- From: Norman Walsh <ndw@nwalsh.com>
- Date: Wed, 28 May 2014 11:01:38 -0500
- To: public-xml-core-wg@w3.org
- Message-ID: <871tvd51ul.fsf@nwalsh.com>
Hello world, On the Core call this morning, David, Paul, Liam and I discussed the xs:numeric type. The background on this type is that XPath (1.0) had a "number" data type. In the XPath 2.x (and following) specs, this was handled in an ad hoc way. For XPath v.next, we'd gain significant improvements with a proper numeric type to use in function signatures and for sequence type matching. The XSLT and XQuery WGs are not asking for a change to XML Schema proper. There is precedent for the XPath Data Model to define new types in the xs: namespace. (We did it for xs:dayTimeDuration and xs:yearMonthDuration before they appeared in XSD.) Assuming, as appeared to be the case on this call, that no one objects to the addition of the new type in the XPath Data Model, the only actual change that the XSLT and XQuery WGs are requesting is that the namespace document for the xs: namespace be updated to indicate that there may be additional types defined in the XDM spec. As I said, that appeard to be acceptable to everyone and I'll report provisional assent to the XSLT and XQuery WGs. If anyone does object, please say so. Be seeing you, norm FYI: the actual XSLT and XQuery proposals are described in these messages: https://lists.w3.org/Archives/Member/w3c-xsl-query/2014May/0053.html https://lists.w3.org/Archives/Member/w3c-xsl-query/2014May/0052.html The XSLT and XQuery WGs adopted them in their meeting yesterday. P.S. The only proposal relevant to this WG is the xs:numeric type. The other proposal, a syntax for defining union types in the sequence type grammar, is just XPath related. -- Norman Walsh Lead Engineer MarkLogic Corporation Phone: +1 512 761 6676 www.marklogic.com
Received on Wednesday, 28 May 2014 16:02:14 UTC