Changing the xn-- prefix

Vint Cerf vint at
Tue Mar 25 13:04:21 CET 2008

James, et al,

May I suggest that we neither rule out nor explicitly suggest that  
there is need for a prefix change in the charter.

If the working group comes to the conclusion that it cannot endorse  
the basic IDNA200X proposal and thinks that a new prefix is needed it  
will report that back to the IETF.

To be honest, I hope that it will not come to that or, if it does, I  
hope that we can implement IDNA200X as an improvement over IDNA2003  
and that a prefix change proposal morphs into a serious "above the  
DNS" exploration that has been set aside in the past for good reasons  
but which might need to be resurrected as a serious piece of research.


On Mar 25, 2008, at 7:52 AM, James Seng wrote:

> I remember xn-- was selected after exhaustive data mining on .COM zone
> file back then and in the process, we found a few others that could be
> use. "xn" basically was then pick "randomly".
> I agree that there are complexity to change xn at this moment.
> However, I am not totally against changing the prefix if the following
> conditions are fulfill
> a) there is a STRONG technical reason that a new prefix is needed
> b) there must be backward compatibility with all existing IDN labels
> with xn prefix.
> For example, and I am not making any proposal, this is really just
> merely an example, that I would consider a change of prefix if it we
> agreed the best way to identify IDNA vs IDNA200X labels is via the
> prefix (and I do not necessary agree at this moment) and that no
> IDNA200X labels with a new prefix can be de-constructed to an xn--
> IDNA label, ie the round-trip conversion between IDN and ACE labels
> must always be consistent.
> Much of the above are already captured somehow in Section 9.3 of the
> issue document.
> Perhaps lets not to mention the prefix change is allowed or disallowed
> in the charter and let the working group sort it out.
> -James Seng
> On Tue, Mar 25, 2008 at 2:29 PM, YAO Jiankang <yaojk at> wrote:
>>  ----- Original Message -----
>>  From: "Patrik Fältström" <patrik at>
>>  To: "Shawn Steele" <Shawn.Steele at>
>>  Cc: <idna-update at>; "John C Klensin"  
>> <klensin at>; "Mark Davis" <mark.davis at>
>>  Sent: Tuesday, March 25, 2008 10:23 AM
>>  Subject: Re: Changing the xn-- prefix
>>> On 25 mar 2008, at 00.13, Shawn Steele wrote:
>>>> I reread your original mail, and in particular I don't want to get
>>>> bogged down in the debate of details while trying to set the
>>>> guidelines, but I'd like to try for "A prefix change MUST be
>>>> avoided" (removing the condition).  If that's going to cause too
>>>> much randomization, then I'd back down, but the repercussions of
>>>> changing the prefix are huge.
>>> I am personally in favor of text in the charter that say "The prefix
>>> xn-- MUST NOT be changed." where MUST NOT is defined according to  
>>> the
>>> IETF definition in RFC 2119 (
>>  +1
>>> This put a constrain on the changes the wg is allowed to do to
>>> IDNA200X (not large so that the prefix has to change) -- at least  
>>> not
>>> without rechartering.
>>>    Patrik
>>> _______________________________________________
>>> Idna-update mailing list
>>> Idna-update at
>> _______________________________________________
>>  Idna-update mailing list
>>  Idna-update at
> _______________________________________________
> Idna-update mailing list
> Idna-update at

More information about the Idna-update mailing list