search

w3c / sdw

Repository for the Spatial Data on the Web Working Group
https://www.w3.org/2020/sdw/
148 stars 81 forks source link

New project proposal: OWL Space #1095

Closed FransKnibbe closed 4 years ago

FransKnibbe commented 5 years ago

In the footsteps of the OWL Time ontology, a domain independent web ontology for spatial data should be developed, allowing unification of the many different models, exchange formats and storage formats for spatial data.

One possible approach is to start with GeoSPARQL and extend it to allow usage for other purposes next to vector geography.

A more elaborate call to arms can be found in this message: https://lists.w3.org/Archives/Public/public-sdwig/2018Oct/0090.html. Also notable is the follow-up discussion on the list.

Other discussion threads on the SDWIG public list in which the idea of a spatial ontology is discussed:

Prior art

Following is a list of web resources that are relevant to the idea of a cross-domain spatial ontology for web data.

chris-little commented 5 years ago

Hi Frans, I think it important to get the scope defined. The 2017 OWL-Time ontology went for full backward compatibility with OWL-Time 2006, which may cause problems, as the previous version has the TimeDate construct which, to me, is a not very clean compromise between coordinates and calendars. So, first question: are both Euclidean ('straight') and Riemannian ('curved') coordinate spaces envisaged? Second question: are both continuous (as in first question) and discrete ('nodes and arcs') reference systems envisaged? Perhaps I have just defined quadrants to think about?

My colleague Mark Hedley has also just asked me to think about the difference between a Coordinate System and a Coordinate Reference System. "More grist to the mill" as we say in English.

Chris

FransKnibbe commented 5 years ago

Hello Chris,

Backward compatibility should certainly be a requirement, or at least something to try to achieve as well as possible. I am hopeful that it will be possible to ensure compatibility with existing standards by including the real mathematical foundations in the ontology. There are many models, standards, storage formats and exchange formats for spatial data. But they all should be compatible at a fundamental level. So with the ability to define everything in fundamental terms, backward compatibility should be achievable (fingers crossed).

To answer your questions: 1) With my background in geography, I think there should at least be support for Cartesian, curved linear, and spherical coordinate spaces, up to three dimensions. That partly includes Euclidean and Riemannian coordinate spaces. So I guess the answer should be yes. 2) By 'discrete references systems', do you mean a system that uses a tessellation to indicate location (as in Discrete Global Grid Systems)?

As for CS vs CRS: At least in the geography domain the definition seems to be that a CRS is a CS plus a datum. Datums are important to define and to share data about, but not required for all spatial expressions. So coordinate systems and datums should probably be separate things in the ontology, but combinable.

Greetings, Frans

chris-little commented 5 years ago

Hi Frans, Re (2), No, I did not really mean tessellations, but that is perhaps a helpful way to think. DGGS is definitely not what I meant. In the time ontology, there is the possibility of the simplest temporal ordering of events, linked to geological or archaeological layers, king lists, ice cores, with reasoning (before, after, during, overlapping, etc). There is no concept of duration, just ordering.

I envisage that there may be a simple spatial 'ordering', but without distance. Two things may be adjacent, or 'further away' with more intermediate adjacent things. Nodes on a network are the obvious example.

Tesellations are not quite the same, as most tesselations involve a proper distance/measure to get precise shapes. They straddle both ideas. As does DGGS if (measured) location is replaced by a (counted) location of grid cell identifier.

chris-little commented 5 years ago

In other terms, I am suggesting that, rather than start with measures/CRS/datum/etc, start with the proximity algebras such as https://en.wikipedia.org/wiki/DE-9IM and perhaps the 3D equivalent .

FransKnibbe commented 5 years ago

Ah, I understand. Yes, starting with topological relationships could be a good idea. An ontology has to start somewhere, and if it is possible to start without touching the difficult subject of coordinate systems perhaps starting there could make a smooth beginning.

A good source of information on that topic seems to be Topological Relationships and Their Use, which describes 69 different relationships in 1, 2 or 3 dimensions.

As a starting point I earlier considered starting with the definition of a point in a 1, 2 or 3 dimensional cartesian coordinate space. As it would probably wise to modularize a spatial ontology, having multiple simple starting points (in separate modules) could also be an option.

