Skip to content

Commit

Permalink
more link fixes
Browse files Browse the repository at this point in the history
  • Loading branch information
situx authored Aug 23, 2023
1 parent 4c71145 commit 4db438e
Showing 1 changed file with 9 additions and 9 deletions.
18 changes: 9 additions & 9 deletions bp/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -1026,12 +1026,12 @@ <h2>Spatial Things, Features and Geometry</h2>
</figure>


<p>First, the domain of <a href="https://www.w3.org/2003/01/geo/wgs84_pos#lat"><code>w3cgeo:lat</code></a>, <a href="https://www.w3.org/2003/01/geo/wgs84_pos#lon"><code>w3cgeo:lon</code></a> and <a href="http://www.w3.org/2003/01/geo/wgs84_pos#alt"><code>w3cgeo:alt</code></a> properties is <a href="http://www.w3.org/2003/01/geo/wgs84_pos#SpatialThing"><code>w3cgeo:SpatialThing</code></a>. While one could interpret these properties as mapping to a <a>geometry</a>, as <a href="http://www.georss.org/simple">GeoRSS Simple</a> does, there isn't conclusive evidence that this is what was intended. Second, <a href="http://www.w3.org/2003/01/geo/wgs84_pos#Point"><code>w3cgeo:Point</code></a> is defined as a sub-class of <a href="http://www.w3.org/2003/01/geo/wgs84_pos#SpatialThing"><code>w3cgeo:SpatialThing</code></a>. As a result, we have inconsistency in how <a href="http://www.w3.org/2003/01/geo/wgs84_pos#SpatialThing"><code>w3cgeo:SpatialThing</code></a> may be interpreted. For example:</p>
<p>First, the domain of <a href="https://www.w3.org/2003/01/geo/wgs84_pos#lat"><code>w3cgeo:lat</code></a>, <a href="https://www.w3.org/2003/01/geo/wgs84_pos#lon"><code>w3cgeo:lon</code></a> and <a href="https://www.w3.org/2003/01/geo/wgs84_pos#alt"><code>w3cgeo:alt</code></a> properties is <a href="https://www.w3.org/2003/01/geo/wgs84_pos#SpatialThing"><code>w3cgeo:SpatialThing</code></a>. While one could interpret these properties as mapping to a <a>geometry</a>, as <a href="http://www.georss.org/simple">GeoRSS Simple</a> does, there isn't conclusive evidence that this is what was intended. Second, <a href="http://www.w3.org/2003/01/geo/wgs84_pos#Point"><code>w3cgeo:Point</code></a> is defined as a sub-class of <a href="http://www.w3.org/2003/01/geo/wgs84_pos#SpatialThing"><code>w3cgeo:SpatialThing</code></a>. As a result, we have inconsistency in how <a href="http://www.w3.org/2003/01/geo/wgs84_pos#SpatialThing"><code>w3cgeo:SpatialThing</code></a> may be interpreted. For example:</p>
<ul>
<li>Because a <a href="https://www.w3.org/2003/01/geo/wgs84_pos#SpatialThing"><code>w3cgeo:SpatialThing</code></a> has lat/lon, some people might equate <a href="https://www.w3.org/2003/01/geo/wgs84_pos#SpatialThing"><code>w3cgeo:SpatialThing</code></a> with <a href="http://www.opengis.net/ont/geosparql#Geometry"><code>geosparql:Geometry</code></a>;</li>
<li>Because <a href="https://xmlns.com/foaf/0.1/Person"><code>foaf:Person</code></a> is defined as a sub-class of <a href="https://www.w3.org/2003/01/geo/wgs84_pos#SpatialThing"><code>w3cgeo:SpatialThing</code></a>, some other people find it natural to equate <a href="http://www.w3.org/2003/01/geo/wgs84_pos#SpatialThing"><code>w3cgeo:SpatialThing</code></a> with <a href="http://www.opengis.net/ont/geosparql#Feature"><code>geosparql:Feature</code></a></li>
<li>Because <a href="http://xmlns.com/foaf/0.1/Person"><code>foaf:Person</code></a> is defined as a sub-class of <a href="https://www.w3.org/2003/01/geo/wgs84_pos#SpatialThing"><code>w3cgeo:SpatialThing</code></a>, some other people find it natural to equate <a href="https://www.w3.org/2003/01/geo/wgs84_pos#SpatialThing"><code>w3cgeo:SpatialThing</code></a> with <a href="http://www.opengis.net/ont/geosparql#Feature"><code>geosparql:Feature</code></a></li>
</ul>
<p>So in summary, it's safer to say that our <a>Spatial Thing</a> equates to <a href="http://www.opengis.net/ont/geosparql#Feature"><code>geosparql:Feature</code></a>, and that it is <strong>not</strong> the same as <a href="http://www.w3.org/2003/01/geo/wgs84_pos#SpatialThing"><code>w3cgeo:SpatialThing</code></a>.</p>
<p>So in summary, it's safer to say that our <a>Spatial Thing</a> equates to <a href="http://www.opengis.net/ont/geosparql#Feature"><code>geosparql:Feature</code></a>, and that it is <strong>not</strong> the same as <a href="https://www.w3.org/2003/01/geo/wgs84_pos#SpatialThing"><code>w3cgeo:SpatialThing</code></a>.</p>
</aside>

