We might be getting a bit off topic, discussing confirming IDs? Not sure. It might actually still be on topic, because frustration with confirming IDs might build when you do more IDs.
Anyway, this thread is not in the “Related topics” suggestions below, and it was a popular post, so you might want to check it out, in order to better understand the reasoning:
https://forum.inaturalist.org/t/in-defense-of-the-thousand-confirming-identifications/
Good luck with addressing that. There’s probably more wrong on the internet than right.
Honestly no (mainly because I constantly forget to set it up). I guess this comment is a sign I should finally do it!
I wonder if I’m able to add browser extentions on my university computer… Our IT blocks a lot of installs. I don’t do all my IDs from my personal laptop; it’s nice to be able to drop in and ID a few things while waiting for code/analyses to finish running.
I have a few thousand IDs, and the original post here & follow-ups are all sentiments I share.
I was heartened to see the recent identifier survey and I hope that it leads to some Quality of Life (QoL) improvements aimed at recruiting, training, and retaining skilled identifiers, which leads to observer retention and satisfaction when they get more IDs faster.
Reading the above discussion, some QoL features come to mind:
A setting that can throttle batch posts aside from setting the order to random (which only partly works). For example, the ability to say “show me a maximum of 5 observations per user per page.” It’s great that people upload huge amounts of observations at once (e.g. every moth from a sheet in one night, every maple tree in a forest) but this moves everyone else to the back of the queue and feels “unfair” to many identifiers. Right now the only choice if you’re concerned about not getting to the other users’ observations while looking at recent observations is to block the observer and not see anything from them, which is silly since these are often high-quality observations, there are just an overwhelming amount at once. Previous advice has been to recruit identifiers for e.g. bioblitzes beforehand to try to manage tidal waves of observations, but even when implemented, this does not cover most of the situations in which large volumes of observations are simultaneously uploaded.
An optional setting that allows identifiers to sort by quality. I am glad that iNat is more chill than sites like BugGuide, where if a photo isn’t deemed “good enough,” it’s unceremoniously “frassed” (i.e. deleted from the ID queue and the website). However, it seems like there’s a middle ground between “review everything” and “the nuclear frass option.” This could create some issues with moderation, but it would be interesting if there were an identifier review option where you could make a hidden, nonbinary vote about data quality besides just saying if something does/does not contain an organism and if something can/cannot be refined any further in ID. After you’ve spent awhile IDing, you get a very quick sense of when something is potentially identifiable or not, and it would be nice to be able to affix a “tricky” label to an e.g. blurry but maaaybe identifiable pic so that it can be optionally filtered from the identifier pool after some threshold (one or multiple ‘hard to ID’ votes). (All observations would still be available to ID if the filter setting is off.) Having it be a hidden category not viewable to the observer could be nice so as to not create unnecessary hurt feelings and conflict, but this is definitely debatable if it fits in with iNat’s ethos of letting people out-vote DQA things in general, so perhaps it would be visible?
Somehow making the “like” feature more prominent. People should not use identification as a method of “liking” an observation–can we get people to use the star button more? Seems like a UI/training/tutorial challenge.
Making a “watch this observation” feature more prominent (ditto my comments about the “like” feature). This is also because some identifiers use the “like” feature to watch observations they may disagree about the ID on later, but do not want to ID yet (if ever). Therefore, the “best”/most “popular” observations get conflated with the ones that are hard to identify as they’re all starred together. This conflation would be nice to prevent, but it’s not the highest priority QoL feature for identifiers.
Overhauling the broken notification system so that mentions, comments, agreements, and disagreements are placed in different streams or otherwise clearly demarcated and accessible by category, and so that clicking on one notification doesn’t make them disappear forever. Perhaps, having an inbox page that can make these easy to review. A lot of frequent identifiers are periodically active (for example, being offline during field seasons).
Along with the notification overhaul, having a native tool/icon/notification stream solely for disagreements, so that people can quickly review where they disagree/are maverick. Right now there are tricks you can do with URL sorting that make this possible.
7. Again with reviewing observations, I’m sure many of us can agree that when starting out we made IDs that were a bit… hasty…, and upon learning better info, it would be nice if there were an easy way to search our IDs by taxon (and others’, so as to learn from them). Again, I think you can compose a URL like this, but it’s not just a button you can push, which would be preferable.
If there’s multiple individual organisms, there technically can be as many observations from one image from one user on one day as there are individuals in it. If however, there is evidence of only one individual in the image, it might be fair in later ones to link to the first one with a note and check the DQA for “no evidence.” I believe something like that was suggested by someone else elsewhere in the forum. I would be reluctant, though, to do things that way if it turned to Casual the one with more agreement.
https://imgs.xkcd.com/comics/duty_calls.png
Agh the image didn’t show up.