As for relative proximity of nodes in a network: isn't graph theory essentially non-spatial? I know graphs are used for spatial problems, but does that warrant inclusion in a spatial ontology?

6a6d74 commented 5 years ago

During the SDW-IG teleconference on 16-Jan (minutes: www.w3.org/2019/01/16-sdw-minutes.html) we said:

OWL Space proposal brinkwoman: I was talking with Jo yesterday about that. Sometimes, people referred to the idea as spatial ontology. Jo joined me as co-chair of the OGC GeoSemantic Working Group.

(Related to the previous item - we’ll all try to review the API Guidelines by the beginning of Feb - we’ll coordinate the input via the SDW group discussion) brinkwoman: We wanted to move the topic to the Geosemantic Working Group. A Domain Working Group from an OGC perspective, which is open to non OGC people. jtandy: That sounds fine with me. It would also give some focus to the Geosemantic Working Group. josephabhayaratna: Having a working group plan to work on it seems a good thing. +1 +1 +1 +1 billroberts: Sounds like a good idea. We discussed in the best practices some of the things missing from GeoSparql. I would concentrate on the simple bits as far as practical. … My general feeling about using ontologies is that most people don't care about the fancy bits. If we can standardize the core things, then that's good. josephabhayaratna: Just to follow up. Driving it from work that is already happening. Trying to keep it focused and relevant. … Having a look at problems that people really have. We want to bring people that are struggling with concrete problems. sounds like the ideal approach to me, Jo - thanks jtandy: Linda or Jo, if you can take an action to update the issue and point people at the right OGC group, that would be good josephabhayaratna: Happy to take that action. brinkwoman: Should we create a proper project for it? jtandy: Yes, move it to a concrete project

So - it looks like we have some concrete proposals, with endorsement from the IG.

Actions / decisions:

  1. Pursue this work within the OGC Geosemantics DWG - it's a DWG, so it's open to non-OGC members
  2. Create a new Project for this activity in the SDW repo

@jabhay - would you be the best person to own this push it forward?

marqh commented 5 years ago

Hello

I think that specification of Coordinate Reference Systems is key to a representation of geolocation within a Space Ontology.

There are recent activities from within OGC & ISO, extending the Coordinate Reference System Model (ISO19111) (docs.opengeospatial.org/as/18-005r4/18-005r4.html) and providing an encoding format for instances of Coordinate Reference System Definitions. (ISO19162) (updated with respect to 19111, not yet published by OGC/ISO, but imminent) (a previous iteration is available here http://docs.opengeospatial.org/is/12-063r5/12-063r5.html)

This is a wide ranging and complicated topic. Initially this may be able to be sensibly handled by referencing out to these standards and encodings, rather than implementing all the details explicitly within an Ontology. This may at least provide a valuable first step

For example, a URI referencing a CRS entity defined in text using Well Known Text for Coordinate Reference Systems might provide a basis for spatial referencing within the proposed Space Ontology.

The complications involved in representing the Coordinate Reference System model fully within an Ontology may be seen as intellectually interesting, but could be a significant activity in its own right, without bringing proportionate benefit.

I am happy to provide some more input on this if it is valuable

mark

FransKnibbe commented 5 years ago

Hello,

It is great to see the idea being moved forward. And to a community group that is free for everyone to join too. I think that's very important. A concern could be that when something actually will be developed, it will be too heavily focused on geographic data. If that happens, the desire of aiding cross-domain interoperability could be at risk. And I really think that should be a core requirement, however small or modest the beginnings might be. So it will be important to reach out to people that by nature do not flock to OGC community groups and might have never heard of the OGC in the first place. Luckily, the OGC is already in a process of expanding its interests wider than basic geography, so that would not be breaking current practice, so it seems.

Greetings, Frans

FransKnibbe commented 5 years ago
  1. Pursue this work within the OGC Geosemantics DWG - it's a DWG, so it's open to non-OGC members

Is the OGC Geosemantics DWG really open to non-OGC members? Here is part of the reply I got after having tried to subscribe to the e-mail list: 'Your request to the GeoSemWeb mailing list Subscription request has been rejected by the list moderator. The moderator gave the following reason for rejecting your request: "This is an OGC member's only list. ...'

To become a member of the OGC as an individual, it seem I would have to pay USD 550 per year.

