Added new Class OldResource and new property wasPartOf

[bellerophons-pegasus]

New class and property for issue #37

[bellerophons-pegasus]

Stiff a few things to fix

[zozlak]

@bellerophons-pegasus acdh:Resource and acdh:OldResource require a common parent class so the acdh:hasPixelWidth and acdh:hesPixelHeight can denote it as its domain - see the issues reported by the automatic check

This common parent will most likely be just yet another helper class. By the way all shared cardinality constraints should be defined on it and only acdh:isPartOf/acdh:wasPartOf cardinality should be defined on the acdh:Resource/acdh:OldResource level.

[bellerophons-pegasus]

Fixed issues. Ready for review

[zozlak]

So the way we proceed now goes as follows:

• I'm adjusting the arche-lib-ingest and/or arche-ingest libraries so they are capable of generating the acdh:oldResource and I'm testing that the automatic single file versioning on ingestion works.
• Then (when we have a live example) we ask Norbert to check if it doesn't break the GUI. Once we confirm/assure it doesn't, we merge and publish.

[sstuhec]

I just have one comment/question:

Currently the acdh:wasPartOf has range "oldResource" and domain "CollectionOrPlaceOrPublication".

I assume this is because acdh:isPartOf has this domain, however, acdh:wasPartOf is specifically for an oldResource and a Resource (new or old) cannot have a range Place nor Publication. Place and Publication also cannot have "versions" (acdh:hasVersion) according to our ontology (we could discuss, if Publications could once have it, but it is currently not the case).

So, should the domain not rather be "TopCollection+Collection" (or whatever is the helper for this)?

[bellerophons-pegasus]

Good point.

To recap:
For acdh:isPartOf the domain and range were intentionally chosen as they are now to allow for such cases where a place is part of a larger place (e.g. a city within a country), or, for publications, a chapter within an edited volume.

So technically, a Resource that is part of a Place would not violate our ontology. This is one of quite a few cases were our ontology is still 'fuzzy' and 'flexible' and not too strict.
To avoid such a case, we have to rely on the curators. If we would like to also enforce this on the level of the ontology we would need two additional properties (one for places and another for publications).

But yes, the domain for acdh:wasPartOf can be narrower

[zozlak]

I prepared and released arche-lib-ingest v5 and arche-ingest v1.5 which should do the metadata management in a way which is in line with the schema changes proposed by this pull request.

The ontology from this request is ingested on the dev instance.

@csae8092 could you please give it a test run on the dev instance?

[zozlak]

@bellerophons-pegasus do you have something against integrating the #40 into this pull request?

[zozlak]

If anyone has any objections, please speak now!
Otherwise tomorrow I'm merging the pull and making a release.