I’ll offer my perspective, as someone trying to get through all the birds in California that still need ID.
I don’t ID unknowns, because I lack the expertise to ID most of them to species. If your goal is to connect your observation with someone that can ID your observation as accurately as possibly, than not adding a general ID is contrary to that goal. It adds extra steps before someone with expertise sees it, and by then it gets shuffled into the middle of a queue that’s over 50,000 observations long (just for birds of California that I have yet to review). Often that means months or even years before someone with the ability to ID the observation to species sees it.
Adding a general ID isn’t a guarantee that it will be IDed to species, but it definitely helps.
I understand your opinion about preferring mis-ID’ing an observation in a very specia group (Chironomidae) over just marking them as insect etc. But this is special in two ways: Chironomidae is a small group of dipterans that most people don’t even know exist (I just had to look them up, I had heard their german name before but I don’t know more), and it is your workflow-based preference.
I sometimes wlakt thru unknonws, but I rarely get more detailed than e.g. “diptera” (there are a few groups for which I’m willing to spend time on learning, but that is currently limited to spiders and brassicaceae). When I see a bird that is neither a duck nor an owl, I mostly just say “Aves” and march on to the next organism. I think bird IDers would be rather unhappy when many people like me would start wrongly dumping thousands of birds obs into e.g. Corvidae just because they look black (which is the same thing that would happen if I were trying to recognize Chironomidae).
I’ve learned here that marking weeds and trees as dikots or even plats would be useless and even harming the workflow of some IDers, and I don’t know the workflow of others (so I leave them alone unless I know the family or better). I still mark unknown monokots and pinopsida until I’m told that is unwanted as well.
I use a similar limit, but I stop at around a month, not just two days (there are a few people in Taiwan who upload thousands of obs and ID them a long time later if at all). Sorting the page by ascending date makes this even easier (and I can just store the URL as a bookmark without having to change it - just pressing Ctrl-r removes the reviewed (most) and ID’d (the few I know) and the next bunch rolls in).
Green stuff to plants is OK ? Then ‘worms’ to animals is also OK.
Remember a wrong ID needs more than 2/3 to convince the CID algorithm. If someone agrees with that wrong ID, you will need FIVE identifiers to agree - and we often do not have ‘five identifiers for that taxon on iNat’
Personally my preference would be that the CV can accurately suggest at least Chironomidae for most Chironomids to begin with. And It would be nice if any coarse fly IDers can learn to ID the family Chironomidae. I know it may sound difficult, but separating them from mosquitos, black flies, phantom midges, etc is not super complicated. Even if a mistake is made, we all make them, I’m happy to explain and correct any.
I think others have probably mentioned this, but I have occasionally had to resort to “Life” as a tag. I’m a biologist and while I try to encourage my students to try and tag the observation to the most specific category possible - if they know it’s an animal or a plant, they should tag it as such. The categories where even “experts” in some areas of biology are still at a loss (at least for me being focused on animals) include the whole slime mold/fungi question and there are certainly others where unless you’re an expert in that area of biology, Life is as good as you can get. I had one just this week that I’d tagged as a slime mold that turned out to be a sponge on the beach.
It didn’t happen to me yet. You have hundreds of thousands of IDs. I’m just at around 17k and never got blamed for saying “spider” or bird", maybe it will happen in the future.
Maybe it is ok to ask fly IDers to recognize the Zuckmücken. Most other people probably never will be able to do that. I’m happy if I get something from unknown or arthropods indo diptera, I hope somebody else will insert the next layers so the animal will become visible to you.
Yes it is very much an over simplification. But I am willing to bet minus algae, seaweeds, and some other niche things. Most people on the planet can tell what a plant it. Certainly not scientifically, but just generally. For almost all people we are surrounded by, or have plants some where near us in our lives. From the densest cities, to deserts, and even some tundras plants are everywhere. Plants are generally quite important to many peoples lives.
The worms to animals thing is more complicated. I don’t expect most people to know something like that. I certainly don’t know the actual reason worms are not in Animalia.
Yes disagreeing IDs can definitely be a problem. But most people are courteous and withdraw their initial ID after a different ID is made, especially when its more focused.
You can prevent this problem by changing “sort by” to Ascending. If you mark all the ones you don’t know as “reviewed” then they won’t show up in Identify for you again. When you do both of those things, the oldest one you have not yet looked at will be first.
I think by “worms to animals” Diana wanted to say that it isn’t useful to ID a lengthy animal as worm, not the other way around. Fortunately there is no taxon “worms”, but I’m sure all worm-like things are animals (and I hope observers take pics of their ventral side so I can see if a non-worm is a butterfly or a sawfly).
