SW and Unicode
sutton at SIGNWRITING.ORG
Sun Jun 22 19:50:44 UTC 2003
June 22, 2003
Dear SW List and Antonio Carlos...
Thanks for this input...Yes...I agree that the numbering system we are
using right now seems to be working and frankly, re-vamping all of
those number codes to fit with the UNICODE standard is very expensive
and time-consuming. Maybe later, if Stuart Thiessen and his
organization can raise the funds ...but it is not mandatory right now...
BUT...that is because of your SWML, Antonio Carlos. Without SWML, we
would have nothing else to turn to, to bridge into the bigger world of
programming...so thanks to you, we can move forward...
Can we focus work on either an SWML conversion program, or a direct
conversion from SignWriter Java, sometime in the next year? That will
help us make the bridge to SignBank. Once I know how that will work,
then we can make the necessary changes to SignBank, to make this all
work in the background...I am hoping to include all SignWriting
software together on one CD in time...so we need to create bridges
between the programs...
I look forward to seeing you in a week!
On Sunday, June 22, 2003, at 04:09 AM, Antonio Carlos da Rocha Costa
> I'm afraid I have to agree with Dan, for yet another reason:
> Valerie's encoding of SW symbols, with that series of numbers
> (category,group,etc.),which evolved from Rich Gleaves original
> encoding of SW symbols for the SignWriter program, is very
> practical in helping programs to understand the symbols, to know
> for example what transformations (rotations, flops) can be done
> with each of them.
> Perhaps Unicode can encode symbols in a way that keep the
> codes as meaningful as those ones, I don't know. But I feel that
> something similar to the original Rich's encoding is mandatory.
> SWML keeps that encoding as it is, and our experience with
> writing programs based on it reinforces that impression.
> But perhaps I'm wrong, and Unicode can work that way. That would
> be great, because it could put SW within the important industrial
> standard that it represents.
> All the best,
> Ant ? io Carlos
>>> Our first project, however, will be working with Valerie and others
>>> the process of getting SignWriting into Unicode and dealing with
>>> rendering issues for Signwriting in Unicode.
>> I keep hearing this idea thrown around, so here's where I run aground
>> when I think about it. Please help me out.
>> Since Unicode is a 2-byte code for representing linear text (i.e., all
>> rotations, placements, etc. are to handled algorithmically), I'm not
>> sure there is any provision within Unicode for representing
>> at any level beyond basic glyphs. AFAIK, there are no native
>> for placement on a grid, rotation, etc. And if there were, all that
>> would have to converted to a some sort of scale relative to the font
>> size of the glyphs, which would mean some sort of floating-point
>> representation (IEEE?). I'm willing to bet that SWML is probably the
>> best vehicle for representing the whole shebang; furthermore, it
>> probably ought to be one of the formats that SWJava can handle without
>> any external conversion (i.e., it ought to work within "Save As...").
>> What did you and yours have in mind?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 3355 bytes
Desc: not available
More information about the Sw-l