Excellent question and points from Albrecht Dreiheller.
[AD:]
So the _receptive vocabulary_ might be pretty big for many people.
[...]
So the _productive vocabulary_ of symbols will always be very, very small.
I was thinking a similar thing, and I'm inclined to agree.
But I know of parallel Chinese input programs that let the user mix
different input methods. I have seen people use input software that lets
them mix in English expressions, and the underlying algorithms seem to
work impressively well. So the user may be able to intersperse a string
of pinyin or Cangjie with English words like "philosophy" (with the
letters typed in just like that), and the input method will display or
let the user select 哲學/哲学. Not surprisingly, such input methods are
popular among children who have spent some time abroad, perhaps in a
boarding school; this will let them continue to type (slowly) despite a
rather passive knowledge of Chinese characters.
It is not hard to imagine a relatively trivial search-and-input
mechanism for a vocabulary of symbols.
I am certainly not suggesting that one should encode all symbols
(whatever that might mean), and I am for sure suspicious of /some/
projects mentioned on this mailing list (though note that not every such
project is asking for encoding, an issue dependent on the respective
project's putative value but therefore also posterior to any potential
success and hence likely moot in many cases), but I really do not see
why a comprehensive effort to classify symbols wouldn't be worthwhile.
If one ends up with a semantic net that looks like an elaborately
layered multitree – well, at least that's interesting.
[KW:]
Essentially we are talking
about entirely different *modes* of symbolic communication here, and
they cannot (imo) be mapped one-to-one into character encodings.
Any particular collection of such signs starts off with the obvious symbols
that are *already* encoded as characters, shades off into stuff we could
have arguments about appropriateness for symbols encoded as characters,
and there very quickly veers into total pictorial crazyland.
[...]
Where do you start? Where do you stop? I contend there is not and cannot
be any definitive answer.
This is somewhat analogous to any natural language's vocabulary. There
is a place for dictionaries, and I can see an evolution of symbols, with
a changing distribution over time.
Whether this works is another question and doubtful, given that the
representations of the two types of natural languages I'm aware of being
alive (ie: spoken and signed, with writing being essentially derived
from the former) seem to work with morphemes, words, and a syntax. It's
unclear how a proliferation of signs would fit in, though I concede that
there might be territory to explore.
Stephan
