Can you report an identifier?

I was wondering if this was going to come up in the forum so I’ll explain what I’ve been doing. I have absolutely no interest in anything other than civil conversation about these issues. To be clear @birdwhisperer and @sedgequeen, my degrees are real last time I checked. I asked for advice with this problem in the following thread, but I was given many conflicting viewpoints and no real solution to what I see as a major problem:

Here was my introduction to this person while doing some identifying:

A bird that is clearly a gnatcatcher, but Black-tailed Gnatcatcher is out of range and I personally cannot be persuaded there is sufficient evidence in the photo to identify it as a Black-tailed Gnatcatcher. I left it at Gnatcatchers and tried to coax out the right identification, but this was ignored so I deleted my comments and identified it as California Gnatcatcher.

Here’s the next one:

A misidentification of a difficult group of Andean flycatchers that I have a lot of experience with, in which the identifier admits that they have zero experience. When I explained why I didn’t think it was the correct genus it was met with another misidentification.

And another one:

A difficult group of tanagers identified to subspecies when species-level identification using the photo provided is only possible by the world’s experts in this genus (if possible at all). Misidentifications are fine, understandable even with these difficult birds, but I noted a lack of interest in contrasting viewpoints and unwillingness to change an identification based on additional information.

There were a bunch of other subspecies-level identifications on other observations with some back and forth, mostly me offering some information without response.

I noticed a possible fall migrant Song Sparrow on the coast of California identified as the subspecies merrilli:

As an undergrad and grad student and now as a collections manager it has been a part of my job for the last 20 years to identify specimens to subspecies so I understand that it is difficult for many groups even with an excellent reference series. I don’t believe that all specimens, even within the breeding range of “Merrill’s” Song Sparrow can be correctly identified to subspecies. Ridgway thought the subspecies was not real and couldn’t be distinguished from intergrades between two other subspecies. So the idea that one could identify a fall migrant to this subspecies based solely on a photo that doesn’t show the entire bird is a difficult one to swallow, even amongst Song Sparrow experts. Note that I correctly identified it to Song Sparrow, as I did with the majority of the observations being discussed here. Also note that the subspecies definition followed by most in ornithology is that they must meet or exceed 75% diagnosability from other populations, so by definition upwards of 25% of a subspecies is not necessarily identifiable, even with a specimen.

So I saw the following observation identified to subspecies and I was a bit shocked:

If I just put Song Sparrow nothing changes and it stays a Research Grade identification of Merrill’s Song Sparrow, so I put Perching Birds as my identification. Note, not incorrect, just too coarse for everybody. I also answered the question that “No, I do not think the evidence in the photo is sufficient to identify it as Merrill’s Song Sparrow” honestly, but I guess that’s why somebody flagged it. I withdrew my identification because I started receiving personal insults (not from @birdwhisperer).

In the photos you mention of the birds being held I can only see part of the bird’s head. I cannot identify a bird based on what I think the observer knows or is seeing but I’m not seeing, I just can’t as a scientist. I have to evaluate the information presented.

What I really can’t understand here is that anybody can join and identify tons of observations, many incorrectly, with little or no comment on the observations. But, I identify ~500 observations at a coarser scale based on the photo I’m looking at (I believe correctly) and I don’t immediately explain myself it’s a major problem. I’m primarily interested in preventing incorrect or inaccurate information from entering the databases and subjecting researchers to weeding these unidentifiable or misidentified observations out of their datasets (as I have to do in my own research).

I had hoped that if @birdwhisperer wouldn’t listen to me about toning down the subspecies identifications then they would listen to others, particularly others that had agreed with their subspecies identification but then had to give it some more thought.

I noted in trying to tone back some of their identifications, particularly of difficult female hummingbirds with terrible photos, that others were doing the same. So, some of the top identifiers on this site spend part of their time correcting the over-precise identifications of a single person.

I’ll stop identifying because this is the third time this has happened, where I run up against an identifier identifying thousands or tens of thousands of observations incorrectly and then I focus on trying to figure out how to stop all of those spurious data from migrating to GBIF.

Apologies if I upset you, @birdwhisperer, I wish you the best of luck and hope you heed some of the advice given by others on this forum and the site about subspecies identifications.


