Interoperability

[jefsey]

Further to the discussion of the IDNA/ML-DNS issue, I would like to 
confirm that the only way that I can see ML-DNS being deployed is by 
being fully IDNA interoperable. This means that ML-DNS must be 
Unicode and punycode compatible on option as well as being based on 
the same ISO 3166 country, script, and language table.

Being multitechnology and user oriented, the ML-DNS must support 
various types of implementations of the presentational layer. It may 
do this by using "semantic perspectives" to be implemented (or not) 
as Internet classes. These perspectives should use ISO 3166 ccTAGs 
extended by an Open Standards 3166-4 table describing all the 
languages that are being used in each ISO 3166-2 area.

Debby Garside has indicated that ISO 639-6 should be published soon, 
but has not replied about the availability of her language entity 
table. As indicated, we have meanwhile pursued the preparation of our 
own table and we will support full interoperability with the ISO 639 
series. This may call for some further work but should be implemented 
without problem since our table is based on the same Linguasphere 
Observatory list as ISO 639-6 is (http://linguasphere.eu). My ccTAG 
I_D http://www.ietf.org/internet-drafts/draft-mltf-jfcm-cctags-01.txt 
will be updated accordingly.

I do hope that this will be satisfactory to all and will help to 
restart the IDNABIS debate, as well as assist in the LTRU LC.
jfc