Regards, Frans

ogcscotts commented 5 years ago

Hi Frans - I am looking into the list moderation as this list is supposed to be public. Scott

lieberjosh commented 5 years ago

Hi,

Thought we’d made the geosemantics list public, but the group is not. The Spatial Data on the Web Interest Group is supposed to be public, I believe, and is in essence also a Geosem subgroup.

-Josh

On Mar 12, 2019, at 1:50 PM, Scott Simmons notifications@github.com wrote:

Hi Frans - I am looking into the list moderation as this list is supposed to be public. Scott

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

ogcscotts commented 5 years ago

The list is now public, so please try again to subscribe, Frans. Scott

FransKnibbe commented 5 years ago

Scott, Josh, thank you for being on the case. I have just tried to subscribe again. And this time the subscription was succesful.

lvdbrink commented 5 years ago

Under discussion in GeoSemantics DWG (OGC)

nicholascar commented 5 years ago

I have just presented on the Australian spatial Linked Data infrastructure LocI at the European Semantic Web Conference yesterday and discussed with a number of attendees possible enhancements to the GeoSPARQL Ontology which could be picked up in OWL Space. There seems to be a common set of requirements from at least attendees here, some of which I’ve placed in my informally presented and open-ended GeoSPARQL Extensions Ontology: https://github.com/CSIRO-enviro-informatics/geosparql-ext-ont. I will soon publish that small ontology, after some additions and a review, since we need to use it now.

kjano commented 5 years ago

Hi,

I am all for these ideas but I have reservations about the name. OGC (and Simon can correct me) has this great distinction between abstract specifications and implementation specifications. I do believe that we need a better and more modern `Space Ontology', and I also believe that we should have an OWL and SHACL implementation of this ontology.

Best, Jano

On 6/6/19 6:03 AM, Nicholas Car wrote:

I have just presented on the Australian spatial Linked Data infrastructure LocI at the European Semantic Web Conference yesterday and discussed with a number of attendees possible enhancements to the GeoSPARQL Ontology which could be picked up in OWL Space. There seems to be a common set of requirements from at least attendees here, some of which I’ve placed in my informally presented and open-ended GeoSPARQL Extensions Ontology: https://github.com/CSIRO-enviro-informatics/geosparql-ext-ont. I will soon publish that small ontology, after some additions and a review, since we need to use it now.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/w3c/sdw/issues/1095?email_source=notifications&email_token=AANMP5R66D6ONTPDWTX77WLPZEDLFA5CNFSM4GDZBJ32YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXCYN5I#issuecomment-499484405, or mute the thread https://github.com/notifications/unsubscribe-auth/AANMP5XXJNAFSE2GH37IKMTPZEDLFANCNFSM4GDZBJ3Q.

-- Krzysztof Janowicz

Geography Department, University of California, Santa Barbara 4830 Ellison Hall, Santa Barbara, CA 93106-4060

Email: jano@geog.ucsb.edu Webpage: http://geog.ucsb.edu/~jano/ Semantic Web Journal: http://www.semantic-web-journal.net

dr-shorthair commented 5 years ago

Indeed.

What notation should be used for an abstract ontology?

(OGC/ISO mostly used UML for the abstract spec, and various XML/JSON/DDL specs for implementations.)

Simon

From: kjano [mailto:notifications@github.com] Sent: Friday, 7 June, 2019 03:20 To: w3c/sdw sdw@noreply.github.com Cc: Subscribed subscribed@noreply.github.com Subject: Re: [w3c/sdw] New project proposal: OWL Space (#1095)

Hi,

I am all for these ideas but I have reservations about the name. OGC (and Simon can correct me) has this great distinction between abstract specifications and implementation specifications. I do believe that we need a better and more modern `Space Ontology', and I also believe that we should have an OWL and SHACL implementation of this ontology.

Best, Jano

On 6/6/19 6:03 AM, Nicholas Car wrote:

I have just presented on the Australian spatial Linked Data infrastructure LocI at the European Semantic Web Conference yesterday and discussed with a number of attendees possible enhancements to the GeoSPARQL Ontology which could be picked up in OWL Space. There seems to be a common set of requirements from at least attendees here, some of which I’ve placed in my informally presented and open-ended GeoSPARQL Extensions Ontology: https://github.com/CSIRO-enviro-informatics/geosparql-ext-ont. I will soon publish that small ontology, after some additions and a review, since we need to use it now.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/w3c/sdw/issues/1095?email_source=notifications&email_token=AANMP5R66D6ONTPDWTX77WLPZEDLFA5CNFSM4GDZBJ32YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXCYN5I#issuecomment-499484405, or mute the thread https://github.com/notifications/unsubscribe-auth/AANMP5XXJNAFSE2GH37IKMTPZEDLFANCNFSM4GDZBJ3Q.

