Don't suggest genus for monotypic genera (computer vision)

Feature Requests
21

Probably just a matter of time before the single polytypic and wide-ranging species in Diadophis gets split.

1 Like
22

Wait if this is true then this issue shouldn’t be happening…

23

They’re in the “We’re pretty sure this is in the genus” section of the drop-down (which is programmatic/not a “cv-trained” taxon), but they’re not included separate as taxa that are listed below that part.

24

A lot of us can relate. One of these days, I’m going to see a “plants” observation of a tree with leaves shaped like a duck’s foot – and just to be That Guy, I’m going to identify it as Class Ginkgoopsida.

3 Likes
25

Haha, yes, there are actually plenty of observations at “Ginkgoopsida” level on iNat. Most fly under the radar because of being casual. The monotypic genera I most commonly bump to species are Galax and Medeola.

26

I don’t really get the point of this for genera unless the CV is 100% accurate at monotypic genera, because if someone blindly clicks ‘we’re pretty sure’ on species it is the CV directly contributing to becoming RG whereas it just makes it easier for human IDers to find if they click the ‘we’re pretty sure’ at genus.

Disallowing recommending ‘pinopsida’ instead of ‘pinales’ seems harmless although I also don’t see how it has any meaningful impact on making anyones life easier because I don’t know how common it is to set up search filters like ‘all tracheophyta currently IDd to order exactly’.

1 Like
28

The current setup seems fine. In your example the species is the top result in the species suggestions, so the AI is doing its job.

I’d rather that the system stays as regular and uncomplicated as possible, and if that means that some people choose genus rather than species, well, that’s also a good thing. Means that they’re being responsible iNat users and not trying to identify beyond their confidence level.

4 Likes
29

It usually means they have no idea what they id and just choose the first option.

30

All the more reason to keep the genus option there then in my opinion.

2 Likes
31

Old topic - but kind of a peeve of mine. I’ve noticed this creates a mildy annoying UX for identifiers of monotypic genera in the bird world (ex: Corthylio and Icteria).

It is easy for an identifier to unintentionally agree with a genus-level id provided by computer vision instead of agreeing with the species id since both options are nearly visually identical in the UI. If there are other species ids, a notification is sent to anyone that previously added a species identification. This would be avoided if genus was not suggested in the first place.

Ex: https://www.inaturalist.org/observations/208993193

image
image1155×505 86.9 KB

2 Likes
32

the recent-ish(?) widespread addition of common names to monotypic genera that are just the single species’s common name pluralized certainly doesn’t help reduce confusion

1 Like
33

Also, I don’t think it’s a guideline anywhere, but no two taxa should ever share the exact same default photo. It is just asking for misidentifications. It would be very obvious that the genus and species above were at least different if the pictures were different. I’ve changed these when I find them for taxa that I’m familiar with, so if someone can judge what good pics are for those taxa and change one, that will help! It will also presumably help with CV suggestions with monotypic genera and their species to show that the two suggestions are not identical.

34

Monotypic genus has only the one species tho.

35

I don’t understand the issue?

36

The genus only contains that one sp. The identifier has to realise - monotypic genus - go straight to species. Feature request asks to ‘default out of sight’ the genus, in favour of CV straight to the sp.

37

Yes, but in @zahnerphoto 's example above, the specific aspect they are discussing is agreeing with an initial CV-based genus ID vs. a more specific species ID. This issue is made more confusing when the taxon photos for the species and genus are the same as they are in the example. The same goes more broadly for the CV suggestions where the genus and species look more similar than they should be in the suggestions drop down if the photos are the same. Having different species and genus pics helps reduce confusion, even without the implementation of a change to the CV suggestion logic/process to only suggest one. So having different pics is a way to alleviate some of the problem in the absence of any change to CV suggestions, which is worthwhile in my opinion.

3 Likes