W3C home > Mailing lists > Public > public-sdw-wg@w3.org > September 2016

Re: Clarification required: BP6 "use HTTP URIs for spatial things"

From: Krzysztof Janowicz <janowicz@ucsb.edu>
Date: Thu, 1 Sep 2016 20:18:09 -0700
To: Frans Knibbe <frans.knibbe@geodan.nl>, Jeremy Tandy <jeremy.tandy@gmail.com>
Cc: Simon Cox <Simon.Cox@csiro.au>, Joshua Lieberman <jlieberman@tumblingwalls.com>, SDW WG Public List <public-sdw-wg@w3.org>
Message-ID: <d69c357f-88fe-97d0-65c5-a89d101f6c1b@ucsb.edu>
> We have to be very careful with recommending owl:sameAs for 
> identifying equivalence of resources (individuals). This page 
> <http://ontologydesignpatterns.org/wiki/Community:Overloading_OWL_sameAs>  
> (Overloading OWL sameAs) summarizes discussion about owl:sameAs in the 
> semweb community years ago and it specifically says that linking a 
> thing with data about that thing with owl:sameAs is abuse of owl:sameAs.
>
> The paper When owl:sameAs isn't the Same: An Analysis of Identity in 
> Linked Data <https://www.w3.org/2009/12/rdf-ws/papers/ws21> goes 
> further and says that when two resources are linked by owl:sameAs they 
> are expected to have the same properties, and "[..] any statement that 
> is given to a single URI is true for every other URI that has an 
> owl:sameAs link".
>
> The reason that a satisfactory solution to asserting resource 
> equivalence did not seem available on the web was the reason the use 
> case Modelling In The Construction Sector 
> <http://w3c.github.io/sdw/UseCases/SDWUseCasesAndRequirements.html#ModellingInTheConstructionSector> was 
> contributed, with its deriverd Subject equality requirement 
> <http://w3c.github.io/sdw/UseCases/SDWUseCasesAndRequirements.html#SubjectEquality>. 
>
>

Yes, see my previous email about this issue. owl:SameAs relates two URIs 
to each other and should be used if both URIs signify the same entity 
(have the same 'identity'). 'A:EddystoneLighthouse a 
VerticalObstruction.' and 'B:EddystoneLighthouse a NavigationAid.'  do 
not cause any issues. These are just two statements and RDF-based Linked 
Data is simply about sets of statements. As long as 
A:EddystoneLighthouse and B:EddystoneLighthouse signify the same  entity 
on the surface of the Earth, we can (and should) use owl:SameAs. What we 
should not do is to use owl:SameAs between A:EddystoneLighthouse and 
C:EddystoneLighthouse if C:EddystoneLighthouse is the URI of a picture 
of the lighthouse, e.g., 
https://upload.wikimedia.org/wikipedia/commons/f/fb/Phare-d-Eddystone-Rocks.jpg 
.

Best,
Krzysztof



On 09/01/2016 08:25 AM, Frans Knibbe wrote:
> [snip]
>
>
>     It is tricky - I've been confusing myself for the last couple of
>     days at least! So if we _do_ conflate the real world thing (e.g.
>     Eddystone Lighthouse) and the discerned feature (e.g. Eddystone
>     Lighthouse seen as a vertical obstruction) then it would be
>     acceptable to use <owl:sameAs> with no need for the
>     "sameRealWorldEntityAs" property; e.g. ...
>
>     <http://example.com/sar/features/vo/EDY
>     <http://example.com/sar/features/vo/EDY>> owl:sameAs
>     <http://example.org/maritime/navaid/2650253
>     <http://example.org/maritime/navaid/2650253>> .
>
>     This would align with the common approach used in Linked Data
>     where (authoritative) identifiers are reused across different
>     domains, datasets and applications with the view to providing
>     common "nodes" in the "knowledge graph". The BP doc actively
>     encourages such reuse of identifiers (assuming that the data
>     publisher / curator can be 100% sure that the identifier
>     identifies the thing they're making statements about!). For
>     example, we might want to encourage folks to reuse the identifier
>     for Eddystone Lighthouse minted by Google for the Knowledge Graph:
>     <https://g.co/kg/m/013qr8> (which I think is derived from an older
>     Freebase identifier)
>
>     If I've interpreted correctly (as above), then I will try to
>     include a Note in the BP document alluding to feature discernment
>     and the related cognitive process.
>
>     Jeremy
>
>
> We have to be very careful with recommending owl:sameAs for 
> identifying equivalence of resources (individuals). This page 
> <http://ontologydesignpatterns.org/wiki/Community:Overloading_OWL_sameAs>  
> (Overloading OWL sameAs) summarizes discussion about owl:sameAs in the 
> semweb community years ago and it specifically says that linking a 
> thing with data about that thing with owl:sameAs is abuse of owl:sameAs.
>
> The paper When owl:sameAs isn't the Same: An Analysis of Identity in 
> Linked Data <https://www.w3.org/2009/12/rdf-ws/papers/ws21> goes 
> further and says that when two resources are linked by owl:sameAs they 
> are expected to have the same properties, and "[..] any statement that 
> is given to a single URI is true for every other URI that has an 
> owl:sameAs link".
>
> The reason that a satisfactory solution to asserting resource 
> equivalence did not seem available on the web was the reason the use 
> case Modelling In The Construction Sector 
> <http://w3c.github.io/sdw/UseCases/SDWUseCasesAndRequirements.html#ModellingInTheConstructionSector> was 
> contributed, with its deriverd Subject equality requirement 
> <http://w3c.github.io/sdw/UseCases/SDWUseCasesAndRequirements.html#SubjectEquality>. 
>
>
> Regards,
> Frans


-- 
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
Received on Friday, 2 September 2016 03:18:41 UTC

This archive was generated by hypermail 2.4.0 : Thursday, 24 March 2022 20:31:25 UTC