- From: Frans Knibbe | Geodan <frans.knibbe@geodan.nl>
- Date: Mon, 23 Feb 2015 11:46:19 +0100
- To: public-sdw-wg@w3.org
- Message-ID: <54EB04FB.4080509@geodan.nl>
On 2015-02-21 19:38, Krzysztof Janowicz wrote: > Hi, > >> This discussion seems to point at an important requirement for our >> work, which may or may not be derived from use cases, but still is >> important: We need clear and universal semantics. > > IMHO, this has been tried in the realm of philosophical ontology for > 2500+ years and we are not there yet. Language simply does not work > that way. In information ontology we want to restrict the meaning of > domain vocabulary towards its intended interpretation. Today's > semantic web technologies, for instance, are meant to uncover > similarities and differences by providing a formal axiomatization of > the used terminology to foster interoperability, not to arrive at some > sort of 'truth' (i.e., universal semantics). Actually, I was thinking about universal semantics in the sense of a widely accepted web ontology. Perhaps we will not succeed in completely harmonizing all the different ontologies for spatial data that exist now, but if we somehow could work towards web semantics for spatial data that describe some basic concepts (spatial resource, location, coverage, ...) that bear the mark of approval from both OGC and W3C, that would be a huge success. All new and existing vocabularies that have something spatial could then make use of those definitions, creating a base level of interoperability. One data publisher might like to publish spatial data as WKT geometry, the other might publish spatial data as post addresses, but still humans and software would be able to combine the data because they would be based on a common definition of 'a spatial resource'. These universal web semantics for spatial data that I am thinking of do not necessarily have to come in the shape of a new vocabulary. It could also be an improvement of an existing vocabulary like GeoSPARQL or the Location Core Vocabulary. I think having a simple glossary, with definitions in human language, a few examples, and links to existing definitions could be a good starting point for formalized universal web semantics. > >> What makes a thing spatial? Does it have to have geometry? Does it >> have to have three dimensions? Does it need to be a terrestial >> object? Does it need to be non-fictional? All these things are >> debatable, but rather than debate them it would be good to have >> agreed upon semantics. > > I tried to make this point during our call by introducing the > platial/placial notion but maybe I was not clear enough. I would > suggest not to confuse spatial with geometry and explicitly include > the notion of place into our scope. Otherwise we would not be able to > address many relevant spatial search problems like vague regions. This > nice paper "Montello, Daniel R., Alinda Friedman, and Daniel W. > Phillips. "Vague cognitive regions in geography and geographic > information science." /International Journal of Geographical > Information Science/ 28.9 (2014): 1802-1820." from my college Dan > illustrated the problems and differences nicely. I agree. It is that kind of potential confusion that led me to propose starting a glossary page. > [snip] > > Personally, I would strongly prefer if we would scope the working > group *methodologically* instead of by trying to define exact borders > for a complex and dynamic topic such as spatial data. Yes, trying to define exact borders would take a lot of effort with low chances of success. But we do need to make sure that we will talk about approximately the same things. Regards, Frans > > Best, > Jano > > > On 02/19/2015 05:14 AM, Frans (Geodan) wrote: >> >> I think that the question whether data are spatial data largely >> depends on the data publisher. The designation “Greater London” could >> be published as a text label, in which case it is not spatial data. >> It could also be published as a toponym, in which it is spatial data. >> >> This discussion seems to point at an important requirement for our >> work, which may or may not be derived from use cases, but still is >> important: We need clear and universal semantics. >> >> What makes a thing spatial? Does it have to have geometry? Does it >> have to have three dimensions? Does it need to be a terrestial >> object? Does it need to be non-fictional? All these things are >> debatable, but rather than debate them it would be