Due to Human Observations in Schools - General - iNaturalist Community Forum and Human Observations with Wild/Captive/Introduced/Native statuses - General - iNaturalist Community Forum; and the fact that humans being marked as “wild”, “captive”, “location inaccurate”, or other DQA votes; I recommend that the DQA should be disabled for observations of humans.
In my opinion, there is no scientific value in marking human observations with DQA votes. All human observations are “Casual” by default, the DQA does not affect that at all. Please share your thoughts.
I think I’d be OK with this if the Community Taxon was a Homo sapiens.
10 Likes
I agree, if the taxon name is Homo sapiens.
2 Likes
Beyond this, I’ve yet to hear a convincing argument why observations of humans are permissible on this website.
9 Likes
Agree! They’re usually non-consensual photos of strangers, or photos of friends posted as a joke. Wasted server space with no scientific value.
5 Likes
you can flag to be hidden - also for the ‘joke’ IDs.
2 Likes
Not all observations with an ID of “Homo sapiens” are photos of humans.
Some are photos of human-made objects that were mistaken by the observer for some non-human organism (chewing gum “lichens”, worn survey whiskers disguising themselves as grasses, plastic beach blobs, artificial flowers, etc.). These are legitimate observations generally made in good faith and I think there are good arguments for leaving them on iNat for reference purposes.
Photos of human artifacts are probably even more likely than photos of humans to get DQA’d to make them casual, since they may start out with an ID of something else. So in the context of this feature request, it may be relevant to consider whether the DQA should be greyed out (as in other situations where specific DQA items are inapplicable to a particular observation) or removed entirely from the observation display as long as the community ID is Homo sapiens. I think the latter would probably be preferable, since I believe that at present previous votes on DQA items that are later greyed out still continue to be displayed, which would mean that observations that were voted “not wild” before the community ID was “human” would continue to appear in Explore filters for non-wild observations
10 Likes
To the extent iNaturalist allows the inclusion of ecological interactions it might be worth including a human entry to show the interaction with another species - pretty sure I have photos on here of mosquitoes feeding on me so it might make sense to also include an entry for Homo sapiens to cover off the other side of the interaction.
5 Likes
Out of curiosity, I checked the Who Eats Whom? project. There are more observations of parasites feeding on Homo sapiens than there are of linked observations of the Homo sapiens in question. For example there are five observations of mosquitoes feeding on humans but none of them have a linked observation, which means they can’t be searched for by host through the project.
I am not sure if there are any projects specifically for parasites feeding on humans; a quick search didn’t bring any up.
Many parasites are very host-specific so being able to search by host like this can be useful in identification. I’ve certainly found it useful for identifying plant parasites and herbivores, which is why even casual observations of cultivated plants have come in handy for me before.
However, there are several observation fields that can be useful for this even without a duplicate entry for the host, such as Host and Parasite Host, which each have a handful of entries for Homo sapiens. I’m not terribly squeamish, but I won’t link directly to the entries for humans because some of those observations definitely have a squick factor.
Overall I think human entries can be useful in cases like this and ones mentioned by @spiphany – saw a rubber snake observation today that I think was uploaded in good faith by the observer, for example! – and I have found the occasional humorous entry of humans to be amusing and not particularly intrusive. I don’t believe disabling DQA for Homo sapiens would affect these use cases negatively for the most part.
4 Likes
After thinking about it more, and discussing it with other staff, I agree with @spiphany that for things like “Evidence of organism” may be useful. I think probably the only DQA section that could really be legitimately used for bullying is probably the “Organism is wild” one. “Location is accurate” or something seems like a non sequitur.
4 Likes
The organism is wild one is the main issue.
Specifically, you can separate observations of “wild” humans and “captive” humans.
1 Like
Humans shouldn’t be marked captive (aka “not wild”) - this has been discussed before. Disabling that choice on observations already rendered casual by having a community ID of Homo sapiens might encourage actually putting an ID on such observations rather than just hitting the “not wild” DQA and leaving it “unknown” for others to sort that out.
4 Likes
Yes, some of the observations IDed as human were made in good faith. Maybe that is reason enough to keep them on the server.
But if you actually look up Homo Sapiens, those are a vanishingly small minority of cases. Most- like around 75%- are just pictures of people. Currently some of the top results are pictures of graffiti, lots of selfies, and human feces(IDed by the observer btw).
We’re reaching into another topic here, and maybe there’s no perfect fix… but if it were my website, I wouldn’t want these on my server.
Given that the ID of an observation can change at any time, I think there are some important logistical problems with the idea that all observations of humans should be deleted (“not kept on the server”).
