[Crm-sig] "Geographycal" CRM

Paul Cripps paul.cripps at southwales.ac.uk
Tue Dec 2 16:43:49 EET 2014


Hi Dan,
As Gerald says, if you use the GeoSPARQL classes you get the Simple Features you would like to use but you also then get all the extra loveliness of GeoSPARQL.
Re your problems:

1.       Circle is supported in WKT through the use of CIRCULARSTRING() or CURVEPOLYGON() – the latter being an areal feature. These are being included in Simple Features but are not in the 1.0 schema you listed. Many geodatabases already support these geometry types. As GeoSPARQL uses Simple Features, circles should also become available in due course (I haven’t tried passing in a WKT circle to see what happens; it may be that some implementations support this already).

2.       WKT is (in my opinion!) a lot nicer, cleaner and easy to use than GML. And there is good support for WKT in GIS. And there also is good GIS support for GML to be fair, but it’s a lot easier to handcraft and work with WKT, and it’s less verbose. Plus it’s portable and standards compliant rather than just eg pairs or groups of coordinates in some custom arrangement which may require additional (custom) parsing to eg plot onto a map or consume in some other web service.

3.       All flavours of Simple Features support z coordinates.

Schema.org is a handy platform for doing simple markup tasks but as Gerald quite rightly points out, it depends what you want to do with your geodata. If you want to eg ask queries like “show me all the bronze age sites within 2km of this study area polygon” you’re going to need something a bit more powerful.

Do have a look at the simple integration of GeoSPARQL I posted previously. Basically just uses Feature and Geometry (using Simple Features) aligned to an existing ontology. So no extra work really ;-)

All best,
Paul.


From: Crm-sig [mailto:crm-sig-bounces at ics.forth.gr] On Behalf Of Gerald Hiebel
Sent: 24 November 2014 09:16
To: 'Dan Matei'; 'crm-sig'
Subject: Re: [Crm-sig] "Geographycal" CRM

Hi Dan,
Just to note that “Simple Features” is part of GeoSPARQL and elevation in WKT and GML is recorded through 3 dimensional coordinates.
I believe the question is always how you want to make use of the geodata.
Best,
Gerald

Von: Crm-sig [mailto:crm-sig-bounces at ics.forth.gr] Im Auftrag von Dan Matei
Gesendet: Samstag, 22. November 2014 14:04
An: crm-sig
Betreff: Re: [Crm-sig] "Geographycal" CRM

Well...

I do not find GeoSPARQL (in toto) very useful to me. However "Simple Features" (http://schemas.opengis.net/sf/1.0/simple_features_geometries.rdf#<http://schemas.opengis.net/sf/1.0/simple_features_geometries.rdf>) is. I'll reuse sf:Polygon and sf:Line.

Problems:

1. I miss "Circle".
2. WKT is not very attractive :-(
3. I need "elevation" for the points (surfaces on slope).

Besides, I expect schema.org<http://schema.org> to be more popular in the next years :-)

Dan

PS. Thanks for the hints.

On 20 November 2014 22:29, Dan Matei <danmatei50 at gmail.com<mailto:danmatei50 at gmail.com>> wrote:
Thanks. I'll look carefully to GeoSPARQL. Of course I prefer to re-use elements from "canonical" ontologies and not to re-invent the warm wheel :-)

Best,

Dan

On 20 November 2014 20:51, Paul Cripps <paul.cripps at southwales.ac.uk<mailto:paul.cripps at southwales.ac.uk>> wrote:
Hi Dan,
Just to second Gerald, GeoSPARQL will probably give you the spatial components you require.

It is also possible to link GeoSPARQL directly to any ontology using the methodology outlined in the GeoSPARQL User Guide (Battle & Kolas, 2012). See the link below for an example of this. The approach I present is not as articulated or have as much potential for full on spatio-temporal use as CRMgeo but such lightweight approaches can still be useful; it’s working nicely for my current research scenario. I’m also using WKT rather than GML as it’s a bit less convoluted.

http://gstar.archaeogeomancy.net/2014/11/extending-crmeh-with-geosparql/

All best,
Paul.


