Expand AssertedDistribution to be polymorphic [Was: Biological association related with Asserted distribution for a specific OTU with the corresponding Source]

[cicadina1]

Feature or enhancement

We would like to be able to have linked Asserted distribution and Biological association of any OTU with the corresponding Source? Example 1: a certain leafhopper species is known to feed on one host plant in Germany (we have a literature source for this), while in Italy it feeds on another plant (we have a literature source for this too). Example 2: the host-parasite catalogue of world Dryinidae (Hymenoptera: Chrysidoidea) is constructed on the basis of hosts as follows: Host (OTU) Parasitoid (other OTU) Country Literature source We have endeavoured to digitise and update this catalogue without losing the links between biological association and asserted distribution. Mat said that is similar to "Time ranges" and "As a generic model extension (move from otu_id in AssertedDistribution to asserted_distribution_object and resolve the (many) consequences".

Location

New Biological association

Screenshot, napkin sketch of interface, or conceptual description

No response

Your role

No response

[mjy]

The idea would be to follow previous generalization patterns and add asserted_distribution_object_type and asserted_distribution_object_id, replacing otu_id.

This gives us a generic way to say:

• The ability to assert that some data originated in some GeographicArea according to some Source
• The other target may be Observations. If on OTU then they are straightforward. If on CollectionObject then inference is something like "when the specimen was in this area, according to X, we observed Y".

There are complications to consider.

• We need to refine the use of Citation to define whether or not it is transitive, and in what ways. It is likely better to Cite both BA and AD here, with the same source (unified decision in the UI), and infer identity when reporting. Alternately we could consider adding a generic "Idea" (Group of triples) that could be cited. Again the meaning of the Citation and the nature of its transitivity would be important to consider.