Re: Missing geometric shapes (vertical text)

From: Philippe Verdy <>
Date: Mon, 12 Nov 2012 20:08:16 +0100

It is a good question when many of these pictograms (including basic
geometric shapes) are coming from former CJK encodings, where vertical
layout is common, AND the ideaphic composition square facilitates a lot
their insertion (and it is the major reason why many emojis were highly
developed there and then encoded, even if they are used, since long, but in
a very different visible form, in alphabetic scripts, using "ASCII art" as

Generally emoticons won't be rotated, so I would expect those stars not to
be rotated as well : if the star does nt rotate, would the the half cut for
showing the half-filled star rotate ? If not, would it keep its mening for
representing an additional half when presenting vertically a series of
stars terminated by an "half star" ?

As you can see, such encodings of pictograms would open a large Pandora's
Box. Unless there's a proven existence of a domain of usage where it is
consistant and well understood, we should not encode them, simply because
they break the basic principals of the UCS standards, which is to separate
characters from styles, and characters from glyphs.

I see this half star for now as a dangerous precedent if it was encoded. Or
its encoding is premature.

For me, encoding this type of symbol is a "would-be-cool-if". It is not
even needed as there are plethoras of symbols and pictograms used (and
constantly invented) for similar usage (but always without any rule of
consistency), that have never needed any sort of encoding as abstract
characters. The UCS should only be there to endorse common usages or proven
standards and norms where their "abstract" identity is defined exactly by
this usage (for an implied common meaning).

2012/11/12 Jean-François Colson <>

> Le 11/11/12 23:25, Frédéric Grosshans a écrit :
>> Le 11/11/2012 23:08, Doug Ewell a écrit :
>>> Personal opinions follow.
>>> It looks like the only actual use case we have, exemplified by the xkcd
>>> strip, is for a star with the left half black and the right half white.
>>> There *might* also be a case for the left-white, right-black star.
>> What is missing in the attachment of Simon Montagu's email
>>**arch/unicode-ml/y2012-m11/**0024.html<>to make it a convincing case for the left-white, right-black star ?
>> I wonder whether similar half-filled stars would be required for
> vertically written text.
> Would a star black above, white below, be required for vertically written
> Japanese, Mongolian, Sutton, Tangut, Phags-pa, etc.?
> Would a star white above, black below, be required for Batak and Hanuno’o?
Received on Mon Nov 12 2012 - 13:11:55 CST

This archive was generated by hypermail 2.2.0 : Mon Nov 12 2012 - 13:11:56 CST