Flawed logic in the comment 'Identifier disagrees this is X' at subspecies level

Bug Reports
1

TLDR: Whether the disagreement with a subspecies by the identifier is intentional or not, the automatically added comment ‘‘Identifier disagrees this is subspecies A” suggests it is, even though that may not be the case, does not correspond with the statement in the orange button but the Community ID currently gets kicked back to species X anyway.

Full post

There seems to be a logical fallacy in the disagreement at subspecies level, especially for species with only two subspecies. It is technically impossible to disagree with subspecies A without automatically identifying the observation as subspecies B. I think the flaw is in the fact that the statement ‘Identifier disagrees this is subspecies A’ gets added as a comment below the ID, but that statement does not correspond with the answer in the orange button to the question whether the evidence provided is enough to confirm it as such.

The identifier does not necessarily disagree with the observation being subspecies A - because in the case of only two subspecies this would automatically mean they identify it as subspecies B - but they disagree with the possibility of identifying an observation to subspecies level based on the evidence. These are two different things.

Let’s take species X with subspecies A and subspecies B in the database. When an identifier pushes the subspecific identification back to species level, they get asked: “Is the evidence provided enough to confirm this is subspecies A of species X?” and is given two options.

Green button: I don’t know but I am sure this is species X.
> Agreement at species level; no comment is added.

Orange button: No, but it is a member of species X.
> Disagreement at subspecies level; comment ‘‘Identifier disagrees this is subspecies A” gets added below the ID.

This extra comment, however, does not correspond to the answer in the orange button to the question that was being asked. The identifier does not necessarily say “I disagree with the subspecies”, they only say ‘No’ to the question whether the evidence provided is enough for confirmation or not. In the case of only two existing subspecies in the database, the identifier should have simply selected subspecies B if they really disagreed with subspecies A. They maybe should have even been forced to do so, but that is a different issue .

Furthermore, in the case where the orange button is selected by the identifier, meaning that according to this person ‘the evidence provided is not enough to confirm it as subspecies A’, iNaturalist currently kicks the ID back to species level. But why should the ID by someone who thinks the evidence is not enough to confirm an ID at subspecies level automatically overrule the earlier ID of someone else (for example the observer) who does think it is identifiable at subspecies level based on the evidenc e?

The easiest solution to this problem would be to change the statement ‘‘Identifier disagrees this is subspecies A’ to something like ‘‘Identifier disagrees this can be identified as subspecies A based on the evidence”. Another solution to this problem would be to introduce a three-way ‘traffic-light’ system, which could go somewhat like th is:

Is the evidence provided enough to confirm this is subspecies A of species X?

Green: I don’t know if the evidence provided is enough to confirm this as subspecies A but I agree with species X and I do not necessarily disagree with subspecie s A:

> Agreement at species level; no disagreement at subspecies level; earlier ID as subspecies A gets preserved as Community ID. No comment a dded.

Orange: I think the evidence provided is not enough to confirm it as subspecies A but I do not wish to challenge the ID as subspec ies A.

> Agreement at species level; disagreement at subspecies level is made public, but earlier ID as subspecies A (for example by the observer) gets respected, prioritized and thus preserved in the Community ID. Comment added: ‘Identifier questions the possibility of identifying as subspec ies A’.

Red: I think the evidence provided is not enough to confirm it as subspecies A and strongly suggest against identifying it as such.

> Agreement at species level; (public) disagreement at subspecies level; the ID gets kicked back to species level and the new ID gets prioritized as the Comm unity ID.

Lastly, another problem for the observer is derived from this issue: if subspecies A is disagreed upon and kicked back to species level and the observer does not do anything, the observation loses its subspecies value, does not get stored in the subspecies account and gets stored as species instead. If the observer opts out of Community ID, the subspecies ID is preserved as such, but the Research Grade is lost because it does not correspond with the Com munity ID.

2

Sorry, I am not sure why the lay-out of my original post got messy from copy-pasting the text from Word. I had wanted to add that introducing a button that suggests selecting a different subspecies could also be helpful: “I disagree with subspecies A and wish to select subspecies B (or C, D, etc.).”

3

This isn’t a bug as the behavior is intentional. You could submit a feature request to change the text to

Although, I have always understood this to be implied so I don’t think the current text is incorrect (but I understand if you think it is confusing).

If you think the evidence does not confirm the subspecies, then you do not think it can be IDed as that subspecies. It’s not that you don’t think it could be that subspecies, but based on the evidence provided you think it can’t be IDed as that subspecies. You may not think it can be IDed to any subspecies based on the evidence, and so you think it should remain at species. In this sense it doesn’t make sense to force an observer to pick a different subspecies if there are only two and they are disagreeing with the existing one.

How do you have both a disagreement and a preservation of the existing ID? That seems contradictory to me.

