Re: Re CJKV typographic variations.

From: Harry Weeks (
Date: Wed Jul 15 1998 - 18:49:38 EDT

> >The ISO/IEC 10646-1 standards document is quite good in showing
> >the variations among the various traditions for each character,
> >something unfortunately left out of the Unicode standards book.

> Actually, what 10646 shows is a "typical" glyph for each typographic
> tradition. This can also be misleading, as it give no indication of what
> is acceptable typographic practice. As a case in point, characters with
> the grass radical can have the radical drawn in either its three- or
> four-stroke form within Taiwanese typography. It all depends on the
> overall typeface style. 10646, however, consistently shows the grass
> radical one way for Taiwan and another for Japan, which creates a false
> impression.
> One reason why we put a visual form of the Unihan database on-line was to
> clarify this point.

Thank you for this clarification. I was not implying that the typographic
forms shown in the ISO/IEC document were necessarily comprehensive nor
even representative for a given tradition, but they do, generally, show
the principal variant forms that are to be found for a given character.
The Unicode on-line database includes glyphs from fonts encoded for CCCII
and for Big Five, providing two more possible alternative forms for the
traditional characters as used for Chinese. However, such a clarification
as yours is still required. For example, although the on-line examples
for the Big Five font appear to always show the "grass" radical with three
strokes, a Big Five font in the clerical script might very well exhibit
a four-stroke "gass" radical.

I was not intending to initiate a digressionary thread, only to point
out a ready hard-copy reference source illustrating some variant forms.

- - - - - - - -
Harry Weeks,
Entelech Consulting Corporation
Elmwood Station, Suite 5029
Berkeley, California 94705-0029 USA
Vox +1 510 548 5175
Fax +1 510 644 3060
- - - - - - - -

This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:40 EDT