I was recently banned from a popular FB bug ID group for mentioning iNat as a resource. Has anyone else encountered this kind of issue? The mod stated that “iNaturalist is know for misidentifications and should not be relied on for ID.” I’m flummoxed.
What resources do they accept? I feel like there aren’t many out there and most of them are even less reliable than iNat…
Did you recommend iNat in general or the computer vision specifically? I wouldn’t recommend the CV for most arthropods, but the community identifications are really helpful.
18 Likes
Seems like a lot of people view iNat as untrustworthy. Heard that from a youtuber aswell.
6 Likes
That’s terrible! Sounds like the mod is bitter 
14 Likes
Do the rules explicitly state that certain resources are not allowed.
6 Likes
Sometimes you can gauge someone’s familiarity with a topic by how strongly they feel about it.
The more they hate it, the less they know about it
45 Likes
As someone who specializes in insects I do believe that you cannot rely on the community ID to be correct, people make errors, and sometimes a chain of people agree with a bad ID.
I love iNat, but I would not rely on the community ID to always be correct. iNat is only a way to find people who might know what the bug is, but you still have to think critically about whether you trust this person’s ID, iNat it is not itself a source.
That said, iNat can be a really good place to find others who know what the bug is, and I do not understand why you would be banned for mentioning it? There is a big difference between “don’t rely on it to be right all the time” and “you are not allowed to mention it”
45 Likes
Perhaps you could suggest they become identifiers to improve the data.
22 Likes
I suspect I know the FB group and I’ve had unpleasant experiences with them concerning iNat as well, which I posted about here a few years ago. Post history doesn’t seem to be recorded for long enough to find the original post and link to it.
A person in the group, a friend from grad school actually, posted an image. I downloaded it, cleaned it up, used iNat to get a few potential IDs, and then researched those, landing on one that seemed right.
I provided a possible ID and explained the process.
One of the group admins flipped out and went on a rant about how iNat was bad and that if I didn’t have the skills to properly ID species I shouldn’t be participating in the group.
Thing is my ID was correct and I pointed out that using iNat to get to a place where you could look into the literature was no different from using any other resource, such as an ID book. The admins didn’t take kindly to that and there was a bit of back-and-forth.
Didn’t get banned at the time, but the gate-keeping anti-iNat sentiment was very, very clear. I suspect it’s only gotten worse over time.
You get just as many, if not more, incorrect IDs using reference books and similar sources, but they don’t seem concerned with that.
Certain FB groups have a very short-sighted and elitist view, and don’t recognize the limitations of their perspectives. And since FB groups tend to be run by one person (or a very small number), individual biases and authoritarian behavior is common.
48 Likes
I think what they’re getting at is that a lot of people on those Facebook ID groups misuse iNat and provide bad IDs. As someone who’s been active on Facebook ID groups before, the sequence of events is often:
-Person posts photo asking for ID
-Person who doesn’t have any idea what the correct ID is uploads it into the iNat app and looks at the top suggested species, usually without enterring any location data
-CV spits out a best attempt at an ID which the person then parrots on the Facebook group
-Actual identifiers have to come in and correct the misinformation
It’s not just iNat either… my wife is working on a PhD in ornithology and knows her birds really well, and she’s had experiences on bird ID groups where she’ll say “that’s a warbler” and then 5 other people will chime in with “I asked ChatGPT and it says it’s a sparrow!!” or “iNat says it’s a reed bunting!!” By “iNat”, of course, they mean the top CV suggestion of the uncropped photo with no location data given, not the community of iNat identifiers.
As a result, many Facebook groups have banned members from taking someone else’s picture, asking an AI/CV what it is, and then just repeating that ID in the group. Just like on iNat, your ID suggestions are supposed to be your suggestions based on your knowledge. If the person asking the question wanted an AI’s opinion, they wouldn’t be posting on a community forum of human experts.
So despite iNat being a great resource, I suspect the FB mods have just become fed up with wrong IDs from outside sources, and decided to implement a zero tolerance policy for any “machine-assisted” IDs. I might not agree with it, but having modded some FB groups in the past, I can see how after deleting the 100th “My AI says this moth is a spruce tree” comment of the day, someone might become jaded toward the whole CV-ID process.