I apologize. I was wrong. (The reported lack of responses was what led me to the error, well, that plus I was really grumpy that day for reasons having nothing to do with iNaturalist.)


Thanks, no worries at all.

My apologies as well. I didn’t realise (although I should have) that things were more complex than they seemed. Your replies to what you considered mis identifications were thorough and polite. I cannot comment on their correctness - I can hardly identify the birds in my region, good picture or not. Although I find the idea of sub species suspect, I don’t encounter it much in moth identification. As far as I’m concerned, with these examples, you have done what I hope most identifiers do!


What you wrote here makes a lot of sense. What you commented on the relevant posts at least wasn’t clear to the reader. Perhaps it would have been useful to write something like, “This bird is clearly a Song Sparrow, but I think that it can’t be identified as Merril’s Song Sparrow from the photo. In fact, there is reason to doubt that Merril’s holds together well enough to deserve subspecies status.” And then if you’re pushing it to “Perching birds” status explain why you’ve done that (peculiarities of iNaturalist rating system), and that you’ll be glad to agree to a Song Sparrow identification once the subspecies identifications have been removed. That’s a lot to write each time, but I recommend a method I use; I have a file of paragraphs I use when changing Timothy (grass) or Teasel identifications, and I copy and paste.


This is just my opinion:

@jmaley Any ID made with the belief in what you are identifying is a valid ID. There will be a lot that are “wrong”, because there are a lot of observers and identifiers that don’t have a good grasp of what is involved in making a “correct” ID… but they are making the IDs with the knowledge and information they have available to them at that time, and so they are valid identifications as well!

This seems to be a common misconception in iNat, that the Community ID either IS RIGHT or IS WRONG, or that it NEEDS TO BE CORRECTED. I don’t think it does… what I think needs to happen, is that the community needs to be educated as to what would be an appropriate ID to make. Forgetting for a moment the problematic question presented when making a coarser ID (the explicit disagreements), any identifier is going to be armed with a certain amount of knowledge, whether that be from having written the paper describing the species or just “Uncle Tom told me that’s what these are”. They hold this level of knowledge in isolation, and as far as they are concerned they are right with how they ID. Theoretically, we all put our IDs as per our personal knowledge, and the system calculates a community ID that best represents what the COMMUNITY thinks. If the calculation is based on the IDs made by the community, then it follows that if you think the community ID needs to be changed, you need to change how those other identifiers are formulating their IDs!

We don’t need to be having arguments, or calling names or attacking identifiers for identifying as they see fit. What needs to happen is to encourage more dialogue, and help people understand the complexities in making certain IDs. Even if it’s just a link to a page that explains well the difficulties. But sometimes making “controversial identifications” can be the stimulus needed to get that dialogue to happen!

A somewhat sideline issue here is whether it matters that the observation is “Needs ID” or “Research Grade”… seriously, I don’t think it does. Any scientific investigation on iNat data relying solely on that “Research Grade” status for data validity is in my opinion instantly suspect. It would be comparable to studies into whether smoking causes cancer by asking people on the street what they think! There would be the need to interrogate that data pool for validity and inherent flaws, which wouldn’t necessarily disqualify it as useful, it just puts constraints on how and when it should be used.

@birdwhisperer You make reference to the “No, this is not so and so…” question, which to me is at the heart of a lot of these arguments over explicit disagreements. The question asks one thing (Do you think there is enough evidence), but then the impact is something different (Identifier thinks it is not that taxon). In the majority of cases, when challenged the identifier will state categorically that they do not explicitly disagree that it is that species, just that they think there is not enough evidence to be sure. In many cases, they will admit that it is highly likely that species, but again, can’t be sure! This problem is going to continue to arise, and cause rifts through the community until it is resolved one way or the other. If we are to EXPLICITLY DISAGREE WHEN WE SEE A LACK OF EVIDENCE, then the identification needs to have that “Joe disagrees it’s a fish” removed, because that starts the arguments. If, on the other hand, we are to ONLY EXPLICITLY DISAGREE WHEN WE SEE EVIDENCE IT IS NOT THAT SPECIES (which is what I personally understand the explicit disagreements to be for) then the question needs to be changed accordingly. Even just a statement from iNat to indicate which use-case is correct, would help resolve the arguments over these quicker.