This is just what happens if someone thinks the observer is incorrect. It’s a democratic system, so I don’t think this is a problem.

4

But when you click the orange button, below your ID it says ‘Identifier disagrees this is subspecies A’. Maybe you don’t disagree that it is that subspecies, but you simply and only disagree with the possibilty of identifying it to subspecies level based on the evidence. In that case, you are correct to click the orange button, as the answer rightfully corresponds to the question (No, I don’t think the evidence is enough), but the statement that you suddenly ‘disagree this is subspecies A’ does not come from you; it is placed upon you in the ID section even if you don’t say so.

5

A ‘new’ identifier can say there is not enough evidence to confirm it to subspecies level based on the evidence, but that should not automatically overrule the opinion of an earlier identifier who says there is enough evidence by having selected subspecies A. Currently, the ‘new’ identifier can only agree on species level or ‘disagree’ on subspecies level overruling the previous subspecies ID, instead of the third option of offering a disagreement on identification at subspecies level without undoing a previous subspecies ID.

6

I understand, so

That seems to completely rework this system of how IDs work on iNaturalist. You want to prioritize existing IDs over subsequent ones? That seems to fundamentally change how iNat works.

7

No, and let’s not make this bigger than it is. The main idea of my post is that the statement “‘Identifier disagrees this is subspecies A” gets automatically added when you select the orange button even though it does not necessarily correspond with the answer “No, but it is a member of species X.” to the question "Is the evidence provided enough to confirm this is subspecies A of species X?”

In short: There may be not enough evidence for a subspecies ID confirmation, but that does not necessarily mean you disagree with an earlier subspecies ID like the automatically added comment suggests.

8

Here is an example. An identifier has clicked the orange button. Why? Probably because according to this person, the evidence from the observer (me in this case) is not enough to identify the observation to subspecies level. (And even though they are probably right, that does not matter for now.)
If this person clicks on the green button, they say that they don’t know if there is enough evidence and only agree with the species. But maybe they do know that there is not enough evidence. Then what? Then they have to click on the orange button. Does this automatically mean they disagree with the subspecies ID, however? Well, not per se, because maybe there is only one subspecies in a certain area and it is correct by default (without evidence…). What you merely say by clicking on the orange button is that the evidence for subspecific identification is not enough, but in the comment it says “Identifier disagrees with subspecies A”. Why does it say so? Why is there no third option of not disagreeing with the subspecies in particular and only disagreeing with the possibility of subspecific identification?

Edit: I am not sure what the rules are of showing someone’s ID on the forum. I am just using this as an example case to clarify which sentence I mean. Huge respect to this identifier in particular who has ID’d a whole lot of shield bugs and more! If this is not allowed, please let me know. :)

https://www.inaturalist.org/observations/384648008

9

Yes, something like “disagrees this contains sufficient evidence to be positively identified as A” would probably be most accurate, but is also pretty clunky. I agree the wording could probably be tweaked a bit to make it more clear, but I don’t think this is a bug, just slightly misleading wording for the sake of conciseness.

This is the current option, as the orange text makes clear.

10

Yes, but again, the issue is that the sentence “Identifier disagrees with subspecies A” then automatically gets added, even if that is not necessarily the situation.

11

Again, I think most prolific users understand this to implicitly mean “disagrees there is sufficient evidence to be identified to A” and that is certainly the intention of the button. I am not denying that the wording is a little misleading or that it could be misinterpreted.

12

Hence the proposal in my initial post to change the sentence “Identifier disagrees with…” into something like this “Identifier disagrees there is sufficient evidence to…” like you said indeed. Then we do not have to understand an implicit meaning and the text corresponds with the orange button, which is currently not the case.

13

Sure, and as Thomas said:

14

It may be worth pointing out that this isn’t just about species vs subspecies, it’s any disagreement at any level (yes, that blur might be species A, but there’s no evidence of anything beyond class).

Personally I’m happy to understand what the buttons achieve and click on whichever one gets the outcome I want, regardless of exact wording on either buttons or ID.

15

Thomas and you are probably right. Sometimes I have a hard time choosing between the two (bugs/feature request), because the text “Identifier disagrees with…” not matching with the text in the orange button does seem to be a bug (unintentional), whereas the new ‘feature’ I suggested was actually just thinking in possible solutions, but not the main point of the post. :grimacing:

16

You are right. I should not have limited the post to subspecies and I am sorry if I have only added unnecessary confusion by doing so. :grimacing:

17

all of this has been discussed many times before. there’s no need for bug reports nor feature requests because those have all been logged before.

18

Thank you! I always search for existing topics, as there are already so many and most issues have been discussed before, but I did not find those. Not sure how to delete a topic (or if users can even delete their own topic) but this one can be regarded as closed then and may be deleted by an admin. :)