I think we are also considering the case where it may be possible to distinguish two species, but not based on the photo(s) provided (ie. photos are poor, or organism partially obscured, or in the case of lepidoptera, the specimen condition is poor).
I don’t see why that case should be treated differently however. If there’s something in the photo that an IDer has missed, folks are free to point it out. I’ve done it many times. And once or twice, someone has pointed out something that I’ve missed.
I post Euglossa all the time but mine are not identifiable to species because my photos are not of good enough quality to distinguish the aspects necessary. Other people post Euglossa that are identifiable to species because they have better equipment and thus are able to capture the aspects required.
Right?
I agree this should be treated the same way. The end result should be Genus, Genus, can it be improved? No. Research Grade.
But does that last part even matter? What’s wrong with a Genus level ID remaining in the “Needs ID” state? Again, as someone who does a lot of IDing for a particular region and taxon, and who uses the resulting observation data, this is a non-issue as far as I’m concerned.
One valuable thing about clicking “No, it cannot be improved” when something is at the genus level is that then it gets out of the way of people trying to identify observations. Very useful!
I greatly appreciate the experts working in the field or labs (or here is inclusive) as I have come to recognize that some of these IDs are quite difficult. As a non-expert in the biological sciences, I usually use iNat’s information - photos, descriptions, maps - in making IDs. However, I have become very shy of doing this because I have made or seconded species IDs and later have been told that one cannot make such an ID without looking at the sexual organs with a microscope, etc. And I greatly appreciate this. My question is then, why doesn’t iNat flag such species and require a separate qualifier that insures the person making the ID has performed the required investigation and analysis.
On some IDs, under a given genus, or taxon level, I spend a lot of time and effort eliminating species (or genus or whatever taxon level I am working at) as I try to come to an ID and still come up blank. I often think it would be nice to be able to say it is not species A, B, E, F, etc. leaving the possibility that it is C, D, etc. This may help others in subsequent ID efforts. I sometimes do this as a comment.
My response to this would be - don’t be shy, as long as you are prepared to be responsive when someone comes back with the feedback you describe. As long as it’s a dialogue, it’s not a problem. As someone who does a lot of ID work, I don’t have a problem teaching a man to fish (as the saying goes). What causes me grief are the drive-by IDs, where somebody plunks an ID down and then doesn’t follow up when I type in a long explanation as to why their ID might not be correct.
I don’t know the mechanics of it, but it’s possible to define a “complex” or a “subgenus” of closely related species. So in your example, you could create a “C/D Complex” that specifically excludes A, B, E, and F.
The problem is knowing whether or not such a taxon exists, and getting folks to use it appropriately. I don’t know if such things will pop up in the suggestions offered by the CV. If not, then it’s a convention you have to inculcate into the community: ie. “this is the taxon we are going to use for C and D when we can’t tell them apart”, and then it needs to be reinforced periodically because there are always new users who aren’t aware that the complex exists and is being used in this way. It’s not impossible, but it does require patience and vigilance. And I should add, I think it’s the right thing to do. We’re currently discussing the creation of one for Pearl/Northern Crescents.
I guess that is up to the researchers in question. For the database I manage, I only download research grade observations on my “first pass”. But then for certain problem taxa, where I bother to maintain observations at a higher level (say, at genus level), I perform a second pass where I grab all the “needs ID” observations that fall into that genus. One can do it either way, but I’d rather not force observations to Research Grade just for the sake of including them in my download.
Ok, if GBIF does something useful with Genus level observations, then that’s a legitimate reason for doing this.
I don’t usually see CV suggest complexes/subspecies/section/etc that often, unfortunately, I usually have to type them in and then they’ll start popping up. Which is definitely unfortunate, I think there are groups that could benefit from only suggesting sections/comlexes because they’re just too hard to ID from photos
Good to know - I haven’t used CV often enough to know what does/does not appear in the suggestions. Overall, I think it works remarkably well (most of the time) for my taxon of interest (butterflies). The biggest problems I’ve seen with the CV are issues like this - interface issues which lead users to second guess what the CV is telling them. Unfortunately, some of these are probably difficult to fix. I often see people choose an ID at the genus level when there’s only one member of the genus that could conceivably be found in our area. I keep thinking “you were there! you had it! why did you balk?”.
This discussion has been drifting off point somewhat …
No (@ItsMeLucy), this thread is not primarily about the situation described in my first post. That situation is simply an illustrative case in point. The actual point is given in the thread title. It’s about logic and critical thinking (or the absence thereof). In a broader sense, it’s an issue of scientific ethics and integrity.
I’m very grateful to the folks who made an effort to listen and understand my concern. You’ve given me a lot to think about. Mostly, that it can be really difficult to generalize issues like this beyond one’s own field of interest and/or beyond one’s region of residence in the world.
I’m also amazed and quite scared to see how many contributors are perfectly fine with clear violations of logic as long as the “end result” fits their needs. There are even declarations in this thread that read something like, “yes, for me the statement ‘it’s species A or species B’ is the same as ‘it’s not species A’”. They simply refuse to accept one of the most basic and otherwise universally accepted rules of logic.
