FW: FVS Assignment Mismatch WrapUp

Thanks again Siqin for your input.

Here are a few comments on Siqin's 6 examples:

MediGa01 - By the rules in the OverRide paper submitted earlier plus Siqin's notes on the S/D + Medial GA case, this does not need the over-ride FVS as it is a medial GA preceded by a SA.
MediGa02 - By the rules in the OverRide paper submitted earlier plus Siqin's notes on the S/D + Medial GA case, this does not need the over-ride FVS as it is a medial GA preceded by a DA.
MediGa03 - ADGAU/ADGAQU - by Professor Quejingzhabu's rulings (see my OverRide rulings) the GA should be undotted after DA/SA. In my opinion, this is a mis-spelling in the dictionary. Badral, Erdenechimeg, Siqin, and others, could you comment on this spelling?
Siqin, if we need to over-ride context to get the medial GA dotted, we already have the FVS1 firmly assigned there, so that should not be a problem.

FinalGa01 - Siqin, could you look at this example again. I am not sure how it applies.
FinalGa02 - I know that Baiti would pass this as a masculine word and therefore would not need the over-ride FVS. We need a foreign word that has no masculine vowels (AOU). Can others comment on how their font would handle this text sequence?
FinalGa03 - QODAL SIG - Good, we can use this example

We have a strong case for the Final GA over-ride. The FVS assignment has been on record since the TR170 plus the MGWBM. Plus we have examples. More examples are welcome.

I am still not sure that we have a case for the Medial GA undotted over-ride FVS. More thoughts here ... ?

Greg


From: siqin [mailto:siqin@almas.co.jp]
Sent: Monday, August 24, 2015 11:29 AM
To: Greg Eck <greck@postone.net>; public-i18n-mongolian@w3.org
Subject: Re: FVS Assignment Mismatch WrapUp

Hi Greg,

*         182D Medial - given the case where the contextual rules for the dual dots must be over-ridden. In other words, the context dictates that the medial GA is dotted, however, the actual shaping of the word is desired without the dots. I have not had the time to track down examples for this.
    I did not face with this case in my font implementation experiment.
    It may be :
    There is a grammar rule which the two dots will be omitted if g(182D) follows s(1830) and d(1833) in Traditional Mongolian.
    ( The most dictionaries spell it as QA and read it as GA. )
        medi_ga_exception1.png
        medi_ga_exception2.png
    But there is a exception
        medi_ga_exception3.png
    So, if over-ridden is needed, the doted GA, not the undoted one. I think.
*         182D Final - given the case where the feminine final GA does not follow the common pattern of sweeping to the left, but however sweeps to the right. In other words, the word is composed of feminine vowels, but carries a masculine right-ward swept tail. From discussions with Professor Quejngzhabu, I understand that there are just a small subset of words (5-6 in quantity) that follow this pattern.
    See
        final_ga_exception1.png
        final_ga_exception2.png
        final_ga_exception3.png (?)

SiqinBilige
On 2015/08/24 0:19, Greg Eck wrote:

I am ready to wrap up the discussion on FVS Assignment Mismatch.



However I am still lacking good examples on two of the over-rides discussed ...

*         182D Medial - given the case where the contextual rules for the dual dots must be over-ridden. In other words, the context dictates that the medial GA is dotted, however, the actual shaping of the word is desired without the dots. I have not had the time to track down examples for this.

*         182D Final - given the case where the feminine final GA does not follow the common pattern of sweeping to the left, but however sweeps to the right. In other words, the word is composed of feminine vowels, but carries a masculine right-ward swept tail. From discussions with Professor Quejngzhabu, I understand that there are just a small subset of words (5-6 in quantity) that follow this pattern.

*         I am attaching two files showing data sets for the non-over-ride cases here.



Erdenechimeg, Siqin, I wonder if you or others can help find some good examples that we can state in this regard? Your examples before were so helpful. We have some good examples for the 1822 medial single-tooth over-ride with NAIMA ("eight"). Also, we have a good set with the 1828 undotted medial over-ride. But we are still lacking for the two cases of the 182D GA as listed above. Anything we can document here will be helpful.



Thanks,
Greg

PS Our next topic will be Isolates - an exhaustive overview





-----Original Message-----
From: Greg Eck [mailto:greck@postone.net]
Sent: Friday, August 14, 2015 6:14 PM
To: Richard Wordingham <richard.wordingham@ntlworld.com><mailto:richard.wordingham@ntlworld.com>; public-i18n-mongolian@w3.org<mailto:public-i18n-mongolian@w3.org>
Subject: RE: Reference Scheme for Mongolian Rendering



Hi Richard,



Attached please find the rules for the four over-rides.

I did this a bit fast, everyone please look over carefully to see if I made a mistake.



Thanks,

Greg





-----Original Message-----

From: Richard Wordingham [mailto:richard.wordingham@ntlworld.com]

Sent: Friday, August 14, 2015 8:56 AM

To: public-i18n-mongolian@w3.org<mailto:public-i18n-mongolian@w3.org>

Subject: Reference Scheme for Mongolian Rendering



Looking at Greg's list of data sets (DS...) in his post of Saturday 8th August ('Mongolian Variation Sequences Missing from Unicode 8.00 Code Chart', http://lists.w3.org/Archives/Public/public-i18n-mongolian/2015JulSep/0248.html

), we are missing two important items:



1) A reference scheme for rendering.  I offer one in the attachment rendering_framework.odt.



2) The rules for contextual forms that may be overridden by variation selectors.  Without these rules, we do not know whether we have an adequate set of variation selectors for rendering connected text.



I am trying to identify the contextual rules, though I am not the best person for the job.  NNBSP has me worried.  Do we need to identify suffix rules for every language that might conceivably be written in the Mongolian script with separated suffixes?



Richard.