Print

Print


>.  However, if METS and MODS are going to use xlink schemas that are
> identical to the point of declaring the same targetNamespace, wouldn't it
> still make sense just to make them one in physical fact?

They might not remain identical.

1.  As to the target namespace, MODS had originally used a different namespace, when a number of users complained that that wasn't compliant with the w3c namespace standard, which mandates: targetNamespace="http://www.w3.org/1999/xlink".  So MODS changed.  I think METS went through the same process, or maybe it started out with the right namespace,  I don't recall, but in any case the result was that both schemas ended up with the w3 namespace, not so that the namespaces would be the same for both schemas, but to comply with the standard.  (And now, MODS  -- I can't speak for METS -- is wondering why we bother complying with that standard......)  

2.  The two schemas themselves are identical now but might not remain so. We (MODS) have encountered a flaw (we think it's a flaw) that would require non-compliance to remedy, and there is some sentiment that non-compliance with the xlink standard is the lesser evil. METS would have to decide for itself.

For background,  look at the MODS listserv:
http://listserv.loc.gov/cgi-bin/wa?A1=ind0505&L=mods
subject:xlink:type attribute and MODS

(if you have problems accessing these files you may have to subscribe to the MODS listserv.)

This discussion resulted in the (tentative) opinion that we  should change:

 <attribute name="type"  fixed="simple"/> 

to:
<attribute name="type" use="optional">
<simpleType>
<restriction base="xsd:string">
<enumeration value="simple"/>
<restriction>
<simpleType>
<attribute>

 for reasons explained in the thread.  

--Ray