Is this the intended behavior of the system? Is there a feature request to change it? Seems to me nobody really actively wants this to work like this, so it’s either a bug or a lack of foresight during design - in both cases it should be changed.
No, the “plant” ID only disagrees with the taxa between the original ID and “plant”. So if the first ID is a vascular plant, a second disagreeing higher ID of “plant” will be disagreeing with all vascular plant IDs. But if the observation is reidentified as a moss later on, the “plant” ID will not be in disagreement with those later IDs. It is working as intended.
6 Likes
I reckon it is working as intended? That is, it assumes (wrongly?) that when ‘hard disagreeing’, people also do their best to actually provide the most accurate and refined ID they can - not just picking “life” hastily for lack of a “I disagree” button.
edit - ok, as bouteloua said
1 Like
I check the CID - for the disagreeing ones - to make sure that my ID is doing what I intend it to. And is NOT disagreeing where I don’t intend it to.
I did eventually win a previous battle, where the CID insisted on Ancestor Disagreement when that wrong ID had been withdrawn.
Or deleted altogether.
1 Like
I am trying to understand how this works.
Say the original ID (1) is Rosa setigera. Someone comes along and takes it all the way back to “Plant” (2). Someone else comes along and says it’s in Rosa (3). Does this count as two votes for Rosa and one vote against?
1 Like
yes, that’s right.
1 Like
Okay, in your example it makes sense to me (because there are very few levels involved), but in my example, it’s very possible (even probable) that the person who chose “plants” did not understand the taxonomy well enough to disagree with every single level of it (because most people don’t).
2 Likes
I’ve certainly seen some occasional odd behaviour where a wrong ID has been withdrawn*, but it’s only occasional and I’ve never been able to track down why. I just assume it’s an odd bug but when it’s not easily replicable, it’s hard to fix.
(*) Example from the other day: I found an observation with community ID at dicot or plant (can’t remember which) which had one withdrawn ID in one plant family; a disagreeing ID to dicot/plant; and a correct ID to species in a different family. I added an agreeing ID to species, and it was suddenly research grade for that species.
There was a fairly recent change in the way ancestor disagreements were calculated. In the past, a high-level disagreement continued to count as a disagree with all child taxa even if the ID that it disagreed with was withdrawn. After the change, it is only counted as an ancestor disagreement as long as the original ID is still active. Once the original ID is withdrawn, the disagreeing ID is no longer counted when calculating the community taxon. However, the community taxon for observations affected by this change is not automatically recalculated; this only happens if you interact with the observation in question (adding an ID, DQA, etc.)
https://forum.inaturalist.org/t/change-to-ancestor-disagreement-implementation/49276
10 Likes
Interesting to know - thanks.
while i think what you’re describing matches what i think was intended, based on my testing from back in the day, they actually implemented something else. before the change, a new ancestor A ID that was a disagreement to X would disagree with any descendant of A, even if it was not a descendant of X. they called this a “branch” disagreement.
the current implementation now effectively makes the same ancestor A ID disagree with only X and X’s descendants. this is what they called a “leading” disagreement.
in other words, “branch” disagreements no longer exist. ancestor disagreements are now “leading” disagreements.
1 Like
But still.
2 disagrees with setigera.
2 has not even been ASKED if they disagree with Rosa.
1 Like
You are welcome to @mention me next time, or send a message if you prefer. I can try to untangle where CID is tripping over its feet - from our merely human perspective.
2 Likes
Yes, that’s the part that is counterintuitive. If I see a species ID that is incorrect, and if I say that it’s a plant instead, did I mean to disagree with every single level of the original ID?
It seems like the assumption is “yes, the person who called it just a plant meant to say that it’s definitely not a vascular plant, or a flowering plant, or a rose, or any of the other levels”. Which I just find a very odd assumption to make. (…I thought I was just saying it was a plant.)
4 Likes
It is odd - and it is monumentally difficult to explain to people - especially if they are scientists who are offended because I (explaining CID) am clearly a fool who missed my turn with the office brain.
4 Likes
I don’t think so. From everything I have seen, the disagreement behaves differently depending on whether the original ID has been withdrawn or not. If the original ID has not been withdrawn, it still counts as a disagreement with all descendants of A (e.g. a disagreeing ID of Insecta is considered a disagreement with Pterygota, so if the original ID was, say, Apis mellifera and someone disagreed by entering Insecta, three subsequent IDs of Eristalis would only get the community taxon to Pterygota – not Eristalis as would be the case if it were a “leading disagreement”).
If the original ID has been withdrawn, it functions as you describe.
1 Like
Yeah, I think a disagree button is unecessary and unhelpful.
BUT the way the two optionsof disagreement are presented are not very sensible to me?
Why is there the green option?
I do not understand it; as far as I can see, it does not change the CID, so is it not just spamming the activity field?
I always choose the orange one when I find something I cannot ID to species; insofar there is an actual lack of evidence. If I cannot tell there is a lack pf evidence (due to not being too familiar with the group, is it not better to leave it).
Hard disagree in forcing IDer to provide explanation when disagreeing every time.
I do not do it and will not do it.
Because in 95% percent of cases I get no response in any way. Why would I waste my time writing my arguments, if it will be ignored?
If you want to learn, I think a modicum of effort, like asking, is not too much.
I am happy to answer why I think so if I am asked!
However I do think if you make an ID on a research grade that makes it not so anymore; it is fair to provide an explanantion.
3 Likes
I mean, having the ‘I don’t know’ as an option is good for people who can ID a species but not a subspecies. Using that option can bring an observation to Research Grade without disrupting the community ID.
5 Likes
I sometimes use the green option. I’ve done it with students in classes I’ve taught, in hopes of providing some encouragement or feeling that they’re getting feedback, even if I can’t ID the species (or don’t want to take the time to ID to species, or don’t want to ID to species because the student should do that). It’s also useful when I’ve been tagged but can’t really help. Lately, I sometimes use it (also explicitly saying that I don’t disagree but don’t know enough to ID further) in that iNaturalist experiment about on-boarding new users. Less often but occasionally, I use it as a kind of a note to myself, saying “I’ve seen it and this is the best I can do,” when I want a more positive statement than simply clicking “reviewed.”
5 Likes
The green option is not actually a disagreement in terms of calculating the community ID at least. This is why the popup is titled “Potential Disagreement” (emphasis mine).
The green option is really saying “I can only ID confidently to level X.”
Of course, many users use that limited agreement to suggest that they think the current ID/CID might be incorrect. Functionally, however, the green option is not a disagreement.
8 Likes