</section>
Expand Down Expand Up @@ -2972,7 +2972,7 @@ <h4 class="subhead">Possible Approach to Implementation</h4>
</pre>
</aside>

<p>As can be seen in the example above, the <a>geometry</a> <a href="http://data.ordnancesurvey.co.uk/id/geometry/30505-10"><code>30505-10</code></a> is an <em>attribute</em> of the <a href="http://data.ordnancesurvey.co.uk/id/7000000000030505">City of Edinburgh</a>. If your intent is to make a statement about, or refer to, the real-world entity then make sure you link to the <a>Spatial Thing</a> rather than the <a>geometry</a>. Furthermore, note that the geometry record may be updated and re-published with a new identifier, for example, if the city boundary was resurveyed and would then result in a broken <a>link</a>.</p>
<p>As can be seen in the example above, the <a>geometry</a> <a href="https://data.ordnancesurvey.co.uk/id/geometry/30505-10"><code>30505-10</code></a> is an <em>attribute</em> of the <a href="https://data.ordnancesurvey.co.uk/id/7000000000030505">City of Edinburgh</a>. If your intent is to make a statement about, or refer to, the real-world entity then make sure you link to the <a>Spatial Thing</a> rather than the <a>geometry</a>. Furthermore, note that the geometry record may be updated and re-published with a new identifier, for example, if the city boundary was resurveyed and would then result in a broken <a>link</a>.</p>
<p>Data publishers should also be aware of a common pattern used in the publication of <a>Linked Data</a>, where the <a>Spatial Thing</a> and the information resource that describes it are identified separately &mdash; often, but not always, using <code>/id</code> as part of the URI for <a>Spatial Thing</a>, and <code>/doc</code> for the corresponding page/document/record. When the URI for the Spatial Thing is dereferenced, a <code>HTTP 303 (see other)</code> response is used to redirect the browser to the page/document/record URL. For example:</p>
<ul>
<li><a href="http://statistics.gov.scot/id/statistical-geography/S12000036"><code>http://statistics.gov.scot<span style="text-decoration: underline;">/id</span>/statistical-geography/S12000036</code></a> redirects to <a href="http://statistics.gov.scot/doc/statistical-geography/S12000036"><code>http://statistics.gov.scot<span style="text-decoration: underline;">/doc</span>/statistical-geography/S12000036</code></a></li>
Expand Down Expand Up @@ -3113,7 +3113,7 @@ <h4 class="subhead">Possible Approach to Implementation</h4>

