Hi @brian_d, the discussion as about the way that iNaturalist displays the leading ID, not really about their taxonomic validity.
In case it wasn’t clear, iNaturalist will not display the correct leading ID when a species-level Research Grade observation gets a finer-grained ID to infraspecies level, because the alternative would be to remove the research grade tag, lest users become confused about which taxonomic level it applies to (species, rather than the displayed subspecies). I hope that’s an accurate summary.
Yes, my comment was purely focused on how iNat deals with subspecies when calculating the community taxon or RG status. It has to to with how iNat’s algorithms function. The merit of subspecies in general is a theoretical question that is unrelated to the computational problem at hand.
When you talk about subspecies, it might be more appropriate to use the term ‘infraspecies’ instead. The rank of subspecies is just one of several infraspecific ranks recognised by the ICN.
I don’t think that is correct. No one is arguing for removing subspecies, just the way that they are handled for calculating obs taxon. They would still be submitted to GBIF, if the observation reaches Research Grade at the subspecies level.
My understanding is that Research Grade on infraspecies essentially works the same as on every other taxonomic level. 2 agreeing IDs and over 2/3 of the IDs have to match.
The only problem, which is what I thought we were discussing, is how Research Grade is displayed to the user in the application front-end, and the effect that it has on the leading ID displayed.
This topic has nothing to do with Research Grade. The request is that if an leading infraspecific ID is added to an observation, the observation taxon (or “display taxon”) should change to reflect that.
My understanding was that it is a problem of not being able to mark a species observation Research Grade when it has been identified to infraspecies rank, when it is RG at the rank of species but not the rank of the leading infraspecies, because it would look like the RG tag applies to the infraspecies rather than the species, and this is not the case.
In any case it is definitely not about how RG is applied to taxa at any rank, only about how taxa names (and indirectly, RG tags) are displayed in the front-end.
Maybe the problem is that some people insist on IDing directly to subspecies when it is likely that the observer was mainly interested in the species anyway.
Speaking for myself, I’m usually satisfied with species level IDs. Too many subspecies (at least in the vertebrates I’m interested in) are too poorly defined to be useful and there’s been a general trend away from recognizing them even if they remain technically valid. If the record is from an intergrade area, no subspecies should be used. Where focusing on the subspecies might be most useful is situations where proposed or pending taxonomic revisions in the near future would elevate those taxa to full species rank. I know of a few cases like that.
If I know the organism so poorly that I’m giving it a very broad initial ID – say, “Gastropods” – then I’m interested in what species it is. Any interest in subspecies only comes in after I know the species.
This fact that users are not interested in subspecies circulates since long time but I think that it may be true just in few cases, for the rest it seems more a tale without any foundation.
A lot of people on the site continue to identify subspecies solely based on location, without any visibility of distinguishing traits (and indeed often there really aren’t any as splitters have elevated any remotely distinct subspecies to the species level at this point as well as various non distinct ones). There are some cases where subspecies level ID makes sense, but so often it doesn’t, and it just messes with the mapping and database crossover integrity of observations (though i’ve mostly given up on database crossover anyway due to the taxonomic policies of the site making it nonviable). But anyway, i think there are indeed a lot of reasons users may not want a subspecies level Id on their observation especially when it is questionable.
@charlie I think you’re confusing the situation with animal (particularly vertebrate) subspecies with that of infraspecies across other kingdoms, where they often have well-recognised meanings and application.
As a botanist, my experience could not be more different than what you describe.
nope, i’m primarily a botanist too. not sure why your experience would be different. however i’m an applied field botanist not a taxonomist so that is probably the difference. Most taxonomists tend to agree with the splitter framework and don’t seem willing to consider the other side of it all.
i guess if we are going to identify subspecies based on location (though i dont think we should)… why would we even allow species level ID for those species? Just automatically put it at subspecies to save us the trouble, and mark it somehow so during ID i know it’s the only subspecies in the area.
I get why the current system is weird, and not ideal either.
Or, people could stop obsessing so much on subspecies and first make sure that the observation gets IDed to species. For observations already at species, then subspecies become relevant.
I did not cast my vote for this one; although in principle, I do like what was originally proposed (Class + Subspecies = Species).
I put in a feature request for this yesterday - not realising that we have asked for this
for
3
years
already.
iNat allows us to ID as subspecies. iNat accepts 2 IDs to subspecies, research grade at subspecies. Why not a display ID to subspecies and Needs ID?
The original suggestion, whatever plus subspecies displays as species, is also a second best. For all the taxon levels accepted by iNat, working as intended, that we observers and identifiers can understand, why is only the subspecies ignored?
37 existing threads as the problem comes up again and again with confused new iNatters.