RE: Two Final Threads - Diphthongs / Final glyph checks

Hi Jargal,

Thanks for the good comments.

We have discussed some of these issues already but not in great length as our objective has been mainly to get font developers to talk together in order to attain more uniformity and ultimately an agreed-upon standard Unicode Standardized Variants Listing (USVL).

There was a discussion on character folding as led by Richard Wordingham some time back related to your comments on helehö/helehu, moNgul/moNgol, and ayil/ail. You can find this in the archives.

There have been allusions to the need for smart IMEs. I think we are all in agreement, that they are essential to the use and adoption of the vertical Mongolian script.

Jirimutu has led the discussion for the most part on diphthongs, so you can follow his posts in this area.

There has been discussion also dealing with the rule taught in early grammar lessons that the U+1823/U+1825 in most, if not all cases, is not found past the first syllable.

We welcome your hearty participation,
Greg

>>>>>
From: Jargal [mailto:bjargal@mail.ru]
To: Greg Eck <greck@postone.net>
Subject: Re: Two Final Threads - Diphthongs / Final glyph checks

I have joined the discussion only a couple of days ago. Still looking through to grasp the tails.

I would suggest dividing input and actual representation (both visual and internal (i.e. encoding level) representations).
The input can be done with IMEs. But the main feature of mongolian script which doesn't normally allow two or more vowels to co-occur without a consonant inserted in-between is to be followed. This for example resulted in transcription of Ch. yuan as WMo yuwan, etc.
I think that it might have profound influence on the way future generations think about the nature of Mongolian Script.
There is a long and established tradition in Mongolian philology to interpret Mongolian diphthongs as ayi, eyi, etc. in non-final and as ai, ei, etc. in word final positions. Hence noqai-yin and not noqay-un. It makes sense to stick to this elaborate interpretation of a scholarly tradition. Because in other positions you have to follow the 'consonant insertion' rule as in qariy_a, gabiy_a (the underscore stands for MVS), or above mentioned yuwan. So to keep consistency of interpretation it is good to interpret diphthongs as VCV clusters.

Summing up I would suggest to interpret the diphthongs as ayi, eyi, oyi, uyi, üyi in non-final position and as ai, ei, ui, üi in final position.
üyi should be dealt with OpenType rules  which would yield a positional variant with two 'long teeth' instead of three.
For the sake of consistency it is good to recommend users to type all non first syllable 'gedes' as u or ü depending on word's vowel harmony.

The raw input will be a bit difficult for most people and it is normal for the script as complex as Mongolian. To make it easier and faster it will be great to have an IME which produces consistently encoded sequences according to a standard. For example both helehö and helehu would result in 182C 1821 182F 1821 182C 1826, moNgul and moNgol in 182E 1823 1829 182D 1824 182F, ayil and ail in 1820 1836 1822 182F.

By interpreting all non first syllable 'gedes' as u or ü we make it much easier to type without IMEs (in case it sometimes might be necessary) resulting in a consistent output which ultimately improves end-user's searching experience.

Best regards,
Jargal Badagarov
>>>>>