The notion that the desired end result justifies the means, irrespective of the abandonment of logic and of the falsification of an identifier’s choice along the way, is unscientific to the core. It’s the stuff that gets folks like Marc Tessier-Lavigne (former president of Stanford University) or Francesca Gino (Harvard professor who, ironically, studies dishonesty) come under scrutiny, lose their jobs, get their papers retracted.
I’ve spent what feels like years of my life pointing out inconsistencies, fallacies, unintended ambivalences, unclear or misleading language etc. to authors of PhD theses and publications in the Life Sciences and in Engineering. To me personally it is unimaginable that someone educated in critical thinking might accept that their (and all others’) agreement to the statement, “There is not enough evidence to confirm this is species A” gets automatically converted to “I disagree that this is species A”. That’s akin to saying, “There is not enough evidence that B was involved in his son’s questionable business dealings”, only to later see this your statement in print interpreted as, “I disagree that B is an honest person”. Would you be okay with this?
To my knowledge, there is a huge difference between ”Needs ID” and “Research Grade” as far as broader consequences are concerned. Numerous incorrect “Research Grade” IDs have seriously polluted other databases like GBIF, SCAN, and so on. As identifiers, we should be much stricter with those than the “Needs ID” observations. I am not sure if external databases also incorporate genus-level RG data. Unless I learn that they definitely do, I don’t see much of a point in making those RG (other than satisfying the ego of sometimes not-so experienced identifiers). But I would be extremely skeptical if I got a scientific paper based solely on iNat data for review. At the very least I would require the authors to include a declaration that either they or a named, well-established expert have personally reviewed every single observation used in the study. iNat has been designed (I’m told) primarily to engage citizens from as diverse as possible backgrounds and to promote their interest in nature, not to be used as the sole basis of hard-core scientific research.
Lastly, I think we should be extremely careful about using the “can the Community Taxon still be confirmed or improved?” option in the kind of situations discussed here. There always is a chance that the original observer’s ID suggestion is based on more than the photo, such as local seasonality, habitat, behavior of the specimen, even—in case of moths—the time during the night at which they first appear at the light. The original observer also might be very experienced and/or more knowledgeable about the relevant taxonomy than what can be found in outdated literature or on non-peer-reviewed WEB sites (which, unfortunately, seem to be the only available resource for moth identification in the USA, with the exception of a few recent “fascicles”). Downgrading such observations could remove very valuable data points.
I think the less variable the organism, and the less necessary angles to get a good id, the better CV works.
Even with something as variable as mushrooms, its sometimes really good (like, it’ll at least get genuses right on a lot of the more common ones,) but sometimes its so wildly inaccurate that I just have to laugh.
I feel like you’re making the assumption here that every photo uploaded to iNat is going to be an in focus photo, with good lighting, that has images of all the macroscopic characteristics necessary to ID something to species. Sometimes its just a bad photo, or not enough photos, and putting it at family/genus whatever and marking that the ID can’t be improved is okay. Sometimes a photo is just insufficient. And if subsequent identifiers thinks that it can be IDed further, they have the option to disagree.
I will say, I do like the way mushroom observer handles community agreements - its more a sliding scale based on how sure a given ider is. So not just a binary choice, but I think 5 options ranging from ‘probably’ to ‘definitely not’
But believe me when I say, there are an insane amount of fungi observations that cannot be IDed by just photo, either because the photo is too good or the differences between species come down to microscopy. Yes there is some onus on identifiers to be aware of these differences, but we have to work in a system where not everyone is an expert with expert level knowledge, so there’s going to be some fuzziness.
And beyond all that, even the concept of what a species is isn’t a binary thing. Its fuzzy, there’s a sliding scale, there’s some genuses where Species A can mate with Species B and Species C, but Species B cannot mate with C and produce viable offspring. Are they the same species? Are they different? Where is the line drawn? If big name scientists are still trying to figure this out - expecting clear binary logic on a citizen science platform is, IMHO, a little much.
edit: Here’s just one example, a 2021 paper on the genus Auricularia https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8625027/ - almost the entire paper is focused on microscopy. To an extent you can base IDs off of collection ranges, but there are still overlapping species that lack macroscopic differences.
I am sorry if I have brought any distress to you and hope it allays you to know I rarely make identifications. I was trying to understand but admit to being, as you feared, unscientific. I am just a lady with a small wild space, no more no less, worth no more or less than any other user here.
Nah, I’m very familiar with the quality range of submitted photos. And I, too, sometimes burst out laughing. But when you try to keep your local iNat moth IDs as accurate as possible, it can get a little frustrating when you comment for the thousandth time something like, “This species is endemic to New Zealand and is not expected to occur in America.”
All I am saying is that we should be careful with the “can the Community Taxon still be confirmed or improved?” option. If you are one of the real experts who have studied the field for many years and really know what they’re doing, then of course, go for the “No, it can’t be improved” button.
But you also have to realize that there are plenty of non-experts and wannabe-experts who make generous use of this button after minimal study or no study at all, even when the evidence for doing so is shaky. Then some potentially important data point gets lost.
