Re: Area of spatial objects

Hello Peter,

Thank you for those insights.

About your first point - measurements of space depend on the reference
system: I believe that a spatial ontology can be used to establish much
firmer relationships between geometry and CRS than is now often the case.
If a geometry has a direct relationship with data describing its CRS the
risk of errors and wrong interpretations should lessen. The CRS ontology
made available by IGNF (http://data.ign.fr/def/ignf) can be an inspiration.
Among other things it defines a class for cartesian reference systems (
http://data.ign.fr/def/ignf#CartesianCS) and a class for ellipsoidal
systems (http://data.ign.fr/def/ignf#EllipsoidalCS). When it is clear to
which class a particular CRS belongs, it will be clearer in which way areas
or distances should be calculated. Likewise, other useful data about the
CRS will become directly and explicitly available, like the units used in
the CRS, or its area of applicability. So yes, it should be possible to
state assumptions about the reference system when a general ontology for
spatial data is used.


About the second point, I would like to ask you the same question I asked
Chris. Do you think a means of stating the accuracy of spatial data will be
sufficient to tackle existing problems? My hope is that rigorously using
significant digits in publication of coordinates as well as calculations
performed on coordinates should go a long way. Besides that it will
probably be necessary to have the ability to make the accuracy of each of
the coordinates (e.g. X, Y and Z) known, because sometimes accuracy is not
isotropic.

Come to think of it, the resolution of spatial data probably is something
that needs to be taken into account for non-real world cases too. Software
systems can use different data types for storage and exchange of the
numbers that make up coordinates (e.g. float, decimal, double, …). This has
an impact on the spatial resolution of the data and should be taken into
account when calculations are performed.

It’s interesting you mention the definition of a footprint of a property.
It chimes in with previous mentions of similar much used classifications of
spatial things, like a centroid, a central point guaranteed to be inside a
shape (is there a single word for that?) and an exit/entry point. It would
be good to be able to share common definitions of those things, and a
general spatial ontology could help there too, I think.

Regards,

Frans


Op za 17 nov. 2018 om 03:23 schreef Peter Parslow <Peter.Parslow@os.uk>:

