Hi,
by putting JCR on top of XML DB
That would be possible. Maybe somebody should start doing that!
re: xml in the (computer) industry XML *is* being widely used and will continue grow in usage for *data interoperability*. You don't use SQL or file systems for data interoperability, it doesn't even apply.
It looks like 'data interoperability' means different things for different people. If the point of data interoperability is "to makes sharing data easier", then (in my view) this includes existing systems. Data in existing systems may not be in XML.
However, the model of the spec is largely based on XML. The concepts are very clearly XML based The Repository Model' shows the XML-centric API: -Hierarchy, with a root node -Siblings -Child Nodes -Properties and attributes -Paths
Those concepts are used by many models, not only XML. The JCR model is hierachical. The XML model hierarchical. The hierachical model was there well before SQL and XML, see http://en.wikipedia.org/wiki/Hierarchical_model. File System, LDAP, Subversion and so on are hierachical as well. Namespaces were invented before XML as well. JCR node can be viewed as XML elements, but don't need to be.
JCR data model are not relational
Does XML define references? The spec doesn't say that JCR is an XML database, or has the same functionality or model as an XML database.
"SQLish" language are not capable to work with graph like structures
In JCR (since 1.0), a query returns 'columns', 'rows', and 'nodes'. SQL returns 'columns', 'rows'. Full XQuery returns XML. This looks like a problem to me.
Either a subset of XQuery could be defined, or XQuery could be an optional part of the spec
I agree. But full XQuery wouldn't make sense in my view. I think we have written enough, let's go ahead and define the XQuery subset! Thomas
