numeric (ascii) labels (was: Re: draft-liman-tld-names-00.txt and bidi)

Vint Cerf vint at
Tue Mar 10 09:06:39 CET 2009

To IDNA participants:

Johh and Andrew S have both suggested moving any further discussion of  
numerics in TLDs to DNSOPS. I concur, not only for the reasons John  
suggests but also because the present IDNA documents apply to all  
labels, not to the TLD ones.

This isn't a moderated list, but as WG Chair, I request we cease  
further discussion of the numerics in TLDs question in IDNABIS.



Vint Cerf
1818 Library Street, Suite 400
Reston, VA 20190
vint at

On Mar 10, 2009, at 2:11 AM, John C Klensin wrote:

> --On Monday, March 09, 2009 22:34 -0400 Vint Cerf
> <vint at> wrote:
>> eric,
>> given the need to program something to answer the question
>> "could this dotted thing possibly be interpreted as an IP
>> address" and the alternative "this could not possibly be an IP
>> address because it is not all-numeric" I would incline towards
>> what I think is the easier to program and not very restrictive
>> choice that simply says:
>> 1. no TLD can be all-numeric ASCII
>> 2. leading (and trailing ?) digits are disallowed in labels
>> I gather you favor allowing all-numeric TLD labels. How does
>> the rest of the WG see this?
> Vint,
> I'm clearly not the "rest of the WG".  However, there is a
> reason why Jon wrote "alphabetic" in 1123 when he could have
> written "won't be anything that can be confused with IPv4
> addresses" or "the original no-leading-digit restriction is not
> changed for TLD names".  While they involve more words, the
> latter options would not have been any harder to write than what
> was written.
> The reason was to avoid any possible confusion with addresses
> --not confusion after applying a series of rules, but confusion
> at all.
> But, more important for this WG, it seems to me that we have
> verified that A-labels cannot contain trailing digits.  Beyond
> that, where RFC 1123 should be modified at all is a problem for
> DNSOPS (or some other WG), but certainly not for this one.  And
> the policy matter about what should be permitted at the top
> level within the constraints set by 1123 or an update to it is
> an ICANN problem.    We have documents posted that are not
> getting adequate review and discussion, in part, I believe,
> because endless speculation on this sort of topic is much more
> amusing.  Could you, as WG Chair, please rule this discussion
> out of order and ask people to take it off the list?
> Just my personal opinion, but, if we were making more focused
> attention on the WG's documents, I'd feel a lot better about
> these many excursions into the weeds.
>    john

More information about the Idna-update mailing list