[gnso-idn-wg] Summary of teleconferences today - 20 June 2006

["Bruce Tonkin" <Bruce.Tonkin@xxxxxxxxxxxxxxxxxx>]

Hello All,

Here is a rough summary of the main points of discussion today:

(1) review the Issues Report,

Cary Karp pointed out that the IDN workshops in Marrakech, along with
the update of the IAB's issues report, may identify further issues to
include in the issues report. 

See: http://www.ietf.org/internet-drafts/draft-iab-idn-nextsteps-06.txt
dated 12 June 2006, and
edited by John Klensin, Patrick Falstrom and Cary Karp.

Werner Staub pointed out that the ICANN issues report recognises that
the issues may be different between ccTLDs and gTLDs, but that the
report should also recognise that there may be different issues
associated with categories within the set of gTLDs.  For example some
strings may have a clear dictionary meaning with a strong expectation
about purpose, and some strings may be far more general in nature, and
may be interpreted in many different ways depending on the context of
the registrant or user.

Olof Nordling reported  that the issues report is still a draft, and is
called a Preliminary Issues Report.  The final issues report will have
an opinion from the ICANN General Counsel on whether the work is in
scope for the GNSO, and will also have some recommendations with respect
to a terms of reference.

It is expected that the Issues Report will be finalised early in July
following the Marrakech meeting.




(2) prioritise the policy issues in the Issues Report for further work,

Cary recommended that the working group members review the report from
the IDN committee chaired by Masanobu Katoh. The final report dated 27
June 2002 is available here:
http://www.icann.org/committees/idn/final-report-27jun02.htm .  The
report did note that:
"general view that procedures for ASCII and non-ASCII TLD should be
harmonized".

There was support for studying some of the issues identified in the
issues report around choice of strings and allocation of strings.  These
general policy issues could be studied in parallel with the further
technical work on IDNs.

There was some discussion around whether IDNs raise any new issues that
have not already been considered within the discussion on new gTLDs.  It
was agreed that there are many issues in common (e.g minimal technical
criteria around DNS services), but that a discussion of specific IDN
cases tends to highlight some of the complexities around introducing new
gTLDs.  In other words the existence of IDN strings magnifies some of
the challenges associated with introducing additional new gTLDs.   

For example, one of the draft recommendations under consideration for
new gTLDs is that
a new gTLD should have a clearly identified purpose that differentiates
it from other existing gTLDs.
If you introduce IDN strings, you may have strings that appear with
different scripts that have various interpretations in different
languages and cultures.  It thus becomes more complex to agree on
whether a particular gTLD is clearly differentiated from other existing
gTLDs.

Another aspect that has not yet been considered with respect to new
gTLDs is that an organisation may request a set of strings (some of
which may use characters outside the Latin script, but not necessarily)
to meet their particular purpose, or an existing gTLD operator may
request additional strings to represent their existing purpose.

The policy work on new gTLDs will be informed by a consideration of the
IDN issues, and the policy issues that are specific to IDNs would
benefit from the analysis of a focussed group of experts.


(3) consider the terms of reference for an initial policy development
process in the area of IDNs within the GNSO

There seemed to be general agreement that the terms of reference for
IDNS should follow the same framework as for the policy work on new
gTLDs - ie selection criteria, allocation methods, and contractual
conditions.

The work on IDNs should identify the specific areas within the new gTLD
terms of reference that warrant additional focussed work with respect to
consideration of the implications posed by the use of IDN strings in the
root.


With respect to ccTLDs, one approach was discussed where the ccNSO and
GAC may work on identifying a new table of strings that identify a
country in a way that is meaningful for the people of that country.
Essentially this would create a table additional to the ISO 3166 table
of two letter country codes from the Latin script.  It was mentioned
that this table was originally chosen to use a common script with a
common string length as part of a standardised approach.  Also each
country was required to use one string only.   One suggestion has been
that a new "IDN" table may also be restricted to one string per country
and that the string length would be two "IDN" characters (which would
comprise multiple ASCII equivalent characters in the root zone as per
the IETF IDN standards).   However it was noted that some countries may
wish to have either shorter or longer string lengths, and some may wish
to have multiple strings per country, particularly if the country uses
multiple scripts and languages.  

The issue of the selection of the operator of a particular cctld string
was seen as a separate topic.  One suggestion has been that the process
for delegating a new IDN string to a ccTLD operator, would be the same
as currently used for the ISO-3166 table strings.

There seemed to be support for encouraging the ccNSO and GAC to work on
a new ccTLD table, but that the GNSO would not specifically comment on
the constraints in terms of number of strings per country and length of
the string, other than that any strings chosen would relate to strings
used to identify the country by geography, and should avoid creating
strings that may be confused with the semantic meaning of gTLD strings.

It would seem to make sense therefore for the ccNSO to focus on:
- creation of an IDN string table for cctlds
- the delegation of these strings to operators


(4) Review timetable of related meetings in Marrakech

There will be a meeting of the GNSO working group on Saturday 24 June at
6pm.   This will form preparation for the IDN workshop on Sunday.  The
workshop on Sunday will inform the community on the current Preliminary
Issues Report, and provide an update on the current thinking of the GNSO
and ccNSO.
There will also be an IDN workshop on Tuesday which will be more
technical in nature.

Regards,
Bruce Tonkin