4366 2007-03-05 20:04:22 +0000 Order relation on dateTime 2012-12-04 00:53:19 +0000 1 1 1 Unclassified XML Schema Datatypes: XSD Part 2 1.0 only All Windows 3.1 NEW http://lists.w3.org/Archives/Public/www-xml-schema-comments/2005AprJun/0054.html P2 normal --- 1 sandygao David_E3 www-xml-schema-comments oldest_to_newest 14313 0 sandygao 2007-03-05 20:04:22 +0000 The ordering between two dateTimes P and Q is defined in 3.2.7.4. It has the following problems: 1. Step A is not necessary. Time zones are already normalized when we convert lexical reps to values. (There is no timezone property in the value, only a boolean timezoned.) 2. How can the situation described in B.1.b ever happen? 3. C.3. Shouldn't it be Q with +14 <= P <= Q with -14 4. Similarly in D.3, shouldn't it be P with +14 <= Q <= P with -14 See the thread started at http://lists.w3.org/Archives/Public/www-xml-schema-comments/2005AprJun/0054.html 14319 1 davep 2007-03-06 03:46:49 +0000 (In reply to comment #0) > The ordering between two dateTimes P and Q is defined in 3.2.7.4. It has the > following problems: > 1. Step A is not necessary. Time zones are already normalized when we > convert lexical reps to values. (There is no timezone property in the > value, only a boolean timezoned.) Yes, there are lots of bugs with dateTime in 1.0. We tried to make 1.1 more sensible. I'm not at all convinced that 1.0 can be made really sensible, and I am convinced that it's not worth the effort. Users who care should definitely switch to 1.1 as soon as it is official.