Apart the need to sort out legitimate observations (which for some users and some uses of iNat may include, say, traces left by human visitors such as litter or footprints), it seems like a mechanism that would mark observations with an ID of “human” as candidates for deletion would a) create a lot of extra work for curators and b) open up a fair amount of room for abuse – it would be easy enough for 3 or 4 users to gang up on someone (as a joke or out of revenge or because they don’t like certain types of observations) and start IDing their observations as “human” in order to get rid of them. There is already abuse of DQA items to make observations casual. If users could cause removal of other people’s observations just by IDing them wrongly, they could cause a lot of hurt and damage before they were stopped.
There is also a problem of trust – as I understand it, the policy until now has largely been that content that violates terms of use may be hidden, but generally not deleted unless the user chooses to do so themselves. Changing that would likely be felt to be a major breach of trust and control over one’s own content and would probably generate concerns that other content might later become eligible for being deleted as well (duplicates, blank observations, poor photos…).
Some people have proposed hiding photos of humans and I think I could support that, at least in cases where the human subject doesn’t appear to have consented to be photographed or where posting someone’s photo is being used to make fun of them. Hiding photos is at least reversible, unlike permanently deleting data. But even this would be rife with challenges (some people dislike seeing people or bits of people like fingers in observations so this function might get abused even for observations of organisms other than humans; if it is restricted to unwilling subjects it may not always be easy to tell whether the subject consented or not) and would again create a lot of work for curators.
3 Likes
We can - for those. Flag for curation, and a curator can hide deliberately offensive pictures or IDs of humans or ‘humans’.
2 Likes
Like I said, I don’t think there is a perfect solution. But I don’t see any harm in thinking through policy improvements.
One strategy that might mitigate the issue:
1- Create a policy that delineates the difference between a person being the primary subject of the observation vs a human being partially in frame.
2- Create a popup that communicates this policy when someone is about to upload an observation with a H.sapiens ID.
3- Refuse to upload images that are pre-IDed as H.sapiens.
Curated deletion would certainly create a lot of work for someone filtering through over 140,000 observations, but you know, we have a lot of nuts on here who might love to help with that kind of thing.
1 Like
To your point about losing trust with people over deleting or hiding observations- I don’t see that happening as much as long as the policy is clearly and thoroughly communicated.
The existence of a policy, however clearly stated, does not mean people will consistently follow it. There are already guidelines on iNat that people choose to deviate from even when they have been asked not to.
Also note that many of the observations of humans on iNat are uploaded by users who are unlikely to care much about the guidelines; these users are also responsible for other types of “garbage” observations that are arguably just as much a waste of server space (observations of the moon/stones/other inanimate natural objects, observations too blurry to ID, dozens of photos of the same subject in separate observations, etc.). Should these observations also be deleted? Where do you draw the line? How do you decide in a fair, consistent, and transparent manner which observations “have value” and which do not?
A major change to the rules of the sort that would be the case if iNat were to suddenly start deleting certain types of observations that were allowed at the time they were uploaded is generally going to greeted with concern and will result in some portion of users becoming upset, no matter how clearly that change is communicated.
1 Like
I understand your concerns. There will always be stinkers, and changing any policy may raise concern among easily disgruntled people.
However, I question the utility of pessimism when talking about these issues. People at large CAN deal with gentle changes.
The stated goal of this website is to help people identify organisms around them while also generating data for science. Any change that steers toward that ideal is probably a good one in my opinion.
If there is a serious aversion to this sort of policy, it is because people have become too comfortable uploading garbage in the first place. I think that’s the real issue here.
The question of what is and isn’t garbage can be hard sometimes. But I think it is very clear cut in the majority of cases. It should be explored, not lamented as if futile.
1 Like
I would ask folks to keep comments here focused on the implementation of the OP’s specific feature request itself, not the general value of human observations or other potential feature requests dealing with human observations that don’t involve the DQA. There are a lot of existing threads on the value (or not) of human observations. Closed ones could be reopened or a new one started (or a new feature request made).
On the topic of the existing request, I wanted to note that I participated in creating a “Not Wild” Human observation the other day. A user flagged an “Unknown” observation, and, as a curator, the correct action was to ID it as Human. Another user had already interacted with this observation, and, without IDing it, had downvoted it for “Wild”, presumably to make it casual. This really shouldn’t be done. As it was, after I added an ID of “Human”, the observation was for “not wild” Human. I wonder if this type of process is responsible for some of the observations with this combination. This isn’t ideal, but also not intended as an insult of course. I also wonder if perhaps, for observations which are Unknown, the “Wild” DQA should be disabled, since it specifically refers to the organism, and the focal organism in an Unknown observation is not defined.
4 Likes