SpeciesFileGroup / taxonworks

Workbench for biodiversity informatics.
http://taxonworks.org
MIT License
87 stars 27 forks source link

Nominotypical subs - discussion of alternate modelling approaches #3650

Open mjy opened 1 year ago

mjy commented 1 year ago

Modelling nominotypical sub (subgenera, species) in nomenclatural/taxonomic databases is notoriously difficult.

In TaxonWorks we currently have some of the following guidelines:

In exploring this issue please remember a guiding principle behind how we are modelling nomenclature, we do not change records, we add facts. When we find ourselves changing records (e.g. removing Protonyms) then this is a red flag that we're not following our over-arching philsophy.

To explore practical improvements to what we do perhaps we can enumerate a couple of Pro/Con lists, organized behind the two core approaches we might take.

A single Protonym

how

Pros

*

Cons

*

Multiple Protonyms

how

Pros

*

Cons

*

proceps commented 1 year ago

I general, I would prefer to have a single protonym for all levels. But please remember, that it is not covers speces and subspecies and genera + subgenera. We also have superspecies, supergenus, and even more complicated for family-group names: superfamily, family, subfamily, tribe, subtribe and many more. The only solution I could see to have a protonym placeholder. So that you have a nominotypical sub, with OTU and all associated relationship, The upper level taxon is like a combination. It does not have anything, just a parent placeholder in a combination.

dhobern commented 1 year ago

I agree on the single protonym choice - it makes more sense to me.