|The set of datatypes in OWL 2 RL is a subset of the set of OWL 2
datatypes. The rationale given for this restriction is that the intersection of any two values spaces should be empty or infinite in order to obtain the desired computational properties. I believe this is not correct. In OWL 2 RL one does not need to consider unnamed individuals. Since IRIs cannot be mapped to objects in value spaces and the interpretation of every literal is fixed, finite (intersections of) value spaces do not require reasoning by case. Therefore, the language keeps the desired computational properties, even with finite datatypes.
It seems to me that the list of datatypes in OWL 2 RL needs to be extended or better motivated.
And from 21:
- in OWL 2 RL I do not understand why the following features are not included -- they are easily implemented the rules:
PeterPatel-Schneider 01:13, 6 March 2009 (UTC)
[Draft Response for LC Comment 20:] JDB1
Thank you for your message
on the OWL 2 Web Ontology Language last call drafts.
You are correct that OWL 2 RL does not need the intersection of datatypes to be empty or infinite. Accordingly the datatypes in OWL 2 RL have been adjusted to include all the OWL datatypes that are restrictions of xsd:decimal and xsd:string and also xsd:boolean.
In response to another comment
on the difficulty of implementing datatypes in rules systems, owl:real and owl:rational have been removed from OWL 2 RL. This possibility was mentioned in Feature At Risk #2.
The diffs for the initial portion of the changes can be found at http://www.w3.org/2007/OWL/wiki/index.php?title=Profiles&diff=18687&oldid=18109
Please acknowledge receipt of this email to <mailto:email@example.com> (replying to this email should suffice). In your acknowledgment please let us know whether or not you are satisfied with the working group's response to your comment.
Regards, Peter F. Patel-Schneider on behalf of the W3C OWL Working Group