External name providers

General
1

How is it decided which names are imported automatically via external name providers?
How can one request that a new external name provider is added?

In relation to this flag.

2

I think it is just if it is CoL 2012. Which can be a bit out of date.

https://forum.inaturalist.org/t/glitch-with-search-external-name-providers/53225/4

https://forum.inaturalist.org/t/external-providers-time-for-an-update/55366

2 Likes
3

I suspect the feature used to be broader, see this other discussion linked below. That also indicates EOL, which tomy understanding was then fusing its taxonomy under CoL. However, while CoL continues to be supported with regular updates and developments, it looks to me like EOL has been comparatively stagnant for a few years. In various taxa i frequently see “SANBI” or “uBIO” as the source, but i’m unclear if they’re staff creations or by users. I’m hopeful if those were available in past, that they’re not anymore, the former problematic, the latter defunct.

https://forum.inaturalist.org/t/what-means-search-external-name-providers/3242

Re the CoL 2012. Thanks Thomas for those links which stem from me, and @sbrushes also saw my context comment in that linked flag above. Just to note, there’s another place on forum with some context discussion but not relevant, also a couple of places where i tried dialog with iNat staff, and i may link to those - but request to update was refused. There was one potential suggestion of the angle to take to move the idea to update that which might be where to go next, but it’s going to need voices from multiple curators.

Anyway, with respect i’d suggest “a bit out of date” is a delicate massive understatement. I’m openly willing to say ‘Absurdly out of Date’.

For context to maybe better explain to iNat staff why i’m saying “absurdly”, i’ve just looked back for overview of the 2012 checklist and it says “contains 1,404,038 species” while the 2025 general release says “includes over 2.2 million extant species” (and 50k added there on the last general update). So, if the outdated CoL 2012 is the essence (or only source), then the difference between what’s available to iNat users for import using the “External name provider” feature and what CoL actually have available is about a million names!

(Side note: Also, raw numbers of names available also says nothing about how the CoL database has also many MANY taxononmic updates (recombinations, synonomies etc) and just general error fixes (misduplications removed, spellings changed etc). Else if simply care about those raw numbers, actually CoL also have an even bigger extended release with many more names, and a recently released expanded development version beyond that)

http://www.catalogueoflife.org/annual-checklist/2012/info/about

vs.

https://www.catalogueoflife.org/2025/07/09/release

1 Like
4

in a South African context ?

https://www.sanbi.ac.za/about/ also manage National Botanical Gardens, like Kirstenbosch

5

Hi Diana, yes - South African - what they have databased now might be awesome, and i maybe shouldn’t have mentioned it. Firstly i don’t know how a horde of names assigned to them (as source, referring http://www.sanbi.org/) got into iNat database - easily it’s via staff action/dedicated import so irrelevant to the “search external name provider”. Secondly, those imported names never seem recent - i can only speak for the animal databases, but seem to generally say 2018 (or around). I just now checked one now, indeed “[Added on Mar. 7, 2018 etc]”. Anyway, frequent curator flags on various South African animals have shown some problem with a taxon name(s) ascribed to that source. Could be only a tiny fraction of their names, i only see the problem cases not the (potentially vast number of) existing taxon names being used without any concern.

1 Like
6

Is there some reason not to use the GBIF list? Since iNat exports data to GBIF it would seem sensible to use their taxon dictionary.

2 Likes
7

GBIF’s taxonomy backbone is already mainly based on COL (“It is updated regularly through an automated process in which the Catalogue of Life acts as a starting point”).

It might make sense to use GBIF’s taxonomy, but it turns out that GBIF and COL are collaborating on a backbone taxonomy together, which is apparently anticipated for a fall release this year – called xRelease or XR COL:

“They are currently testing and have two public versions of the xRelease in Checklist Bank, with plans to update the taxonomy during the fall. The goal is to provide a more comprehensive and accurate taxonomic classification”

I would argue the two clearest options are to either completely remove the name importer, or to hook it into the new COL XR when it’s in a more final state.

7 Likes
8

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.