Talk:Sublanguage

[Untitled]

This whole article seems to be a little odd and definitely not NPOV. First, without a definition of complete language, this really doesn't make clear what a sublanguage is. Second, is it really true that sublanguage only relates to database theory, as the article implies? Finally, can it be accurately stated that Oracle is the only source for a complete language based on the SQL "sublanguage"? I should think Microsoft, IBM, Sybase, and several other companies would disagree with this statement. -- Jeff Q 09:42, 9 May 2004 (UTC)[reply]

I can now say with some confidence that this article is very POV, slanted toward a interpretation that serves Oracle Corporation's marketing interests. The use of the term sublanguage in database theory seems to have originated with E.F. Codd in his paper "A Relational Model of Data for Large Shared Data Banks" (Communications of the ACM, Volume 13, Issue 6, June 1970). He apparently uses the term to describe a language incomplete from a mathematical point of view, not one that needs extra "features" to make it useful, as the Oracle description would suggest.

Furthermore, sublanguage is not exclusively a database term. I easily found several uses in information theory, relating to subsets of natural language for more effective machine translation, as well as non-computing uses like linguistic subsets of human languages and even as a term for industry-related jargon (e.g., "immunology sublanguage").

Therefore, I have two conclusions. First, this article should be generalized. I don't know enough about database or information theory to expound correctly upon their proper use of sublanguage, but I can provide a general definition:

A sublanguage is a subset of a human language or computer language, defined either by a specific vocabulary and grammar (e.g., the SQL database sublanguage) or by restrictive description (e.g., an "immunology sublanguage").

It could certainly use some work.

The second conclusion is that this article should be moved to and formatted for Wiktionary, as it is a basic definition. I have added the Wiktionary tag to signal this. I have left the factual dispute tag for now. I would like people to comment on what I've come up with here before I change the text. — Jeff Q 06:14, 21 Sep 2004 (UTC)

Rubbish

From googling I've seen this term used in a semi-illiterate fashion on computer science courses, but in terms of computer language theory it seems to be utter gibberish. The example given, SQL, is a complete language in and of itself. Nothing provided by a RDBMS alters that. --Tony Sidaway|Talk 06:35, 26 Apr 2005 (UTC)

Rewrite

I've rewritten the thing from scratch with reference to Codd's usage. --Tony Sidaway|Talk 11:22, 26 Apr 2005 (UTC)

Thanks, Tony! I was despairing of ever getting a solid article out of this mess. — Jeff Q (talk) 11:55, 26 Apr 2005 (UTC)