mdoering
commented
2 years ago Redundant taxa with sourceId 0 are likely an editorial bug/artefact that needs manual curation. They are likely remains of some imperfectly deleted former sector...
The duplicates page for the COL22 release shows quite a few accepted duplicates of various kinds: https://www.checklistbank.org/dataset/9837/duplicates/overview (warning! this is a slow page that takes time to load)
There are more than 3000 accepted homonym uninomials within the same nomenclatural code. That is pretty bad. If they cannot be removed we should mark some of those at least as provisionally accepted.
In general there are even >6500 redundant uninomials which are accepted or provisionally accepted in COL. For the vast majory this comes directly from the sources and COL only marks some as provisionally accepted because it is difficult to remove them because they still contain "provisional" species which are known to not belong to the genus anymore, but for which there is no better combination currently. If they would have been shared with COL according to best practises we could handle them better. But that only rarely happens so far unfortunately.
yroskov
aoern
There are 960 redundant Decapoda taxa (genera, families and superfamilies) in CoL. These taxa (and the contained low taxa) are listed by WoRMS Brachyura. But, they are also listed directly under Decapoda by sourceId 0 without any low level taxa. Example: Acanthilia Galil, 2000 (WoRMS Brachyura) Acanthilia (sourceId 0)