Hi there, I’m mostly an IDer that does a little bit of observing, so I hope I can share some of my thoughts.
Firstly, I totally agree that identifying observations over and over again can get tedious. For captive and cultivated plants, which is what I do often, it can even feel a little Sisyphean, especially since many observers don’t respond to corrections.
However, based on my relatively small experience with uploading observations, I notice that while my vertebrate observations get to Research Grade relatively quickly (thank you!), many of my plant observations are stuck for months or even years. I keep this in mind as I identify other people’s observations, as I want to share the same feeling of satisfaction that I get when my observations reach Research Grade.
The other thing that motivates me is the possibility of helping the Computer Vision. According to this page, taxa with 100 complete, verifiable or “would be verifiable if not captive” observations and 50 with a community ID are eligible for training. Right now, I know of lot of plants (often commonly cultivated) that fall into this category, but aren’t recognized. I’m maintaining such a list here and I get excited whenever I see an observation of any of these taxa, because hopefully they will be added in the next version and it’ll save us work in the long run.
I’ve been trying to keep my number of identifications higher than my number of observations as away to help out a bit. As an amateur it can be frustrating though. I try to only do identifications for things I have seen that seem “easy” for my level and still need identification. I always consult BugGuide to see if they deem a species identifiable from photos and what range, identifying marks, and similar looking species are noted. When BugGuide also shows a lot of similar quality photos already identified to species level I figure it is safe to do so here too. I also check the similar species tab and range maps here, as well as any other sources I might find.
It can be quite disappointing when after doing hundreds of ids of a species, someone more knowledgeable than me comes along and posts that the species can’t really be identified! It makes my stomach sink. Sometimes it’s based on new information, other times there is seemingly just a lack of good documentation online. I then stop identifying the species and wonder if I should figure out how to go back and retract all my past ids. Sometimes then I see other people continue to identify them to species level anyways. It’s hard to know what to do. I know the main purpose of the site isn’t for collecting scientific data, but it’s still getting used that way. With the large number of observations being added, it seems very unrealistic to just rely on experts to do ids though.
Yes! It will learn the Mandevilla yet! It’s been a while since I went through those, I have 10 pages of backlog for “identified as Mandevilla but community taxon is not Mandevilla”
A description is often a really helpful thing. It sort of lays things out in black and white, so to speak. I don’t know what ‘bugs’ you work with, but Bugguide often has information on identification features, information about similar species, and a list of resources which may or may not be accessible. I work with moths - often I need to go to other sources both for images and descriptions. Google Scholar is a great resource - just be sure to check the links on the R side. If there is no link there, the paper is paywall protected. Fortunately, there are a number of good resources for Noctuid moths. Just do your best - I’ve had lots of ID’s corrected, and it can be a little demoralising. At least keep in the back of your mind that you are learning (though that can be hard to do), and follow up as much as possible. I wish you good luck!!
I just got done with Penstemon in my county. Did I get them all correct? Probably not. However, my motivation was for the ‘computer vision’ to stop suggesting Rocky Mountain Beardtongue (P strictus) on every single observation of the genus!
For example, this is obviously not P strictus (white-throat and light pink-purple petals), but that’s probably what ‘computer vision’ recommended. https://www.inaturalist.org/observations/84263817
I am not great with many types of ID’s and as I just recently began using this app on my phone while on primitive survival training walks with my 9 and 10 year old’s we have posted I believe 89 observations in the last few weeks. We try to get a PID from the seek app, but that seems to be hit or miss on accuracy. Since our observation of a Yellow-billed Cuckoo we now refer to as Yellow Bill my children and I have begun compiling as much data as we can and making video’s in hopes of inspiring conservation efforts from the average person. I hope we can find a way to get those members who are good at making PID’s to continue what they do as those ID’s have sparked a passion in my children, and I am certain they would not be the only children inspired by the app. I know from an observer standpoint I am thankful when I open the app and see another ID added to my observations.
I think about the large number of people who join iNat and casually log only one or a few observations. They may be superabundant species, poor quality photos, etc; but they joined in good faith hoping to contribute. It would be disappointing for them to receive no IDs whatsoever for weeks or months on end despite how many experts there are on the site. I’m by no means a higher level taxonomic expert in many insect groups, but even putting down a family name might be the only acknowledgement they need to maintain their interest in the site; and hopefully inspire deeper learning.
Funny, but when I think about it from that perspective it makes me more stressed about identifying! It makes me feel like the world will end if I don’t push myself to keep doing what isn’t enjoyable anymore. Researchers shouldn’t feel safe about using GBIF data in their research…so I shouldn’t feel responsible for delivering “research grade” iNat data to them. I’m a researcher myself, and use iNat observations which I examine with my own eyes, but would never use mass GBIF data without some way of verifying its legitimacy–especially the data that’s pulled from museum databases as those are as bad as iNat data.
For me, if I stop getting gratification from identifying commonly posted stuff, I just stop identifying it and move to something that captures my interest. Then I’ll come back to it later if I want to. I do keep in mind that many folks appreciate the identification, and since I get gratification for helping others, that does help me maintain an interest.
Also, to keep things varied, I go about identification using different approaches. There are the taxa that I focus on (a particular genus, for example) within an area I’m comfortable with. For these, sometimes I’ll display only the research grade observations of each species individually to quickly look at the thumbnails to see if any might be misidentified. Sometimes I’ll display only the ones that aren’t research grade and provide IDs for those. Sometimes I’ll display the entire genus to catch those that aren’t IDd to species yet. Sometimes I’ll bump it up to a higher taxonomic level and go about it that way.
Then there are people that I’ve become friends with (either in person from bioblitzs or online via iNat). For that approach, I’ll only display their observations (not focused on any taxa) and see if they need anything identified. This way, I get to see what my friends have been observing and comment on anything that I think’s especially interesting–to keep the friendship bonds alive.
And sometimes I’ll look at the observations map of a particular species to check if there’s anything identified as that species outside it’s expected distribution (those are almost always misidentifications).
For the taxa that I focus on, I’ve captured the ID modal URLs (narrowed down by a particular region) and I keep those links on my profile. This makes it easy to get to the particular ID modal that I want without having to reset the specific taxa/location combination.