- From: Andy Seaborne via GitHub <noreply@w3.org>
- Date: Tue, 03 Feb 2026 15:32:22 +0000
- To: public-shacl@w3.org
afs has just created a new issue for https://github.com/w3c/data-shapes:
== Should symmetric RDF be allowed during rule set evaluation? ==
[Symmetric RDF](https://www.w3.org/TR/rdf12-concepts/#section-symmetric-rdf) is RDF where the subject is not restricted to a URI or a blank node. Teh subject can also be a literal or a triple term, same as the object. The predicate position is still required to be a URI.
It naturally arises in inference - for example, in RDFS:
[**rdfs2**](https://www.w3.org/TR/rdf12-semantics/#dfn-rdfs2)
```
:p rdf:range xsd:integer .
:x :p 123 .
->
123 rdf:type xsd:integer .
```
or from "inverse" (not just owl:inverseOf, which is restricted to individuals). SPARQL has reverse paths (if we allow that path form):
```sparql
RULE { ?x :p ?z } WHERE { ?x ^:q ?z }
```
For working with RDF software, the outputs of ruleset evaluation must be a legal RDF graph.
The question is whether symmetric RDF would be allowed during evaluation.
See also:
1. SPARQL CONTRUCT
2. [**rdfd1**](https://www.w3.org/TR/rdf12-semantics/#dfn-rdfd1)
```
ex:a ex:p "123"^^xsd:integer .
->
ex:a ex:p _:x .
_:x rdf:type xsd:integer .
```
3. [Generalized RDF](https://www.w3.org/TR/rdf12-concepts/#section-generalized-rdf)
Please view or discuss this issue at https://github.com/w3c/data-shapes/issues/761 using your GitHub account
--
Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config
Received on Tuesday, 3 February 2026 15:32:23 UTC