“Withdrawing wrong IDs is not trivial” Yeah, It can be a problem. There’s probably like 1000 Chironomid observations stuck because of it. Unfortunately there isn’t any single perfect solution to funneling every obs to where IDers can ID them. My ideas certainly aren’t any where near sound. Luckily there’s been plenty of good points and ideas brought up by many people here.
Not much to add that hasn’t been said but just chiming in to say that this African beetle is such a great example and made me happy! : )
And I’ve gotten lucky, I guess, since I ID multiple Unknowns from Los Angeles per day and have for years w/o a complaint. But then again, the vast majority of newbies adding things as UNK probably wouldn’t bother complaining - they’ve probably forgotten all about iNat by the time I see their photos.
I’ve had exactly one “Yes, it is a plant.” It was very startling and caused me to mildly dislike that observer (who has more observations than I do, so perhaps it was a workflow thing, but I don’t think so). That was a few years ago and it hasn’t happened again.
I’m a newbie but did take invertebrate and vertebrate zoology many (many many many… you get the idea) years ago. I did a little refreshing on the iNat taxon pages and feel fairly comfortable doing higher level IDs (Kingdom, Phylum, Class and Order) of unknowns and that is helping me identify genus and species that look familiar. I search for unknowns in my area/state and try to give them a higher level classification hoping that leads to a more specific one. It’s good training for me and gets me involved in a way that (I hope) is helpful.
However…. I do get frustrated with the process of adding an observation sometimes because of the way observations are added. On the IOS app, you add your photo and are then asked for a ID then the date, location etc. If the key pieces of info to use the CV model are the photo and the location, then location should be entered before the identification. If you try to do it before a location is added you sometimes get too many options and it’s overwhelming. If location is added then it narrows down the options considerably and makes an educated guess possible.
Also, if the CV gives me a genus level ID that I’m not sure about it would be great to see other genus options or have the ability to go one level higher. For those who never learned “Kings play chess on fine grain sand” or “Kids prefer cheese over fresh green spinach” this would be really helpful. Most people know at least that their observation is a plant, animal, or fungi and their insects from their mammals, etc. But when given the choice of a very specific genus or species they aren’t familiar with the best option is often to leave the id as unknown. I guess what I’m saying is the id process could be both more interactive and education at the same time.
Are those not in yoru photo’s metadata? The app is designed for importing photos taken with smartphones, which should always have date and time, and usually have location data, so those are normally just filled in automatically.
Honestly it doesn’t bother me at all because I will almost always put a starting id on all my observations which helps the observation fresh new eyes from someone else on the identification page. I also found just helping others out by applying a starting id on other’s posts without a starting id, just to get them started.
I may get into hot water here (you did ask for honest opinions), but I simply don’t understand why people do not put at least a general ID of plant, animal, fungi, etc. I have no expectation of a fine-level ID, but surely an ID to plant vs. animal can be added. It takes little effort and does a lot to speed up the process of getting the observation to a place where it actually can contribute to science. At unknown, it cannot. It also requires an extra level of identifiers to get the observation out of “unknown”. Creation of data on iNaturalist is a collaboration of those who put in the work to find the organism and those who work to curate the raw data.
To be clear, I do on very rare occasions encourage observers not to leave things at unknown and do so politely. As far as I know, I have never taken a chastising tone nor do I approve of those who do. But, I don’t at all consider it problematic to suggest an observer add some kind of ID on their observations, even if the identifier hasn’t quite learned how to apply the friendliest tone. Remember, these users are putting in the work of turning the observations into something that can be used by scientists.
Generally speaking, I’m far more worried about reducing the workload of identifiers and curators which are in the minority than reducing the workload of observers that are in the majority (I am both an observer and an identifier).
I love UNKNOWNS because they have been my pathway to learning. They are now also a very small fraction of the newly submitted IDs in my area. I do add a comment to most of time. I tweaked it over time based on reflecting on about what was triggering the occasional negative responses.
This isn’t really about trying to shape the observer’s behavior; it is trying to help them understand why I’m adding the ID. I don’t worry about the experience* of the user and I stay on top of new unknowns in my area. The vast majority of people don’t seem to mind. I do try to avoid giving the same user too many of the same comments at once, but if they see it a few times, it makes the point that it is copy-paste not specific to their observation.
edit: *experience level/length of time on iNat. I do care about their having a positive experience with iNat.