- From: Bjoern Hoehrmann <derhoermi@gmx.net>
- Date: Tue, 13 Jul 2004 04:23:04 +0200
- To: public-rdf-dawg-comments@w3.org
Dear RDF Data Access Working Group, From http://www.w3.org/TR/2004/WD-rdf-dawg-uc-20040602/ it is not clear to me whether you think that users will desire to store queries as separate resources, i.e., as files or HTTP delivered resources. It seems however likely that they will, e.g. to provide users of software products based on such a query language the ability to access the queries elsewhere, to modify them or to provide their own queries. Or they just want to get the queries out of their sight. If you consider this desirable, this seems to motivate a requirement to provide a MIME type for query documents and to think about the digital representation of queries, i.e., software that reads queries from external digital resources would need to determine the character encoding of the resource. Considering that queries likely get rather lengthy due to long URIs and such, I would certainly favour the possibility of externally stored queries and would thus propose it as a use case which then motivates a Requirement to register a new MIME type for the query language and unless the syntax is XML-based to sort out character encoding issues. For the latter I would then encourage the Working Group to avoid allowing any other character encoding than UTF-8 just like Notation3 did so far, with the exception that I would further encourage you to allow such a document to start with a byte order mark. See http://www.w3.org/2002/06/registering-mediatype for details on W3C's MIME Type procedures and http://www.w3.org/TR/charmod/#sec-Encodings for more information on choosing a character encoding for new formats. regards.
Received on Tuesday, 13 July 2004 00:40:04 UTC