-- Krzysztof Janowicz

Geography Department, University of California, Santa Barbara 4830 Ellison Hall, Santa Barbara, CA 93106-4060

Email: jano@geog.ucsb.edumailto:jano@geog.ucsb.edu Webpage: http://geog.ucsb.edu/~jano/ Semantic Web Journal: http://www.semantic-web-journal.net

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/w3c/sdw/issues/1095?email_source=notifications&email_token=AAEUQL5OFM6TQ3JE6GV23ULPZFBM3A5CNFSM4GDZBJ32YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXDRZ4I#issuecomment-499588337, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AAEUQL2CA7KHBG4HR5QQ6VTPZFBM3ANCNFSM4GDZBJ3Q.

lieberjosh commented 5 years ago

It’s not just OGC, but is referred to as the Cook & Daniels methodology by Werner Kuhn and others. RDF seemed to be this great opportunity to have less of a stark difference between concept / type and instance / tuple. Perhaps because RDF can also be used for implementation or just because of familiarity, OGC has been reluctant to move from UML towards RDF / OWL as a language of abstraction despite dissonance between UML and many implementation languages. With domain conceptual models, type catalogs, etc., the abstract <-> implementation dichotomy really isn’t what it used to be anyways, even in OGC.

That said, there is at least an update of the spatial schema (geometry, topology) abstract spec (AS Topic 1 / ISO 19107). Plenty of concepts to drag into OWL/RDF as long as it isn’t done robotically.

As far as an update to GeoSPARQL, I’ve tried to push this for a while, but my sense is that there has not been much agreement on what the requirements for such an update might be. I tried in SDWWG days to modularize it (sdwgeo) and bring a core closer to GeoRSS (ogeo), but there wasn’t much interest. Some concerted effort to figure out what is needed might get past the pitfalls of special needs that most other spatial ontologies seem to fall into.

The drafts are still here: https://geosemweb.org/ogeo https://geosemweb.org/ogeo and https://geosemweb.org/sdwgeo https://geosemweb.org/sdwgeo. We will have the opportunity to discuss GeoSPARQL update prospects in the Geosemantics DWG in Leuven and try to progress in understanding what the needs might drive this.

—Josh

On Jun 6, 2019, at 11:17 PM, Simon Cox notifications@github.com wrote:

Indeed.

What notation should be used for an abstract ontology?

(OGC/ISO mostly used UML for the abstract spec, and various XML/JSON/DDL specs for implementations.)

Simon

From: kjano [mailto:notifications@github.com] Sent: Friday, 7 June, 2019 03:20 To: w3c/sdw sdw@noreply.github.com Cc: Subscribed subscribed@noreply.github.com Subject: Re: [w3c/sdw] New project proposal: OWL Space (#1095)

Hi,

I am all for these ideas but I have reservations about the name. OGC (and Simon can correct me) has this great distinction between abstract specifications and implementation specifications. I do believe that we need a better and more modern `Space Ontology', and I also believe that we should have an OWL and SHACL implementation of this ontology.

Best, Jano

On 6/6/19 6:03 AM, Nicholas Car wrote:

I have just presented on the Australian spatial Linked Data infrastructure LocI at the European Semantic Web Conference yesterday and discussed with a number of attendees possible enhancements to the GeoSPARQL Ontology which could be picked up in OWL Space. There seems to be a common set of requirements from at least attendees here, some of which I’ve placed in my informally presented and open-ended GeoSPARQL Extensions Ontology: https://github.com/CSIRO-enviro-informatics/geosparql-ext-ont. I will soon publish that small ontology, after some additions and a review, since we need to use it now.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/w3c/sdw/issues/1095?email_source=notifications&email_token=AANMP5R66D6ONTPDWTX77WLPZEDLFA5CNFSM4GDZBJ32YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXCYN5I#issuecomment-499484405, or mute the thread https://github.com/notifications/unsubscribe-auth/AANMP5XXJNAFSE2GH37IKMTPZEDLFANCNFSM4GDZBJ3Q.

