- From: Tim Bray <tbray@textuality.com>
- Date: Wed, 10 Jul 2002 14:48:47 -0700
- To: www-tag@w3.org
On the basis of further investigation, I don't believe that there is an architectural issue here. My raising of the issue was based on an erroneous impression that XQuery has a normative dependency on the PSVI, which is not the case. Should a situation arise in which W3C or other work does develop normative dependencies on the PSVI, it may be the case that we want to consider the architectural implications. But for now, I don't think it should be taking up space on our (crowded) agenda. -Tim PS: The relationship between XQuery and the PSVI is as follows: XQuery sections 1 and 2.1.2 make it clear that the information required to implement XQuery is that contained in the XQuery Data Model, which could be constructed using the PSVI as input. It further notes (especially in 2.1.2) that a Data Model instance "might also be synthesized directly from a relational database, or constructed in some other way... XQuery is defined in terms of operations on the Data Model, but it does not place any constraints on how the input Data Model instance is constructed". The data model does include types identified by qname, and uses XSchema terminology to describe this, but as noted above, such qnames could be generated in a variety of ways. The XQuery Data Model draft provides instructions for how one might construct the XQuery data model using the PSVI as input, which seems quite sensible and useful. But the syntax and semantics of XQuery, as seen by implementors, have no normative dependency on the PSVI.
Received on Wednesday, 10 July 2002 17:48:48 UTC