Yes, thank you.
What I saw much more often while working on observations stuck at “Birds” was misidentifying Accipitriformes, Falconiformes, and Cathartiformes as each other.
2 Likes
Actually, I sometimes might do these kinds of disagreements as well… it’s often a kind of laziness or not seeing a big deal… e.g. I might disagree and suggest “insect” instead of “pterygota” even if almost all insects will fall into this category. But I guess IDers doing a broad sweep will filter for insect anyways rather then winged insects.
Btw. 222k sitting on Araneae, if anyone wants to help work on that too 
1 Like
For what it’s worth, I do have the impression that observations given an initial ID of “Insecta” tend to sit there longer than if the initial ID is “Pterygota”, at least in Europe – I suspect more IDers focus on Pterygota because this is where the order disagreements end up; since there are so few insects not in Pterygota they don’t regularly check Insecta even though for the non-entomologist this is probably the more familiar category. (Springtails also create some confusion, so I’ve started looking at Hexapoda through Pterygota instead of just Pterygota to catch all the broadly ID’d 6-legged arthropods, but I don’t know how many other people do the same).
So if it has wings it may be worth the extra effort to put it in Pterygota instead of Insecta.
3 Likes
iNaturalist’s blog is celebrating 10 years of the City Nature Challenge on iNaturalist. Now, I really enjoy participating in the CNC and am one of the local organizers for it. But in my (perhaps excessive) need to “finish” identifying everything on iNat, I will note that there are more than 20,900 true Unknowns still remaining from the 2024 CNC. I know that last year and this year, the global organizers stressed concentrating on wild organisms (and it would be interesting to know if the percent of captive/cultivated observations declined last year), but they haven’t stressed having observers give an initial ID when an observation is uploaded. Personally, I think it’s completely understandable that the global organizers don’t stress giving initial IDs; they have an enormous job wrangling the CNC as it is.
But is there anything we identifiers can do? Sure, if we’re local organizers or leading a CNC field trip, we can emphasize giving an initial ID, but let’s fact it - we few won’t make a big difference.
We could try pasting a message about adding initial IDs to Unknown observations as we see them come though during the CNC, but again, I don’t think that will make a noticeable difference.
We could try ignoring the CNC Unknowns. Mostly, that is what happens already. For lots of reasons, that’s not a particularly satisfying end point. But is there anything else we can do? Suggestions, anyone?
2 Likes
TLDR version: I think your point about leaving an ID so people can learn is excellent, but I think it’s also on the IDer to follow-up.
I looked at the link on not deleting misidentified species, but I too have deleted observations mis-id’d by overconfident IDers that have sat there too long or that I strongly disagree with. I did not repost, but I actually wouldn’t have any compunction about doing so in most cases. The way I look at it, with respect to deletion, I don’t owe iNat everything I observe (and you wouldn’t want to see it! I promise!). And if the IDer is engaged, they will (1) add a comment listing their reasoning on a disagreeing ID and (2) check back on their work. I usually do when I disagree (“You’ve ID’d a Bald Eagle, but your photo shows a Yellow-rumped Warbler. Wrong photo?” or “Leg and bill color doesn’t look right and a Snowy would show some yellow in the lores in all age classes.”) I should tag disagree-ers and engage with them more, and I will try to do so, but see below: if they wait too long, I’ve moved on.
I love iNaturalist for the community, and I try to be a good citizen. I follow quite a few people who have been especially nice about IDing my posts or who post about things I’d like to learn more about. I ID birds a lot, and I’m an excellent birder. But every time I log in, I check my very full feed, which disappears if you click, which is maddening, and I have a link to check for disagreements. I check anything I had any uncertainty about (usually but not always in other taxa), with a special scan for non–research grade observations as well. I have definitely learned and made corrections based on that practice. But adding a disagreeing ID with no comments about fieldmarks / your decision and never checking back is not learning. I have deleted under five observations out of roughly 1250, so I’m not a rampant deleter. But I’m usually ok with it. I definitely leave observations where disagreement could be warranted (poor photo, species hard to distinguish in that plumage, species I have limited knowledge on, etc.).
