2.4.2 Item types 'The word "element" or "attribute" alone refers to the wildcard types for any element or any attribute.' Not quite, not any more. ElementNameOrWildcard does not derive the empty string (unlike the 'ElementName?' that it replaced), so element is no longer a valid ElementType. Instead, you have to say element * (which is not so good, since people will look at it and sometimes misinterpret the '*' as an OccurrenceIndicator, and read it as "zero or more elements"). This change has not been propagated throughout the inference rules. Specifically, the following judgments need to have a '*' inserted for ElementNameOrWildcard. 4.7.3.1 / STA / rule 1 / premise 3 4.7.3.1 / STA / rule 2 / premise 3 4.7.3.1 / STA / rule 3 / premise 5 4.7.3.1 / STA / rule 3 / conclusion 4.7.3.1 / STA / rule 4 / premise 5 4.7.3.1 / STA / rule 4 / conclusion 4.7.3.1 / DEv / rule 1 / premise 9 4.7.3.1 / DEv / rule 2 / premise 11 4.7.3.3 / STA / rule 1 / premise 2 4.7.3.3 / DEv / rule 1 / premise 3 4.7.3.3 / DEv / rule 2 / premise 4 4.13 / STA / rule 1 / premise 2 7.1.5 / STA / rule 1 / premise 1 7.1.5 / STA / rule 1 / conclusion 7.2.5 / STA / rule 3 / conclusion 7.2.5 / STA / rule 4 / conclusion 8.1.3 / Sem / rule 3 / conclusion 8.1.3 / Sem / rule 4 / conclusion 8.1.4 / Sem / rule 3 / conclusion 8.1.4 / Sem / rule 4 / conclusion 8.2.2.1.5 / Sem / rule 1 / conclusion 8.2.2.1.5 / Sem / rule 2 / conclusion 8.2.2.1.5 / Sem / rule 3 / conclusion 8.2.2.1.5 / Sem / rule 4 / conclusion 8.2.2.1.5 / Sem / rule 5 / conclusion 8.2.2.1.9 / Sem / rule 1 / conclusion 8.2.3.1.1 / -- / rule 4 / conclusion 8.2.3.1.1 / -- / rule 8 / conclusion 8.2.3.1.1 / -- / rule 12 / conclusion 8.2.3.1.2 / Sem 2 / rule 3 / conclusion 8.2.3.1.2 / Sem 2 / rule 4 / premise 1 8.6.1 / Sem / rule 1 / conclusion Similarly for AttributeNameOrWildcard: 4.7.3.1 / STA / rule 1 / premise 3 4.7.3.1 / STA / rule 2 / premise 3 4.7.3.1 / STA / rule 3 / premise 5 4.7.3.1 / STA / rule 4 / premise 5 4.7.3.1 / DEv / rule 1 / premise 9 4.7.3.1 / DEv / rule 2 / premise 11 4.7.3.2 / STA / rule 2 / conclusion 7.1.5 / STA / rule 1 / premise 1 7.1.5 / STA / rule 1 / conclusion 7.2.5 / STA / rule 3 / conclusion 7.2.5 / STA / rule 4 / conclusion 7.2.6 / STA / rule 10 / premise 3 8.1.3 / Sem 2 / rule 3 / conclusion 8.1.3 / Sem 2 / rule 4 / conclusion 8.1.4 / Sem 2 / rule 4 / conclusion 8.1.4 / Sem 2 / rule 5 / conclusion 8.2.2.1.3 / Sem / rule 2 / premise 1 8.2.2.1.3 / Sem / rule 3 / premise 2 8.2.2.1.3 / Sem / rule 4 / premise 2 8.2.2.1.4 / Sem / rule 2 / premise 2 8.2.3.1.1 / -- / rule 17 / conclusion 8.2.3.1.1 / -- / rule 21 / conclusion 8.2.3.1.1 / -- / rule 25 / conclusion 8.2.3.1.2 / Sem 3 / rule 3 / conclusion 8.2.3.1.2 / Sem 3 / rule 4 / premise 1 (semi-wildcard types) Moreover, ElementTypes and AttributeTypes with semi-wildcarded names (e.g., foo:* or *:bar) don't exist any more. So I think the following rules need to be discarded: 8.2.3.1.1 / -- / rule (2|3) 8.2.3.1.1 / -- / rule (6|7) 8.2.3.1.1 / -- / rule (10|11) 8.2.3.1.1 / -- / rule (15|16) 8.2.3.1.1 / -- / rule (19|20) 8.2.3.1.1 / -- / rule (23|24) And these rules need to be modified: 8.2.3.1.1 / -- / rule (8|12|21|25)
Re-classified as 'minor' for editorial. - Jerome
This is not "resolved fixed" in the 2006-06 CR. It looks like you discarded the rules that needed to be discarded, but haven't done any of the other fixes.
The fix for this bug does not appear in the Recommendation of 23 January 2007. It will be considered for a future publication (either an Errata document or some possible future version of the specification).
The fix for FS erratum E018 has resolved some of the remaining items of this issue, specifically all the ones in section 8.2.3.1.1.
The remaining items of this issue have been entered as FS erratum E023, whose fix has been committed to the source files for the next edition of the FS document. Consequently, I'm marking this issue resolved-fixed, and also closed.