-- Krzysztof Janowicz

Geography Department, University of California, Santa Barbara 4830 Ellison Hall, Santa Barbara, CA 93106-4060

Email: jano@geog.ucsb.edumailto:jano@geog.ucsb.edu Webpage: http://geog.ucsb.edu/~jano/ Semantic Web Journal: http://www.semantic-web-journal.net

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/w3c/sdw/issues/1095?email_source=notifications&email_token=AAEUQL5OFM6TQ3JE6GV23ULPZFBM3A5CNFSM4GDZBJ32YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXDRZ4I#issuecomment-499588337, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AAEUQL2CA7KHBG4HR5QQ6VTPZFBM3ANCNFSM4GDZBJ3Q. — You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/w3c/sdw/issues/1095?email_source=notifications&email_token=ABGFNBUCK4UT456ODHWIU23PZHHOPA5CNFSM4GDZBJ32YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXEXSYY#issuecomment-499743075, or mute the thread https://github.com/notifications/unsubscribe-auth/ABGFNBWHQ4N3NNFAUCWA3ODPZHHOPANCNFSM4GDZBJ3Q.

kjano commented 5 years ago

On 6/6/19 8:17 PM, Simon Cox wrote:

Indeed.

What notation should be used for an abstract ontology?

(OGC/ISO mostly used UML for the abstract spec, and various XML/JSON/DDL specs for implementations.)

I like something like the 'Simple Event Model' (SEM) [1]. We could even name the ontology LPPS for 'Locations, Positions, Places, and Spaces' and cover all of them as they are constantly confused in the literature and even more so in other ontologies. My main point is not to confuse the ontology with one potential language to represent it. If we want to make impact, we may be better served with a schema.org-like, and OWL, and a SHACL version to suit different needs and communities. IMHO, what matters most is the common model and less whether one implementation can express a bit more or less of the underlying model in terms of classical  reasoning or constrain checking.

Best,

Jano

[1] https://homepages.cwi.nl/~hollink/pubs/vanHage2011SEM.pdf

Simon

From: kjano [mailto:notifications@github.com] Sent: Friday, 7 June, 2019 03:20 To: w3c/sdw sdw@noreply.github.com Cc: Subscribed subscribed@noreply.github.com Subject: Re: [w3c/sdw] New project proposal: OWL Space (#1095)

Hi,

I am all for these ideas but I have reservations about the name. OGC (and Simon can correct me) has this great distinction between abstract specifications and implementation specifications. I do believe that we need a better and more modern `Space Ontology', and I also believe that we should have an OWL and SHACL implementation of this ontology.

Best, Jano

On 6/6/19 6:03 AM, Nicholas Car wrote:

I have just presented on the Australian spatial Linked Data infrastructure LocI at the European Semantic Web Conference yesterday and discussed with a number of attendees possible enhancements to the GeoSPARQL Ontology which could be picked up in OWL Space. There seems to be a common set of requirements from at least attendees here, some of which I’ve placed in my informally presented and open-ended GeoSPARQL Extensions Ontology: https://github.com/CSIRO-enviro-informatics/geosparql-ext-ont. I will soon publish that small ontology, after some additions and a review, since we need to use it now.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub

https://github.com/w3c/sdw/issues/1095?email_source=notifications&email_token=AANMP5R66D6ONTPDWTX77WLPZEDLFA5CNFSM4GDZBJ32YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXCYN5I#issuecomment-499484405, or mute the thread

https://github.com/notifications/unsubscribe-auth/AANMP5XXJNAFSE2GH37IKMTPZEDLFANCNFSM4GDZBJ3Q.