Its frustrating, but that’s just the nature of the site. People are illogical, and they do things without thinking too hard about it, and hopefully they learn but a lot of times they don’t.
Personally, I only click the data quality ‘can it be improved’ option for things I’m sure on, and even then there’s some local observers that I won’t, because I know they’re sending things in for DNA testing. But I definitely do use it. I will say, I feel like most casual users probably don’t even realize its an option, you have to scroll way down on the website and its a little hidden on the app.
Agreed - this is definitely a YMMV kind of issue. I am kept so busy reviewing observations for my taxon/region that I have no idea what’s going on in other taxa/regions. I admit that I have a very narrow view.
I think you’re greatly exaggerating the problem. Yes, the text that iNat displays when you bump an ID up to a higher taxonomic level says “A disagrees that this is species X”. How about if the text were changed to “A disagrees that this is DEFINITELY species X”. Would that satisfy your rules of logic? If so, I humbly submit that your objection is trivial. It’s not that I believe that the ends justify the means, it’s more that I don’t expect absolute perfection from iNat. Having seen how badly some other citizen science platforms perform, I recognize that it’s a miracle that iNat works as well as it does, and so I’m willing to shrug off a minor issue like this. I’m sure the developers have bigger fish to fry, and I’d rather they focus their limited resources on those other issues. I can live with a poorly worded bit of text.
As have I, but you have to pick your battles. I don’t know about you, but I already spend enough time standing out in the yard, shaking my fist at the clouds.
As I pointed out earlier, would inserting the word “definitely” solve the problem?? As in, would the statement “I disagree that this is definitely species A” not be equivalent? If so, can we not just taken it as given? (just to save the developers a bit of time)
I have no doubt of this. My only response to this is “not on my watch”. That’s why I asked others why it’s necessary to push IDs to Research Grade when they can be left at Needs ID.
Speaking for myself, I think it’s more likely to be to satisfy the egos of not-so experienced observers. In my experience, it’s observer who are more likely to be obsessive about having their observations all ID’d to species level and research grade. I’m perfectly content to leave IDs at Needs ID. It doesn’t hinder me in the least.
Your tone suggests that you don’t believe that this could ever happen. I suspect that you are largely correct. Not because I have any supporting evidence, but simply because I don’t see very many people who are thorough enough to do something like that. But in the case of the database I manage, I can confidently say that pretty much all the iNat observations that go into it have been reviewed by me personally. For 2023, I’m considering making it explicit, now that the download function offers the option of filtering based on whether or not the data has been flagged as “reviewed”. I’m considering doing so because it’s a convenient way of excluding dubious observations (ie. I simply never mark them as reviewed, and then I filter for only reviewed observations). After the data is downloaded, it is filtered several more times based on various criteria, so effectively, I’ve reviewed it again, outside iNat. But I doubt that many other databases are so rigorous. Nobody could ever afford it.
If this evidence exists, shouldn’t it be included in the observation? If it’s excluded, and a reviewer downgrades the observation in response, shouldn’t the observer respond with the evidence that should be considered? I’ve had lots of people submit a single photo with an observation, only to discover after some back-and-forth that they actually have additional photos that they didn’t bother to include which in fact allowed for positive identification. You seem to want to cast these people who withhold information as some kind of victim. I prefer to think of them as some kind of knucklehead. But I try to be patient and diplomatic.
Again, I can’t speak for what happens in your region for the taxa that are of interest to you, but in my world, far more often than not, I’m the guy who pulls the rabbit out of the hat. I’m the guy who looks at the blurry photo that only shows part of a wing and tells the person exactly what species they have photographed (with an explanation of how I arrived at that conclusion, with links to supporting photos, etc.) That’s what happens far more often than my having to say “well, we can’t ID this to species so it will have to remain at the genus level”. And yeah, I’m not a professional entomologist, but I do communicate with one regularly, so I’m usually up on at least some of the most recent research. And I bounce the tricky observations off him - always conscious that he doesn’t have a lot of time to devote to iNat (and weird as it may seem, he bounces stuff off of me from time to time). How do you know that a given Website hasn’t been peer reviewed? ( not that peer review guarantees accuracy - I’ve found errors in publications that have been reviewed …)
Ok, if you’re working with Moths, then you definitely have my sympathies. Moths be tough. I can see IDs being punted up to genus level being a much bigger problem there. But still, aren’t you just quibbling over that one missing word? Maybe my perspective is distorted because I’m working towards a very concrete goal - the data I vet goes directly into a database that can be publicly accessed and is used by many people. So for me, yeah, worrying about that one little word is a minor thing. I am actually trying to get something done, something concrete. Did iNat use the best wording ? Maybe not.
This is the biggest problem. You seem to be worried that IDer’s may downgrade a few IDs. I think that should be the least of your worries. What do I worry about? I worry that many of the locations and dates of observations I’m verifying are completely fabricated. I see enough that I’m reasonably sure are incorrect to make me suspect that I’m only seeing the tip of the iceberg (ie. I’m only picking up on the ones that were done sloppily). A downgraded ID is easily reversed. We’ll never know if the locations/dates of many observations are accurate.