In drafting an RDF representation of MRTG[1], I am struggling with how close to stay to the DC and DwC minimalist style of ontology architecture. I find this well justified by the sentence "For example, though the data types and constraints are not provided in the term definitions, recommendations are made about how to restrict the values where appropriate" in the DwC introduction[2].
What I'm struggling with is how much to slant MRTG---which uses a lot of DwC--more toward OWL or more towards only RDF. See [3] for discussion of the dillemma raised when you put any of the DC or DwC RDF files into Protege4 or WonderWeb, the Manchester OWL Validator[4]. Roughly speaking, I took [3] to be the Manchester and Stanford axis of OWL to be saying, "In the future, don't expect to talk about OWL without stronger typing." {As I understand DwC/RDF, this all is accomplished by declaring terms to be mainly rdf:Property with the only typification coming from the assertions using the predicate dwcattributes:organizedInClass }.
So my questions are possibly:
1. What committment does TAG have to OWL 2. Has TAG examined the applicability to TDWG of the W3 OWL 2 recommendations recently advanced to the Proposed Recommendation stage [4], [5], or its relatives. 3. If no to 2, is there a plan to do so? Against what use cases?
A few months ago, Roger seemed to wax enthusiastic about Knowledge Representation and reasoning on biodiversity data, but I haven't seen much traffic about what are the modeling requirements to support that, or whether TAG has a plan to move that way.
Bob Morris
[1] MRTG 0.8 http://www.keytonature.eu/wiki/MRTG_Schema_v0.8 [2] DwC intro http://rs.tdwg.org/dwc/index.htm [3] p4-feedback thread on rdf:Property https://mailman.stanford.edu/pipermail/p4-feedback/2009-October/002448.html [4] WonderWeb http://www.mygrid.org.uk/OWL/Validator [5] OWL2 Overview http://www.w3.org/TR/2009/PR-owl2-syntax-20090922/ [5] OWL2 rdf semantics http://www.w3.org/TR/2009/PR-owl2-rdf-based-semantics-20090922/
Hi Bob,
I don't know if you are going to Montpellier or not (wouldn't be TDWG without you) but I am hoping to discuss this there in an informal way.
The management of the TDWG ontology has become a pain as no one is doing it in any coordinated way and no one is keen to pay for it to be done. I have been talking to various people off line and have a cunning plan that I outlined in a blog here:
http://www.hyam.net/blog/archives/643
I was waiting to hear back from Donald (as TDWG Chair) and David Remsen (as GBIF access point) to see if they would support this course of action. I didn't want to advocated an approach that didn't have the high level support of the organisations involved.
I have now heard back from them both ( but have been travelling and not had a chance to post to this list) and throw this open to discussion before leaving for a week to a place with limited email contact (family).
Yes I do love OWL and OWL 2 - if anything for the clarity of thinking they impose. I don't wish to impose this on others, particularly taxonomists, who (according to my calculations using OWL) do very little clear thinking. See http://www.hyam.net/blog/archives/598 for one set of argument but there are man more. BTW Taxacom have already done this to death.
Hope this helps,
Roger
On 15 Oct 2009, at 21:52, Bob Morris wrote:
In drafting an RDF representation of MRTG[1], I am struggling with how close to stay to the DC and DwC minimalist style of ontology architecture. I find this well justified by the sentence "For example, though the data types and constraints are not provided in the term definitions, recommendations are made about how to restrict the values where appropriate" in the DwC introduction[2].
What I'm struggling with is how much to slant MRTG---which uses a lot of DwC--more toward OWL or more towards only RDF. See [3] for discussion of the dillemma raised when you put any of the DC or DwC RDF files into Protege4 or WonderWeb, the Manchester OWL Validator[4]. Roughly speaking, I took [3] to be the Manchester and Stanford axis of OWL to be saying, "In the future, don't expect to talk about OWL without stronger typing." {As I understand DwC/RDF, this all is accomplished by declaring terms to be mainly rdf:Property with the only typification coming from the assertions using the predicate dwcattributes:organizedInClass }.
So my questions are possibly:
- What committment does TAG have to OWL
- Has TAG examined the applicability to TDWG of the W3 OWL 2
recommendations recently advanced to the Proposed Recommendation stage [4], [5], or its relatives. 3. If no to 2, is there a plan to do so? Against what use cases?
A few months ago, Roger seemed to wax enthusiastic about Knowledge Representation and reasoning on biodiversity data, but I haven't seen much traffic about what are the modeling requirements to support that, or whether TAG has a plan to move that way.
Bob Morris
[1] MRTG 0.8 http://www.keytonature.eu/wiki/MRTG_Schema_v0.8 [2] DwC intro http://rs.tdwg.org/dwc/index.htm [3] p4-feedback thread on rdf:Property https://mailman.stanford.edu/pipermail/p4-feedback/2009-October/002448.html [4] WonderWeb http://www.mygrid.org.uk/OWL/Validator [5] OWL2 Overview http://www.w3.org/TR/2009/PR-owl2-syntax-20090922/ [5] OWL2 rdf semantics http://www.w3.org/TR/2009/PR-owl2-rdf-based-semantics-20090922/
-- Robert A. Morris Professor of Computer Science (nominally retired) UMASS-Boston Associate, Harvard University Herberia email: ram@cs.umb.edu web: http://bdei.cs.umb.edu/ web: http://etaxonomy.org/FilteredPush http://www.cs.umb.edu/~ram phone (+1)617 287 6466 _______________________________________________ tdwg-tag mailing list tdwg-tag@lists.tdwg.org http://lists.tdwg.org/mailman/listinfo/tdwg-tag
participants (2)
-
Bob Morris -
Roger Hyam