Jirimutu,
Yes, this is a very good point. I have this in my database, but had never pushed the envelope to see that the shaping without the NNBSP does present a problem. In the context of the NNBSP, the form is very predictable. Your OT rulings can handle it easily. However when the form is not suffixed with the NNBSP, then I don’t know how I would be able to form [cid:image001.png@01D0EB58.730A5210] properly. I think this may be important.
Greg
>>>>>
> Can you think of a case where one would actually spell a word CVxxxxxxx, where V=OE/UE,
> and the V takes the default shape without the long tooth ?
Yes, if we don't see the word ᠳ᠋ᠦᠭᠡᠷ( D+UE+G+E+R) without long tooth as the Mongolian Suffix, see it is one separated word,
It is the case. What I mean is we do not use NNBSP to control the form, but use the FVS to override the first syllable UE.
It is not very convincing example, but we cannot say that there will be no this needs been raised in the future in some of special requirements.
My opinion is we can ignore this kind of things to use other work around solution, not using extra override encoding to make things complicated.
Jirimutu
>>>>>
