RE: Sub-properties for locn:geometry? (was: RE: ISA Core Location Vocabulary)

Frans Knibbe wrote:

On 2014-01-10 11:55, John Goodwin wrote:

Hi all,



I am still working my way through all the recent activity on this group so apologies if this issue has already been discussed. One requirement Ordnance Survey has in its administrative geography linked data (http://data.ordnancesurvey.co.uk/datasets/boundary-line) is to specify different geometries. In this case we have two: a 'representative point' and an 'extent'. So I was wondering if we should create a few subproperties of locn:geometry for a few common geometry types, e.g. "hasCentroid" [1], "boundary", "extent", "representativePoint", "mbr" etc.



Thoughts?
An interesting idea. Aren't all these geometries derived geometries? Somehow a result of a calculation performed on the original geometry? I would

Yes they are derived geometries – though I would argue that most/all published data will contain a derived geometry. Certainly most of our geometries will be derived as the result of generalisation or some other calculation from the source content.

expect at least a need for (SPARQL) functions that can return those derived geometries. But that puts extra demands on SPARQL endpoints, so why not simple make those data directly available? I am sure it will be helpful in some cases.

Well in the case of the Ordnance Survey linked data we want to publish both a representative point of an administrative region (allows the region to be centred on a map, and simple searches/queries) and we also publish the ‘extent’ which is a polygon or multisurface indicating the region the administrative area covers. As an aside one of our non-linked data products also publishes the boundary of administrative areas (a polyline).


By the way, what is the difference between 'extent' and 'mbr'?

In my view an extent will be a polygon/multisurface representing the area covered by a place. An MBR is a minimum bounding rectangle of the extent.


As all the subproperties are geometries themselves, why not define them as subclasses?

The simple feature ontology does subclass geometry to polygons, points etc. Are you suggesting having new classes for things like ‘centroid’ (e.g. subclass of sf:Point) etc?

John

This email is only intended for the person to whom it is addressed and may contain confidential information. If you have received this email in error, please notify the sender and delete this email which must not be copied, distributed or disclosed to any other person.

Unless stated otherwise, the contents of this email are personal to the writer and do not represent the official view of Ordnance Survey. Nor can any contract be formed on Ordnance Survey's behalf via email. We reserve the right to monitor emails and attachments without prior notice.

Thank you for your cooperation.

Ordnance Survey
Adanac Drive
Southampton SO16 0AS
Tel: 08456 050505
http://www.ordnancesurvey.co.uk