an IDNgTLD group is technically necessary at ITU, IGF or ICANN

[Marie-France Berny <mfberny@xxxxxxxxx>]

I participate to the IETF WG/IDNABIS. The technical aspects involved in the
IDNA are not simple. However, what strikes me is that IDNA is only
considered from a technical point of view by the IETF and from a political
and deployment point of view by the two ICANN representatives.

Someone in the IETF, ISOC, ICANN, ITU, IGF, etc. must discuss it from a
usage (not users but practical, operational usage) point of view. For
example, I observed that Affilias made a real and positive effort in
supporting the ASIWG. There is the need for other language communities to
copy them. And also to support the two ICANN representatives there with
practical rationales and global community support. Only a few ccTLD
participated. The ccNSO does not.

The technical target is that the same natural language is presented in the
same manner by all the registries and registrars. Someone is to take care of
that.

For example: there was is a difficult technical conflict to solve over the
arabic TATWEEL character. This character only serves as a typographic
extension and has usually no meaning in Arabic. IETF technicians agreed in
general to block the character at network level for security and stability
reasons. The iucg@xxxxxxxx people (contributing users) opposed in saying
that such a decision is to be taken by the concerned parties. They also said
that if there is a blocking/mapping it must be carried at user level (this
is what the WG Charter implies). Arabic language and Persian experts said:
we need first to discuss with every language community which uses the Arabic
script (and may not speak Arabic) to know what use they make of TATWEEL.

Only an IDNgTLD common WG or constituency can offer the necessary forum and
permit a consensus in cases like this. This is very important because it
affects the whole Internet naming capacity. What if a gTLD candidate wants
to use TATWEEL for Arabic caligraphic art, or if a TM includes a TATWEEL?

Will TATWEEL be removed from the Arabic Whois documents ?
Who is otherwise going to consensually agree about the text and format of
the Arabic script Whois page?

If the GNSO is to be 25.000 languages in two houses as I read from Mr.
Shepard, these two houses must have 25.000 windows. Who is to agree but an
IDNgTLD which colour each window is to be painted. If we want users to feel
at home, this cannot be decided by the BC, the IETF or the ITU. But all of
them will be welcome in expanding the houses. Constituencies should be
specific to types of large tasks. Internationalization is a large task.
Everyone is involved in.

Marie-France Berny
Chercheur