What is my reasoning? I am human, and I am annoyed. I choose to give myself grace on that. I’ve worked so hard (with great joy, that is true!) to gain skills, and having someone with little investment in birding or iNat derail an ID and never check back irritates me. It becomes virtually impossible to find most bird IDs relatively quickly due to the number of observations. They’ve moved on, and I wish to do so as well.
Tagging people doesn’t always work, and I know, because I ID all the time, that people are doing other important work. At least one of the top IDers in birds is sadly deceased.
Yours in pettiness, perhaps, but trying very hard, always, Nancy
4 Likes
I agree. Out of seeing my local birds every day, I can now id quite a lot of common birds here in india, but I know quite little about other, non-indian species.
1 Like
Shouldn’t these be swept up in the Phylogenetic Projects for ‘unknown’ observations? Those don’t actually move the observations out of the “Needs ID” pile, but at they provide a suggested ID for any IDers who pursue these, more efficiently than if we added high-level IDs to each of them.
I just spot checked a few of these unknowns. Some had been added to these projects and some hadn’t. I don’t know the protocol for what gets included and what doesn’t.
P.S. Continuing my spot checking turns up examples of all of the problems that always plague us identifiers. There are some unknowns with multiple conflicting IDs that ought to be in the Pre-Maverick project. Looking for true unknowns by adding &identified=false to the URL turns up some cases where the Phylogenetic Project quite reasonably added observations to the Unknown / Vertebrata project…but the humans prominently in the photo probably intended the observation to be the plant shoot or insect they’re holding in their hand. I suppose we ought to add an ID to indicate disagreement with that guess.
Sigh.
1 Like
That is updated every 6 months. Since it uses API calls. Would be better if iNat did the Pre-Mavericks - since it does the outright Mavericks.
And the same API calls issue for all the Phylogenetic Projects.
We try. Our best.
CNC projects need more emphasis, across the iNat world, on IDing long after the project deadline. But it is part of the - we need more identifiers - we need identifiers to ID more - we need more iNatters to ID.
But - difficult when identifiers gets hostility because their ID is not a ‘perfect at sp’.
Ultimately we choose slices (or a slice at least) to ID which is our pain point. I am painfully slowly going thru Good as Can Be - where we can now apply the DQA for Multiple Subjects instead (and irritating another busy identifier with notifications - sorry - not sorry - we have muted each other)
Please tell me that other identifiers did not ID for your deleted obs? And that you don’t expect them to ID again if you upload your picture again?
In that spirit of cleaning up before CNC.
The African ones are done.
395 obs for the Rest of the World
https://www.inaturalist.org/projects/multiple-species-per-observation
Since it is a project it can be used as a filter with your location.
Do you by any chance have a link that’ll help filter for these? I rarely work through true unknowns but could spend some time chipping through it.
Edit: you might have a more targeted one, but for anyone interested, here’s the filtered unknowns based on last year’s dates
https://www.inaturalist.org/observations/identify?iconic_taxa=unknown&d1=2024-04-26&d2=2024-04-29
There are two ways you can refine your link even further. The first is to cull out the bacteria, viruses, Life, etc. by adding &identified=false to the URL, like this: https://www.inaturalist.org/observations/identify?iconic_taxa=unknown&d1=2024-04-26&d2=2024-04-29&identified=false
The second is by adding the three umbrella CNC projects (Eurasia, North and South America, and Global), like this: https://www.inaturalist.org/observations/identify?iconic_taxa=unknown&d1=2024-04-26&d2=2024-04-29&identified=false&project_id=189155,181467,195872
Thanks for chipping away at this!
2 Likes
Thanks so much! Because yes I certainly can’t manage bacteria and viruses!
I worry that I have unwittingly contributed to a perception of the City Nature Challenge as the main contributor of Unknowns to iNaturalist. Certainly, lots of CNC observations are uploaded without an initial ID - but so are many, many observations throughout the year, completely separately from the CNC.
I also worry that people will get the idea from my posts here that identifying Unknowns is THE most important contribution an identifier can make to iNaturalist It’s not; there’s a very strong case that making IDs that (justifiably) boost observations to Research Grade is more important. It’s wonderful that @hcoste may make time to work on CNC Unknowns, but perhaps her considerable talents are better spent working on vascular plants in northeastern North America, where she’s already done so much.
