This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
1. F+O 3.0 refers to the ISO 8601:2000, which has been superseded by ISO 8601:2004 2. Section 8.1.1 of F+O contains an explicit reference to section 5.5.3.1 of ISO 8601, which does not appear to contain any pertinent information. The most likely intended target is 4.3.2.1 in the 2000 edition, which becomes 3.2.1 in the 2004 edition. 3. Section 8.1 of F+O defines xs:dayTimeDuration and xs:yearMonthDuration ab initio, and speaks of XSD 1.1 only in a non-normative note as something that might reach Recommendation status in the future. However, XSD 1.1 is now at Recommendation status, and most of the material in 8.1 could be replaced by a normative reference to XSD 1.1. 4. Section 9.1 states that XSD extends the basic ISO representation YYYY-MM-DD to allow -Y*YYYY-MM-DD. This is confusing, because in ISO the basic format is YYYYMMDD, the extended format is YYYY-MM-DD, and the extended format of the expanded representation is -Y*YYYY-MM-DD. (ISO also allow a leading plus sign). 5. The fact that ISO 8601 appears as a non-normative reference is defensible, but the case for it is weak. Many of the references appear in text (generally introductory or explanatory text) that is not explicitly described as non-normative, and is only non-normative if one can show that the actual function specifications do not depend on it. The references from format-date/time in particular would be better treated as normative.
For the frst draft of F+O 3.1 I am addressing these issues as follows: 1. Replace the whole of section 8.1 with a statement that the types xs:dayTimeDuration and xs:yearMonthDuration as defined in XSD 1.1 shall be available even in an implementation that does not otherwise support XSD 1.1. 2. Delete the first paragraph of 9.1.1. 3. For the sake of format-date/time, make the reference to ISO 8601 normative, and update it to the 2004 edition.
The changes have been applied to the F+O 3.1 specification, and have been noted in the draft 3.0 errata maintained by the editor for potential future publication (in changes.txt); the bug is therefore being marked as resolved and closed.