<p>By way of example, let's explore some data for Edinburgh.</p>

<p>The <em>City of Edinburgh Council Area</em> (e.g. the geographical area that Edinburgh City Council is responsible for) is identified by the <a href="https://www.ons.gov.uk">Office for National Statistics</a> (the recognized national statistical institute of the UK) using their GSS code (a 9 character alpha numeric identifier) <code>S12000036</code> and the URI <a href="https://statistics.data.gov.uk/id/statistical-geography/S12000036"><code>http://statistics.data.gov.uk/id/statistical-geography/S12000036</code></a>. At the same time, the devolved government in Scotland, operating under its own jurisdiction, retains the GSS code but uses the URI <a href="http://statistics.gov.scot/id/statistical-geography/S12000036"><code>http://statistics.gov.scot/id/statistical-geography/S12000036</code></a>. Furthermore, the <a href="https://www.ordnancesurvey.co.uk">Ordnance Survey</a> maintain yet another URI for the City of Edinburgh Council Area as part of its 'Boundary Line' service that contains administrative and statistical geography areas in the UK: <a href="http://data.ordnancesurvey.co.uk/id/7000000000030505"><code>http://data.ordnancesurvey.co.uk/id/7000000000030505</code></a>. Similarly, Geonames identifies Edinburgh, a <em>second-order administrative division</em>, as <a href="http://sws.geonames.org/2650225/"><code>http://sws.geonames.org/2650225/</code></a>. All of these URIs refer to the same <a>Spatial Thing</a> and are equated using <code>owl:sameAs</code>.</p>
<p>The <em>City of Edinburgh Council Area</em> (e.g. the geographical area that Edinburgh City Council is responsible for) is identified by the <a href="https://www.ons.gov.uk">Office for National Statistics</a> (the recognized national statistical institute of the UK) using their GSS code (a 9 character alpha numeric identifier) <code>S12000036</code> and the URI <a href="https://statistics.data.gov.uk/id/statistical-geography/S12000036"><code>http://statistics.data.gov.uk/id/statistical-geography/S12000036</code></a>. At the same time, the devolved government in Scotland, operating under its own jurisdiction, retains the GSS code but uses the URI <a href="http://statistics.gov.scot/id/statistical-geography/S12000036"><code>http://statistics.gov.scot/id/statistical-geography/S12000036</code></a>. Furthermore, the <a href="https://www.ordnancesurvey.co.uk">Ordnance Survey</a> maintain yet another URI for the City of Edinburgh Council Area as part of its 'Boundary Line' service that contains administrative and statistical geography areas in the UK: <a href="https://data.ordnancesurvey.co.uk/id/7000000000030505"><code>http://data.ordnancesurvey.co.uk/id/7000000000030505</code></a>. Similarly, Geonames identifies Edinburgh, a <em>second-order administrative division</em>, as <a href="http://sws.geonames.org/2650225/"><code>http://sws.geonames.org/2650225/</code></a>. All of these URIs refer to the same <a>Spatial Thing</a> and are equated using <code>owl:sameAs</code>.</p>

<aside class="example" id="ex-linking-equality-edinburgh" title="Asserting equality between URIs for the City of Edinburgh Council Area (TTL format)">
<pre>
Expand All @@ -3131,7 +3131,7 @@ <h4 class="subhead">Possible Approach to Implementation</h4>

<p>Also note that in this [[TURTLE]] snippet one could easily include additional properties to help users determine whether the <a>link</a> is worth traversing, such as providing human-readable labels and specifying the <em>type</em> designated by each data publisher.</p>

<p>In contrast, the resource identified by <a href="https://data.os.uk/id/4000000074558316"><code>http://data.os.uk/id/4000000074558316</code></a> defines the <em>named place</em> Edinburgh &mdash; a colloquial definition for the city itself. This is not the same as the <em>City of Edinburgh Area</em> and therefore use of the <code>owl:sameAs</code> relationship is inappropriate.</p>
<p>In contrast, the resource identified by <a href="https://www.ordnancesurvey.co.uk/id/4000000074558316"><code>http://data.os.uk/id/4000000074558316</code></a> defines the <em>named place</em> Edinburgh &mdash; a colloquial definition for the city itself. This is not the same as the <em>City of Edinburgh Area</em> and therefore use of the <code>owl:sameAs</code> relationship is inappropriate.</p>

