I absolutely understand your frustration regarding users who are not particularly interested in data correctness and education. I think those users are the same people who post a photo of an insect to every online identification website they find and simply wait 'til someone (expert or otherwise) identifies it.
I believe those individuals are an entirely different type of user than most of the people erroneously identifying organisms. The site (iNat) does not require users to complete a tutorial before uploading images, thus, people excitedly begin sharing observations no doubt believing the magic of iNat has identified their photographs. Even worseā¦there is a population using the site as photo backup.
Personally, I wish to see who is identifying my images, so I click on the usersā profile picture. I feel more confident accepting their ID if I see a Curator tag, orā¦pertinent credentials. If any aspect of the process should be improved - to improve on the limitations of the AI, too - it should be the process that allows someone to be a trusted IDer. Perhaps some minimum number of Curator/expert-reviewed IDs that must be met before agreement by any new user is provided the credentials to be the final confirmation on an observation going to Research Grade? Sort of like the criteria (50 verifiable observations) required before a user may create a Place.
If - in many cases - only two individuals must agree on an ID for it to become Research Grade (assuming the location, etc. are also filled in)ā¦couples, friends, student peers, untrained bioblitzers, etc. are probably constantly (unintentionally) misidentifying organisms and pushing them to Research Grade. I donāt think those individuals are intentionally being malicious, they are simply mutually misinformed and ignorant to the consequences of their misidentified images being considered Research Grade.
I do the same. Additionally, if there is only a standard generated phrase xxxx is a naturalist! I look into their identification page. If they ID only certain group, this is a good sign. If they ID certain group and something more from their own region, this is good, too. But if I see an IDer with species-level identifications from plants to diptera and all the other organism groups, I am cautious. If I see the IDer with comparatively few and diverse IDs, I ask about the ID, usually āWhy is this xx species and not yy?ā. Almost invariably I do not receive any answer,or the answer is āI checked the picturesā.
I am here with you absolutely. I think,it would be good idea to add some banner for Bioblitz or City Challenge projects,or School projects where two important things would be seen as a blazing message: 1. Short explanation about what is a meaning and the consequences of Research grade; 2. Reminder to mark cultivated plants and domestic/zoo animals as captive/cultivated.
I understood the AI was the hallmark feature that separated iNat from other systems. I think itās a mistake to expect people to avoid using it. They wonāt remember and itās too easy right now. Maybe add a button called ID help instead of having it autofill with suggestions. I realize this will mean many more unknown because folks just wonāt bother at the time. But itās a question of which is more important right now. The data integrity or the expansion of adoption. Then work on getting the AI to be more location specific, request a justification comment if the choice is rare for that area and possibly a refer for additional ID if the ID is questionable.
Am I abusing the system? Iām mostly cataloguing moths in my area. Even Iām certain of the ID, I donāt type it in. I let the system pick the name because I then avoid typos and I often donāt remember the scientific name anyway, but I know it when I see it. I guess, bottom line, does an ID entered this way differ from one typed in manually?
Interestingly, lately I entered a bunch of moth images I identified manually using various field guides and
online sources. I was happy to discover that my manual IDs matched the system generated IDs in almost all cases. This has led me to trust the system, not invariably, but most of the time.
I think itās noted that the AI was used but it doesnāt really affect how the observation is listed. And yeah, the IDs for butterflies and moths, plus beetles and honeybees, have been very accurate in my experience. I personally try to catalogue other arthropods though and the AI doesnāt always hold up. Most flies I end up having to just list them as ādipteraā since none of the AI suggestions match, or they match too well and I canāt tell them apart. Same with a lot of spiders, small crustaceans, fungi, etc.
Agreed. I also use the AI feature as a way not to have to type. I avoid switching back and forth from mouse to keyboard as much as possible. I also use it as a check on what I think the organism is. Does it agree or is it suggesting something that I should be considering? And I use the AI to give me a direction to investigate if Iām not sure where to start.
But I always (unless Iām absolutely sure on id, likeā¦ Northern Cardinal) let the suggestions pop up and use a right click on all options that look feasible or interesting and open them in a new tab. Then I go to each and take a look. Does it look similar to me? Then I often look at the map and see if itās found widespread in my area, not at all in my area, or maybe Iām on the fringe of the area in which itās found. If Iām just looking to back up my hunch that Iām right, I might stop there but if Iām truly seeking out new knowledge, I then go to the internet. I have my favorite local websites for in depth identification help for wildflowers, dragonflies, and butterflies. I have field guides for my area of the US I might reference. When I feel confident that I have a good ID, I then hit the AI suggestion to autofill in the field.
