I think that schema writers would expect the same control over derived complexTypes whether the base type contributes attributes * individually (with <attribute> element info items), or * with a named attribute group (with <attributeGroup ref="groupName">). So it seems a little asymmetrical that one is allowed to write <attribute name="a" use="prohibited"> but one is not allowed to write <attributeGroup ref="groupName" use="prohibited"> to mean: none of the attributes which are part of that group should appear when this derived complexType is used. Question: Was this an oversight, or a conscious design choice by the working group? Thank you, Bob Schloss IBM Thomas J. Watson Research CenterReceived on Monday, 19 February 2001 16:45:25 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 23 October 2007 06:13:20 GMT