Re: BP & SSN: feature of interest vs spatial thing

Given that the term feature is already understood by some communities in a different sense from that of OGC/ISO, it would be good not to mint yet another sense. There seem to be multiple ways that FOI could be differentiated from GF_Feature. It could be a subclass of GF_Feature (itself equivalent to SpatialThing). Or it could be any feature individual that is distinguished by having an object property relationship with an observation. Presumably not every feature becomes a feature of interest (although the number is increasing daily), making it a subset in practice. It is not clear to me, though, that features of interest have any other distinction from SpatialThing except that expression of interest via observation.

Whether FoI is equivalent to Feature / SpatialThing or a subclass, the important articulation to preserve, I think, is the transformation by which a sensed stimulus is interpreted in the context of an observation to estimate a feature property. If we are not clearly asserting and documenting that a sensor is telling us something specific about the world, we will have lost much of the power of the vocabulary to communicate what we know and how we know it. 

As for the case of “type of lightbulb”, in the RDF/OWL world, an average power consumption measured across a population sample of lightbulbs is a statistic derived from a number of individual observations of the observable properties of actual / spatial lightbulbs used to predict the observable property of any other specific members of the population. Any individual of a feature type is still a feature with a location in space at any specific time, whether or not it’s observable property has been individually measured. Interestingly, a feature “type" in the UML sense may not itself be a feature, but in OWL/RDF set theory it is a collection of features, which is itself a feature. So the set of lightbulbs of a particular type may be scattered across the globe, but the set is still “spatial”.

Josh


> On Feb 26, 2017, at 4:43 PM, Krzysztof Janowicz <janowicz@ucsb.edu> wrote:
> 
> Hi all,
> 
> The term FOI also aligns very well with POI (Point of Interest) and so forth. In addition, spatial feature does not have the same discriminative power as FOI. For instance, an actuator is a spatial feature but the thing it is acting on, e.g., a window, is a FOI (and a spatial feature). This does not mean that there is no such case where an actuator cannot be a FOI itself. In fact, when its acting is observed by a sensor (like a human eye), said actuator becomes a FOI wrt this observation and so on.
> 
> Cheers,
> Jano
> 
> 
> 
> On 02/26/2017 02:57 AM, Rob Atkinson wrote:
>> Foi  is not necessarily spatial... i might be observing average power consumption of a type of light bulb... make it Thing but not SpatialThing if you must.
>> 
>> 
>> On Sun, 26 Feb 2017, 5:35 PM <Simon.Cox@csiro.au> <mailto:Simon.Cox@csiro.au> wrote:
>> Please note that ‘Feature’ was never an O&M (ISO 19156) concept per se. It comes from the ISO 19101 Reference Model.
>> 
>>  
>> The relevant terms in the O&M spec were
>> 
>> -          A property (rolename) ‘featureOfInterest’ to link an Observation to the thing whose property-value is being estimated
>> 
>> -          A Class GFI_Feature, which is an instantiation of the meta-class GF_Feature, taken from ISO 19109.
>> 
>>  
>> The name of the property feature-of-interest was a topic of considerable discussion during the development of O&M. It was originally called ‘target’ but this was changed in discussions around 2005 at the specific request of some of the project sponsors, whose affiliation with the D&I industry and community made them a little sensitive around that term …
>> 
>>  
>> “feature-of-interest” was chosen to align with the 19101 terminology, but seems to have                   stood up well in consultations with many communities.
>> 
>>  
>> Simon
>> 
>>  
>> From: Kerry Taylor [mailto:kerry.taylor@anu.edu.au <mailto:kerry.taylor@anu.edu.au>] 
>> Sent: Sunday, 26 February, 2017 16:36
>> To:  <mailto:public-sdw-wg@w3.org>public-sdw-wg@w3.org <mailto:public-sdw-wg@w3.org>
>> Subject: BP & SSN: feature of interest vs spatial thing
>> 
>>  
>> SDWers,
>> 
>> I note that BP has moved away from the O&M concept of “feature” towards “spatial thing” instead – yet in SSN we are using  O&M’s  “Feature of Interest “ with the following 2 descriptions:
>> 
>>  
>> (1) sosa:FeatureOfInterest:  “The thing whose property is being estimated or calculated in the course of an Observation to arrive at a Result or whose property is being manipulated by an Actuator”
>> 
>> AND
>> 
>> (2) Ssn:FeatureOfInterest: “A feature is an abstraction of real world phenomena (thing, person,  event, etc)”.
>>  
>> Formally, ssn traditionally defined it simply as an Event or Object.
>> 
>>  
>> What do those in the BP space think about this? In ssn I don’t think we even care whether the thing being observed  has a geometry, but indeed “Feature” used in the context of ssn inherits all the same problems that the BP documents. Should we use “Spatial thing” as a way of lining up with BP?
>> 
>> A short extract from current BP draft follows:
>> 
>> To avoid confusion, we adopt the term “spatial thing <http://w3c.github.io/sdw/bp/#dfn-spatial-thing>” throughout the remainder of this best practice document. “Spatial thing <http://w3c.github.io/sdw/bp/#dfn-spatial-thing>” is defined in [W3C-BASIC-GEO <http://w3c.github.io/sdw/bp/#bib-W3C-BASIC-GEO>] as “Anything with spatial extent, i.e. size, shape, or position. e.g. people, places, bowling balls, as well as abstract areas like cubes”.
>> 
>> The concept of “spatial thing <http://w3c.github.io/sdw/bp/#dfn-spatial-thing>” is considered to include both "real-world phenomena" and their abstractions (e.g. “feature <http://w3c.github.io/sdw/bp/#dfn-feature>” as defined in [ISO-19101 <http://w3c.github.io/sdw/bp/#bib-ISO-19101>]). Furthermore, we treat it as inclusive of other commonly used definitions; e.g. Feature from [NeoGeo <http://w3c.github.io/sdw/bp/#bib-NeoGeo>], described as “A geographical feature, capable of holding spatial relations”.…….
>> 
>> Looking more closely, it is important to note that geometry is typically a property of aspatial thing <http://w3c.github.io/sdw/bp/#dfn-spatial-thing>.
>> 
>>  
>> -Kerry
>> 
>>  
>>  
> 
> 
> -- 
> Krzysztof Janowicz
> 
> Geography Department, University of California, Santa Barbara 
> 4830 Ellison Hall, Santa Barbara, CA 93106-4060 
> 
> Email: jano@geog.ucsb.edu <mailto:jano@geog.ucsb.edu>
> Webpage: http://geog.ucsb.edu/~jano/ <http://geog.ucsb.edu/~jano/>
> Semantic Web Journal: http://www.semantic-web-journal.net <http://www.semantic-web-journal.net/>