-- Krzysztof Janowicz

Geography Department, University of California, Santa Barbara 4830 Ellison Hall, Santa Barbara, CA 93106-4060

Email: jano@geog.ucsb.edumailto:jano@geog.ucsb.edu Webpage: http://geog.ucsb.edu/~jano/ Semantic Web Journal: http://www.semantic-web-journal.net

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/w3c/sdw/issues/1095?email_source=notifications&email_token=AAEUQL5OFM6TQ3JE6GV23ULPZFBM3A5CNFSM4GDZBJ32YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXDRZ4I#issuecomment-499588337, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AAEUQL2CA7KHBG4HR5QQ6VTPZFBM3ANCNFSM4GDZBJ3Q.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/w3c/sdw/issues/1095?email_source=notifications&email_token=AANMP5R34CC4VGJHHXFXQSDPZHHOLA5CNFSM4GDZBJ32YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXEXSYY#issuecomment-499743075, or mute the thread https://github.com/notifications/unsubscribe-auth/AANMP5R4VDC5EZSQBJVQT3LPZHHOLANCNFSM4GDZBJ3Q.

-- Krzysztof Janowicz

Geography Department, University of California, Santa Barbara 4830 Ellison Hall, Santa Barbara, CA 93106-4060

Email: jano@geog.ucsb.edu Webpage: http://geog.ucsb.edu/~jano/ Semantic Web Journal: http://www.semantic-web-journal.net

FransKnibbe commented 5 years ago

Hi all,

I have just added Nick's and Josh's ideas for improvement of GeoSPARQL to the list of prior art in the top message of this thread.

As for the name: at this point I think a name like 'OWL Space' should be considered to be no more than a placeholder label. However, I do think it is important for the ontology to be about space and not only geography. Having 'space' or 'spatial' in the name should help to make that clear.

Separating abstract and implementation specifications seems like a very good idea to me. An abstract specification is the thing that is really needed. And if it's is any good (clear, simple, modular, ...) it should be possible for anyone to come up with interoperable implementation specifications.

As for notation: I agree that the main effort should be to arrive at a clear model/knowledge representation, irrespective of notation. It is likely that the skeleton of the model will consist of related classes and properties, which can be done in either OWL or UML. That way of modelling knowledge has been popular at least since Aristotle. It will probable last for some time yet, while OWL and UML may turn out to be short-lived fads. Developing OWL and UML simultaneously could be an option and might even be a way to improve quality? I also agree that any translation between notations should not be done robotically.

Aside from OWL or UML, my gut feeling is that mathematics should be a main language in this model. Fortunately, maths is universal and chances are high that it will also stand the test of time.

A further consideration: one thing I really like about having at least a model in RDFS/OWL is that all definitions will have deferenceable and linkable URI's. That should help users and implementers a lot.

Greetings, Frans

FransKnibbe commented 5 years ago

The proposal to start a general ontology for spatial data from GeoSPARQL has been added to the OGC issue tracker: http://ogc.standardstracker.org/show_request.cgi?id=592

kjano commented 5 years ago

Very good. KJ

On 9/18/19 5:44 AM, Frans Knibbe wrote:

The proposal to start a general ontology for spatial data from GeoSPARQL has been added to the OGC issue tracker: http://ogc.standardstracker.org/show_request.cgi?id=592

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/w3c/sdw/issues/1095?email_source=notifications&email_token=AANMP5TARRC22DVGZYT5CMTQKIPETA5CNFSM4GDZBJ32YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD675XYQ#issuecomment-532667362, or mute the thread https://github.com/notifications/unsubscribe-auth/AANMP5QSSTSAS6KPJ57OFATQKIPETANCNFSM4GDZBJ3Q.

-- Krzysztof Janowicz

Geography Department, University of California, Santa Barbara 4830 Ellison Hall, Santa Barbara, CA 93106-4060

Email: jano@geog.ucsb.edu Webpage: http://geog.ucsb.edu/~jano/ Semantic Web Journal: http://www.semantic-web-journal.net

lvdbrink commented 4 years ago

GeoSemantics DWG is working on a white paper exploring the idea that semantic and graph technologies will increase the value that can be extracted from (geo)spatial data.

This paper has two purposes. Firstly, to describe the benefits of representing geospatial data using semantics and graph technologies. Secondly, to outline some shortcomings of the existing GeoSPARQL implementation specification that, if addressed, would unlock its potential to a greater extent.

@jabhay is leading this work.

See https://github.com/opengeospatial/geosemantics-dwg/tree/master/white_paper

6a6d74 commented 4 years ago

This issue is now incorporated into the OGC GeoSPARQL Change Request. Consequently I'm closing this "proposal". Further discussion should take place in the relevant OGC forums. Thanks all!