At 16:35 12/08/2011, Cary Karp wrote:
I had previously posted a support brief prepared for the Latin study to the general VIP list. I have now posted an update to that text which you will find at:
http://mm.icann.org/pipermail/latin-vip/attachments/20110812/2b5bb0c7/latin-...
Dear Carry, "The code points available for use in IDNs are all taken from the Unicode Character Code Charts. The Latin script is divided there into nine blocks.". This clearly gives the environment of your work. Thank you for it as it seems a very good reference. It shows that you definitly consider Unicode as a partner of the Variant debate. This means that the work I carried to consolidate all the post-IDNA2008 problems resolution vocabulary has to include the Unicode Glossary. I added it under http://iucg.org/wiki/IDNS_Common_Glossary. This is an fastitdious task that is nearly completed. It gives us all the vocabulary that we may need from the existing technologies and SDOs (I may still introduce some additional concepts from the IUTF side). However, I am afraid there are two generic prerequisites before considering "locale" problems (like for the Latin script). These two prerequisites are the use side architectural responses to orthotypography and homography requirements. For years of stubborn debate Unicode has shown they have no simple architectural response to any of them. In the Latin case there are two lack of response: (1) French majuscules and (2) Roman script letters confusions with Greek, Cyrillic, etc. scripts. IMHO this logically precludes Variants to be documented and supported through any Unicode related algorithm. Actually, the true difficulty is that by nature (one sign, one meaning): - the Unicode's offer is "Saussurian", i.e. a dyadic semiology; - while the whole digital ecosystem use demand (e.g. Variants) is "Piercian", i.e. triadic semiotics. (different signs, many people, several meanings or "denotata"). The very idea of "variant", implies a semiosis somewhere. This is why the way I see IDNA2008 (and its consequences) has not much to do with Unicode, except as an accidental [it can change] convenient pointer set to the signs we refer to when talking of scripts. For me IDNA2008 has to do with the DNS we had to preserve and the network simplicity to be supported along RFC 1958 recipes when we have to support semiosis in its four (semantic, syntax, pragmatic and multilinguistic) components. And IMHO we do not know (yet?) how to make it. This definitely is what Fast Track should have tested. I know scripts are not a IETF, ICANN, ITU, etc. area of expertise. This is however the area of the users' demands (e.g. variants). IDNA2008 has proposed what IETF could propose: a stable, reliable and Unicode independent proven basis with the DNS. Vint proposed that ICANN could take over, you accepted. Now you explain how you see that support, in the Latin locale case For example you write: "Various terms have been used to designate this concept of equivalence but none has yet been provided with a definition that is adequate in all the contexts where it is needed". I am afraid this is not correct : "Variants are to be people's and DNS equivalent, i.e. all of the variants must resolve the same IP address". This better defines what we want. Best. jfc