It will take an enormous amount of work for iNaturalist to implement better onboarding, and even more work for all of us to teach observers how best to photograph this taxon versus that one. No one can expect us identifiers to accomplish all of that, especially in response to the short, four-day CNC every year. Maybe we should just ignore Unknowns in favor of learning to ID more taxa to the species level, where possible.
4 Likes
Aw thank you Lynn, but if it helps to ease your mind, how you are worried it came across is not how I took it at all. I saw it and went ‘hmm bet I could make some headway on that’ as a nice change of pace to the more typical IDing I do. Whacking down at numbers of familiar plants in the north east is deeply satisfying, but sometimes its also fun to look at organisms I am less familiar with as well, even if those I can’t get as far as the ones I know best!
5 Likes
Hi, @DianaStuder, regarding whether other identifiers ID’d, yes, another identifier ID’d, and it was incorrect. I’ve deleted 3 out of 1,251 observations. I know one was a Gray-cheeked Thrush near Tern Bar Slough in southwest Indiana, USA, and I got up for the lunar eclipse so can’t recall the other two atm.
Regarding whether I expect them to ID again, I do not—the photos have been deleted. Regarding whether I will upload the photos again, as I said in my comment, I won’t repost / upload the observations again. I generally take many nature photos in a week and don’t upload a fraction of that. I’ve long since moved on and wouldn’t know where to begin to look, although I could probably cross check with my eBird account.
And I don’t expect people to ID my observations at all. Having said that, I usually don’t have trouble with that because I try to ID one hundred or more observations per day (sometimes I have other commitments, but that’s my goal). I also keep an eye on ID’s per personal observations and try to keep the ration at 50 to 1 or more. I work hard to make my photos IDable. I upload poorer quality photos to eBird for documentation of rarities and because they use poorer quality media to train their AI, but I try to include only clear photos including necessary field marks for ID here—I’m sure I miss that mark in taxa I’m not familiar with. Because I’m a frequent IDer; because I mostly post birds, which are easily ID’d by sight or vocalizations; because the birders here are an active community; and because I follow a fair number of people in my area, I have a very high rate of excellent IDers reviewing my photos. I think I have mushrooms and plants that are not IDd. My birds and mammals are all correctly IDd except the three or so out of 1,250 that I’ve deleted, I think, but I’ve not checked.
That doesn’t mean I don’t care about new birders or naturalists in general—I very, very much do. I often lead walks, often take out new birders, try my best to explain any disagreeing IDs myself, and try to think of something encouraging to say regardless.
4 Likes
Thank you. Way ahead of me!
Diving back into Eastern Cottontail (Sylvilagus floridanus) vs Swamp Rabbit (Sylvilagus aquaticus) this week. @jeanphilippeb very kindly populated two yellow projects to help me find erroneous RG IDs.
RG Sylvilagus floridanus with CV Sylvilagus aquaticus - RG observations identified as Sylvilagus floridanus, with highest CV score for Sylvilagus aquaticus
and
RG Sylvilagus floridanus with no CV suggestions
4 Likes
Last week, I followed @lynnharper’s lead in trying to identify the CNC unknowns from last year. I was able to help with some of them, but it wasn’t really a productive use of my time. Moreover, the issues I wrote about here made me sound grumpier than I intended. Because, really, the approach of spring in the northern hemisphere has me in a pretty good mood, with the return of the wildflowers that I’ve been able to identify the most. They may not be blooming around me, yet, but observers are certainly finding them to the south.
Another of my favorite species, Velella velella, not only experiences a seasonal peak in observations between April-May, but it also goes through a multi-year bust and boom cycle. As seen on that species’ history chart, there was a boom in 2018-2019 and a much larger boom in 2023-2024. (Part of that may be explained by the general increase in iNat observations for all species). So far, for the first two weeks of March, it looks like the start of another boom season.
Finally, I’ve been working on the pile of Arisaema observations in eastern North America, over 17,000 of which are still at Needs ID. My amateur efforts usually can’t improve the ID beyond Complex Arisaema triphyllum, but I’m doing what I can. (Indeed, by the key I’m studying, many IDs have been made to species level when those aren’t justified.) When I can’t advance the ID, I’m at least adding annotation which should help the next IDers.
And based on my rate of progress, this project should distract me from my household tasks for a couple months, at least! 
8 Likes