<aside class="note">
<p>The mechanics of determining whether the information provided when resolving two or more URIs does indeed describe the same <a>Spatial Thing</a> is a complex topic all in its own right and way beyond the scope of this best practice document. Tools such as <a href="https://openrefine.org">Open Refine</a> and the <a href="http://silkframework.org">Silk Linked Data Integration Framework</a> are designed to work with, transform and integrate heterogeneous data sources. Their documentation may provide further insight regarding these challenges.</p>
Expand All @@ -3151,7 +3151,7 @@ <h4 class="subhead">Possible Approach to Implementation</h4>

<p>All of the properties list above, are concerned with equality or similarity about resources themselves. However, we often want to talk about the similarity of Spatial Things in terms of location or <em>place</em>. Spatial relations (see <a href="#link-type-spatial-rels" class="sectionRef">above</a>) can be used to describe how locations are related &mdash; either using rigorous topological relationships derived from geometry, such as <a href="http://www.opengis.net/ont/geosparql#sfEquals"><code>geosparql:sfEquals</code></a>, or ones without formal mathematical underpinning, such as <a href="http://www.geonames.org/ontology#nearby"><code>geonames:nearby</code></a>. But <em>place</em> is a social concept that reflect how we humans perceive the space around us, often with a vague or imprecise notion of location; you can’t always define a boundary for a place like <a href="http://sws.geonames.org/2212709/">The Sahara</a> because not everyone agrees where its edge lies!</p>

<p>Talking of places, the <em>City of Edinburgh [Administrative] Area</em> and <em>Edinburgh</em> the <em>named place</em> are strongly related; you might say that they are the <em>same place</em> if that makes sense for your application. This also provides an example where it is worthwhile to provide multiple relationships between Spatial Things: Ordnance Survey uses the <a href="https://data.ordnancesurvey.co.uk/ontology/spatialrelations/within"><code>within</code></a> <a>link</a> relation type to relate the <em>named place</em> <a href="http://data.ordnancesurvey.co.uk/id/4000000074558316">Edinburgh</a> and the <a href="http://data.ordnancesurvey.co.uk/id/7000000000030505">City of Edinburgh</a> administrative area. <a href="https://data.ordnancesurvey.co.uk/ontology/spatialrelations/within"><code>within</code></a> complements a qualitative <em>same-place-as</em> relation between two places.</p>
<p>Talking of places, the <em>City of Edinburgh [Administrative] Area</em> and <em>Edinburgh</em> the <em>named place</em> are strongly related; you might say that they are the <em>same place</em> if that makes sense for your application. This also provides an example where it is worthwhile to provide multiple relationships between Spatial Things: Ordnance Survey uses the <a href="https://data.ordnancesurvey.co.uk/ontology/spatialrelations/within"><code>within</code></a> <a>link</a> relation type to relate the <em>named place</em> <a href="https://data.ordnancesurvey.co.uk/id/4000000074558316">Edinburgh</a> and the <a href="https://data.ordnancesurvey.co.uk/id/7000000000030505">City of Edinburgh</a> administrative area. <a href="https://data.ordnancesurvey.co.uk/ontology/spatialrelations/within"><code>within</code></a> complements a qualitative <em>same-place-as</em> relation between two places.</p>

<p>However, while we see people wanting to assert such qualitative <em>same-place-as</em> relationships based on human perception of place, there is no evidence of a best practice in how to achieve this; see <a href="#c-sameplaceas" class="sectionRef"></a> for more details about possible approaches that could be adopted.</p>

Expand Down Expand Up @@ -3188,7 +3188,7 @@ <h4 class="subhead">Possible Approach to Implementation</h4>
</pre>
</aside>

