Re: ISA Core Location Vocabulary

Hello,

Please see my comments inline...

Regards,
Frans

On 2013-12-24 0:41, Andrea Perego wrote:
> On Mon, Dec 23, 2013 at 12:36 PM, Frans Knibbe | Geodan 
> <frans.knibbe@geodan.nl <mailto:frans.knibbe@geodan.nl>> wrote:
>
>     Hello Andrea,
>
>     It is nice to know that the JRC is seriously researching Linked
>     Data for INSPIRE. Are you also trying to coordinate the different
>     EU activities? Or could we expect some kind of shared repository
>     of INSPIRE RDF vocabularies? Creating ontologies based on
>     specifications of INSPIRE themes is a lot of work, so it would be
>     a pity if many people or organisations are doing that independently.
>
>
> (Wearing my JRC hat now...) This is exactly one of the motivations 
> behind such activity. The first objective is not to create "official" 
> RDF vocabularies for INSPIRE, but rather methodologies for their 
> definition, to be agreed upon within the INSPIRE community. The next 
> steps will depend on the outcome of such work. This includes the 
> possibility of following a collaborative approach to the creation of 
> RDF vocabularies for INSPIRE.
That would be great! Is there already something to be found on the web 
about this activity?
>
>     That brings me to another question: Let us assume that in the near
>     future there will be full-fledged vocabularies for each of the
>     INSPIRE themes, and for the basic semantics that are shared
>     between the themes. What would then be the purpose of the Location
>     Core vocabulary? Not only geometry would be defined externally,
>     but geographical names and addresses too.
>
>
> I think this is something to be decided by the LOCADD CG.
>
> As I mentioned earlier in this thread, originally, the LOCN voc was 
> meant to be a "core" vocabulary to support cross-sector 
> interoperability on location information. Which also means that it can 
> be extended depending on the requirements of each specific sector, by 
> defining appropriate profiles. An external and INSPIRE-compliant 
> definition of notions like "geographical names" and "addresses" would 
> be consistent with this approach. Of course, INSPIRE vocabularies may 
> re-use and/or extend what is already defined in the LOCN voc (e.g., 
> locn:Address, which is modelled on the INSPIRE AddressRepresentation 
> datatype).
>
> So, the question is whether the LOCN voc should be still considered as 
> a "core" vocabulary, or it should evolve otherwise.
>
> Personally, I would be in favour of a design strategy that would allow 
> a wide re-use of the LOCN voc, not only limited to INSPIRE, and 
> possibly also outside Europe.
Yes, I think that make sense. The LOCN vocabulary could provide a light 
weight entry-level vocabulary for location data, to be extended when 
needed. I think it will be useful in many cases that way.

But doesn't that mean that LOCN should have (simple) definitions of its 
own? Take the geometry class for example. Using LOCN as it is now, 
people could publish geometries in many different ways. These ways would 
not interoperable. And defining an extension in the direction of a more 
advanced vocabulary (one from INSPIRE, for example), would be difficult 
because the starting point is not clearly defined. Alternatively, let's 
say that LOCN says that a geometry should be encoded as WKT. That would 
provide  a simple and interoperable way of expressing geometry, and 
extensions to other vocabularies could be easily facilitated.

Or take the case of geographical names... If LOCN makes clear that 
locn:geographicName maps directly to inspire:SpellingOfName, it would be 
easier to extend data publication to fuller INSPIRE specifications, and 
to other specifications as well. And from the perspective of INSPIRE, it 
would be clear how LOCN can be incorporated in yet to be developed 
vocabularies.

>
> Cheers,
>
> Andrea