SignWriting List Forum| Message 3361 | Previous | Next [ Up Thread ] | Message Index |
|
SignWriting List Forum |
|||
|
|
From:
"Angus B. Grieve-Smith" Date: Mon May 8, 2000 1:00 pm Subject: Re: SW in Databases | |
|
One thing to keep in mind is the difference between "flat" databases and "relational" databases. In flat databases, there is a series of records, each with a number of fields containing information. So in a signed-language dictionary, you could have fields for "handshape," "location," etc. You would probably have one record for each sign, and the fields would contain information specific to that sign. The database can be indexed by any of the fields, so that you could look up all the signs with a particular handshape, for example. The SW4.3 dictionaries are essentially flat databases with two fields: the SignWriting entry and the entry in a spoken language. The field for the spoken language entry is indexed, so that you can look up signs that have a particular translation in the spoken language. A relational database is more complex than that. It has fields and records, but the records can contain links to other records in other sub-databases. This would be helpful for people creating complex semantic dictionaries that would link various frames and schemas, but it's a lot more complicated than is necessary for the simple two-language dictionaries. Most of the database formats that Trevor and Valerie talked about (FileMaker Pro, Oracle, Microsoft Access) are designed for relational databases, so using them for a bilingual dictionary is overkill. Trevor and Bill mentioned some simpler methods; I'll describe some of them in my next message. -- -Angus B. Grieve-Smith Linguistics Department University of New Mexico |
|