Common model or not?

3.2 Model for data interchange and open access

reads in part:

*****
This specification defines a common MODEL for data interchange and open access to the deliberative bodies’ documentation, such as parliamentary, legislative, and judiciary texts.
*****

Is directly contradicted by:

*****
At the same time, the Akoma Ntoso model considers the differences that exist in individual document types, that are derived from using different human languages, and that are implicit in the legislative culture of each country. Therefore the common open access model is designed to be flexible, to support exceptions, and to allow extensions far enough to provide support for all individual characteristics that can be found in a complete document set covering different cultures and countries.
*****

The unlimited extensions allowed by Akoma Ntoso are a virtual guarantee that no XSLT schema that works with one local version of the schema will work with any other version of the schema. No XSLT stylesheet can allow for unlimited permutations of a base schema. If you are interested, I did a paper on that subject, the # of variations when using a very small tag set in XML.

That isn't to say that all is lost. If the Akoma Ntoso schema were recast in terms of value patterns in RELAX-NG and then defined solely a a vocabulary, to which profiles could conform, then the profiles for each jurisdiction/locality could have enough specifics to enable meaningful interchange and exchange between applications.

Simply using XML is not a guarantee of interoperability.