Standardizing on IDNA 2003 in the URL Standard

John C Klensin klensin at
Mon Aug 26 13:54:26 CEST 2013

--On Monday, August 26, 2013 14:27 +0800 Jiankang Yao
<yaojk at> wrote:

>> With reference to your comments below, I think that many
>> people's views have evolved in the last four years. I'm sure
>> that Unicode Consortium would be glad to work together on
>> improving UTF46. 
> Is it possible to standardlize UTS46 based on IETF process?
> for eample, making  the main contents of  UTS46 as the RFC.

If it could get consensus, only one problem in principle: the
IETF normally insists on change control over its standards, so
the mapping tables that are effectively part of UTR46 would
either need to be frozen on the IRND2003 mappings or made part
of the IDNA2008 review and update process.  Of course, I have no
idea whether The Unicode Consortium would want to give up their
change control.

However, and with the understanding that this is my personal
opinion only, some parts of UTR46 effectively say that one
should ignore the specifications of IDNA2008 and do what it says
instead.  That is the issue I referred to earlier as really
being a replacement for IDNA2008 that draws on both IDNA2003 and
IDNA2008 rather than being just a supplement to IDNA2008.  I
think that would need to be clarified in order to be acceptable
to the IETF (at least without reopening IDNA2008, which I don't
think anyone wants to do and which would considerably disrupt
the ICANN side of things.

Recommendation: let's see if we can focus on the possible
changes to UTR46 that Mark, myself, and others have been
pointing toward and whether consensus on them is possible rather
than the details of who standardizes what.  The formalities of
standardization are much less important than who is actually
adopting what (in essence, that is the issue that started this
thread) and getting stuck on them at this point would, I
believe, not be in anyone's best interests.


More information about the Idna-update mailing list