Proposed Charter for the IDNAbis Working Group

Martin Duerst duerst at
Sat Mar 29 06:12:33 CET 2008

Hello Leslie,

At 07:09 08/03/29, Leslie Daigle wrote:
>I may well be missing something key in your & John's points, but what I hear is:
>You are disagreeing with John's premise of the need to provide strict scoping of allowed symbols in IDNs, and are suggesting that there is no need to exclude particular symbols, because they are so impractical to use as to be unlikely to appear in registered domain names.

My main point was that for the majority of symbols (let's take smileys, snowmans,
the math square root symbol,... as examples), it doesn't really match much at all
what the charter says (or implies) about them, because the vast bulk of these
symbols (with a few very notable exceptions, such as everything that resembles
a hyphen) won't be that usable, and therefore won't be that used, even if they
look cute.

>The difference I see is in the protocols using, and software manipulating (for purposes other than user presentation) these labels:  those protocols/that software must be built to handle the universe of allowable symbols, and deal with the potential impacts (for matching) across that universe.  In that light, I'm not understanding how the argument that certain symbols are unlikely to be registered is really that manageable.

The procotols should be doing just fine. The WG will end up with a clear definition
of which symbols are allowed and which not, and protocols will be able to take that
into account where necessary, and otherwise will just pass through strings of bytes
in the relevant encoding (be that punycode, UTF-8, or something else, according to
the circumstances).

Although some users might find some such symbols cute and therefore apply for them
if they were available, and some registries might find it good business and therefore
support the inclusion of such symbols, for the vast majority of these symbols
(again, there are a few very special exceptions), there is absolutely no need for
them in order to be able to write names as they might appear in the languages of
the world.

Hope this helps. If there are any questions remaining, please feel free to ask again.

Regards,   Martin.

>But, perhaps I missed your point?
>--On March 28, 2008 2:27:34 PM +0900 Martin Duerst <duerst at> wrote:
>> At 21:35 08/03/27, Gervase Markham wrote:
>>> Martin Duerst wrote:
>>>> This is different from some very specific, small subset of symbols
>>>> that may be highly similar to e.g. dot or hyphen or so, which indeed
>>>> can create problems.
>>> Are you willing to evaluate all the non-letter characters for
>>> confusability and come up with an exceptions list of banned characters,
>>> then? :-)
>> I wasn't arguing for or against disallowing all symbols, I was
>> mainly making the point that for most symbols, it doesn't matter.
>> In terms of charter, and for the resulting spec, that would mean
>> that for most symbols, whether they are in or out in the end isn't
>> a high priority issue.
>> Regards,    Martin.
>># -#-#  Martin J. Du"rst, Assoc. Professor, Aoyama Gakuin University
>># -#-#       mailto:duerst at
>> _______________________________________________
>> Idna-update mailing list
>> Idna-update at
>Idna-update mailing list
>Idna-update at

#-#-#  Martin J. Du"rst, Assoc. Professor, Aoyama Gakuin University
#-#-#       mailto:duerst at     

More information about the Idna-update mailing list