Path: utzoo!attcan!uunet!ginosko!uakari.primate.wisc.edu!ctrsol!cica!iuvax!rutgers!columbia!cs!d-yang
From: d-yang@cs.columbia.edu (David Yang)
Newsgroups: comp.cog-eng
Subject: Re: Re^2: What to know & universal icons
Summary: Chinese radicals
Message-ID: <345@cs.columbia.edu>
Date: 24 Sep 89 18:46:37 GMT
References: <768@cogsci.ucsd.EDU> <3490@rtech.rtech.com> <446@uwslh.UUCP>
Distribution: comp.cog-eng
Organization: Columbia University Department of Computer Science
Lines: 39

(preface:  The following is based on what I know from experiences with my
(warning?) parents, who speak and write Chinese, and a few years of classes
	   plus informal learning.)

In article <446@uwslh.UUCP>, lishka@uwslh.UUCP (Brain-fried after too much hacking) writes:
>      No.  *Written* Chinese is composed of tens of thousands of
> "characters" (i.e. symbols).  Most of the time, these characters have

I believe Robert Brady was referring to the radicals that make up
the tens of thousands of characters.  The radicals do give hints in meaning
(like knowing the roots of words) and pronunciation and help in both memorizing
characters and telling someone how to write a character, but the translation
is still non-trivial.
  Also, though this may not be absolutely necessary, it seems that the icons
people are thinking of are hopefully something almost anyone would approximate
if asked to come up with an icon for the same thing.  But Chinese characters
still seem to be just memorized.

>      Furthermore, whereas current icons graphically represent ideas
> with fairly "understanable" pictures, Japanese/Chinese characters do
> not pictorally represent their respective ideas anymore.  Even though

This is true-- actually, I'm not that sure the original characters were
all that understandable.  The character for big (da1) was supposed to
represent a big man, which doesn't seem inherently universal.

>      I prefer icons that are not tied to *any language*.  Certainly
> all icons are somehow tied to some culture(s).  I do not think using
> Japanese/Chinese characters as "standard" icons would work, although
> it is an interesting idea. 

I agree-- e.g., one might be able to pick up that a big man represents
"big," but I think this requires some knowledge of the way Chinese words
are built up to know that the character didn't mean, say, "adult."
And I think you might still have problems with the meanings of symbols
in different contexts.

David
d-yang@cs.columbia.edu