Re: [Minutes-BP] 2016-04-20

Thank you Linda and Andrea for providing access to the ontology, and thank
you Simon for putting it in perspective.

I have tried to look up two things that I think are essential corner stones
of a spatial ontology: the definition of a point (in 1/2/3-dimensional
space) and the (associated) definition of a coordinate reference system. I
did not find something useful, but perhaps I did not look wel enough.

My gut feeling is that to arrive at a sensible general spatial ontology an
automated transformation of ISO191** models can not provide the desired
result. There are several reasons for that. One is that the ISO191** models
are models for geographic data, not for spatial data in general. Another
one is the baggage from earlier models that Simon mentioned. A third one is
that there is less freedom to define logical modules or layers and to be as
simple as possbile.

A spatial ontology based on ISO191** models could be very useful, but I
think such an ontology would be a lot more useful if it was hand-crafted,
starting from the basic premises.

About the work program resulting from the Linking Spatial Data workshop
that Simon mentions: could it be that the SDWWG Best Practices deliverable
is that very work program?

Regards,
Frans

2016-04-25 1:13 GMT+02:00 <Simon.Cox@csiro.au>:

> Yes - this is the official ISO/TC 211 OWL implementation, overseen by the
> ISO/TC 211 Group on Ontology Management, under chair Jean Brodeur (Canada).
> These ontologies follow the rules defined in ISO 19150-2, and are generated
> automatically - using automation tools in Sparx Enterprise Architect UML
> tool - from the UML originals in the so-called Harmonized Model repository
> hosted on behalf of TC 211 by JRC.
>
> The OWL implementations were being developed successively, since there was
> a need to do significant checking of each one after generation and in some
> cases non-visible pieces of the UML model (e.g. tagged values) required
> adjustment in order to get a conformant OWL. However, the work appears to
> have stalled about a year ago, incomplete with respect to the set of
> standards in the ISO 19100 series that actually have UML models.
>
> Linda, Clemens and Stuart Williams (Epimorphics) did a study of OWL
> implementations for INSPIRE a couple of years ago. Since INSPIRE is based
> on the OGC/ISO stack, this necessarily included an evaluation/critique of
> the ISO 19150-2 patterns and process. Linda can say more about this, but my
> impression was that there was significant scepticism about the brute-force
> rule-based transformation, since what came out could best be characterized
> as UML-in-OWL, therefore not very idiomatic OWL/RDF [1][2]. [OTOH, there
> are lots of ontologies out there that display as much or more baggage from
> earlier frameworks - look at all those based on BFO in the biomedical
> community for example!]
>
> [1]  My own successive OWL implementations of O&M addressed the same issue
> - the one presented at the 2013 Semantic Web conference largely followed
> ISO 19150-2, while om-lite started from scratch - see
> http://www.semantic-web-journal.net/content/ontology-observations-and-sampling-features-alignments-existing-models-0
> for more details and references.
>
> [2] However, a conversation with Clemens at the time suggested that a
> bigger concern than particular OWL-style was a more fundamental 'why even
> do geospatial in OWL?', particularly at the level of detail represented in
> many of the UML models which are implementation-level with lots of typed
> attributes. I guess the SDW Working Group addressed this question in the
> original London workshop when the decision was made to set a work program
> that follows the RDF/semantic web path? (I wasn't there so not sure how
