Re: Decision needed for issue-38: about different models of spatial things.

"stuff we don't know but someone needs to address"

+1 to this being a useful part of the BP doc.

We can't give a solution because there is no practice for us to refer to.
Also this problem strays into the "dragons" of object reconciliation- which
as we commented (much) earlier is a hard problem.

Jeremy
On Wed, 22 Jun 2016 at 13:36, Ed Parsons <eparsons@google.com> wrote:

> I can certainly see there is a real use case, and resulting requirement(s)
> from this, and while it sits mostly obviously within the BP deliverable my
> guess is there is no obvious solution to the issue. So we need to ask
> ourselves is it worth going through the process to document all of this
> without a potential solution in mind?
>
> I personally think we should as it gives insight to others in terms of
> future work... in fact I think a really useful part of the BP deliverable
> will be the "stuff we don't know but someone needs to address"
>
> Ed
>
>
> On Wed, 22 Jun 2016 at 13:26 Frans Knibbe <frans.knibbe@geodan.nl> wrote:
>
>> Hello,
>>
>> One UCR issue that really deserves some urgent progress is issue-38
>> <https://www.w3.org/2015/spatial/track/issues/38>. It is about a
>> possible new use case and possible new requirements as a result from a
>> submission from the public. See the thread Additon to use case & new req
>> <https://lists.w3.org/Archives/Public/public-sdw-wg/2016Jan/0093.html>.
>>
>> The heart of the matter seems to be that people are missing a way to
>> express that two different sets of data about a real world thing are about
>> the same real world thing, when the two sets of data use different
>> information models, one of which could be geospatial. An example is a
>> building. There could be a batch of data describing the building as a
>> spatial feature, using spatial semantic standards, and there could be
>> another batch of data describing the building as a collection of building
>> materials, using some other semantic standard. The submitters of the
>> problem feel that there is no way to express that both batches of data are
>> about the same subject and that the two batches of data could complement
>> each other. Properties like owl:sameAs
>> <http://www.w3.org/TR/2004/REC-owl-semantics-20040210/#owl_sameAs>,
>> rdfs:seeAlso <https://www.w3.org/TR/rdf-schema/#ch_seealso>, umbel:isLike
>> <http://wiki.opensemanticframework.org/index.php/UMBEL_Vocabulary#isLike_Property>
>> , bbccore:sameAs
>> <http://www.bbc.co.uk/ontologies/coreconcepts#terms_sameAs> and
>> http://schema.org/about are regarded as insufficient  or inappropriate.
>>
>> The submitters admit that the problem is not in the spatial domain, but
>> feel that the broader W3C communities have yet failed to address the
>> problem and that addressing the problem bottom-up, with a clear use case
>> and with a certain (spatial) scope could lead to a solution that could
>> later be applied more generally.
>>
>> I am not sure what we should do. On the one hand I think we should guard
>> our scope, and not take on things that are really of a broader nature than
>> spatial data. On the other hand, this is a real world problem that exists
>> with spatial data and that hinders using combining traditional geospatial
>> data with other sorts of data. And identifying a requirement is not the
>> same as promising to meet that requirement.
>>
>> If we were to accept this use case, I guess the resulting requirement
>> should be something like "It should be possible to express that spatially
>> modelled data are about the same subject as data using other information
>> models". And I guess that would be a BP requirement.
>>
>> What do you think? How can we resolve this issue?
>>
>> Regards,
>> Frans
>>
>>
>> --
>
> *Ed Parsons *FRGS
> Geospatial Technologist, Google
>
> Google Voice +44 (0)20 7881 4501
> www.edparsons.com @edparsons
>