I forgot to say one thing.
The Chinese GB character set should have a code point for what I called the
"dot-like hyphen". Peter, why don't you check what Unicode character the GB
character has been mapped to in the official Unicode mapping files?
(ftp://ftp.unicode.org/Public/MAPPINGS/EASTASIA/GB/GB12345.TXT)
That mapping could be sort of an implicit semi-official answer to your
question.
_ Marco
> -----Original Message-----
> From: Cimarosti Marco
> Sent: 1999 December 12, Sunday 18.27
> To: 'Chris Pratley'; Unicode List
> Subject: RE: EA width, Latin punctuation and fonts
>
> Chris Prately wrote:
> >Actually, Marco's explanation is for Japanese only. Peter, you are
> >describing a standard glyph variant for Traditional Chinese of
> IDEOGRAPHIC
> >FULL STOP (U+3002). The Traditional Chinese style is to centre the glyph
> in
> >the character box. The Japanese style is to place the glyph at the bottom
> >left in horizontal layout, and upper left in glyphs rotated for vertical
> >layout. The glyph does not need to move in Chinese vertical layout. The
> same
> >is true for IDEOGRAPHIC COMMA U+3001.
>
> No, it applies to both Chinese and Japanese (not sure about Korean or Yi).
> Both languages use both characters, for the same purposes.
>
> 1) CJK full stop.
>
> It has the same meaning as Western full stop, but it looks like an *empty*
> circle. In horizontal writing it is normally in the lower-left side of the
> square cell; it does not need a blank after it because it already has
> plenty white space on its right side. In vertical writing it is normally
> near the top of the cell and horizontally centered or, alternatively, on
> the left side. An alternative glyph, that is used in both horizontal and
> vertical writing, has the mark centered in the cell. As you suggest, all
> this is true also for commas (there are 2 in CJK) and most other
> punctuation marks ("!", "?", "...", ";", ":", etc.).
>
> --Encoding--. Definitely seems to me that it should be U+3002 (IDEOGRAPHIC
> FULL STOP). If Peter wants his font to be usable in both vertical and
> horizontal layout, he could decide to center the glyph in the cell.
> However this centered form is quite obsolete, especially in horizontal
> writing. The best thing, if possible, would be two have two contextual
> glyphs: one vertical (top, centered) and one horizontal (bottom, left).
> Same applies for commas and other punctuation.
>
> 2) Dot-like hyphen, or raised dot.
>
> It is used to separate words in foreign person and place names (as CJK
> scripts use no spaces between words). This is a *filled* circle and is
> *always* in the center of the cell, regardless that the text is horizontal
> or vertical.
>
> --Encoding--. Both U+00B7 (MIDDLE DOT) and U+30FB (KATAKANA MIDDLE DOT)
> seem reasonable candidates. However, in an average CJK font, U+00B7 would
> be half-width, so it would not be a handy choice (BTW, how does Unicode
> define this character now? Narrow, wide or ambiguous?). Moreover, the
> semantics and history of U+00B7 is totally unrelated with far eastern
> script (I for one tend to see it as a Catalan diacritic mark, as seen in
> the name of one of the most famous streets in Barcelona, the Avenida
> Paral·lel). I would rather suggest U+30FB because it seems to have exactly
> the required semantics, despite its misleading name, and despite the fact
> that it has been put in the katakana block, rather than in CJK Symbols and
> Punctuation, where it logically belonged.
>
> Chris Prately also wrote:
> >If you apply the right fonts to this text, you'll see what I mean:
> >æ-¥æoe¬èªzãEUR' (Japanese, display in Japanese Mincho)
> >ä¸-åoe<ãEUR' (Traditional Chinese, display in Ming Li)
> >The katakana middle dot is a Japanese thing unrelated to what you are
> seeing
> >in those Chinese documents.
>
> What I see in these two string is the CJK full stop. I thought that
> Peter's question was rather more about the other character being
> discussed.
>
> If my email does not kill the poor characters before they reach you, see
> examples of "dot-like hyphens" (or KATAKANA MIDDLE DOTs) as they would be
> actually used in Chinese and Japanese to transliterate a Western name and
> surname ("Marco Cimarosti", in this case :-)
>
> - Chinese (simplified): ???????? (Make Qimaluositi)
> - Japanese (katakana): ?????????? (Maako Timarosutei)
>
> Ciao. Marco
>
>
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:56 EDT