I do get laughable suggestions, Iām sure due to some feature of my photo. When I do, I might choose another photo of the organism to upload, or maybe crop a photo differently.
Using the AI is quite helpful for lots of reasons. But I donāt let it make my decisions for me.
I will add, I have taken to labeling all of my photos pending for upload with numerical numbers. Theyāre usually sorted into folders by organism and they get labeled as 1a, 1b, 1c, etc in order of how I want them to appear in the slideshow. Thatās to prevent it filling in the id field from my file name. After I upload them, I add more info in the file name. Even putting in the name of the location will cause issues.
Reputation systems for Identifiers have been suggested many times on both the old Google Groups forum and this one. I donāt think itās going to happen.
When I created the observation in the screenshots, the app suggested āfrogsā. I have to admit that a blurry cell phone pic of a snake where only the head is above the water, surrounded by lily pads, does look a bit like a frog.
I should do screenshots of them when it happens. I canāt remember any but itās kind of like : I have a fungi and it identifies as a mole or something.
I think the computer vision is a brilliant tool, and what sets iNat apart from any other apps Iāve used for logging nature observations. If I didnāt have the AI, so many of my observations would be logged as āplantsā, āinsectsā, etc. I donāt think that would be more helpful. I tend to only select the genus or group that itās āalmost completely certain aboutā, and only species when it looks really likely from the images (though thatās hard to do on a phone in the sunshine). If it doesnāt suggest an āalmost certainā ID, I tend do choose a genus of one or more of the top 10 species that I think it looks most likely to be, but if nothing seems likely I might still log as kingdom level IDā¦
I think instead of discouraging use of the ID, we should discourage anyone from agreeing with an ID that isnāt already research grade, unless theyāre really certain. That seems to be where the real issue is.
Using the ID to get ācloseā makes it much more likely that the observation is looked at by an expert of that area, instead of just logging at kingdom level, where observations tend to sit for a very long time (if theyāre not trivial) before an expert happens to find them.
Coarse ids are great and helpful, so donāt think theyāre less worthy than what AI suggests, in many cases adding ādicotsā is better that adding a random species (yes, they often look right like the photo) and someone else agreeing with it. You shouldnāt agree with RG if youāre not 100% certain too, but thereās a whole long thread about how to deal with adding ids.
I agree on short informative clips how to use inat being visible and easily accessible.
I wonder if there is a way to exclude or mark taxa requiring microscopy to ID with a special warning / icon.
Another idea would be for identifiers to include verifiable credentials on their profile if they have any, and based on that, mods couId add a sticker near their name so it displayed while giving ID saying something like āverified expert on X (insert family here).ā Kids do not argue with stickers :D
Another one: calculate a rate of misses from identifier (based on corrected IDs) and use it to implement stronger ID policies.
iNat is great for some taxa and not great for others (yet). In my experience, there are a lot of knowledgeable people correctly labelling lepidoptera, herps and birds, so the database iNat has to draw on for these taxa is pretty robust. For most species of plants and fungi itās less reliable, but it is getting better.
iNat serves a lot of important purposes. It adds to an open access database of global biodiversity: that is immensely valuable. There are other databases, and some are more strictly curated. They require less āclean upā, and they generally contain much less data.
Another immensely valuable service that iNat provides is a meaningful way to engage people in exploring their local biodiversity. I work in educational research, and I know that one major obstacle that keeps many educators from bringing children into nature is their own lack of confidence in their knowledge about nature. I am certain that iNaturalist has encouraged a lot of teachers and classes to go exploring - worldwide. The citizen science part of iNaturalist is important in this context: Itās not just some school assignment, there is real, authentic value in the observations. Invariably this means increased noise. It also means vastly increased amounts of data. A trade off between data quantity and quality is almost a law of nature. In my view, messy data can be cleaned up, but missing data is entirely useless. Discouraging new users from using AI suggestions would lead to a massive drop in recruitment of new users, slow down the training of the AI (which is an AI and consequently improves with use). It might eliminate a source of annoyance for a subset of our (esteemed and valuable) IDers, but at a cost!