From: Crm-sig [mailto:crm-sig-bounces at ics.forth.gr<mailto:crm-sig-bounces at ics.forth.gr>] On Behalf Of Gerald Hiebel
Sent: 18 November 2014 09:15
To: 'Dan Matei'; 'Øyvind Eide'
Cc: 'crm-sig'
Subject: Re: [Crm-sig] "Geographycal" CRM

Dear Dan,
Looking at your extension CPOT I find some classes like line, polygon, box... that are modelled in GeoSPARQL (http://www.opengeospatial.org/standards/geosparql) which uses GML. You could check there if you find other cpot elements like first_vertex,... . To my knowledge it should be possible to represent all schema classes in GeoSPARQL.
We build the CRMgeo extension for CRM to link to GeoSPARQL (http://www.ics.forth.gr/isl/index_main.php?l=e&c=661). It’s probably more effort to get into it, but you would be standards conform and there already exist some triple stores that can handle GeoSPARQL queries hoping to become more.
Best,
Gerald

Von: Crm-sig [mailto:crm-sig-bounces at ics.forth.gr] Im Auftrag von Dan Matei
Gesendet: Dienstag, 18. November 2014 10:38
An: Øyvind Eide
Cc: crm-sig
Betreff: Re: [Crm-sig] "Geographycal" CRM

That's the source, dear Øyvind:

http://schema.org/docs/schemas.html

Did I understood your question ?

Cheers,

Dan

PS. They just realeased an upgrade: http://lists.w3.org/Archives/Public/public-vocabs/2014Nov/0093.html


On 18 November 2014 10:24, Øyvind Eide <lister at oeide.no<mailto:lister at oeide.no>> wrote:
Dear Dan,

Thanks a lot for sharing!

A stupid question first: what does "schema" refer to in this context? The one sadly missing elements because you have stolen them, I mean?

Best,

Øyvind

On 14. nov. 2014, at 18:39, martin wrote:

>
>
>
> -------- Original Message --------
> Subject:      "Geographycal" CRM
> Date: Fri, 14 Nov 2014 17:23:29 +0200
> From: Dan Matei <danmatei50 at gmail.com<mailto:danmatei50 at gmail.com>>
> To:   crm-sig-owner at ics.forth.gr<mailto:crm-sig-owner at ics.forth.gr>
>
> Friends
>
>
> (Still) working on converting my legacy data to CRM+ I arrived to geography :-)
>
>
> I felt the need to steel some elements from schema and I came-up with the model (very informally) depicted here: http://www.culturalia.ro/img/CPOT-geo-model-2014-11-12.jpg.
>
>
> I could use some criticism, having enough doubts.
>
>
> So, please give me advice.
>
>
> Dan
>
>
> PS. CPOT stands for "CRM Properties of Things" and identifies my local extensions.
>
>
> -------------------------------------------------------------------------------------------------------------------
> Dan Matei
> Institutul Național al Patrimoniului (National Heritage Institute) - București
> Fundația Gellu Naum
> TermRom - Asociația Română de Terminologie
>
>
> _______________________________________________
> Crm-sig mailing list
> Crm-sig at ics.forth.gr<mailto:Crm-sig at ics.forth.gr>
> http://lists.ics.forth.gr/mailman/listinfo/crm-sig


_______________________________________________
Crm-sig mailing list
Crm-sig at ics.forth.gr<mailto:Crm-sig at ics.forth.gr>
http://lists.ics.forth.gr/mailman/listinfo/crm-sig



--
-------------------------------------------------------------------------------------------------------------------
Dan Matei
Institutul Național al Patrimoniului (National Heritage Institute) - București
Fundația Gellu Naum
TermRom - Asociația Română de Terminologie



--
-------------------------------------------------------------------------------------------------------------------
Dan Matei
Institutul Național al Patrimoniului (National Heritage Institute) - București
Fundația Gellu Naum
TermRom - Asociația Română de Terminologie



--
-------------------------------------------------------------------------------------------------------------------
Dan Matei
Institutul Național al Patrimoniului (National Heritage Institute) - București
Fundația Gellu Naum
TermRom - Asociația Română de Terminologie
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ics.forth.gr/pipermail/crm-sig/attachments/20141202/6c37dbef/attachment-0001.html>


More information about the Crm-sig mailing list