From: Jeremy Carroll <jjc@hpl.hp.com>

Date: Fri, 16 Nov 2007 12:16:10 +0000

Message-ID: <473D8A0A.6030204@hpl.hp.com>

To: Carsten Lutz <clu@tcs.inf.tu-dresden.de>

CC: ewallace@cme.nist.gov, public-owl-wg@w3.org

Date: Fri, 16 Nov 2007 12:16:10 +0000

Message-ID: <473D8A0A.6030204@hpl.hp.com>

To: Carsten Lutz <clu@tcs.inf.tu-dresden.de>

CC: ewallace@cme.nist.gov, public-owl-wg@w3.org

As far as I am aware, xsd:integer and the mathematical set Z have no important divergences i.e. there is no gain in supporting both. string is somewhat harder - notionally it does correspond to the mathematical set of finite sequences of unicode code points. however, to some extent the set of unicode code points is a moving target; and there is always the danger that some specific code point has some semantics that makes some applications treat a particular string differently, .... and this is further complicated by user expectations which are more couched around the displayed format of the string, rather than the mathematical sequence. Since two or more sequences can correspond to identical displayed formats this is an area of concern. As far as I recall, none of the other datatypes have anything approaching a rigorous mathematical analog. JeremyReceived on Friday, 16 November 2007 12:16:40 GMT

*
This archive was generated by hypermail 2.2.0+W3C-0.50
: Tuesday, 8 January 2008 14:13:27 GMT
*