It does seem a serious situation, especially if the misinformation is with malicious intent.
Leaderboards are a double-edged sword to be sure and there is nothing strange about posting 2+ pages a day, especially if there is a rainy afternoon moment with a good cup of java.
And we all make posting ID mistakes, even if it is accidentally picking the wrong suggestion with a clumsy finger. We learn as we go.
I wonder if jbecky might not have touched on something with the idea of feeding the machine learning program the wrong information…you know…just for the pleasure of chaos.
You also need to account for the fact there are some users (hopefully not with this species, but you never know) who believe no ID is valid until they sign off on it, hence they review and confirm every record.
That is right Just today I saw an example of somebody who started “signing off” certain very recognizable and easily IDable fungus. And, as the species is found worldwide, they are clearly aiming for a top in the leaderboard :-)
“Plantae” is useful - I filter for it all the time when adding IDs for my area. I don’t usually add it myself but it makes it easier for me to find stuff I may be able to identify more narrowly.
I’m grateful for all the “plantae” identifications! They feed directly into my identification queue (a local search of plantae). Very useful, not at all worthless.
Why not? Sometimes starting an interpersonal dialog is helpful, and as long as @jbecky were to keep the tone of the PM as polite and respectful as they would in a public comment, it should be fine.
Plus, if this person is that prolific with their IDs, they probably get a lot of notifications and a comment could be overlooked. PM notifications are separate (under the envelope icon instead of the speech bubble icon) and rarer, so they can be a bit more attention grabbing.
I know it’s not that unusual for power identifiers. I’m sure I’ve done more than a page of Nandinas at least once. But that’s Nandina, which is easy. Dandelions are not.
It didn’t help that the first identification I looked at didn’t have any plant that resembled the ID.
I know this guy. He is “a button-clicker”. I think, that new system of ID counts proposed not long ago by @kueda could be a wise solution for solving that sort of issues.
Disabling of the “Agree” button was also a step in this direction, but it was hotly disputed by the community and finally retracted.
I do that a lot with newcomers, who sometimes seem to click Agree as a Thank You. I ID to family or genus so a third party has to take it to species.
But Plants is useful. I try to go to order or below when I can, rather that using just Plants, because I usually know that much. It gets them seen by subject matter experts sooner. But it keeps those Unknown plant observations from taking up the time of non-plant identifiers.
Someone mentioned that they use an app that makes it easy to mark a bunch of things as Plant or Animal, etc. I sometimes wish there were an iNat interface like that (batch high-level ID) to help sort the unknowns. But I’d never want it to go below family.
It didn’t seem malicious for this user. He wasn’t inventing bad IDs, just confirming them. Trying to be helpful? But it does mess up the machine learning and causes those species to be listed as “Seen Nearby” on suggestions.
I periodically go through the IDs of “Green Mistletoe” in North America because that’s found around Australia. It came up in CV suggestions when I was trying to ID some mistletoe in Texas, but I quickly saw it was out of place by looking at the map.
I like having a way to know the difference between whether I agreed with the ID or have no clue.
I Agree with previous IDs if I had to do research to determine if it was correct, because I know everyone makes mistakes. Just the other day, I identified a Unknown 6-legged Harvestman as a True Bug :-) But I withdrew my ID when corrected.
I check Reviewed if I have no clue or get partway through an ID and realize I’m out of my depth. I mark those as reviewed so I won’t see them again in the Identify interface. It doesn’t mean I agree.
I only came across iNaturalist a month ago. Meanwhile, I looked around a bit. I am very surprised that statistics and many leaderboards are given such a prominent weight. Is there a competition being created that makes serious work (photographing and determining) more difficult or even hinders it? Doesn’t this promote mass instead of quality?
I know the European Adelidae well. I am particularly interested in this family of Microlepidoptera. That’s why I’ve been working my way through the unspecified images over the last few weeks. Most of the time I have only been able to determine a few pictures safely, sometimes a little more. For many, I wasn’t 100% sure. That’s why I didn’t make the identification.
What I learn from this discussion: I have to explain much more often what characteristics lead me to a species. And I should never use the “Agree” button again.
There’s no reason not to ever use the agree button. If you feel the identification is right, the end outcome is identical if you hit agree or type in the species name.
There are varying opinions on the need to supplement a record which is already at research grade with another vote if you believe it is correctly identified
OTOH it’s annoying that the leaderboard doesn’t include your identification as the observer. For a number of my taxa I was only able to identify them because I was there and measured them or collected and IDed later, but someone else (often someone I’m 100% sure isn’t familiar with the group) is listed as the top identifier because they came along and agreed.