[ga] RE: IDNs & the GNSO New TLD PDP

[Danny Younger <dannyyounger@xxxxxxxxx>]

Vint,

I will admit to a degree of discomfort with the idea
of different registries choosing among several
possible tables.  To a layman it would seem that a
standardization body should be taking the lead in this
area.  For the benefit of those of us who are neither
linguists nor engineers, could you perhaps comment on
the current standardization work that is being done in
this area by either the IETF or by the ITU?  Are there
any reasons that mitigate against the use of
harmonized tables?

--- Vint Cerf <vint@xxxxxxxxxx> wrote:

>  Danny,
> 
> I think the IDN matter splits into several parts.
> For gTLDs, IDNs are
> potentially much harder to solve. There might be
> more than one French
> restriction table developed by France, Senegal,
> Canada where potentially
> different dialects of French are spoken. To take
> Canada for example, it
> could conceivably have a single TLD (.CA) and have a
> restriction table for
> French (if one is needed) as it is written in Canada
> for registrations in
> .CA. The restriction table might differ from the one
> developed by France or
> Senegal. A gTLD, such as .info, might have to choose
> among several possible
> tables or develop its own for registrations in
> French. 
> 
> For ccTLDs, then, it would appear valuable for all
> those ccTLDs who wish to
> accept registrations in a given script and language,
> to work together as the
> Chines, Japanese and Koreans have. They developed a
> common CJK plan. 
> 
> I did not intend in my earlier remarks to suggest
> that the ccTLD operators
> has to comply with the ICANN contract for gLTDs. I
> only meant that the new
> gTLD process should include provision for non-Roman
> scripts (ie IDNs). I
> agree with you that the ccNSO should take up the
> question of IDNs for
> themselves. I also think that it is important to
> place an upper bound on the
> allowed number of additional TLDs using non-Roman
> scripts. For the present
> the suggested number is one. 
> 
> This won't quite satisfy all needs since some
> countries have more than one
> official language and many may want registrations in
> other unofficial
> languages that are widely in use. In the US, where
> English is the official
> language, there is a considerable spanish speaking
> population, for example.
> 
> 
> 
> Vinton G Cerf
> Chief Internet Evangelist
> Google/Regus
> Suite 384
> 13800 Coppermine Road
> Herndon, VA 20171
>  
> +1 703 234-1823
> +1 703-234-5822 (f)
>  
> vint@xxxxxxxxxx
> www.google.com
>  
> 
> -----Original Message-----
> From: Danny Younger [mailto:dannyyounger@xxxxxxxxx] 
> Sent: Monday, December 12, 2005 9:11 PM
> To: Vint Cerf
> Cc: ga@xxxxxxxxxxxxxx
> Subject: RE: IDNs & the GNSO New TLD PDP
> 
> Vint,
> 
> Let me pose a somewhat delicate question.  We are in
> the midst of discussing
> selection criteria for new gTLDs.  As we start
> thinking about allowing
> additional TLDs for each countrycode TLD where
> scripts other than Roman
> characters are needed, are we not asking for a lot
> of trouble if we take up
> the work of such IDNs within a gTLD discussion
> framework as you seem to
> suggest?
> 
> I know that you gingerly sidestepped the
> jurisdictional question earlier
> (and I hate to press the point if you're
> uncomfortable talking about it),
> but historically, gTLD selection criteria winds up
> becoming imbedded into
> contract language, and yet we know that less than 5%
> of all ccTLDs are
> currently willing to work on the basis of either a
> contract or MOU with
> ICANN.
> 
> Can we really expect ccTLD operators that will
> launch such IDN services to
> comply with a RFP that spells out typical selection
> criteria (such as proof
> of ample, firmly committed financial resources), or
> expect such ccTLD
> operators to sign a contract with ICANN as a
> precondition to entering an IDN
> in the root?
> 
> Don't you think that it would be more appropriate to
> refer these type of
> considerations for vetting by the ccNSO before the
> GNSO takes up such
> issues?  
> 
> 
> --- Vint Cerf <vint@xxxxxxxxxx> wrote:
> 
> > Danny,
> > 
> > I am not a fan of alternate roots and recognize
> the challenge as 
> > regards IDNs. In the Vancouver meeting it seemed
> to me that we made 
> > some progress on thinking about allowing one
> additional TLD for each 
> > countrycode TLD where scripts other than Roman
> characters were needed. 
> > In that sense, we might move ahead of the new TLD
> general questions.  
> > I continue to believe, however, that the general
> treatment of IDNs 
> > ought to be taken up in the work on new TLDs.
> > 
> > Vint
> > 
> > 
> > 
> > Vinton G Cerf
> > Chief Internet Evangelist
> > Google/Regus
> > Suite 384
> > 13800 Coppermine Road
> > Herndon, VA 20171
> >  
> > +1 703 234-1823
> > +1 703-234-5822 (f)
> >  
> > vint@xxxxxxxxxx
> > www.google.com
> >  
> > 
> > -----Original Message-----
> > From: Danny Younger
> [mailto:dannyyounger@xxxxxxxxx] 
> > Sent: Monday, December 05, 2005 12:54 PM
> > To: vint@xxxxxxxxxx
> > Cc: ga@xxxxxxxxxxxxxx
> > Subject: IDNs & the GNSO New TLD PDP
> > 
> > Vint,
> > 
> > Perhaps you could weigh in on these
> questions...Your
> > answers might help to
> > clarify my thinking as to whether IDNs should or
> > shouldn't be considered
> > within the GNSO New TLDs PDP.
> > 
> > In the IDN forum, Abdulaziz Al-Zoman, Chairman of
> > the Steering Committee of
> > the Arabic Domain Name Pilot Project, reported
> that
> > "We have two alternate
> > root servers, one in Saudi Arabia and one in the
> > United Arab Emirates".  He
> > also stated, "The duration of the project is open.
> 
> > So until we have a
> > worldwide implementation or recognition of Arabic
> > domain names, this pilot
> > project will continue."
> > 
> > My first question is as follows:  within the
> context
> > of the GNSO PDP on new
> > TLDs, are we to regard such a namespace as a
> future
> > sTLD that is awaiting
> > ICANN approval (in that like .cat it appears to
> have
> > a sponsoring community
> > that has created it for cultural and linguistic
> > reasons), or are we to treat
> > it as a namespace that awaits a process to be
> > instigated by the ccNSO?
> > That's the jurisdictional question...
> > 
> > The second question pertains to my worry that if
> > ICANN fails to act quickly,
> 
=== message truncated ===


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com