Re: [libreoffice-l10n] traditional Mongolian script support

On 10.01.2012 12:49, Michael Bauer wrote:
> Have you ever tried typing (Latin) text turned 90 degrees? Not really a
> sensible undertaking.

Actually, traditional Mongolian has nothing to do with Arabic :) So it's 
a joke, I think :)

> It would be good, perhaps, to hear what it would require to enable that
> sort of thing, both for typing and the UI, perhaps with a view of
> building a taskforce with Peter's help?

1. UI

I think, it is really can not be done (with contemporary UI toolkits). 
Each text-oriented UI element has horizontal orientation. Traditional 
Mongolian has not. End of story.

2. Typing.

2.1. ISO standards.

The first problem is that ISO standard is quite strange. ISO 639-2:

bua — Buriat
xal — Kalmyk, Oirat
mon, mn — Mongolian

but this classification almost irrelevant.

2.1.1. Buriat mongols (bua) writes in Cyrillic, live compactly, easiest 
case. At first glance. But they have great cultural heritage — 
chronicles and historical records in traditional Mongolian.

2.1.2. Kalmyk mongols (xal) writes in Cyrillic, and they are *one* of 
Oirates. Oirates live in Mongolia (write in Cyrillic, but uses Mongolian 
language), in Inner Mongolia (write traditional Mongolian), and in 
Russia, where they speak Kalmyk language and write in Cyrillic. How the 
standard's authors made one category for all Oirates? It's almost the 
same as to put all Slavic languages (Czech, Russian, Polish etc.) into 
one category. Even worse, 'cause Slavic languages use only «Western» 
alphabets, though different ones.

2.1.3. Mongolian (mon) people speak in many dialects of Mongol language. 
But they can be divided into two categories: people that use Cyrillic 
(Mongolia) and that use traditional Mongolian script (Mongolia, Inner 
Mongolia [China]).

And, after all, there are several different forms of Mongolian (худум 
бичиг for Mongolia, тодо бичиг for Oirates, now almost not used, and 
other). But all traditional Uyghur (not Arabic Uyghur, but Mongolian 
Uyghur), Mongolian, Oirates scripts share same features:

        * only vertical orientation
        * top-to-bottom, with rows left-to-right

2.2. Different scripting support vs. Languages.

As you can see, one «language» (in terms of ISO, and, as I understand 
right, in terms of LibreOffice) can use different scripts. These 
different scripts are not only different alphabets. These are quite 
different *languages*, that share many syntax rules, but use different 
grammatic. E.g., one word «a mighty warrior»:

        * in contemporary Mongolian: баатар, roughly == «baatr»
        * in translit from traditional Mongolian, it'd be «baghatur»

So, the spell checking would be different for these scripts, despite one 
«language».

The big problem here is that these scripts can be *mixed* in one 
document. I can easily imagine an attempt to create a materials for a 
school lesson in Mongolia, where all service text will be done in 
Cyrillic and exercises will be in traditional script (now they do a lot 
to revive traditional script in Mongolia).

I found a related work in AOO:

https://issues.apache.org/ooo/show_bug.cgi?id=91226

<skip />

Maybe, the «right thing» for me would be to look into the code. But I'm 
scared by C/C++ :) I write only in Python, and there can be not so much 
help from me, I'm afraid. But I'll try, anyway.


--
Peter V. Saveliev

--
Unsubscribe instructions: E-mail to l10n+help@global.libreoffice.org
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://listarchives.libreoffice.org/global/l10n/
All messages sent to this list will be publicly archived and cannot be deleted