1. Attendance
2. Draft Panel Formation Proposal
a. Work schedule v.1 (.xls) (.pdf)
b. Language list v.4 (.doc) (.pdf)
3. A look at the Latin part of MSR2 with attestations v.6 (.docx) (.pdf)
4. AOB
5. DoNM
Notes
1. Ahmed Bakht, Chris Dillon, Sarmad Hussain, Mert Saka, Ngô Thanh Nhàn, Alireza Saleh and Mirjana Tasić
2. Chris had drafted some sections of the panel formation proposal. He used examples from European languages, some of which would be better replaced with examples from a wider selection of languages. He encouraged colleagues actively to participate in the drafting of further sections of the document, for example, descriptions of the use of the script in various parts of the world.
2a. Chris listed factors affecting the schedule, comparing it with the Arabic Generation Panel's: experience gained from the AGP (faster), larger repertoire (slower), smaller team (slower), etc. The schedule presumes that in-script variants need to be declared. If that is not necessary, several months may be saved.
This led to a discussion on whether variants should be defined for Latin.
The conclusion of Considerations in the used of the Latin script in variant internationalized top-level domains was that in-script variants should not be defined for Latin. There are some possible candidates for in-script variants, for example, LATIN SMALL LETTER A and LATIN SMALL LETTER ALPHA (U+0251 ɑ).
There is, however, evidence (e.g. the proposal for the Armenian LGR) that there is an expectation that cross-script variants should be defined. For Latin this would mean strict homoglyphs shared with Cyrillic, Greek, Armenian (possibly only two letters: O and H) and possibly other scripts. The letter A would be an example of a strict homoglyph with Cyrillic. Letters such as H and P are visually identical to Cyrillic letters, but are not variants, as they are actually other letters (pronounced /n/ and /R/ respectively). Moreover, an historical relation on its own is insufficient for letters to count as variants. For example, D and Cyrillic Д are related, but not variants, as there is little visual similarity.
2b. The working group thanked Mirjana for her Herculean work on the language list.
Some minor edits were suggested.
The large number of EGIDS level 5 languages means that code points only used in languages of level 5 or higher require substantial research before inclusion in our code point table.
3. Chris borought up the issue of the status of Romanizations. It is not clear, and so, until it is, code points in the Latin part of MSR2, if only used for Romanization, should be marked red in the list, i.e., requiring further research. Currently this affects several code points used for Pinyin, a common Romanization for Mandarin Chinese. It was agreed that the status of Romanizations would be on a list of questions for our next meeting with the Integration Panel.
There was a discussion about combining marks and precomposed forms. The Latin part of MSR2 includes some combining marks which may be included, probably with WLEs, in our code point list, if they are required for languages in modern use. The issue of the status of forms not in official orthographies was also raised. An example would be LATIN SMALL LETTER Z WITH STROKE which is used in Polish as an allographic variant of LATIN SMALL LETTER Z WITH DOT ABOVE. See: https://en.wikipedia.org/wiki/Z_with_stroke
4. Chris encouraged colleagues to send him short CVs, as the panel formation proposal requires CVs which are accessible to the public.
5. Chris is unable to chair the meeting on Jan. 19.