completing the  
present specifications not least in order to see what a solution along  
the charter lines looks like in detail.

Vint



Vint Cerf
Google
1818 Library Street, Suite 400
Reston, VA 20190
202-370-5637
vint at google.com




On Feb 19, 2009, at 8:59 PM, Paul Hoffman wrote:

> At 7:41 PM -0500 2/19/09, Vint Cerf wrote:
>> Your proposal to extend 2003 is in effect a mapping proposal since  
>> mapping is part of the IDNA2003 structure, is it not?
>
> Yes, exactly.
>
>> If you don't want to discuss your proposal at the next meeting  
>> that's ok with me but I was trying to accommodate it.
>
> I'm happy to speak about it; I never said anything to the contrary.
>
>> If you think a re-chartering is needed then I would recommend that  
>> we complete the current idna2008 vector and separately consider  
>> whether a new effort is needed.
>
> As I have said before, rechartering is now needed even if we keep on  
> the "current idna2008 vector". Given that we need to recharter,  
> maybe we should recharter to a much simpler work path.
>
>> With regard to esszet and final sigma, my sense of the email and  
>> face/face discussions is that the most directly affected parties  
>> (german speakers particularly in .de and .ch were proponents on the  
>> esszet matter and .gr favored inclusion of final sigma as a pvalid  
>> character despite earlier mapping behaviors).
>
> That may be true, but it is also irrelevant. If we made decisions  
> like that in 2003, we would have language tags in IDNA2003 for CJK  
> characters (and lots of other changes as well). What is relevant is  
> whether the changes that are proposed are (a) the consensus of the  
> Working Group and (b) within our charter.



More information about the Idna-update mailing list