On Monday, April 19, 2010, 10:22:26 PM, Cyrille wrote: CB> On Monday 19 April 2010, Kai-Uwe Behrmann wrote: >> Can a XML based format be properly implemented without the accompanying >> schema? Would it render a great disadvantage to be not able to use and >> distribute the available schema because of license issues? CB> Not really. The main point of schema in an application is to validate the XML CB> file. I might be wrong, but I don't think many applications tries to validate CB> a XML file before reading it.
No application can do so. It must first be parsed, then schema canformance can be determined. (This is different from DTDs, which are (if used) read before the document instance is parsed). CB> (I tend to try to be able to parse even invalid CB> XML files). Parsing and validity are orthogonal concepts in XML. An XML file is parseable if it is well formed. Otherwise it is not XML. Whether it is also valid (to a DTD, or schema, or whatever) is a separate issue. A document may be valid to more than one schema, also. Its not generally considered good practice for the namespace URI to be that of the schema or RNG or whatever. A namespace URI may be defererencable, but need not be. -- Chris Lilley mailto:[email protected] Technical Director, Interaction Domain W3C Graphics Activity Lead Co-Chair, W3C Hypertext CG _______________________________________________ CREATE mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/create
