1. Attendance
2. Presentation (for newcomers)
3. Panel formation proposal v.5 (.pdf) (.docx)
4. Latin repertoire v.8 (.pdf) (.docx) and List of languages using the Latin script v.4 (.pdf)
a. Volunteers
b. Advisors
5. How best to deal with Cyrillic and Greek cross-script variants
6. Any other business
7. Date of next meeting
Notes
1. Ahmed Bakht, Marc Blanchet, Chris Dillon, Hazem Hezzah, Gustavo Lozano, Meikal Mumin, Phan Thi Nhung, Alireza Saleh, Michel Suignard, Andrew Sullivan, Aysegul Tekce
2. Chris made a presentation aimed at newcomers, giving the current status of the work.
3. He tabled version 5 of the panel formation proposal, drawing attention to the following points:
- The group has poor coverage of languages outside Europe. This and the lack of expertise in XML had prevented the formation of the panel. It was suggested that ICANN may have XML experts who could help. Chris would continue to reach out to language experts at conferences such as Unicode and Association of Asian Studies and colleagues were encouraged to do the same. Authors of key publications would also be approached.
- The group had not yet discovered characters in EGIDS 1-4 languages without pre-composed Unicode forms. These are likely to exist in African languages, Montenegrin, etc.
- Reasons for the exclusion of Unicode ranges Latin Ligatures and Full-width Latin letters from the MSR were elucidated – they are compatibility characters, not part of IDNA 2008.
- Code points required exclusively for Romanization and unofficial orthography (e.g. LATIN SMALL LETTER Z WITH STROKE for LATIN SMALL LETTER Z WITH DOT ABOVE in Polish) were likely to be excluded.
- It was noted that Cyrillic, Greek and Armenian shared characters with the Latin Script.
Some terminology in the report, e.g. letter and character, required attention.
Some issues remain: What constitutes an adequate attestation for a code point?
4. Inclusion criteria (e.g., EGIDS 1-4 and “in modern use”) were discussed. Some issues remain:
- How many speakers should a language have to be declared “in modern use”?
- What factors could exclude a code point which falls under the inclusion criteria?
The case of the EGIDS 5 language, Esperanto was mentioned.
The group had discovered some in-script variants (e.g., LATIN SMALL LETTER I, LATIN SMALL LETTER DOTLESS I and LATIN SMALL LETTER IOTA).
The implications of the following sentence in the Procedure were discussed: “Finally, in investigating the possible variant relations, Generation Panels should ignore cases where the relation is based exclusively on aspects of visual similarity.”
5. Options were discussed for cross-scripts variants:
- Declare them unilaterally.
- Confirm the declarations of the relevant panels.
6. There was a discussion about allocatable and blocked variants. The group was encouraged to follow the guidelines in the Procedure and declare the non-LDH variants as blocked.
7. 22 March at 17:00 UTC
