align geologic feature defn with guidance

[nicholascar]

Guidance as per https://github.com/ESIPFed/sweet/wiki/SWEET-Annotation-Convention. This PR makes it clear that it was the definition that was contributed to by GSQ, as opposed to a general contributor note on the class as a whole.

This sets a precedence for the future (lots of - 50+) definitions that we'll be pushing up shortly where the definition provider (creator) is some academic author with GSQ playing the role of contributor.

Also, the node for the GSQ is called out with its own URI, as opposed to using a Blank Node, since we will need repeated references to GSQ and this will therefore better link things back than a series of un-related BNs all with identical sdo:name & sdo:identifier properties.

[nicholascar]

This PR's changes are now included in PR #187

[pbuttigieg]

Tricky to handle in a computable ontology - metaphysics and things in the conceptual (let alone hypothesised conceptual) realm can have great difficulty mixing when logical coherency needs to be preserved.

[dr-shorthair]

I believe this all goes back to the model originally introduced by the NADM team in the early 2000s, and subsequently adapted for GeoSciML. The idea is that there is a notional 'geologic feature' which is primarily associated with the events and processes which led to its formation. The full extent of its actual geometry will usually not be known, but there are mappable 'occurences' - which are what used to appear as coloured blobs on maps and x-sections.

It is conceptually rather similar to the split between taxon and organism-occurrence.

[nicholascar]

@pbuttigieg if you want to break with the conceptual feature from GeoSciML, can you suggest an alternative definition via a PR? The Geological Survey of Queensland staff and others really require written definitions as well as ontology relations so just removing the definition wouldn't be great.

[graybeal]

OK, this risks getting into ontolog-type, angels-on-pins discussion, but I'm not sure the conceptual/hypothetical nature of the feature is critical for our ontology's purposes.

What breaks if you say "A coherent structure in the Earth resulting from geological processes"?

I understand the purists will argue about whether the structure actually exists, but for practical purposes, that argument takes place in the performance of the science. (I claim it exists, Joe claims my science is wrong, we have two views of whether this feature is real or not—but we can be referring to it as a feature during the argument. Just like we refer to taxa as entities, and then argue about them, but still assign organism-occurrence to them. Great analogy, @dr-shorthair!)

[smrgeoinfo]

The other thing to bear in mind is that 'GeologicFeature' was defined in GeoSciML with an eye towards the xml implementation, in which GeologicFeature is an abstract element. From an ontologic point a view the concrete subclasses of GeologicFeature (geologic structures, Events, geologic units) are things in the Earth someone wants to describe and relate to other things, so from that point of view of that observer, they are real and that's all that matters as far as a computer information system is concerned.