Software

[Mia Kalish]

Hi, Sue, 

 

I checked the page out that you gave me. . . . :-) 

 

Yesterday, we had a discussion about creating materials for Communities to
use . . . and, I'm not sure this is it, at least, not as a final output. I
think that if it has an export function, that it could be really useful for
the middle-level data collection that would then be used to create the
Community materials. 

 

My first observation of the site was that it looked very much like the
search and retrieval system our library uses here at NMSU. These systems are
an outgrowth of the Retrieve/Browse paradigm of early computer systems when
there really was a space/functionality constraint. (My first invoice/cash
application had 32K of computer memory. That's 1/100th of what I have today
on this laptop; which I don't have to share). 

 

First, I searched for one of my favorite polysemic words, "object", and
didn't find anything.

 

Then, I read that the words were about the Southwest, so I searched for
"cactus". "Cactus" showed up at the top of the retrieved list which then
went on to show such things as California Redwoods. Every once in a very
great while I find this kind of option useful, but usually only to confirm
the bad news, that what I was looking for isn't there. I thought perhaps the
alphabetic listing might be useful for helping people correct spelling
errors, so I went back and looked for "cactus". I didn't find anything,
which surprised me. This means that the application is using a database that
supports browse after keyed access. IBM's DB2 used to have that facility. .
. . Most non-commercial applications aren't coded this way; this kind of
functionality is useful when you have multiple items coded under an
identifying key, as for example Customer #/Invoice 1, Customer #/Invoice 2,
and so on. It might also be useful for Cactus:Cholla, Cactus:Sequaro,
Cactus:Rainbow. But it doesn't appear to be working that way . . . . back
soon . . . I put in "cactus" again and selected the "Each term followed by
its relationships" radio button. Again, I got the StartKey:Browse List
configuration. 

 

Interestingly enough, you can input "c" and get a list of items beginning
with that letter. You can input "ce" and get the list commencing with "ce".
. .  but if you input "cectus", you get item not found. . . . I wondered if
the function was perhaps dependent on input string length. I input "cec" and
again got a Not Found. 

 

 

 

I have some issues with the list of metadata types. These are all highly
abstracted, culturally embedded terms. I'll bet there is only one entry for
"preferred term". . . Having spent several to many years in an institution
that is famous for its critical pedagogy, I have to has, Preferred By WHOM?
It also is limited in that to make the categories work for different
disciplines, you have to identify each one of them as a language. Lastly, I
tracked through these categories long enough to realize that the
relationships form a Network. Both the search keys and the outputs are all
terms. I did find a place where there was a description. I found it kind of
frustrating because I felt like I was "always on the road" never "at the
Inn". In other words, I was always searching, always searching, and because
the responses were so broad, and crossed so many categories, I never felt
like I was done. 

 

     PT - Preferred Term
     SY - Synonym
     BT - Broader Term
     NT - Narrower Term
     FR -  Functionally Related
     CR - Conceptually Related
     TR - Temporally Related
     PS - Physically/Spatially Related

 

You mentioned that this was pretty easy to use. . . there is a kind of Rule
of Interaction in Computer Science that says that the easier something is to
assemble on the front end, the less sophisticated the functionality will be
on the back end.  I was a bit worried that there was a Search Everywhere
option, that allowed you to go to Google. IMHO, this is an hypertext
application without the graphics and sophisticated search engine tools. It
probably doesn't even allow you to embed special fonts for languages that
require characters other than those supported even by Unicode. (like
Athapascan languages). Given the sophistication of today's technologies, I
don't think users will want this for anything more than the mid-level
interface tool . . . 

 

A final, final note: This is very Modern (one view, one meaning) as opposed
to PostModern (many voices) and PostStructuralist (lots of pieces,
diachronic component). 

 

It seems you might like a Wiki better. . . it gives you places for people to
participate. The people I have worked with always seem to think there is
only one word for something in their language, and one way to say that word.
The battles that ensue around this idea tend to halt revitalization rather
than facilitate it. . . Oh: SWT doesn't seem to have a place for sound. . .
. no pictures? All Text? 

  _____  

From: Indigenous Languages and Technology [mailto:ILAT at LISTSERV.ARIZONA.EDU]
On Behalf Of Susan Penfield
Sent: Sunday, February 26, 2006 8:29 AM
To: ILAT at LISTSERV.ARIZONA.EDU
Subject: [ILAT] Software

 

Hi,
Yesterday I mentioned that I have been looking at a new software which might
be usable 
for creating thematic dictionaries (a very user-friendly option) . I would
like some others to take a look
to see what they think of it . Check out www.swt.arizona.edu. 
Best,
Susan

-- 
Susan D. Penfield, Ph.D.

Department of English
Affiliate faculty: Department of Linguistics 
and the Second Language Acquisition and Teaching Program 
American Indian Language Development Institute
Phone for messages: (520) 621-1836 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/ilat/attachments/20060226/39ab966f/attachment.htm>