[sw-l] FOR SANDY: New Fan Handshapes

Sandy Fleming sandy at FLEIMIN.DEMON.CO.UK
Wed Nov 24 17:56:53 UTC 2004


Hi Charles!

Yes, it seems to me that the main point is that it would no longer be
necessary for Val to create the graphics individually.

Another useful thing would be that programmers would have the choice between
having their program generate the graphics from the data dynamically or just
using a fixed set of gifs. Also that if, in future, there was some
unexpected development regarding the preferred appearance of handshapes in
general, the whole IMWA could be re-issued by changing the relevant
algorithms rather than having to do a whole set of new gifs by hand.

My data representation (which, incidentally, arose from studying the
handshapes of SW rather than being invented from scratch) involves making up
a table with a row and column for each finger and listing the properties of
each finger in the diagonal cells and properties describing how each _pair_
of fingers relate to each other in the other cells. Here's an example of the
"Spock" handshape, which is simple (use a fixed font to read this!):

           Thumb      Index      Middle      Ring      Baby
Thumb      straight   spread
Index                 straight   unit
Middle                           straight    spread
Ring                                         straight  unit
Baby                                                   straight

Reading the diagonal first, this says "all the fingers are straight". The
Thumb-Index cell says "the thumb and index are spread apart from each
other", the Index-Middle cell says "the index and middle fingers are a unit
(ie they're together), and so on.

Although this is a simple example, more complicated handshapes are just as
easy to work with once a system of values for the cells has been devised.

Note that it's not necessary to list any entries below the diagonal as they
relate precisely to the corresponding entries mirrored in the diagonal (eg
if the pad of the thumb is touching the nail of the middle finger then the
nail of the middle finger must be touching the thumb, or if the middle
finger crosses over the index finger then the index finger must cross under
the middle finger).

Other values for the diagonal can be cup, bent, hinge &c. Other values for
the cells can describe where one finger touches another and so on. So you
might, in a more complicated handshape, express that the tip of the index
finger touches the middle bone of the middle finger &c.

Sandy

-----Original Message-----
From: owner-sw-l at majordomo.valenciacc.edu
[mailto:owner-sw-l at majordomo.valenciacc.edu]On Behalf Of Charles Butler
Sent: 24 November 2004 16:46
To: sw-l at majordomo.valenciacc.edu
Subject: RE: [sw-l] FOR SANDY: New Fan Handshapes


Query Sandy, and everyone,

On generating a handshape from data, is you still have to select it.  How
could you do that quickly, as a word processor does, from data driven alone?

>From Valerie's IMWA rules, I could see a program being able to assemble from
the IMWA codes a handshape, an orientation, and a palm facing, as they are
directly linked to the code and one would need a Graphic Interface to
manipulate the various parameters that make up the IMWA.

Since her rules are clearly defined, it might be possible to make the IMWA
self-generating for a new handshape, if one can work out, in principle, that
"x finger" on "y" base can only be written one way in the IMWA.  If we can
agree to that in principle, then a "generating" system would work that would
enable Valerie to no longer have to actually create picture by picture 96
drawings for any new handshape.


Sandy Fleming <sandy at FLEIMIN.DEMON.CO.UK> wrote:
Hi Val!

Thanks, these look great!

Your lessons on how handshapes were made were very interesting. A while ago
I tried devising a non-graphical scheme where any handshape could be
expressed using up to 16 pieces of data - it wasn't very hard to do but I
imagine that it's already been done.

I'm wondering, do you think it's possible for a computer program to be
written that would generate handshpe graphics from data? This would be a
better solution for wordprocessors, instead of having to have the IMWA
downloaded, which is huge. But are the rules for drawing handshapes
consistent enough to make this possible?

Sandy

> -----Original Message-----
> From: owner-sw-l at majordomo.valenciacc.edu
> [mailto:owner-sw-l at majordomo.valenciacc.edu]On Behalf Of Valerie Sutton
> Sent: 20 November 2004 16:55
> To: SIGNWRITING List
> Subject: [sw-l] FOR SANDY: New Fan Handshapes
>
>
> SignWriting List
> November 20, 2004
>
> Dear SW List and Sandy:
> A long time ago we discussed some new fan-like handshapes, as shown in
> the attached diagram. Numbers 1 and 2 in the diagram are new handshapes
> that will be placed in the IMWA in the future, but number 3 is already
> in the IMWA. Do these seem readable, Sandy? Thanks for your
> suggestions for new handshapes - Val ;-)
>
>
>



More information about the Sw-l mailing list