This has already been answered by staff and is clearer in the wording in the pop-up than the misleading text attached to the IDs. Explicitly disagreeing based on a lack of evidence is a normal use case for the “orange button”. More on that topic here: and


@kiwifergus thanks for your input but I couldn’t disagree more. I think the idea that there is no such thing as a correct identification is absurd and coddling. I believe that this encourages, or does nothing to discourage, seriously bad behavior. I think you overlooked one of my main points in the initial thread and this thread: it’s fine if users do this occasionally but not tens of thousands of times. I periodically make incorrect identifications and I always appreciate it when somebody corrects my mistake, but when somebody is going through and massively altering the data with identifications that are not supported by evidence then it is a problem. I agree that a study relying solely on iNat data would be suspect, but if iNaturalist wants scientists using their data for research then they have an interest in providing the best data possible. Allowing problems to persist affecting huge numbers of observations in the interest of “training” potentially untrainable identifiers is not acceptable in my opinion. I’ve also heard the argument about the top priority of iNaturalist as connecting people to nature, but it is also a citizen science portal providing observations to one of the largest natural history observation databases in the world, so maximizing data quality should be of high interest to everybody.


Unfortunately that is the plus and the minus of iNat. Lots of contributors, both as submitters and reviewers, and a huge range of variation in experience and knowledge (and opinion) about what can be discerned from a photograph.

In a research museum, species IDs of difficult-to-identify specimens are typically made by experts in those taxa. Those IDs might or might not be accompanied by annotations on the specimen label indicating why that ID was made. Sometimes the reputation of the IDer is enough (not that an expert is infallible), at least until the next expert comes along. The museum doesn’t invite anyone and everyone off the street to offer their opinions. But, in essence, iNat does. So we’ll always have such disagreements for some records. Which isn’t necessarily a bad thing since many attempts to determine a taxonomic ID involve a lot of back-and-forth, even occasionally among the experts, and – let’s face it – some records will not be adequately resolved for everyone.


This is the somewhat split-personality nature of iNat that might never be fully resolved. Is it mostly about connecting people to nature, with sound science being a secondary consideration? Or is it a major database for documenting life on this planet, with scientific applications? I think it’s trying to be both, in which case it is used differently by different participants. That leads to some tension, which I think is reflected in this thread.


There is, of course, a compromise solution. Say, allowing individuals with proven taxonomic bona fides (like @jmaley) to have a greater say in the identification process through some sort of “Reputation System”. Every time this suggestion is brought up, there’s a handful of vocal users to shout it down… but implementing it would solve a lot of problems on here and undoubtedly encourage more participation from researchers.

From the iNat staff:


“ … but it is also a citizen science portal providing observations to one of the largest natural history observation databases in the world, so maximizing data quality should be of high interest to everybody.”

I believe this should be the main purpose of iNat. I am a biologist and came to iNat after using eBird for many years because I wanted the eBird experience with other taxa that interest me. I have been surrised to find that iNat does not prioritise data such that it can be reliably used in the way that eBird data is regularly used.

It would be helpful if all suggested initial IDs from we users, and suggested disagreements/changes were required to give some reason for the identification. Just saying “I think this is X” is inadequate without adding “because” to it.

WhileI am at it - lets get rid of zoo observations and domestic pets.


@joe_fish - I think the idea has merit, although there’d be a lot of challenges in implementing it, assuming it even got out of committee. But I think iNat has to make some decisions about what it wants to be.

1 Like


This conversation has strayed from the topic’s initial post and is entering into areas of discussion that are already in progress elsewhere on the forum. For a nice selection (there’s more if you search for terms or phrases like, “disagreement” or “zoo animals.”) see here: (a search results page for the many places a reputation system has been mentioned)

I’m setting a timer to close this topic in a few hours. If you have any topic-specific last thoughts or responses chime in soon.

Thanks, all.


Communication! Better communication could have headed this particular problem off before it came to involve hundreds of specimens. When in doubt (or when you’re not in doubt, but obviously the readers aren’t getting it), explain more!


