Since I’ve been doing a lot of demos of our new SP7 instance to our community of collection users, I’ve noticed some persistent weird behaviors in the Taxon Tile display that I was hoping to get some clarification about. I can’t find anything in the forum related to how taxon tiles work, but perhaps some of these are known issues that are slated for fixing at some point in the future. Aside from pointing out the the bugs, I have a couple related suggestion for new features of the taxon tiles and stats panels (some of which may be possible at present, some I know are not). I’ll post those in the New Feature Requests subforum and link here.
-
The tiles are not always proportional to the number of specimens identified as the taxon displayed on the tile. e.g. in our database, there are two tiles of roughly equal size, both displaying the name “Lampsilis siliquoidea” on hovering.
Upon clicking, you can see that the two linked queries are for different taxa: the first of the queries, for siliquoidea, results in over 2000 lots, and the second query, for an obscure subspecific name “siliquoidea rosacea” returns 0 lots. The name L. s. rosacea has only been used a single time in the whole collection and is not the current determination for any lot), so in principle it should not show up in the tiles anywhere, regardless of whether the preferred or current taxon is used to count occurrences. -
As seen in the example above, the tiles do not always display the name of the taxon that the query accessible by clicking the tile is associated with. What displays when you hover over a tile is often the preferred name of that queried determined taxon in the taxon tree, but sometimes, the determined taxon name itself displays.
This is the case for the tiles in our database for two junior subspecific synonyms of the species “Fusconaia subrotunda”: “F. subrotunda subrotunda”, and “F. subrotunda leseuriana”. This case is notable as well because it appears that the preferred name, F. subrotunda, does not exist in the taxon tile graph at all despite being used as the current determination for more lots than its invalid synonyms.
Has anyone else noticed these or other seemingly related issues before? And does anyone know how the tiles are working in the backend?