[9]
ジョージア文字は1つの均質な用字系ではなく、少なくても4種類の文字の体系があります。
大文字と小文字の関係性と似ていますが、事情はかなり複雑です。
[10]
Unicode 以前の7ビット符号や8ビット符号は、
1組だけ符号化し、フォントの切り替えによって対応してきました。
[11] Unicode は当初2種類を用意していました。 Unicode 側は統合されているのでフォントの切り替え等で対応できるという立場でしたが、 ジョージアの実情には合わず無視されていました。
[12] 現代に大文字に相当して使われるものが Unicode に存在しなかった (と理解された) ため、現代で使わない古い大文字 (に相当するもの) の符号位置を使ったり、 フォント切り替えによったり、 PUA を使ったりと、 フォントごとに異なる対処をしていました。
[13] 平成時代末期になってようやく Unicode は方針転換し、 現代の大文字相当のものにも独立の符号位置が与えられました。
[2] console-setup Georgian fonts, need test. მიშველეთ გეთაყვა.[BPG&GIA], BPG Kolumbus, 2007/06/14 22:59:50, https://groups.google.com/g/geofonts/c/lxme64FtS60/m/I_Afcz9SicwJ
4. Representation of the "de facto" (not existed in Unicode *) Capital/Title case "Mkhedruli Mtavruli" instead of practically unused anient "Asomtavruli" **
* Unocode standard for Georgian now is not up to date comparing to new trends in Georgian mass media and the web - sharply influenced by Western media
** Example of extensive usage of Geogian "Mkhedruli Capitals" in the range of Unicode old "Asomtavruli" is the web site of the Parliament of Georgia where Georgian "Capitals" are used same as in English and other Latin based scriptures. See: http://www.parliament.ge
[2] console-setup Georgian fonts, need test. მიშველეთ გეთაყვა.[BPG&GIA], Ben Laenen, 2007/06/16 5:09:04, https://groups.google.com/g/geofonts/c/lxme64FtS60/m/HdVqMdnDJJUJ
* Georgian fonts "hijacked" the Asomtavruli code points since that script isn't used anymore (due to the lack of the feature mentioned above). They even used code points that aren't defined by Unicode.
* Not following the standards is *really* bad. Using undefined code points is even worse, because those can lead to rendering problems.
So now we are between two walls: as a Unicode font we can't afford to put Mtavruli in the area of Asomtavruli, since the latter may be added in a distant future, and then we would really have a problem.
On the other hand, we cannot deny the current usage. I could evangelize a lot here about how bad it is, it doesn't take away the fact that numerous text documents already use the Asomtavruli code points for Mtavruli.
In short: Unicode has a serious problem here, because a variant of the standard is in use now, and I really doubt Unicode will change that standard, unless some serious work is done by Georgian people for that to happen, with a *lot* of good arguments why to change the code points (mentioning current usage is different from the standard would be a good argument, I'd think, but it's going to need a lot more than that)... I'd like to get the opinion of other people that know about the issue first as well before taking a decision on what to do with Mtavruli.
[4] Re: Georgian range for "DejaVuSerif" by BPG-InfoTech, BPG Kolumbus, 2007/09/08 23:57:12, https://groups.google.com/g/geofonts/c/gAhM_5JP2GI/m/Sbv7L1gejUQJ
2. On Georgian Parliament web site they are using "BPG Glaho" where Mkhedruli Mtavruli is inside range of AsoMtavruli (and are using Dynamic, Web Font for MS Internet Eplorer) -- here is font whuch can solve this problem for you.
Actually - De-Facto standard in Georgia is Mkhedruli Mtavruli in the range of AsoMtavruli of Unicode - this is a case when practice9 practical necessity commands...
