Re: [unicode] Character for generic human figure

2012/2/19 António Martins-Tuválkin <tuvalkin_at_gmail.com>:
> On 2012/2/18 suzuki toshiya <mpsuzuki_at_hiroshima-u.ac.jp> wrote:
>
>> Excuse me, you are asking for a character codepoint that
>> ISO/IEC 10646 or Unicode spec puts a glyph something look
>> like "generic human figure"?
>
> You’re excused. And, yes, that's what I’m asking.
>
>> I don't think these Hangul characters are intended to
>> be a generic human figure.
>
> Either your irony is too suble, or mine is.

Anyway, the Hangul characters that were cited are not supposed to look
like a person. This may be true with some styles, but you would be
very dissatisifed if those Hangul characters were (correctly) rendered
in a cursive handwritten style, notably those drawn with a brush and
real ink on paper by very respectable typographers and artists.

The UCS encodes in a single codepoints all allowed variations that
convey the intended semantic of the languages and scripts for which
those characters were encoded. Nothing else.

So unless these characters were intended and have been encoded to have
an geometrically exact glyph definitions, you cannot use them safely
for something else. This also explains why there are many characters
that "look" similar (even in their "representative" glyphs displayed
in chart maps).

And in my opinion, it is a good reason for why "emojis" have been
encoded separately, to avoid quirks such as "ASCII art" (like
emoticons, that are only meaningful in a too much restricted set of
possible renderings and page layouts, with encodings that have been
unified in usages for which they were initially intended): instead of
conveying the semantics of the component characters that were used in
ASCII art, another new character is encoded to convery the intended
semantics (and then a renderer may render them using ASCII art, or
icons, depending on the available fonts and technical constraints in
which it runs).
Received on Mon Feb 20 2012 - 16:44:36 CST