I am not (by far) a major contributor of IDs, mostly because I am not a specialist of any taxon. I try to contribute - mostly birds and locally common species that I am familiar with - and because I mostly do local and common species IDs, I come across a lot of newbies and misidentifications. I am happy with that, it means I can contribute.
I can see why something very obviously wrong to an expert can seem silly and annoying, but it might very well be an honest and curious attempt by someone, who years along might be the next expert themselves. If you invest a bit of patience into guiding the observer in the right direction instead of telling them off, you are both nudging someone upwards on a learning curve, and probably doing iNaturalist a huge favour in the long run.
If we donāt use the suggestions from our pics, then how is anyone who is not a world authority on the taxon in question going to identify anything, or learn to identify new species? I have learned a great deal based on the suggestions presented. What qualifies someone as as IDer? Maybe what would be better would be more guidance to new users. When I started I assumed I had to identify objects as tightly as possible, otherwise they would just sit in limbo. If Iād had some guidance saying that it was OK to leave it vague, and that others would suggest more specific identifications I wouldnāt have made this mistake. We all started somewhere, and iNat is a fabulous learning tool. Without the AI it would just be a database for āqualified IDersā to argue over the finer points of taxonomy.
Itās not a mistake if you thought you were correct. That is, if you have some knowledge or have done some research and you think you can identify at a more specific level, itās okay to be that specific.
When you make an observation, identify the organism as best as you can, even if that is just āplantā or ābird.ā (emphasis mine)
Add identifications at a level that you are confident of and that is supported by the evidence provided in the observation.
But I wouldnāt want you to think that only experts get to suggest species level ids. Many of us ānon-expertsā can identify certain types of organisms just fine (Iām pretty good at birds from my area, for example) and can do pretty well with guides on how to differentiate one species of wildflower from another species in the same genus. So donāt hesitate to identify at the level you think you are right and youāve got reason to think youāre right.
For myself, I view identifying as I did my high school math with Mrs. Gilbert. Prove my work - or be able to. I often will write out why I think my photo is a particular species of spiderwort including what details Iām looking at to differentiate from other spiderworts. And if I donāt specifically write it out, I feel I could if asked to. (For example, I donāt try to provide details that identify a photo I submitted as a Northern Cardinal but I could if asked to)
If I privately canāt feel like I can provide these detailsā¦ prove my thesis, so to speak, Iām much (much) less willing to add a species level id (or even higher levels).
I am late to the party and must admit Iāve only been able to read about a third of the initial replies to this thread but I did have thoughts I would like to add. As someone who considers himself one that falls under the category of good with bird ID but only dabbles in everything else I often find it very difficult to decide what to post as an identification for the organism Iāve photographed, and it frequently leads me to saying heck with it and not posting anything or posting to an incredibly vague classification which will likely never see further identification (such as āplantsā or ādicotsā). Alternatively, occasionally I know it is one species of plant or another, think itās one but am not positive. Sometimes this is easy like if itās a cinquefoil I can just label it as cinquefoils, other times the two species are not closely-related enough to do that and again it goes to an overly-vague classification.
My thought process is this:
iNaturalist wants as many users to contribute to their citizen science initiative as possible.
In order to do so it needs to be intuitive and allow those who may not be expert naturalists in a particular field to submit their observations.
Beyond their observation submission it needs to be classified enough so that other users will actually take the time to provide or verify the identification in order to become usable (research grade) data. Otherwise, the observation will sit there forever needing identification not being usable data and over time this may lead users who are not experts in identification to stop submitting observations because they never get to find out what it is theyāre seeing around them.
The million dollar question is how do we best achieve all three of these goals. I donāt have the answer specifically but I do think keeping things as quick and intuitive as possible is the way to go. Personally, a feature I would love seeing would be the option of illustrating the observation submitterās confidence level in their observation. This could range from āAI-based guessā to āFairly certainā to āHigh level of confidenceā (and could obviously be expressed a bit more poetically), and using some sort of system like icons or color-coding of observation for these different levels of confidence would make it instantly recognizable whether an observation might require higher scrutiny by an identifier. I know there is the computer-generated icon but often times I use auto-ID anyway because the species Iām submitting is often on of the top suggestions and itās the fastest way to ID it, in which case Iām using the feature for speed not auto-identification.
