The vast majority of data in databases is relational, which means no hierarchies which are what "come naturally" to XML. Consequently, databse vendors (thats relational database vendors) are focussing on mediating data access via XML rather than data storage. There are various XML-and-database tools available for example Oracle's XSQL Servlet,and I know that DB2 offers storage support for XML documents (as in store an XML document in the database, not use XML to store relational data)but a lot more work needs to be done on data modelling and querying with XML in mind before it'll be ready for prime time as a database format (Speaking of which you might want to take a look at Lore and XML-QL )
This isn't free but there are free preview chapters available and all reports I've read of it have been excellent (I have no association with Crane except as a user of the previews)
no features for reuse or modularity I'm not sure I get exactly what you mean here; XSL-T allows a stylesheet to include other stylesheets thereby making the included stylesheets templates available to the including one. At a finer level of granularity you also have things like named-templates which can be called by name as required. Did you have something radically different in mind ?
Prof. Jon Kleinberg of Cornell did something similar; theres a paper of his titled "Authoritative Sources In A Hyperlinked Environment" available here Its also available as an IBM Research Report so maybe it has overlap with what you has in mind.
Slashdot Mirror
The vast majority of data in databases is relational, which means no hierarchies which are what "come naturally" to XML. Consequently, databse vendors (thats relational database vendors) are focussing on mediating data access via XML rather than data storage. There are various XML-and-database tools available for example Oracle's XSQL Servlet,and I know that DB2 offers storage support for XML documents (as in store an XML document in the database, not use XML to store relational data)but a lot more work needs to be done on data modelling and querying with XML in mind before it'll be ready for prime time as a database format (Speaking of which you might want to take a look at Lore and XML-QL )
The Zvon XSL Tutorial
Crane Softwrights' Practical Transformation using XSLT and XPath
This isn't free but there are free preview chapters available and all reports I've read of it have been excellent (I have no association with Crane except as a user of the previews)
no features for reuse or modularity
I'm not sure I get exactly what you mean here; XSL-T allows a stylesheet to include other stylesheets thereby making the included stylesheets templates available to the including one. At a finer level of granularity you also have things like named-templates which can be called by name as required. Did you have something radically different in mind ?
Prof. Jon Kleinberg of Cornell did something similar; theres a paper of his titled "Authoritative Sources In A Hyperlinked Environment" available here Its also available as an IBM Research Report so maybe it has overlap with what you has in mind.