- From: Marc de Graauw <marc@marcdegraauw.com>
- Date: Wed, 30 May 2007 21:45:56 +0200
- To: "'John Cowan'" <cowan@ccil.org>, <noah_mendelsohn@us.ibm.com>
- Cc: "'Norman Walsh'" <ndw@nwalsh.com>, "'David Orchard'" <dorchard@bea.com>, <www-tag@w3.org>
John Cowan: | In shorter, a language is a (mathematical) set of XML documents. Which means giving an extensional definition of a language, and that solution suffers from the same problems as all extensional definitions: while superficially clear (a set is a mathematical notion after all), it is in practice impossible to give me say, all members of the set of UBL 1.0 or of HL7v3 documents. So the extensional definition doesn't help in practice: I cannot determine whether doc X is an UBL 1.0 or HL7v3 document given an extensional definition. Therefore an intensional definition is a necessary evil: though, as Noah pointed out, not necessarily a formal grammar: in most XML languages, certainly in the two mentioned above, natural language, with all it's imprecision and expressiness, plays a major part, and - as far as I am concerned - rightly so. Marc de Graauw www.marcdegraauw.com
Received on Wednesday, 30 May 2007 19:46:40 UTC