> Frans, Chris,
>
> in addition to the coastline paradox, which I guess is what Chris is
> referring to, there is the general point that any measurement depends on
> the geometry characteristics of the space in which the measurement is
> occurring, in two kinds of ways (that I can think of!)
>
>
> - the underlying reference system - the shortest distance between two
> points on a sphere does not look like a straight line when drawn on a flat
> projection of that sphere. And the world not being a sphere, there are a
> lot of horrible gotchas in the real world related to this, some of which
> have serious political or legal consequences. But the basic maths point
> applies in any non-real "world".
>
>
> - the length of a line, or the 'area' of an area / polygon - in the real
> world, it depends on how bumpy it is, as well as any general mathematical
> curvature. Therein lies much legal & financial debate, for example when it
> comes to farm payments.
>
>
> I do support the idea of a general ontology - but it would have to be
> clearly & explicitly stated what its assumptions are, e.g. for use only in
> planar spaces, or spherical ones, or whatever.
>
>
> Note, this is also independent of the semantics of the relationship
> between the feature/spatial thing & the geometry, as raised by Simon. A
> "footprint" semantic relationship between a "property" (house, land parcel,
> flat,....) and a polygon geometry will have a specific semantic - is it the
> intersection of the property with the actual surface of the earth, or with
> some "imagined" surface (flattened at some datum). The
> measured/estimated/calculated area of that polygon is then effected by the
> characteristics of the spatial system in which it is calculated.
>
>
> Of course, there are many simple / well defined / closed / well used
> systems in which this doesn't matter. Or to be more precise, the degree of
> error caused by ignoring these issues is not relevant to the question being
> addressed (use case).
>
>
> But that can't be assumed in any general ontology - the assumptions have
> to be explicit.
>
>
> Peter
>
> (OS Principal Data Architect & Open Standards Lead)
> ------------------------------
> *From:* Frans Knibbe <frans.knibbe@geodan.nl>
> *Sent:* 16 November 2018 15:06:00
> *To:* Chris Little
> *Cc:* public-sdwig@w3.org
> *Subject:* Re: Area of spatial objects
>
> Hello Chris,
>
> You are referring to the coastline paradox
> <https://en.wikipedia.org/wiki/Coastline_paradox>, right? That problem
> only occurs for real world phenomena, I think a general spatial ontology
> should not be limited to those.
> But I wonder: wouldn't a specification of spatial accuracy (either
> explicitly or by using significant digits) be sufficient to express the
> idea of having different sized rulers?
>
> Greetings,
> Frans
>
> Op do 15 nov. 2018 om 13:40 schreef Little, Chris <
> chris.little@metoffice.gov.uk>:
>
> Rob, Frans
>
>
>
> And I think we should extend the question to: “What is the length of the
> boundary of an area/geometry/polygon etc?”
>
>
>
> First answered convincingly by Lewis Fry Richardson: it depends on the
> length of your ruler. This is an intrinsically different answer to the (2D)
> areal problem.
>
>
>
> I agree they could be good contributions to Frans’ spatial ontology.
>
>
>
> Chris
>
>
>
> *From:* Rob Atkinson <rob@metalinkage.com.au>
> *Sent:* 14 November 2018 21:11
> *To:* Krzysztof Janowicz <janowicz@ucsb.edu>
> *Cc:* Car Nicholas <Nicholas.Car@csiro.au>; Frans Knibbe <
> frans.knibbe@geodan.nl>; public-sdwig@w3.org
> *Subject:* Re: Area of spatial objects
>
>
>
> IMHO this is a candidate for discussion in the context of a general
> spatial ontology.
>
>
>
> There is an unhelpful proliferation of solutions to the semantics of the
> relationships between objects/features and geometries in the Linked Data
> world.
>
>
>
> For example it common practice to have a point representing an address,
> but that may semantically actually represent many possible things:
>
> * centroid of dwelling
>
> * centroid of land parcel
>
> * centre of road frontage of land parcel/property
>
> * actual entry point to property
>
> * entry point to dwelling
>
> * something unspecified
>
>
>
> or for a city:  location of main post office, centroid of designated area,
> centroid of observed area, etc. etc.  Same goes for an area - what is the
> area of a city? many possible answers depending on semantic context.
>
>
>
> and each geometry has a range of metadata - precision, accuracy, area
>
>
>
> is area a property of a specific geometry, or a spatial attribute (again
> with semantic qualification) of a thing.
>
>
>
> On Tue, 13 Nov 2018 at 14:16, Krzysztof Janowicz <janowicz@ucsb.edu>
> wrote:
>
> On 11/7/18 4:54 AM, Car, Nicholas (L&W, Dutton Park) wrote:
>
> Hi Frans,
>
>
>
> What if I wanted to publish Features’ areas without also publishing
> geometries? What about features with a point geometry and an area, no
> polygon. Also, I have data where an area is given and also a polygon but I
> don’t know for sure if the area was calculated from the polygon. In fact I
> have data with an area and an Albers area and a geometry and don’t really
> know what happened to make which.
>
>
>
> So, I want to be able to represent area as a spatial property of a
> feature, independently of any geometry. For this I will try using
> Observations and Measurements-style mechanics for this.
>
>
>
> I think I might use subclassing of a sosa:ObservableProperty (
> https://www.w3.org/TR/vocab-ssn/#SOSAObservableProperty) to express
> spatial properties such as area and then relate then to a geo:Feature which
> would also, by SOSA logic, be a sosa:FeatureOfInterest. I may invent
> modelling to relate that spatial property to a geometry, but this wouldn’t
> be required, just nice to have if known.
>
>
>
> Yes, you would subclass sosa:ObservableProperty. Just as a tiny note, and
> sorry for being overly picky, there is a certain risk with properties such
> as area in conjunction with geometries, namely that the are not in sync. In
> fact, we published a paper that shows that this happens in many, if not in
> most, cases. Of course, you could argue that the defined area is the
> 'correct' observation and the area computed from a geometry just some
> approximation but it is worth keeping this in mind as we have no
> uncertainty model linked to sosa.
>
>
>
> Cheers,
>
> Jano
>
>
>
>
>
> Cheers,
>
>
>
> Nick
>
>
>
> *From:* Frans Knibbe <frans.knibbe@geodan.nl> <frans.knibbe@geodan.nl>
> *Sent:* Monday, 5 November 2018 6:22 PM
> *To:* public-sdwig@w3.org
> *Subject:* Re: Area of spatial objects
>
>
>
> Hi Nick,
>
>
>
> I wonder: Doesn't the fact that multiple areas for a single spatial thing
> are published mean that the areas are calculated from different geometric
> representations of that spatial thing? That would logically make the area a
> property of a geometry. Besides, the geometry instance could be used to
> link to the CRS (e.g. Albers), ideally by URI.
>
>
>
> CRS data linked to a geometry could also give access to the basic unit of
> the CRS (e.g. meter) and through that provide information on the units of
> properties derived from the geometry (like the area), but that would demand
> a fair amount of reasoning on the part of the data consumer.
>
>
>
> 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
>
>
>
> This email is only intended for the person to whom it is addressed and may
> contain confidential information. If you have received this email in error,
> please notify the sender and delete this email which must not be copied,
> distributed or disclosed to any other person.
>
> Unless stated otherwise, the contents of this email are personal to the
> writer and do not represent the official view of Ordnance Survey. Nor can
> any contract be formed on Ordnance Survey's behalf via email. We reserve
> the right to monitor emails and attachments without prior notice.
>
> Thank you for your cooperation.
>
> Ordnance Survey Limited (Company Registration number 09121572)
> Registered Office: Explorer House
> Adanac Drive
> Southampton SO16 0AS
> Tel: 03456 050505
> http://www.os.uk
>