- From: Miles, AJ \(Alistair\) <A.J.Miles@rl.ac.uk>
- Date: Wed, 7 Dec 2005 16:43:47 -0000
- To: <public-swbp-wg@w3.org>
Hi all, Here is my review of the document: [1] http://lists.w3.org/Archives/Public/public-swbp-wg/2005Nov/att-0022/QCRs-revised-01.html I am aware of no serious issues that should prevent this document being published as a Working Draft, or as a Working Group Note. General Comments ---------------- I am satisfied that the use cases presented are compelling, and justify the documentation of a work-around solution. I am also satisfied that the work-around solution is both sufficiently inconvenient, and lacks the required expressivity with respect to the use cases described, as to justify the documentation of a 'non-endorsed' extension to OWL DL. If it is true that, as Alan says [2], QCRs are a standard construct in DLs (and therefore do not affect decidability), and that they have a publicly documented model-theoretic semantics, I see no objection to a non-endorsed extension to OWL DL to support the expressiong of QCRs being described in [1]. I would like to see references to the elaborated model-theoretic semantics for QCRs present in the references section. Given that, as I understand it, the main purpose of OWL DL is to provide a decidable subset of OWL, the reader needs to know that the documented OWL extension does not in principle affect decidability. Specific Comments ----------------- ... Simple part-whole relations in OWL Ontologies ... This is the wrong HTML title. ... (NOTE: Some Italian restaurants ... This is supposed to be a joke, right? A lot of people (esp. outside US) won't get it. Suggest you remove it, or if there is an important point, explain the point better. ... It therefore cannot deal with the other use cases or with generalisations of 3 ... It's not at all clear what is meant by 'generalisations of 3'. ... ObjectProperty(has_desert, super(has_course) range(Desert)) ... Why does this assertion have a comma in it, when others don't? I suggest you use a consistent style. ... Class(Minimal_Italian_Dinner partial intersectionOf( Restriction( has_starter cardinality(1) )) Restriction( has_main_course cardinality(1) )) Restriction( has_desert cardinality(1) )) )) ... Are there too many closing brackets here? (I don't know the OWL abstract syntax well enough.) I suggest you run some sort of syntax checks on all snippets of OWL abstract syntax. ... restriction( has_dessert someValuesFrom(Tiramasu_course))) restriction( has_course someValuesFrom(IceCream_course))) ... Again, too many closing brackets? ... The syntax looks like this [3] ... This is confusing. Reads as: the extension looks like this (see below) OR the extension looks like this (see reference [3]). I agree with Jacco's suggestion [4] that you introduce the extension to the abstract syntax, and give examples in the abstract syntax, before explaining the mapping to RDF triples. ... In the RDF ... This is pretty loose wording. Typos ----- The following typos should be addressed before publication: s/piatoo/piatto/ s/valujes/values/ s/undertainty/uncertainty/ s/lline/line/ s/hve/have/ s/has_couse/has_course/ s/has_main_couse/has_main_course/ s/Desert/Dessert/ s/incoporatd/incorporated/ s/represtation/representation/ s/cardinalty/cardinality/ ---- That's all. Cheers, Alistair. [2] http://lists.w3.org/Archives/Public/public-swbp-wg/2005Nov/0044.html [4] http://lists.w3.org/Archives/Public/public-swbp-wg/2005Nov/0042.html --- Alistair Miles Research Associate CCLRC - Rutherford Appleton Laboratory Building R1 Room 1.60 Fermi Avenue Chilton Didcot Oxfordshire OX11 0QX United Kingdom Email: a.j.miles@rl.ac.uk Tel: +44 (0)1235 445440
Received on Wednesday, 7 December 2005 16:45:14 UTC