[5] Re: Georgian range for "DejaVuSerif" by BPG-InfoTech, BPG Kolumbus, 2007/09/18 9:40:03, https://groups.google.com/g/geofonts/c/gAhM_5JP2GI/m/OvVosuE5HswJ
Here are some other examples of using "Mkhedruli Mtavruli" on Georgian governmental (!) web sites:
Central Election Commission web site http://www.cec.gov.ge/
Ruling (Pres. Saakashvili's) "United National Movement" web site http://www.unm.ge/
[7] Georgian Supplement 2D00-2D2F for DejaVu Sans, https://groups.google.com/g/geofonts/c/dV4Ls_pYvcA/m/lsTa39hy9ckJ
[6] ახლა მაქვს „უბუნტუ გნომით“ და გადავხედავ ბპგ დეჟავუ-ს, https://groups.google.com/g/geofonts/c/tI-lwjIJtQc/m/R3-c9HKwFNwJ
[3] DejaVu - BPG Georgian 2009 Version, Besarion Gugushvili, 2009/02/21 16:43:50, https://groups.google.com/g/geofonts/c/ac5MG0GzuOY/m/pavKACA-YuYJ
Includes: Unicode Standard • Asomtavruli (Archaic) - Unicode Standard 10A0-10C5 • Mkhedruli (Modern) - Unicode Standard 10D0-10FC • Nuskha-Khutsuri (Archaic) - Unicode Standard 2D00-2D25 Private • Uppercase/Titlecase (Modern) - Non Unicode Standard, Private Use Area (PUA) F400-F426 • Dilaectics and AddOns - Non Unicode Standard, Private Use Area (PUA) F428-F447 Nomenclature: Unicode Standard 10A0-10C5 ႠႡႢႣႤႥႦႧႨႩႪႫႬႭႮႯႰႱႲႳႴႵႶႷႸႹႺႻႼႽႾႿჀჁჂჃჄჅ 10D0-10FC აბგდევზთიკლმნოპჟრსტუფქღყშჩცძწჭხჯჰჱჲჳჴჵჶჷჸჹჺ჻ჼ 2D00-2D25 ⴀⴁⴂⴃⴄⴅⴆⴇⴈⴉⴊⴋⴌⴍⴎⴏⴐⴑⴒⴓⴔⴕⴖⴗⴘⴙⴚⴛⴜⴝⴞⴟⴠⴡⴢⴣⴤⴥ Private F400-F426 F428-F447
[8] n47xx-georgian.qxp_n3531-oldhungarian - 16081-n4712-georgian.pdf, , https://www.unicode.org/L2/L2016/16081-n4712-georgian.pdf#page=7
And then there are the semi-UCS fonts. BPG’s
“Paata New” font gives Mkhedruli glyphs at 10D0..10FF and Mtavruli glyphs at 10A0..10CF. And yet BPG’s “Paata New Caps” fonts gives gives Mtavruli glyphs at both 10D0..10FF and at 10A0..10CF. The implications are clear: In both of the Paata fonts Mtavruli has been encoded as glyphs for Asomtavruli uppercase, and in one of them, Mtavruli has also been encoded as glyphs for Mkhedruli small letters. Yes, this is the wrong thing to do. But since the encoding fails to support Mtavruli, this is BPG’s workaround. A number of
UCS fonts exist for Mkhedruli, but Mtavruli may be encoded as Mkhedruli, or as Asomtavruli, or indeed in the PUA. Michael Everson
recently received the text of both Alice and Looking-Glass by prize-winning translator Gia Gokieli—the text was in a Latin 8-bit-based encoding and had to be transformed to UCS text. The issue of Mtavruli also occurs in the text, where Carroll’s use of all caps in “ORANGE MARME- LADE” and “DRINK ME” and “W. RABBIT” and “THE KING AND QUEEN OF HEARTS” had to be kludged in the typesetting. The only relevant
OpenType options, it would appear, would be character variants or a stylistic set. Neither of these options is particularly portable, however. Monotype GmbH is the one and only font developer we are aware of which has added OpenType features which mimic the ordinary uppercase/lowercase relationship behaviour, but obviously this does not work in all environments and is not a plain-text solution.
[15]
ジョージア文字の配列順序は少なくても2つあります。
[17] 付の Unicode の原案には、 MKHEDRULI が現在用いられており、 KHUTSURI は既に用いられていないとして、 MKHEDRULI 用の1組のみが含まれています。 >>16
[18] 配置は、
となっており >>16、 GEORGIAN-ACADEMY と同じ現代文字を優先する並べ方です。
[19] これ以上の詳細は不明ですが、古い文字は GEORGIAN-ACADEMY と同じ文字数なので、 順序も同じでしょうか。
[20]
句読点は1文字のみなので、その後の版と比べると჻
と推測されます。
前後に隙間がある不思議な位置に置かれていますが、
その後の追加や既存の8ビット符号と比べても、
この位置とした理由はよくわかりません。
[21] このジョージア文字の出典について、興味深い記述があります。 >>16
Standards: 2nd DP ISO 10646: Unicode departs from the 10646 arrangement for Georgian. In
Unicode, the archaic letters are placed together in a group after the modern letters. In 10646,
these two groups of letters are sorted together (in an order that is open to question).
[22] つまり ISO/IEC 10646 草案が出典とされていながら、 敢えて文字の位置を入れ替えて、 新旧文字両方を含む並び順から、現在の文字の後に古い文字を置く並び順に変更したというのです。
[23] すなわち同時期の ISO/IEC 10646 案では現行配置と異なる、 新旧文字両方を1つに並べた順序を採用していたということです。
[24] するといくつか疑問が出てきます。
[30] Unicode と統合された DIS 10646-1.2:1992 では、 2組が配置された現行のものになっています。 >>29
#
[1] FAQ, https://www.georgian-language.com/learn_georgian/faq.php
[31] CodeCharts2.pdf, , https://www.unicode.org/versions/Unicode1.0.0/CodeCharts2.pdf#page=104