I recently made a feature request that is not-unlike this:
Originally, the two “queues” were called “watch” and “ignore”, but I changed it to “queue1” and “queue2” to make it more generic. Queue1/Queue2 can mean whatever you want them to mean.
So if this is a feature you’d like to see added, you can vote for it.
IDing made me appreciate annotations. Get the scat. tracks.nests and eggs to the people that like that stuff!
Trying to annotate life stage and sex makes you learn so much about taxons.
Maybe the reason for umpteenth confirming ID is that an annotator has passed through while working on the Identify page and giving an ID makes the observation reviewed whereas simply annotating it does not.
I find I am good at IDing my taxons in my area but it can be harder in the next state.
So many undescribed species in my area too. And I know just enough to know that something looks likes Species A but something is off but not enough to know which of Species B-G it might be.
You get to know who the gurus are. You ask silly beginner questions re morphology. You start to leave comments and tag people. Bugs are not normally my field but I have a handful of taxons I can do. But because I am no expert, my comments are in plain English. I hope that helps other noobs.
I am a fan of sorting by geoscore. Get the outliers but it also groups taxons together so you can see local variations.
You can make an image appear by editing the link so that it begins bang, left bracket, right bracket, left banana. I enter an image as a link with the text being a bang (which would produce a link that looks like just a blue bang), then move the bang leftward out of the brackets.
im sorry but wtf is a bang? I can not figure it out, other than maybe referring to an exclamation mark but that doesn’t entirely make sense here
as well as left banana? assuming you mean ( commonly known as left parentheses or round brackets
I also don’t understand the terms being used here, but the syntax for adding an image is:
ok so the normal html code, just didnt get the weird phrasing, thanks
You may not use ‘no evidence’ if it is not true. We do not have a DQA which we would be allowed to use in iNat’s guidelines. We need a DQA for Duplicate = Casual.
Yes we need a DQA for duplactes, but if anyone casualed a duplicate observation of mine like that, I would welcome it and then delete it. It’s all unintended clutter that we’re keeping around on a technicality.
I personally mark one of them as no evidence of organism and leave the other one RG until they delete or merge the observations.
If you come across a duplicate observation, please do not flag it or treat it differently. If you want to identify it, add an identification to the most accurate level you can and include a polite comment explaining to the observer that they should make one observation per encounter with an organism and, if possible, remove extraneous observations. This is usually enough to prevent them from doing it again.
Guidelines yes - but not helpful to identifiers - who have to ID all the extras. Or at least they remain in Needs ID till enough identifiers knock them down. Often by the ‘wrong use’ of a random DQA = Casual.
I linked to the official guidelines because people here were suggesting abusing the DQA to make duplicates casual. If you disagree with the guidelines, take it up with staff, not with me.
I don’t understand the fuss or the stubborn determination I see from many users that they MUST be made casual by any means. Am I annoyed by duplicates? Yes. Are they less than ideal from a data quality standpoint? Yes.
But of all the problems encountered as an IDer, the issue that they pose is really quite minor. I can ID them or not as I choose. If I choose to ID them and there are links to the other observations, it takes all of a few seconds to check the other observations and ID them if needed. True duplicates (identical media with identical metadata) are about the easiest thing for end users of the data to find and remove.
I’m dangerously close to my first 1k! Here’s what changed:
1 – I started by identifying “Unknowns”, but I got bored with it. For the one exciting bug, there were lots of blurry images, objects, pets, and garden plants. There are also far more plants than animals, and I’m still a beginner with plants. Most of the time, I couldn’t get them past Angiosperms. A while ago, a forum user referred to angiosperm IDs as “the green bin”, since they tend to get overlooked. Moreover, in the “Unknown” category there’s a lot of content I can’t even get past Life: microscopic marine organisms, things that look like algae, slime-mold-fungus-like things, tree parasites, etc. “Unknowns” aren’t for me, and that’s okay.. other identifiers are really good at handling those.
2 – I’ve now moved on to insects, especially Diptera, which are my main interest. I’m not confident identifying them to species yet, but I can sort them into broader taxa (Nematocera, Brachycera, Calyptratae, Oestroidea, Muscoidea), which feels really rewarding. I feel like I’m handing off really valuable fly observations to specialists, especially when clear photos have been waiting for years. After a few months, I’m starting to recognize some families, combined with a very thorough review of the literature. I feel like I’m learning, even if I make a lot of mistakes.. which I’m more than happy to correct. I think I’ve found my main focus here \o/
and maybe try Pterygota ? - which is where I dump a lot from Unknowns.