- From: <Patrick.Stickler@nokia.com>
- Date: Mon, 5 May 2003 12:21:54 +0300
- To: <jjc@hpl.hp.com>, <w3c-rdfcore-wg@w3.org>
My preferences, in order, most prefered to least preferred: Option 4, 1, 3, 2 > Option 4 in my mind is simply incorrect - there are XMLLiterals for which the > language is semantic meaningful. Well, why is it not unreasonable to require that, where an xml:lang tag is relevant to an XML literal, that it be specified *within* the XML literal. Why do we have to do it for *every* XML literal automatically? Let those that need them specify their own wrapper elements. Not everyone want's their literals infected by the xml:lang of the RDF/XML instance. Likewise, one can use other means, such as reification, to assert a language scope for the assertion (which applies to simple literals as well). This is the approach I am beginning to take, which has the added benefit of making the language knowledge explicitly visible to reasoners without having to parse node labels. So, while option 4 might require some test cases to change, it's not necessarily "incorrect". It may be, in fact, the most correct and optimal option. Patrick
Received on Monday, 5 May 2003 05:22:03 UTC