Hi Mirjana,

Attached is the promised set of test cases. I mostly focused on

a) Excluded glyphs -- to test that they actually get rejected

b) glyphs which are in MSR, but not in our repertoire. (I confess that I ran out of gas after the Latin Extensions A and B.) Again, to test that they actually get rejected.

In the course of this, I came across the Serer language (ISO 639-3 srr). It isn't in our list of languages included, but it is EGIDS 5 and has roughly 1.2 million native speakers so it would appear to fit our criteria. Also the Kpelle language (ISO 639-3 kpe), which is also EGIDS 5 and has about 1.4 million native speakers.

Adding Serer would entail adding 3 code points for Serer to the repertoire:

U 0188 Small Letter C with Hook ƈ

U 01A5 Small Letter P with Hook ƥ

U 01AD Small Letter T with Hook ƭ

For Kpelle we would also need to add:

U 0260 Small Letter G with Hook ɠ

We might pick up a couple of Variants as well. For example, a P with Hook (ƥ) resembles a Thorn (þ) far more than a P alone does.

Alternatively, of course, we could simply ignore them. And hope we can skate thru Public Comment without anybody noticing -- which, given the small number of ICANN members from West Africa, seems moderately likely.

Bill