Improvements in Sign Writing dictionaries.
chazzer3332000 at YAHOO.COM
Wed Nov 25 14:37:51 UTC 2009
Stop, you are misconstruing my point, I'm not saying anything about Bing or anything other search engine, but SW in its own right. Signed languages need dictionaries, just like any other language, and the ordering system needs to be consistent across the board so that two signed languages can be compared in print and one knows where in the corpus to find a particular sign. When I started a 4-language dictionary in Brazil, comparing ASL and LIBRAS in print makes no sense if one is relying on English or Portuguese as the primary language.
Stokoe has its ordering system, SW has had an ordering system since its creation, the Sign-Symbol-Sequence which does go handshape, orientation, location, movement, facial expression, it is within the 10 larger groups that I am looking to impose order so that I know, for example that:
one hundred Used for people, places, things, not money
Modified: November 14, 2009 11:38
Puddle page 10501
one hundred page number
Searching purely for the dominant hand, orientation, and rotation.
If I am going to print a dictionary, it has to have some order to it, not everything is in a database searchable by everything when one is printing. That's my point. If it is to be entirely in Sign Language first and then spoken language second, following the alphabetic rendering of the Roman Alphabet does not lend itself to this kind of search and comparison.
From: Trevor Jenkins <bslwannabe at gmail.com>
To: SignWriting List <sw-l at majordomo.valenciacc.edu>
Sent: Wed, November 25, 2009 9:00:52 AM
Subject: Re: [sw-l] SignWriting on Bing - Improvements to Ordering Signs
I think the exact opposite! It is not that SignWriting (or HamNoSys or Stokoe) needs to accommodate Bing, Google, Wolfram Alpha or so later search engine. Instead the search engines need to change to accommodate SignWriting (and everyother non-Latinate script). We should not change the order in which signs are transcribed -- we do not alter the order of written lexemes so that search engines can retrieve web pages or emails. What we do need is for the present and all future search engines to be capable of searching on inflected sign forms (for example using the Stokoe classification of handshape, orientation, location, movement, repetition). It is us as users who impose order on lexems whether signs or words.
We could be consistent in the way that we write each SignWriting symbol in the same way that there is a convention for how Stokoe is written generally following.location, handshape, movement, orientation, repetition and alterations as we describe the full sign.
The ISWA will prove sufficent for Bing, Google, Alpha, Yahoo!, etc to retrieve on because it is part of Unicode. But let's not make their lives easier at the expense of making our own more difficult. We have better things to do than help Microsoft, Google, Wolfram or Yahoo! fleece us.
On Wed, Nov 25, 2009 at 11:50 AM, Charles Butler <chazzer3332000 at yahoo.com> wrote:
>Although I don't know how to change Bing, I'm glad that the system footnotes my article, written back in 2001, on an ordering system for Sign Writing.
>I believe that we need to fine-tune the system so that the order of handshapes follows logically, not simply as they are put into the system, as having the articulated fingers starting straight, then together, then curved, then bent, then crossed, seems logical but because of the order of creation of a given handshape in the historical progression of the ISWAthat sometimes does not follow.
>Ordering of the system now simply follows the order of the coding, so that signs using the same articulators can be put into a system. The sign-shape-sequence which I have been trying to include or edit all the signs I find to include, follows the glyphs in sequence order internal to a sign.
>1) Right hand (by hand group, sub-hand group, orientation, rotation)
>2) Left hand (by hand group, sub-hand group, orientation, rotation)
>3) Right hand contact (touch, grasp, brush, rub, in-between)
>4) Left hand contact (touch, grasp, brush, rub, in-between)
>5) Right hand location (include face or body) (location on the face, location on the body)
>6) Left hand location (include face or body) (location on the face, location on the body)
>7) Right hand movement (straight, curved, compound)
>8) Left hand movement (straight, curved, compound)
>9) Right hand speed (prosody) (slow, fast, smooth) There are signs in LIBRAS where the only difference is the speed of the sign)
>10) Left hand speed (prosody) (slow, fast, smooth)
>11) Facial expression (I have no idea how to order facial expressions)
>12) Body posture (there are signs in LIBRAS where the only difference is a posture)
>Now that we have a sufficiently large corpus, I would propose we use this system for some experiments to see how clearly it actually works. The only change I would put in might be in defining 1) as "Dominant Hand" and 2) as "non-Dominant Hand" but there are many signs such as "WITH" in ASL that have no clearly dominant hand, so that it might be simpler to continue with "right-hand dominant".
>SW-L SignWriting List
>SW-L at majordomo.valenciacc.edu
>List Archives and Help
>Change Email Settings
<>< Re: deemed!
-------------- next part --------------
An HTML attachment was scrubbed...
-------------- next part --------------
SW-L SignWriting List
SW-L at majordomo.valenciacc.edu
List Archives and Help
Change Email Settings
More information about the Sw-l