I18n WG comments on XQuery 1.0 and XPath 2.0 Data Model

Version reviewed


Main reviewer

Felix Sasaki fsasaki@w3.org


These are so far personal comments, NOT on behalf of the I18N WG.


ID Location Comment Additional information Accepted
1 Sec. 3 (about the data model construction) Language information is not part of the data model. It would be good if one could deal with pcdata with an associated language information, or document fragments with associated inherited attributes in general. RDF might provide an appropiate inheritance mechanism, see for example rdfs:subClassOf . Based on Martin's review, comment [7]. See Bugzilla -
2 The accessors of the data model The accessors sometimes rely on the data type xs:anyURI, i.e. baseURI (5.2), document-uri (5.4), unparsed-entity-system-id (5.16). xs:anyURI itself in its latest version is defined in terms of IRI. I would be good if you could refer to IRI itself, in addition to anyURI. Based on Martin's review, comment [20]. See Bugzilla -
3 S3c. 6.1.5 (infoset mapping) You write at the end of the section that the [character encoding scheme] property has no value. You should add s.t. like "Its value is determined during the serialization of the data model." and a link to the section which defines the encoding parameter of the serialization, e.g. in sec. 3. Based on Martin's review, comment [21]. See Bugzilla -
4 On namespace-bindings accessors and namespace nodes In these sections you refer to URI in a rather colloquial way, e.g. "a set of prefix/URI pairs" (sec. 5.7) or "prefix, uri, parent" (sec. 6.4.1). Please refer to the definition of IRI (or anyURI, which is defined in terms of IRI) instead. Based on Martin's review, comment [30]. See Bugzilla -
5 Examples (sec. E), the attribute "currency" at the type "price" in the xml schema Defining a default currency in the schema is bad design practice. Without the schema, the data is basically useless. Please choose something different for an example of default attribute handling. Based on Martin's review, comment [35]. See Bugzilla -
6 Examples, the data type "monetaryAmount" The monetaryAmount type works well for some currencies (USD, EUR,...), but does not work for others (Yen,...). Please generalize. The number of fractional digits needed currently is 0, 2, or 3. So instead of "/d+/./d{2}" please write s.t. like "/d+/./d{2-3}?" Based on Martin's review, comment [36]. See Bugzilla -

Version: $Id: xq-xt-datamodel-review.html,v 1.7 2005/05/25 03:18:24 fsasaki Exp $