shapes-ISSUE-142 (loose terminology): SHACL spec is too loose with its uses of terminology [SHACL Spec]

shapes-ISSUE-142 (loose terminology): SHACL spec is too loose with its uses of terminology [SHACL Spec]

http://www.w3.org/2014/data-shapes/track/issues/142

Raised by: Peter Patel-Schneider
On product: SHACL Spec

The SHACL spec is much too loose with its terminology.

The spec uses different terminology from SPARQL where it is using concepts
from SPARQL.  For example, the SHACL spec uses "row" where SPARQL uses
"solution", and "value" where SPARQL uses "binding".

The spec uses different terminology from RDF where it is using concepts from
RDF.   For example, the SHACL spec uses "resource" where RDF uses "node".

The spec uses terms that are not defined.  For example, in Section 6.2.3
there is "value" and "count", but these terms are not defined.  In multiple
places in the spec there is "value type", but that term is not defined.

The spec uses definite descriptions that do not have referents..  For example, in
Section 6.2.2 there occurs "the constraint node", but there is no
specification of what that might refer to.

The spec uses IRIs from the RDF and RDFS namespace in a way that appears to
be different from their use in RDF or RDFS.  For example, rdfs:Resource is
used in contexts (e.g., Section 3.1.1) where it appears that the intent is
to restrict to IRIs, but that is not what rdfs:Resource means
in RDFS.  However, there is no specification of what rdfs:Resource is
supposed to mean tthere.


There needs to be a comprehensive attempt to clean up the use of terminology
in the spec.

Received on Wednesday, 30 March 2016 19:32:06 UTC