From: "Peter Kirk" <[EMAIL PROTECTED]>In fact it should be considered a variant of g. Its pronunciation is that of small caps G (U+0262) or small caps inverted R (U+0281).
This letter looks nothing like U+01A3, which is also visible in the
bottom line of the attachment.
You exhibit a glyph that ressembles to the two ligated digits "01", where the second goes below the baseline. This type of glyph variant is well known in some fonts that do not display all digits aboge the base line (notably the 4, 7 and 9 digits often have descenders). The right part of the glyph in Arial Unicode MS is rounded rather than angular like a "1" for this second part. This can be seen as a matter of style. But the first time I saw this "LETTER OI", I first identified it as a variant of "g" or "9", then the Unicode name uggested it would read as a ligature of a "o" and a dotless "i" or iota...
Are we talking about two different characters? There is one in the bottom row of the picture I sent before, a consonant, which is U+01A2 (because it's a capital) and which looks very like the Unicode reference glyph, and not very different from the Arial Unicode MS and Code2000 glyphs; according to the data I have, this was used in Azerbaijani from 1923 to 1939. And there is one in the third row, right hand side (above M) which does not look like any Unicode reference glyph that I know of, and which is a vowel pronounced like Turkish dotless i; this was used from 1923 until 1933 when it was replaced by something looking very like U+042C, the Cyrillic soft sign.I don't know which language really uses this character, but given the various way it appears in fonts, your proposed character really is within the variations admitted in some fonts to represent U+01A3.
The representative glyph for this character seems to be good. But, given that the name is so misleading but cannot be changed, it is good that there is a note "= gha" in the Unicode character charts.If there's a bad name for "LETTER OI", then how can we interpret the Unicode assignment, if neither the representative glyph, not available glyphs in common fonts, nor the noramtive Unicode name will reveal abstract character identity?
But in the light of naming errors like this one implementers should be advised not to use character names, because they are not reliably helpful.
-- Peter Kirk [EMAIL PROTECTED] (personal) [EMAIL PROTECTED] (work) http://www.qaya.org/
