Michael Everson wrote:

>
> At 16:17 -0500 2001-11-08, Peter T. Daniels wrote:
>
> > > This was actually argued out at some length not long ago on the Unicode
> >> mailing list. The upshot of it all was that yes, we know "ideograph" is
> >> wrong, but it's what's been used in the west for centuries and there
> >> really isn't a potential replacement which is significantly better and
> >> simple.
> >
> >Well, that certainly says something about the Unicode gang: use what's
> >convenient instead of what's right. I gather that's been the approach to
> >the whole project!
>
> That's not at all fair to "the Unicode gang", nor is it accurate, I
> am afraid. We strive to encode the world's writing systems accurately
> and correctly so that the world's data. We try to give characters
> good names. Sometimes it gets done wrong.

I recently found out that there's some sort of "free space" where anyone
can stick anything -- and someone has stuck in there the Dr. Seuss
"letters" from *On beyond Zebra* -- except that they are, obviously, not
"letters," but ligatures of various Roman letters, and they are,
obviously, not components of any sort of writing system. That's when I
became inclined to take Unicode less than seriously.

> My favourite bugaboo is
> LATIN LETTER OI, which if course is Turkic GHA. We had to use
> "Myanmar" instead of "Burmese" for political reasons. The first is an
> error, people naming the character without knowing what it was. The
> second is a practical compromise which got the Myanmar national body
> to sign up to the encoding more easily, and that was a good thing for
> millions of Burmese. Our approach is, indeed, to be practical. We are
> also adding in terms like abugida and abjad in the next version of
> the standard, I believe.
>
> >The notion of "ideogram" was debunked as long ago as 1838, by Peter
> >Stephen Duponceau, so there's really no excuse for its hanging on.
>
> For good or for ill, the standard uses, for various reasons, the term
> "ideograph". (**Not** "ideogram".) And the standard was approved a
> decade ago, and it's not possible to change certain aspects of it
> (such

(Why -graph rather than -gram? Isn't the interference from "phonograph"
'record player' enough to decide it?)

> That said, it should be pointed out that those responsible for the
> CJK characters in the standard are those who belong to the IRG, the
> Ideographic Rapporteur Group, which reports to ISO/IEC JTC1/SC2/WG2.
> This group is made up of the national bodies and/or industrial groups
> of China, Japan, North Korea, South Korea, Taiwan, Singapore, and the
> U.S. Work continues to encode missing Hanzi characters, whether they
> are denotated as "ideographs" or as "logograph" or whatever else one
> may prefer.

Well who invented their stupid name? If CJK characters encoded "ideas,"
Leibniz would have had his "universal language" centuries ago. As we've
just been discussing at sci.lang, "ideogram" puts bad ideas in people's
heads.
--
Peter T. Daniels grammatim@...