I think the more thoughtful process that @earthknight suggests is perfectly reasonable, but I also see how a Facebook mod could just be worn out and exhausted by the awful IDs coming from people who don’t know how to use the CV, and could have a gut negative response to seeing anything about iNat suggestions. Again, having been a mod on one of those groups before, the number of ridiculous ID suggestions that we would clean up and delete before the general usership even saw them is higher than one might expect.
40 Likes
So, one thing I see a lot of on bird indentation FB groups:
User A posts a recording of an unknown bird asking for ID help.
User B comments, “I played the audio for Merlin and it says it’s species x” (misidentification).
I imagine this sort of thing happens with insect ID groups all the time. A user asks for ID help for a photo. Another user screenshots and looks at the iNat Computer Vision suggestion, and then comments that ID without having any expertise in identifying that taxa.
So obviously, if you’re a moderator that doesn’t use iNat yourself, you assume that iNat made a misidentification, when in reality, the computer vision made the misidentification, but the community is ultimately what actually makes the ID on iNat (when people use iNat properly).
15 Likes
That’s authoritarian, intolerant behaviour from the group admin. 
Whatever their reasons be, banning someone from a group for this is not warranted. iNaturalist is a great resource. It’s not perfect, like democracy isn’t perfect. But it is what it is, and it is citizen science at its best; something that FB groups can never be.
4 Likes
I love iNaturalist, but there are some issues with community ID. And I do know from experience some in the academic community would prefer students to learn to ID by a key instead of posting their observations on iNaturalist.
4 Likes
The fact you were straight up banned for mentioning iNaturalist seems like a pretty extreme measure. I wonder if it’s less about iNaturalist being inaccurate, and more about it being a rival for activity to that Facebook group in general? Censoring any mention of iNaturalist within your group makes sense if you don’t want your Facebook group members taking their activity elsewhere.
You find these types of people everywhere online. They have extreme beliefs and mistreat others to impose their own control because nobody else can hold them accountable for the bad they do. You see it on Facebook, Reddit, Discord and niche tropical pet forums. It’s best to avoid these kinds of places and only use websites that have more professional moderation and mods who treat people fairly and with respect. I’ve found iNaturalist mods to be very understanding and respectful. I’ve had a few bad experiences online, but never on iNaturalist.
14 Likes
Exactly, the CV suggestion is a great starting point when trying to ID something, but too many people thoughtlessly repeating it without any further consideration on a group that’s intended to facilitate discussion of ID pointers is not helpful. I’m not suggesting that’s what OP was doing, but having to delete the comment “an AI tells me this is [wrong ID]” literally hundreds of times per week is going to result in some mods who are understandably annoyed by the very idea of the use of machine-learning to aid in IDs.
Reminds me of some trucking companies that banned the use of onboard GPS navigation aids because truckers were relying on them too heavily and getting stuck back unnavigable roads. There’s nothing inherently wrong with using the GPS to help navigate, but they allowed some people to be lazy and make poor choices, and the result was they got banned for everyone, even those who knew how to use them responsibly.
10 Likes
What FB group is this (so I can be sure to avoid it)?
2 Likes
Both iNat and FB ID groups have their pros and cons. If I need a quick plant ID, the “Plant Identification” FB group can provide me with an ID way faster than iNat (typically just a couple of minutes), which is what I need when I have datasheets to submit for my job. (I almost always post the photos to iNat as well). Part of the reason they are so fast is they have a strict set of rules that users must adhere to for optimal workflow. There are a number of plant ID experts and IDs are validated by that community.
FB groups have posted rules and likely OP violated a specific rule prohibiting mentioning iNat. Mods of high-use FB groups have a lot to manage and it is pretty typical to have a zero-tolerance policy regarding rule violations, though, being muted for 30 days is a more typical repercussion for a first offense.
7 Likes
I think part of the issue is why such a rule would exist
3 Likes
This!! I’ve brought it up to several professionals I’m volunteering with at the moment. There are so many taxons that are in need of more experts to help with, and that’s coming from someone who mainly IDs fish. I’m sure there are orders of magnitude more bug taxons in that situation…
6 Likes
Sure the CV gets a lot of stuff wrong on here right now, but the experts, when they weigh in, are flippin’ good at what they do. I suppose, having read through the thread, I can see the FB group’s point to an extent, but doesn’t iNat look toward the future? iNat, Merlin, insert-site-of-choice here, they are, at the end of the day, helpful tools that will be refined and become even better tools moving forward. But, even then, they are just tools. With a hammer, I can smash my thumb, but with the same hammer, a carpenter can build a house.
6 Likes