[dns-operations] Karl Auerbach on adding 'millions' more TLD - what do folks think about the operational impact?

[Patrik Fältström]

On 9 jan 2007, at 04.05, Douglas Otis wrote:

> On Jan 8, 2007, at 2:18 PM, Patrik Fältström wrote:
>> 6000 languages times 300 TLDs = more zones than today.
>
> Limited to languages where ASCII is problematic reduces this number.

Not much. I do not know of many languages (<10) where ASCII is  
sufficient.

We still talk about M*O(1000) times n*O(100) => m*n*O(10^5). Still a  
number that is possible to handle (I claim) although not as easy when  
introducing mechanisms like the one

Especially when(!) the root zone is signed, there is a possibility  
to, as Mark suggests, in a secure and trusted manner let anyone on  
any IP address do whatever they want with the zone. Especially pre- 
loading into resolvers etc will be extremely interesting.

This will definitely not be as easy if the root zone is the largest  
zone we know of in the DNS hierarchy.

> Catering to these languages may invite ACE labels as substitutes  
> for existing gTLDs, ccTLDs and SLDs that redirect the application.   
> Ideally this translation should take place within the application  
> using established conventions.  Without translational conventions  
> at the application, a desire to cater to non-ASCII languages may  
> create an N factorial expansion of TLDs.  ICANN probably finds  
> itself wandering down the road of internationalization to remain  
> relevant.  One must wonder where this road leads.
>
> http://www.icann.org/announcements/announcement-05dec06.htm

It is definitely not clear the solution will be "aliases" in the root  
zone when creating IDN versions of TLDs. Strong lobbying say that a  
local language version of for example .COM is to be run by a registry  
different from Verisign. This lobbying (that mostly is after becoming  
this registry for monetary reasons) of course uses technical  
arguments like "DNS people say DNAME is not possible to use in the  
root" etc.

I personally agree the only solution possible (for administrative  
reasons, not (only) technical) is to have "aliases" for as many TLDs  
and languages as possible, and then create new TLDs when that is the  
only possible solution.

> When there is money to be made, it seems likely registries will not  
> go away easily.

Agreed.

    Patrik