good to have >> agreed upon semantics. >> >> At the moment, there are some definitions out there on the semantic >> web. For example, the Location Core Vocabulary >> <http://www.w3.org/ns/locn> defines the concept 'location'. >> Unfortunately the definition is self-referencing: “any location, >> irrespective of size or other restriction”. In other words, it is >> very open to interpretation. Is “Paris” a location (knowing that >> there are multiple locations with that name)? Is Atlantis (fictional) >> a location? Is Olympus Mons (on Mars) a location? >> >> GeoSPARQL <%28http://www.opengis.net/ont/geosparql> has definitions >> for the concept 'SpatialObject': “..everything that can have a >> spatial representation” (unfortunately the 'spatial representation' >> part is undefined) and 'Feature': “..equivalent to GFI_Feature >> defined in ISO 19156:2011”. Unfortunately GFI_Feature as defined in >> ISO 19156:2011 is not a web resource and ISO 19156:2011 is not an >> open standard (because you have to pay for it). But it's a start... >> >> Greetings, >> Frans >> >> >> On 2015-02-19 11:50, Ed Parsons wrote: >>> This is a great discussion and I think it is central to the >>> potentially difficult overlap between the two community perspectives. >>> >>> I'm sure Josh will chip in but I do think we need to recognise that >>> we need to include spatial information for which it is not possible >>> to define a geometry or have linked to as an attribute - This I >>> think is what Josh means by context, I am writing this email from a >>> location within "Central London" although there is not a canonical >>> geometry that represents the shape of central London. >>> >>> This is an example of what Mike Goodchild calls a Platial Problem ! >>> >>> This must be in scope, does the current wording around spatial >>> information accommodate it ? >>> >>> Ed >>> >>> >>> On Thu Feb 19 2015 at 10:26:55 Andrea Perego >>> <andrea.perego@jrc.ec.europa.eu >>> <mailto:andrea.perego@jrc.ec.europa.eu>> wrote: >>> >>> Andreas's mail gives me the opportunity to explain the objection I >>> raised during the call [1] about the proposal of adding "spatial >>> context" into scope question #1 [2]. >>> >>> My main concern is that the use of "spatial context" in the scoping >>> question may be confusing, and probably unnecessary. >>> >>> In my understanding, spatial context is specified through >>> spatial data >>> - i.e., it denotes one of their possible uses. So, "spatial data" >>> should be inclusive enough - it would cover spatial data as a whole, >>> irrespective of their use. >>> >>> Thanks! >>> >>> Andrea >>> >>> ---- >>> [1]http://www.w3.org/2015/02/18-sdw-minutes.html >>> [2]http://www.w3.org/2015/spatial/wiki/Scope_questions_and_Requirements >>> >>> On Wed, Feb 18, 2015 at 10:10 PM, Andreas Harth <harth@kit.edu >>> <mailto:harth@kit.edu>> wrote: >>> > Hi, >>> > >>> > the issue I had with the term "spatial context" is that I did >>> not know >>> > what the "context" part was supposed to mean. >>> > >>> > If I understood Josh correctly, he mentioned that a geometry, >>> > a place description or a spatial feature should be referenceable >>> > in data. >>> > >>> > If "spatial context" does mean that, I'm fine with the phrasing of >>> > the scoping question. >>> > >>> > Cheers, >>> > Andreas. >>> > >>> >>> >>> >>> -- >>> Andrea Perego, Ph.D. >>> Scientific / Technical Project Officer >>> European Commission DG JRC >>> Institute for Environment & Sustainability >>> Unit H06 - Digital Earth & Reference Data >>> Via E. Fermi, 2749 - TP 262 >>> 21027 Ispra VA, Italy >>> >>> https://ec.europa.eu/jrc/ >>> >>> ---- >>> The views expressed are purely those of the writer and may >>> not in any circumstances be regarded as stating an official >>> position of the European Commission. >>> >> >> >> >> ------------------------------------------------------------------------ >> <http://www.avast.com/> >> >> Dit e-mailbericht bevat geen virussen en malware omdat avast! >> Antivirus <http://www.avast.com/> actief is. >> >> > > > -- > 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 ------------------------------------------------------------------------ Frans Knibbe Geodan President Kennedylaan 1 1079 MB Amsterdam (NL) T +31 (0)20 - 5711 347 E frans.knibbe@geodan.nl www.geodan.nl <http://www.geodan.nl> | disclaimer <http://www.geodan.nl/disclaimer> ------------------------------------------------------------------------
Received on Monday, 23 February 2015 10:46:55 UTC