Synonymized nodes are not counted in the tree

[grantfitzsimmons]

Synonymized terms are not counted. In the tree, next to each term "(0)" records are always displayed, despite the fact that they have been used as a determination for a collection object. If these synonyms are children of a preferred term, the indirect records of the parent are not counted.

Records can be obtained, however, from the query that is generated directly (button at the top of the tree).

Reported By: Jordi at MCNB (Trello)

[realVinayak]

This discrepancy is because counts use preferred taxon and the 'Query' button uses taxon (and not preferred taxon). Not sure if that's a bug or a design decision. In this case, probably what happened it the nodes were synomized later. The preferred taxon of determination was changed (as it should have been) to point to the new taxa, but since taxon id of determination remained same, the 'Query' still gets the rows before synomize action, but the tree count won't count these (as it looks at preferred taxon).

"If these synonyms are children of a preferred term, the indirect records of the parent are not counted." Do we really want any synomized node to instead get counts from the accepted node? I don't think it would have enough value to the user. For example, if every species is synomized to a genus in a particular branch, then after implementing this, all the species in that branch would have the exact same count (which would be of determinations directly assigned to genus, and to every other species in that subtree). Not sure if that'd be of any useful value to the user.

instead, it might make more sense to just report determinations through taxon id in case a node is not accepted (which is what 'Query' does). This would then represent the count of determinations which were previously assigned to this node before it got synomized to anything else. This might confuse the users though because the count will have different meaning depending on whether node is accepted or not.

[grantfitzsimmons]

This was brought up again today on Trello:

If we have correctly interpreted how the number of records associated with each taxon should be shown in the tree, we think that this example should be displayed as follows:

I think the tree behavior is what we should expect when the query is run. I don't think the synonyms should not be counted as you mentioned.

The query button should run the same query (so it should use Preferred Taxon instead of Taxon. This is just a bug in my opinion that has to do with the query being constructed incorrectly when the "query" button is clicked.

Let's discuss this with them and others to hear their fedback!

[grantfitzsimmons]

RBGE says it would be useful to show the synonym counts to find the filing location.

Looks like it should be a preference.

[grantfitzsimmons]

This is exactly what you mentioned at the end. It would be logical for the number shown in the tree to match the search results using the query button magnifying glass. We believe that the number should show how many CollectionObjects use the valid term (isCurrent=True). Our museum's collection database dates back to 1800, and pieces are often requested by their original name. Therefore, the history of determinations and synonyms can be long.

Jordi on Trello

[specifysoftware]

This issue has been mentioned on Specify Community Forum. There might be relevant details there:

https://discourse.specifysoftware.org/t/inconsistencies-in-the-taxon-tree-concerning-the-number-of-collection-records-associated-with-synonymized-and-preferred-terms/1686/4