The introduction of character encodings for sign writing is an interesting challenge, though, and the Unicode encoding is rather challenging. I hope browsers and other applications will eventually support it and also find a good font to represent it. It would be rather interesting to see a browser with sign writing!

Comparing such a transcription to the written form of a spoken language is largely irrelevant. For one thing writing sign is far less abstract. With speech you have to change the phonological sounds to visual units, which is arbitrary and complex. With a visual language the phonological units are already visual. Sign language written with SignWriting is processed in the brain the same way as actual conversation.

There is no need to change the phonological units (signs), just transfer them to a surface, and omit extraneous details. Microsoft Paint will let you do ordinary handwritten ASL, although it’ll be sloppy and ugly.

SignWriting can also be animated, which is even more like actual sign language. It’s debatable whether that can be called writing. It’s more like an avatar–a better one in my opinion, since it shows nothing except meaningful morphemes.

The only real problems that remain to be solved concern doing all this with computers, and it’s basically the same problem presented by tone languages. Chinese has five vowels, each with one of four tones, and possibly a diacritic mark. So programmers have to choose: 11 symbols in three sets, or 40 symbols in one set. We’ve chosen separate symbols sets for the tones, which works but it’s a nuisance.

ASL just presents the same choice way magnified: lots and lots of symbols, in a half dozen sets. Using one set of symbols and lining them up in a row is not an option since you would need a nearly infinite number of symbols. ASCII is out, and I’m not geek enough to offer intelligent alternatives, but I have faith you guys will, or are.

However, there are a few things that make this solution hard to carry out. Firstly, Sign Language which is a “movement language” is very different from speech/written language. Sign Language often means visualizing the shape and motion of different objects and how they co-relate. This is acted out in the Sign Language tellers imaginary “scene” e.g. placed in front of the body of the person who is signing.

For example: The sign ”big” is not signed the same way in the meaning “big elephant” as in “big squirrel”. The sign “moving fast” will also be very different depending on whether you are talking about the elephant’s or the squirrel’s movements. The text “The elephant was watching the squirrel” should be a sign downwards if the squirrel is sitting on the ground or maybe upwards if it is sitting in a tree. If the squirrel is sitting on the top of the elephant’s back the dialog will be completely different. This may seem as a trivial problem but the consequences could be devastating if the text is i.e. information about tsunamis and neither size, shape, position nor speed is correctly translated by machine.

The Sign Language uses the possibility to visually explain the position, relations and movements to gain ultimate information transfer. It is the core of Sign Language and contributes to make the Sign Language an extremely rich language. The Sign Language often exceeds the written form when it comes to transferring information, especially when it is about positions in the room and shapes. If a machine (automatic?) translation shall reach the same level as a manual translation you need to consider many variables since the computer needs to know about the consequences of the size, movement and position that occurs in the text.

The Sing Language ”signs” will, for the same reasons, be hard to transfer to a written form. The precision in the description will be less adequate than when it is carried out by a balanced and well adjusted movement. There is a risk that it will be abstract in a written form. The profit of the visualizations that exist in Sign Language would be lost.

To remove the sharpest tool from the Sign Language is not the ultimate way if the goal is understanding for the target group of Sign Language users that cannot understand written form.

Each step is a step closer to the goal regardless of the distance and I encourage investigation of the conditions in order to try and solve the existing problems. Blogs like these are a good start. It must be done with great respect for the Sign Language and with a clear delimitation of who the target group is, and what their conditions are. It should also be carried out in close co-operation with a group of Sign Language linguists.

Your computer-scientist obsession with boiling everything down to characters (I note you misrepresent even Chinese in this regard) will serve you ill.