[sw-l] IMWA & Language-Specific Symbolsets

[Stuart Thiessen]

I think your specific symbol set is a good idea.  However, from what I
understand about Unicode and its philosophy, they do not group things
by language, but by writing system.  For example, even though most
European languages use the Latin alphabet plus a certain range of
diacritic markings, there is only one encoding for all of those
symbols. In other words, Unicode is a repository of symbols that any
language can utilize to express itself and everyone knows what symbol
you are talking about. If my understanding is correct, it would be
better as far as Unicode is concerned to have the whole IMWA.

Now that does not mean that we couldn't set things up so that we have a
designated range for IMWA based on a mapping like what Tomas suggested.
Instead of closing things off entirely, we could map some "blank rows"
in each group so that the Unicode set has some room to grow. It may
also be possible that we could set up different ranges to cover
SignWriting versus DanceWriting.  Even though the IMWA has it in one
numbering scheme, we could map the SignWriting from IMWA to a specific
Unicode range, and then map DanceWriting or any of the other movement
writing systems to another range in Unicode. As long as it doesn't
seriously affect the sorting process, it should not be a problem.

As far as the issue with rotations, it may be possible to treat
rotations within Unicode as a set of 8 "diacritics" which the renderer
processes to create the actual symbol like we would expect.  This can
get around Steve's concern about all the rotations being mapped to the
base symbol, while still keeping the actual count of symbols lower. It
shouldn't be too difficult to remap a two character sequence back to
the IMWA scheme. Theoretically, we could do the fills the same way, but
maybe not worth it.  This approach puts less work on the font side
itself and more work on the input process and the renderer.  Of course,
we can leave all the work to the font itself and just have a code point
for every symbol.  We just have to figure out what is best when the
time is right to begin working on Unicode.

Thanks,

Stuart

On Jun 22, 2005, at 14:32, Valerie Sutton wrote:

> SignWriting List
> June 22, 2005
>
>> Steve Slevinski wrote:
>> If we keep using the SSS ID numbers to identify the symbols, we'll
>> never have to close the IMWA. While that idea may make Val groan, the
>> flexibility it offers is worth it.
> Hello Steve, Tomas and Everyone!
> Just the opposite...I applaud the flexibility and need it and request
> it, at least for the next decade. If writing full dance, sports, mime
> and general gesture are to be programmed in the IMWA, as Stuart feels
> is a good idea, then I must have the time to enter all those large
> amounts of amazing details. So I need the flexibility.
>
> I know this can cause problems for those who want complete stability.
> There is one way to have BOTH worlds at the same time. Keep the
> general IMWA flexible and open for me to add more symbols..... but
> create language-specific symbolsets from the IMWA, for those signed
> languages that have been written for a long time. Those
> language-specific symbolsets would become rock solid and never change,
> even though new symbols are added into the IMWA monthly...they woudl
> remain stable.
>
> That is stability for specific SignWriting of Sign Languages, coupled
> with flexibility for the generic Movement Writing.
>
> That was my idea behind this web page:
>
> Language-Specific SymbolSets
> http://www.movementwriting.org/symbolbank/symbolbank.html
>
> The original SSS-US was my first attempt at this concept. I believe
> that that SSS-US could be improved now, by using a Symbol-Frequency
> test to find the specific symbols needed and then the revised SSS-US
> could become our first test-language-specific symbolset...
>
> By the way, the SSS means the Sequence of Sign-Symbols within the
> confine of one criteria (symbolset)...so the SSS-US is the sequence of
> Sign-Symbols used in sgn-US (American Sign Language).
>
> So I hope this helps you all see my personal vision on all this...That
> is why I suggested before, that once the SSS-US is finalized, we could
> start Unicode just for the SSS-US first, and that at least would give
> the world some Unicode for SignWriting, even though it is not the
> entire IMWA in Unicode...
>
> Val ;-)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/enriched
Size: 4418 bytes
Desc: not available
URL: <http://listserv.linguistlist.org/pipermail/sw-l/attachments/20050622/54e288b3/attachment.bin>