Include alternate names, or IDs of merged inactive taxa, to taxa data in open data taxa.csv

Platform(s), such as mobile, website, API, other: Other

URLs (aka web addresses) of any pages, if relevant: https://github.com/inaturalist/inaturalist-open-data

Description of need: I’m using iNaturalist open data for a project where I’m using another model to identify species in iNaturalist images. Sometimes, that model gives me a taxa that is no longer valid (e.g. Agraulis vanillae). That taxa is present in the data, but there’s no way to tell whether, say, Agraulis vanillae and Dione vanillae are synonyms, as that information is not encoded into the data. Currently, my workaround is to check if the taxon that matches a name is inactive, and if it is, do an API request (with all_names set to true) to find a taxon that includes that synonym in its names field.

Feature request details: There are two possible ways to address this that I can think of:

  1. Include a column in the data that lists taxon IDs that have been merged into that taxon.

  2. Include the data present in the all_names API search in the taxa.csv data.

Using the API as a workaround is fine – it’s not really bottlenecking anything – but it would be nice to not have to use it. I also feel like having alternate names in the data would be good anyway, but that’s a separate point. I totally understand if this isn’t something that is seen as important enough to include, though.

Not sure this request will fully address your use case, since it is possible that somewhere back through iNat’s taxon change history, different versions of the same inactive taxon name were associated with different active taxon names, and/or represented different taxonomic concepts.

iNaturalist is not, and never will be, a rigorous taxonomic database where names can always be resolved unambiguously.

2 Likes

Makes sense, thanks. It’d still be nice to have that information outside of the API but this helps me better understand why it’s not.

1 Like

At present, though, iNaturalist is using a taxonomy, and there is no good way for people to know what that taxonomy is except by manually navigating the UI and trying to piece it together. It won’t be perfect, it could be much better.