“I’ve seen it with two accounts in particular, both of young users who can’t possibly identify many of the observations they’ve identified, either because of poor quality photos or lack of identifying features. Both observers focused on birds but also “identified” other taxa. One has almost 57,000 identifications and the other has over 14,000 identifications.”

-@jmaley on his topic of “Overzealous Identification”

I thought I’d bring this up to offer an apology and to explain how @jmaley and I have had the same common goal, did it in different ways and obviously had the same feelings of anger or frustration towards the other user.

I’m taking the hint that I’m the second user you referenced in that thread. You also stated, “where it would appear a user has gone through and just agreed with the identification of every observation that isn’t research grade for certain taxa.”

Is that statement true in the case of me? Yes and no. Yes because I do look up certain taxon (or my local state) and I identify what I can, but I don’t just open the page and go click, click, click the agree button. I take my time looking at each photo and then agreeing but if I do like a page (so 30 observations) a day, if someone checks the taxon each week that’s 210 (probably less because I do not agree with all the observations) observations that a user might say “hmm, are you sure?”

And I look up these taxon because (1) I’m really familiar with that species, (2) I learned how to identify the species thru iNat, example being the Carolina Grasshopper, and there’s not enough identifiers to check up on the hundreds or thousands of observations that need ids and (3) because of iNat, I can at least give a general id for local observations because I probably have seen it already.

So now that you know my intentions and I know yours, it seems @jmaley that we have the same goal, to provide accurate information on iNaturalist but our methods of doing this are not aligned. And I guess you can say that I get on the defensive often and I personally blame that on big name local birders who have harassed or shut me down in the past because I’m a kid/college student and they’ve been birding for like 30 years or so, they can’t possibly be wrong. And that’s what I like about iNaturalist, because I get to talk to awesome people like some people in this thread because they correct with the benefit to help build up newer users instead of breaking them down. I really don’t mind if you correct me on sightings or identifications, all I’m asking for is a little explanation so that way I get better as a birder.


How do you interpret my statement that:

to mean that there is no such thing as a correct identification?

I don’t think it has been made clear at all, and if anything supports my point of view. From that first link you give, the case I am talking about is exampled:

What are Ancestor Disagreements?

So what are Ancestor Disagreements? If one person adds an identification of one node and another person thinks it’s not that but can’t provide an alternative on another branch, they might add an identification of an ancestor of that node. For example, I might add an identification of Seven-spotted Lady Beetle, but you might add an identification of the family lady beetles, which contains that and many other species.

You’ll notice that I bold and italicise the point I am addressing here… if you think it’s NOT that species (either you can see evidence or evidence can be inferred) then that is the use-case for explicit disagreement. If you can’t see why it’s NOT then I don’t think you should be explicitly disagreeing.

The language is clear, it talks about “disagreeing with the community taxon” and “the identification is labeled accordingly”, which very much implies that the wording of the identification is correct more-so than the question as asked in the popup. It explains how implicit disagreement was assumed, and that it was a change to become explicit disagreement because the implicit disagreement of an ancestor level ID was problematic in many cases, meaning that you now have to choose whether you think it could or can’t be that species when you ID at an ancestor level.

In the use case that I am talking about, the identifiers almost always confirm that they are not saying it CAN’T be that species, just that there is not enough evidence to be certain.

There is more support for my interpretation further into that first link. There is an image showing the current question as asked, and how it “could have been worded” to more precisely capture how the community taxon was being calculated:

**image **

In the case of the explicit disagreements that are causing the arguments, this change in the wording of the question itself would have led to a different choice being made. When asked “is there enough evidence to confirm it is seven spotted” they would reply “No”, and so would select the orange option. If asked “are you disagreeing that it is seven spotted” they would reply “No”, and would select the green option. THIS IS THE SOURCE OF THE CONFUSION THAT LEADS TO THE ARGUMENTS!

Note also that the blog article is largely about the confusion over leading vs branch disagreements (terms as defined in that blog), and doesn’t directly address the issue of whether we should be explicitly disagreeing when we can’t see evidence to support it NOT being that species, which is where the confusions and arguments are arising.


This topic was automatically closed 3 hours after the last reply. New replies are no longer allowed.