On 24/11/2017 12:18, J. Alan Bird wrote:
> (minor edit in attached version)
i'm attaching a couple of replacement images. One is easier to read on
the slide. The other is an update to the matrix showing some additional
colours: i updated Urdu, and added orange to Bengali and Devanagari for
places where grapheme clusters don't necessarily cover syllables and
that may affect the behaviour in that category (you don't need to
understand that ;-), but the Indian folks in cc probably will). The main
thing is that it adds a bit of colour, and shows that the indic stuff is
not nearly done.
You should, however, mention (and perhaps add a rider to the slide) to
say that these are only preliminary guesses, and we need experts to
thoroughly research the many unknowns and validate the current assumptions.
hth
ri
