I like this, however I think there should be a flagging option for possible bullying or privacy violations that hides the photo pending curator review, since there seems to be a backlog of flags and this particular type of rule breaking content can cause harm in a short period of time (just like copyright)
2 Likes
I feel indifferent about this because itâs similar to inappropriate comments now. Curators have the ability to hide those when absolutely needed, but a curator has to take it upon themself to get involved. It can be upsetting if a flag lingers and the comment is something that shouldâve been hidden immediately but wasnât. On the other hand, if anybody could hide anything, like comments or pictures, it can abused. Which Iâve seen happen a lot with copyright infringement flags, where theyâll be used as a retaliation.
I guess my concern is just about not overdoing it and hiding things in a reckless way, which happens already, accidentally or not, and what youâre suggesting might invite an increase in that. Unfortunately. Obviously. In a perfect world Iâd agree with you but I tend to be weary after seeing how some interactions go down on the site, and in the world. Though, I guess it could be argued that itâs better to be cautious and have something not needing to be hidden to be hidden vs something inappropriate that should be hidden not being hidden quickly. But then again, itâs just as unfair for somebodyâs content to be wrongly hidden. I dunno. Nothing will be perfect, I think.
5 Likes
There actually isnât really much of a backlog of observation flags; although there is over a hundred pages of âduplicate observationâ flags, since the Jun 2022 policy change to stop flagging duplicate observations, there are only about 60-70 observations that are still flagged. Most of those are the kind of human observations that this thread is about addressing.
There is an actual huge backlog of taxon flags though.
4 Likes
I think this is a great solution and I appreciate the staff being willing to add in extra tools to allow for case-by-case review on flags.
I donât know if this has been brought up, but would it be possible to have a few curators assigned specifically to handle these sorts of flags? Science curators may not have the knowledge or time to handle stuff like interpersonal disputes.
3 Likes
I donât see how this would create an increase in malicious flags, since it wouldnât be any easier to flag than the existing copyright flag
But then again, itâs just as unfair for somebodyâs content to be wrongly hidden. I dunno. Nothing will be perfect, I think.
I think the impact of leaving bullying or privacy violating content up is far greater than the impact of temporarily hiding a real observation, the damage done by such content is often not reversible once it is seen by certain people (like the victim or other harassing individuals), whereas the negative impact of a wrongly hidden observation is less serious an dis reversible by un-hiding it
That said I donât want to just hide everything and make the site unusable
5 Likes
Creating different curator roles has indeed been discussed before, and the current status (4 years old) is summarized in this post and topic:
https://forum.inaturalist.org/t/remove-all-curators-and-start-from-scratch/140/25
As with many other things, it has since taken a back seat to just keeping up with the scaling issues of the site. Although since that time, things did change to require prospective curators to apply directly to staff, instead of being able to be created by other curators.
3 Likes
I donât think anyone here wants to remove photos of animals, plants, or fungi just because there is a person in the photo. What we are discussing is removing photos where the primary object of the photo is a human, not any animal, plant, or fungus photos with a human in the picture
I am not calling criticism of scientific error or fraud bullying, what I am referring to is primarily teens and preteens posting classmatesâ photos with derogatory captions, thereby publicly shaming them for no reason, this kind of thing may be a joke or serious abuse, but it is not at all similar to adults being called out for demonstrable wrongdoing or incompetence
10 Likes
As misidentifications of humans is really getting a problem, I wonder if it might be technically possible to limit the possibility to ID humans as something else other than as human.
@insectobserver123 put it really well. This really is for cases of school-age bullying, which unfortunately definitely does happen.
4 Likes
There were different proposals, including using AI to blur human faces, and a proposal by me to hide photos of all observations with a community ID of human. What we on staff decided to work on is functionality that allows curators to hide photos and IDs that could be bullying.