<p>The example snippet above, provided in [[TURTLE]] format, shows the relationships between the <a href="http://data.ordnancesurvey.co.uk/id/7000000000030505">City of Edinburgh</a> <a href="https://data.ordnancesurvey.co.uk/ontology/admingeo/District">district</a> and the <a href="https://data.ordnancesurvey.co.uk/ontology/admingeo/Ward">electoral wards</a> it contains. Also note that complementary use of <a href="http://www.opengis.net/ont/geosparql#sfTouches"><code>geosparql:sfTouches</code></a> to relate the <a href="http://data.ordnancesurvey.co.uk/id/7000000000030505">City of Edinburgh</a> to its adjacent districts; Midlothian, West Lothian etc.</p>
<p>The example snippet above, provided in [[TURTLE]] format, shows the relationships between the <a href="https://data.ordnancesurvey.co.uk/id/7000000000030505">City of Edinburgh</a> <a href="https://data.ordnancesurvey.co.uk/ontology/admingeo/District">district</a> and the <a href="https://data.ordnancesurvey.co.uk/ontology/admingeo/Ward">electoral wards</a> it contains. Also note that complementary use of <a href="http://www.opengis.net/ont/geosparql#sfTouches"><code>geosparql:sfTouches</code></a> to relate the <a href="https://data.ordnancesurvey.co.uk/id/7000000000030505">City of Edinburgh</a> to its adjacent districts; Midlothian, West Lothian etc.</p>

<p>A second domain where relationships between <a>Spatial Things</a> and non-spatial resources occur is earth observing. The example below, provided in [[GML]], relates a monitoring point at Deddington on the Nile River, Tasmania, to the sensor that is deployed there (using the <code>sams:hostedProcedure</code> property) and relates that monitoring point to the waterbody whose properties are being measured (using the <code>sam:sampledFeature</code> property). Here, the <a>links</a> are defined using [[XLINK11]].</p>

Expand Down Expand Up @@ -4636,7 +4636,7 @@ <h3>Helping software understand units of measure</h3>
</section>
<section id="c-sameplaceas">
<h3>Defining that two places are the same</h3>
<p>Unlike administrative areas and other topographic features that have clearly defined boundaries, places often have ill-defined, fuzzy boundaries that are based on human perception of ‘place’; you can’t always define a boundary for a place. For example, <a href="http://data.os.uk/id/4000000074558316">Edinburgh</a> the <em>named place</em>, published by <a href="https://www.ordnancesurvey.co.uk/">Ordnance Survey</a>, is described using only a notional point <a>geometry</a>; information is not provided about the geometric <a>extent</a>. Other examples of places with ill-defined, fuzzy <a>geometries</a> include <a href="http://sws.geonames.org/2212709/"><em>The Sahara</em></a>, the <em>American West</em> and <em>Renaissance Italy</em>. The relationships between places, with their ill-defined (or even absent) geometrical extents, defy description using the topological relationships which are computed mathematically from geometry.</p>
<p>Unlike administrative areas and other topographic features that have clearly defined boundaries, places often have ill-defined, fuzzy boundaries that are based on human perception of ‘place’; you can’t always define a boundary for a place. For example, <a href="https://www.ordnancesurvey.co.uk/id/4000000074558316">Edinburgh</a> the <em>named place</em>, published by <a href="https://www.ordnancesurvey.co.uk/">Ordnance Survey</a>, is described using only a notional point <a>geometry</a>; information is not provided about the geometric <a>extent</a>. Other examples of places with ill-defined, fuzzy <a>geometries</a> include <a href="http://sws.geonames.org/2212709/"><em>The Sahara</em></a>, the <em>American West</em> and <em>Renaissance Italy</em>. The relationships between places, with their ill-defined (or even absent) geometrical extents, defy description using the topological relationships which are computed mathematically from geometry.</p>

<p>Given the lack of existing best practice, we propose the use of a <em>qualitative</em> assertion based on human perceptions to relate places that are deemed to be the same: <strong>samePlaceAs</strong>.</p>

Expand Down

0 comments on commit 4db438e

Please sign in to comment.