Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Extended examples #161

Merged
merged 8 commits into from
Jun 30, 2021
Merged

Extended examples #161

merged 8 commits into from
Jun 30, 2021

Conversation

nicholascar
Copy link
Collaborator

Examples that are too long for the Spec Annex. Please note that the examples are considered another profile resource and thus are listed in the profile declaration.

@nicholascar nicholascar marked this pull request as ready for review June 16, 2021 03:08
@nicholascar nicholascar requested review from situx, FransKnibbe and dr-shorthair and removed request for situx June 16, 2021 03:08
@nicholascar nicholascar mentioned this pull request Jun 16, 2021
Copy link
Collaborator

@dr-shorthair dr-shorthair left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The examples are not consistent. .auspix and .gml are pure geometry.
The others all have the geometry nested under some other structure. Could they be made more uniform?

@nicholascar
Copy link
Collaborator Author

nicholascar commented Jun 16, 2021

The examples are not consistent.

I agree. We have confined the equivalent RDF literals to just the geometries but I thought I'd supply full Feature instances here. Problem is, there is no standard way of describing an AusPIX/DGGS Feature, other than... GeoSPARQL 1.1!

I have fixed the GML one, but AusPIX might just have to stay like that.

@nicholascar
Copy link
Collaborator Author

@dr-shorthair @FransKnibbe @situx can you please all review so we can merge this in. The intent is to extend the examples perhapps continuously so please allow this PR in then add to the examples. This PR will tidy up some of the files in the repo too (e.g. removing old rdf-egs.html docs)

Copy link
Collaborator

@situx situx left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks ok to me

@FransKnibbe
Copy link
Collaborator

FransKnibbe commented Jun 23, 2021

Hi @nicholascar. it looks good at a glance (and knowing it is a work in progress).
Two questions though:

  1. Why use the ^^xsd:anyURI notation instead of (and next to) turtle's <IRI> notation? I would opt for using <IRI> always.
  2. locn is one of the default namespaces. Why not use locn:geographicName instead of dcterms:title for geographical names?

@nicholascar
Copy link
Collaborator Author

Why use the ^^xsd:anyURI notation instead of (and next to) turtle's notation? I would opt for using always.
locn is one of the default namespaces.

<IRI> and xsd:anyURI are for completely different things:

  • <IRI> indicates the node is an RDF resource - the IRI acts as the identifier and, in the best case, the IRI can be dereferenced to get more RDF thus it may extend the Linked Data web
  • xsd:anyURI is a datatype specialised for a URL, so the thing it applies to is a literal, not a resource, and while it's a URL, as opposed to a number, generic string, token, etc., it's not expected to be able to be dereferenced to get more RDF. It's not extending the Linked Data web

e.g. rdfs:seeAlso "https://github.com/GeoscienceAustralia/AusPIX_DGGS"^^xsd:anyURI ; - here the object is a literal - a URL, not a resource. We can expect a person to "look up" this web address (URL) as opposed to a machine using it as an ID for an RDF node and perhaps trying to resolve it for more RDF. This URL, in GitHub, can never return RDF.

locn is one of the default namespaces

Default for what, a particular system you are using? It's hardly widely used and I only come across it in DCAT things and I'm hoping to see it phased out in favour of GeoSPARQL, now that GeoSPARQL has things that DCAT requires, like Bounding Box, Centroid etc.

Why not use locn:geographicName instead of dcterms:title for geographical names?

  1. Because I make no use of locn
  2. I use dcterm:title widely (for Datasets, for FeatureCollections etc.) so it's just more use of something widely known to continue using it

@nicholascar nicholascar merged commit ed0c513 into master Jun 30, 2021
@nicholascar nicholascar deleted the extended-examples branch June 30, 2021 15:17
@FransKnibbe
Copy link
Collaborator

locn is one of the default namespaces

Default for what, a particular system you are using? It's hardly widely used and I only come across it in DCAT things and I'm hoping to see it phased out in favour of GeoSPARQL, now that GeoSPARQL has things that DCAT requires, like Bounding Box, Centroid etc.

Actually, locn is a prefix defined in demo-dataset.ttl. Your own file :-). That is what I meant by "default namespace".

I think of locn as an existing ontology for spatial referencing by identifiers. The upcoming terms & definitions will mention three types of spatial referencing: by coordinates (CRS), by predefined structured geometry (DGGS) and by identifiers (e.g. geographic names). With CRS and DGGS covered by GeoSPARQL, LOCN could be considered complementary to GeoSPARQL.

@nicholascar
Copy link
Collaborator Author

Actually, locn is a prefix defined in demo-dataset.ttl. Your own file :-).

I've been caught out! So it's not true that "I make no use of locn", but I certainly don't use it much! Oh, well, I don't think the single use of it for locn:geometry is preblematic: this extended example is showing locn side-by-side with GeoSPARQL for geometry lining, so that's useful.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants