Subject: [LC response] To Jeremy Carroll
Thank you for your comment
on the OWL 2 Web Ontology Language last call drafts.
To address your points in turn:
- According to a normative part of RDF Concepts,
"Each member of the value space may be paired with any number (including zero) of members of the lexical space .... RDF may be used with any datatype definition that conforms to this abstraction, even if not defined in terms of XML Schema. ... RDF may be used with any datatype definition that conforms to this abstraction, even if not defined in terms of XML Schema."
So here RDF datatypes can have an empty lexical space and a trivial lexical-to-value mapping, as in owl:real.
This is contradicted in RDF Semantics, <http://www.w3.org/TR/rdf-mt/#dtype_interp>:
"Formally, a datatype d is defined by three items: 1. a non-empty set of character strings called the lexical space of d; ...."
There does not appear to be any technical reason for this prohibition on empty lexical spaces in RDF Semantics. For any new datatype, RDF systems must be upgraded to recognize which strings are ill-formed for that datatype. For an empty datatype, they must simply treat any string as ill-formed which is the desired behavior here. Thus, owl:real is eminently compatible with RDF systems and technically in compliance with one of the definitions supplied by the specifications. The Working Group will send an error report about this contradiction to the appropriate list.
- There are no negative effects from requiring the OWL 2 domain of discourse to be uncountable. The fundamentals of the semantics are unchanged; there is no need to significantly change implementations.
- The working group has discussed your document extensively and addressed the issues it raised. It is in fact partly in response to the concerns mentioned there that OWL 2 has owl:real (and owl:rational) as support for avoiding the problematic use of floating point numbers, not just for n-ary data predicates but also as a matter of modeling cleanliness.
Given these considerations, the Working Group has decided to make no change in response to your comment.
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, and whether you would like us to record you as Formally Objecting to the advancement of OWL 2 along the W3C Recommendation Track. (Note that according to the W3C Process, Formal Objections are made by individuals, not organizations.)
on behalf of the W3C OWL Working Group
CUT AND PASTE THE BODY OF THE MESSAGE (I.E. FROM "Dear" TO "Group") INTO THE BODY OF AN EMAIL MESSAGE. SET THE To:, CC:, AND Subject: LINES ACCORDINGLY.
PLEASE TRY TO REPLY IN A WAY THAT WILL ALLOW THREADING TO WORK APPROPRIATELY, I.E., SO THAT YOUR REPLY CONTINUES THE THREAD STARTED BY THE ORIGINAL COMMENT EMAIL
[[ owl:real: The new numeric datatypes specific to OWL 2 have been added partly to support reasoning with n-ary datatypes. Unions of other datatypes are not adequate for this purpose.
And indeed it is still in the current WDs
This is a forward compatibility hook, which is in general a bad idea. It breaks RDF conformance, since it is not a datatype (no lexical space, no lexical to value mapping). It is a fundamental change to OWL since it means that the size of the domain of discourse (including the literals) depends on the Continuum Hypothesis, and is certainly not countable.
And the forward compatibility for which it is a hook has known issues that need addressing: http://www.hpl.hp.com/techreports/2007/HPL-2007-37.pdf
TopQuadrant also intend to formally object to this feature
Jeremy Carroll, AC Rep., TopQuadrant