In defence of philosophical languages (was: RE: Comparison ofphilosophical languages

>Absolutely. However, the different part of the word must be *different*,
>or even *unique*. The shape of the word should serve as a *mnemonic*, not
>a *taxonomic definition*, of what it refers to. I like your term,
>iconicity. The brain seems to like to associate icons with meanings; but
>it is the uniqueness of the icon, what makes it stand out, that defines
>its usefulness. Suppose there were 15 operating systems with logos
>identical to MS Windows except with different colors on the four panes.
>The mind would have great difficulty sorting them apart. However, as we
>see today, each operating system carries a very different, unique logo,
>which is memorable by the very fact that it is different from the rest.
>
>Of course, we could go one step further, and say, hypothetically, that all
>OS logos share some common characteristic, such as the letters "OS"
>somewhere in the logo. This would correspond to the "similar" part of a
>word, and the rest of the logo would correspond to the "different" part of
>the word.
>
>I guess what I'm trying to say is that one should not represent more than
>one level of taxonomic hierarchy in a word. I'd even posit that the
>"different" part of the word ought to constitute its bulk, and the
>"similar" part constitute a minor addendum to hint at its possible
>relations with other words having the same "similar" part. This is the
>difference I'm trying to draw between OS icons that all derive from MS
>Window's icon, and OS icons that have the letters "OS" in common but are
>otherwise completely unique. I posit that the latter is much easier to
>learn, and much more memorable than the former.
U know what this reminds me about? Chinese writing. Basicly, the Hanzi works this way.
And u can see it in a double way :

1st way(semantic):
most characters are based on a radical carrying a general cathegory (semantic
mnemonic). metals include the metal radical, birds include the bird radical,
diseases have also a common part, and so on (there are a bit nore than 200
radicals). If the system had been devised in a technological time, you would
maybe have a radical for OS, or more likely a general one like software, or
even computer. the other part of the characters makes the diversity. from a
semantic point of view, this second half is almost random.

2way:
although on a semantic point of view the second part is meaningless, it is actualy not
random. because chinese writing system refers to a spoken language, there are
also references to pronounciation. the second part is a tip (not an exact
description) on how to pronounce the character. this part you put next (or in,
or under...) to the radical is actualy another charcter with a similar (not
necesarly identical) pronounciation. so in this point of view, the non radical
part is the mnemonic, and the radical creates diversity.

so of course, you can not know the meaning of all characters just by logic. but
when you know quite a lot of them, and you encounter a new one, you not only
have the context to help you understand, but also the two tips of the
character.

when you try to guess, you can think like : it is a word in the field of XXX which
sounds similar to YYY.

an example out of chinese : it is a word in the field of disease/health problem,
which sounds like winter (dong) => teng (pain/painfull)

let's make imaginary ones in english :
*related to "water", sounding like "she" could be "sea"
*related to "people/human", sounding like "sun" could be son

again this system does no pretend to be completly explicit, but give some help.