> rubust the discussion was on this point.)
>
>
> Simon J D Cox
> Research Scientist
> Land and Water
> CSIRO
> E simon.cox@csiro.au T +61 3 9545 2365 M +61 403 302 672
>    Physical: Reception Central, Bayview Avenue, Clayton, Vic 3168
>    Deliveries: Gate 3, Normanby Road, Clayton, Vic 3168
>    Postal: Private Bag 10, Clayton South, Vic 3169
> people.csiro.au/C/S/Simon-Cox
> orcid.org/0000-0002-3884-3420
> researchgate.net/profile/Simon_Cox3
>
> ________________________________________
> From: Linda van den Brink <l.vandenbrink@geonovum.nl>
> Sent: Saturday, 23 April 2016 4:38 PM
> To: Andrea Perego
> Cc: Phil Archer; SDW WG Public List
> Subject: Re: [Minutes-BP] 2016-04-20
>
> More on the status of this: Status is official tc211 ontology
> implementation.
>
> > Op 21 apr. 2016 om 10:55 heeft Andrea Perego <
> andrea.perego@jrc.ec.europa.eu> het volgende geschreven:
> >
> >> On 21/04/2016 10:51, Linda van den Brink wrote:
> >> Hi all,
> >>
> >> The ISO 19107 and other ISO19xxx are available as OWL ontologies here:
> >>
> https://github.com/ISO-TC211/GOM/tree/master/isotc211_GOM_harmonizedOntology
> >>
> >> This is what I was referring to yesterday. As far as I know these are
> work in progress. Certainly relevant to look at in the context of defining
> a spatial ontology.
> >
> > Many thanks, Linda.
> >
> > I've tried to use LODE for a human-readable preview, but the import URLs
> > (http://def.isotc211.org/iso19107/*) are not working, so the modules
> > need to be visualised separately.
> >
> > In case it may turn to be useful, I include the relevant links below.
> >
> > Andrea
> >
> > ----
> >
> > SpatialSchema
> >
> >
> http://www.essepuntato.it/lode/https://raw.githubusercontent.com/ISO-TC211/GOM/master/isotc211_GOM_harmonizedOntology/19107/2003/ISO19107_2003SpatialSchema.owl
> >
> > Geometry
> >
> >
> http://www.essepuntato.it/lode/https://raw.githubusercontent.com/ISO-TC211/GOM/master/isotc211_GOM_harmonizedOntology/19107/2003/iso19107Geometry.owl
> >
> > GeometryRoot
> >
> >
> http://www.essepuntato.it/lode/https://raw.githubusercontent.com/ISO-TC211/GOM/master/isotc211_GOM_harmonizedOntology/19107/2003/iso19107GeometryRoot.owl
> >
> > GeometricPrimitive
> >
> >
> http://www.essepuntato.it/lode/https://raw.githubusercontent.com/ISO-TC211/GOM/master/isotc211_GOM_harmonizedOntology/19107/2003/iso19107GeometricPrimitive.owl
> >
> > CoordinateGeometry
> >
> >
> http://www.essepuntato.it/lode/https://raw.githubusercontent.com/ISO-TC211/GOM/master/isotc211_GOM_harmonizedOntology/19107/2003/iso19107CoordinateGeometry.owl
> >
> > GeometricComplex
> >
> >
> http://www.essepuntato.it/lode/https://raw.githubusercontent.com/ISO-TC211/GOM/master/isotc211_GOM_harmonizedOntology/19107/2003/iso19107GeometricComplex.owl
> >
> > GeometricAggregates
> >
> >
> http://www.essepuntato.it/lode/https://raw.githubusercontent.com/ISO-TC211/GOM/master/isotc211_GOM_harmonizedOntology/19107/2003/iso19107GeometricAggregates.owl
> >
> > Topology
> >
> >
> http://www.essepuntato.it/lode/https://raw.githubusercontent.com/ISO-TC211/GOM/master/isotc211_GOM_harmonizedOntology/19107/2003/iso19107Topology.owl
> >
> > TopologyRoot
> >
> >
> http://www.essepuntato.it/lode/https://raw.githubusercontent.com/ISO-TC211/GOM/master/isotc211_GOM_harmonizedOntology/19107/2003/iso19107TopologyRoot.owl
> >
> > TopologicalPrimitive
> >
> >
> http://www.essepuntato.it/lode/https://raw.githubusercontent.com/ISO-TC211/GOM/master/isotc211_GOM_harmonizedOntology/19107/2003/iso19107TopologicalPrimitive.owl
> >
> > TopologicalComplex
> >
> >
> http://www.essepuntato.it/lode/https://raw.githubusercontent.com/ISO-TC211/GOM/master/isotc211_GOM_harmonizedOntology/19107/2003/iso19107TopologicalComplex.owl
> >
>
>
>