RE: SKOS in GeoSciML

No bites on prototype/holotype/exemplar then ...

Another potential application we have in GeoSciML is representing the Geologic Timescale - which is pretty much the key classification system in general geology.

The timescale is an 'ordinal reference system' whose core is the use of names for time-periods or eras, like Paleozoic, Silurian, Wenlock. It is straightforward to encode the hierarchy of eras using SKOS (e.g. see https://www.seegrid.csiro.au/subversion/GeoSciML/trunk/instances/Timescale_ICS_SKOS.xml ) but a key property of these is their ordering relationships. SKOS provides OrderedCollection, but the elements of the timescale are ordered at all levels of granularity, so in general an era is immediately followed and preceded by multiple eras of different rank - e.g the Paleozoic Era is followed by immediately by the Mesozoic Era, the Triassic Period, the Lower Triassic Epoch, and the Induan Age - see http://www.stratigraphy.org/upload/ISChart2008.pdf. So it begins to look like each era is an ordered collection, rather than a concept.

Can OrderedCollections be the members of an OrderedCollection?


[Of course the real way to resolve this is to recognize that, although it is the eras (intervals) that carry names and get the attention, the key concepts are actually the boundaries between them - which are the nodes of multiple temporal edges in a temporal topology. This has in fact been the focus of the practice of stratigraphy for about the last 40 years - see
https://www.seegrid.csiro.au/twiki/bin/view/CGIModel/GeologicTime for more discussion if you are interested.]

______
Simon.Cox@csiro.au  CSIRO Exploration & Mining
26 Dick Perry Avenue, Kensington WA 6151
PO Box 1130, Bentley WA 6102  AUSTRALIA
T: +61 (0)8 6436 8639  Cell: +61 (0) 403 302 672
Polycom PVX: 130.116.146.28
<http://www.csiro.au<http://www.csiro.au/>>

ABN: 41 687 119 230