13 Likes
No, and in fact as I think about there is at least some danger the opposite could happen if human IDs are automatically hidden; people IDing things in observations that are present but obviously not the focus of the pictures because they donât want it to be hidden. Like a group photo IDâd as the grass they are standing on, or a picture of a stop sign IDâd as âconifersâ because that is what kind of wood the sign post is made of. Then theyâd probably get voted âcaptiveâ, but not hidden. And that kind of thing wouldnât violate site policy per se because there is no rule that says the focus of an observation has to be the most obvious thing in the picture and there are plenty of perfectly valid observations where it is not; see, for example, this project: https://www.inaturalist.org/projects/ignore-the-elephant-seal
2 Likes
To clarify I do remember a small number of users (1-3?) here suggesting that using AI to blur humans automatically may be a good idea, but as I understood them they wanted to selectively blur human faces while leaving the adjacent organism in the same picture visible. I expressed strong opposition to this on the grounds that the AI would often mistake non-human organisms for humans (my cameras face detection will lock onto an antâs face sometimes)
There was also a lot of discussion of proposal auto-hiding photos with a community ID of human, but this should not apply to good faith observations of organisms next to humans, as those should have a community taxon of whatever organism is next to the human
Now there is one situation where I actually do want photos of humans next to organisms removed, and that is when the photo is being used to abuse the human and the organism nearby is a bad faith tactic to prevent photo removal (eg person leaning on oak tree identified as oak, with a caption like âlook at this dorkâ etc)
I honestly donât see a problem with this so long as the group appears to be in on the photo taking, and there is no derogatory language about them in the notes
Since this is really about protecting humans from bullying I donât see the issue with leaving the stop sign up, that could already happen under current policy and marking as captive takes care of it
2 Likes
To be clear about what I was saying earlier, I was NOT suggesting that the copyright flag could be used for this. Instead, I was suggesting a feature similar to the copyright flag.
The way I understand the copyright flag works is this: it hides the image and brings it to the staffâs attention. Staff members can then confirm that it breaks the rules and keep it hidden.
This means that itâs technically possible to implement a feature similar to this one, but for the purpose of observations that involve bullying, breaches of privacy, etcâŚ
Thatâs disappointing (and very weird that itâs impossible to remove hidden photos from the site or at least unclutter the list).
Later in this thread staff said they were working on fixing this
1 Like
I think bc some people might use iNat as a way to track what extinct species they have seen? so I am guessing that allowing ppl to make H. neanderthalensis obs is for the ppl who like taking a pic of a taxidermy passender pigeon or dinosaur fossil and uploading it? thats my only guess, is that if someoneâs at a museum and wants to record bones or whatever.
Related to this, Iâve seen many students who include their photo ID inside of observation photos. The worst case was a set with a complete passport data page. I suspect it was a poorly thought out instruction by their teachers in order to ensure they donât just re-post images found online or taken by a classmate. I tried raising the issue with a flag, was told:
âCurators cannot hide or remove observations, so there is nothing we can do.â
But that same inclusion of personal data or human faces could be used to harass or endanger others. It would be good to have a privacy flag to help mitigate the risk.
3 Likes
It seems like the idea Tony suggested theyâre rolling out would alleviate this. Though, FWIW, I think hiding with a copyright flag wouldâve been totally appropriate in the situation youâre describing. It could be dangerous for that information to be easily accessible and even though itâs not a perfect answer, I think that wouldâve been the appropriate response for a curator to take. I really donât think there was ânothing they could doâ, but everybodyâs got different perspectives, so some curators wonât want to do that. Or some people donât know that thatâs an option at all.
But again, hopefully what they implement takes care of all these concerns.
3 Likes
Iâve definitely removed photos that include stuff like credit card info or passport info. Itâs only happend a few times in my recollection.
5 Likes
iNat could try using a dedicated email account. I have used some myself on rare occasions.
@ThePublicAtLarge tiwane and I have swapped email addresses. I suggest no further comment at this time
Sound good @tiwane?