ISSUE-17
role punning
RESOLVED: Object and datatype role punning
- State:
- CLOSED
- Product:
- Raised by:
- Jeremy Carroll
- Opened on:
- 2007-10-24
- Description:
Reported by jlc415, Jun 11, 2007 The current model-theoretic semantics document allows punning between all signature elements, including punning between object and datatype role names. It is extremely difficult to imagine a case in which punning between different types of roles would be anything other than user error. Furthermore, this semantic model is incompatible with that of OWL-DL. Under OWL-DL a role name was required to be given a single interpretation as object or datatype across all use in the ontology; under OWL 1.1 this is no longer the case.
- Related Actions Items:
- No related actions
- Related emails:
- Re: Agenda for Washington F2F (from pfps@research.bell-labs.com on 2008-03-23)
- Re: Punning discussion (from jjc@hpl.hp.com on 2008-01-23)
- ISSUE-17 (role punning): REPORTED: Object and datatype role punning (from sysbot+tracker@w3.org on 2007-10-24)
Related notes:
2007-11-20 09:37:55: See also ISSUE 18 [Ian Horrocks]
2008-04-14 07:53:09: Resolved at F2F2 [1]: Property punning forbidden as per Boris Motik's proposal [2], the reason being that technically we don't know how to do it. NOTE that this includes punning of AnnotationProperties (i.e., no punning of *any* types of property).
[1] http://www.w3.org/2007/OWL/wiki/F2F2_Minutes#Issue_17
[2] http://www.w3.org/mid/000001c89659$6d8508f0$2a12220a@wolf [Ian Horrocks]
Display change log.