Description of need:
I am an avid user of the dynamic life list, which means I am familiar with the default taxon pictures. In my mind, these pictures have to be clear and informative: they should show the organism in a somewhat neutral pose, with its main ID features clearly visible and with a “typical” life stage / color morph / regional variation.
For most taxa, these are perfectly good and well chosen. Now, I notice quite often that users would change a perfectly good default picture with another one that is less useful in that context: most of the time it will be an “artistic” picture that is pretty but not very informative.
Since the “historic” feature has been added, it shows that often the user that puts a given picture as default is the very user that uploaded the picture. That to me is a problem.
For some “popular” taxa, such as birds and mammals, the default pictures are sometimes changed multiple times in the span of a few days. I am not giving specific examples since I don’t want to call out anyone in particular, but I’m sure you’ve already encountered this kind of behaviour.
I have also seen changes that pertain to vandalism, with borderline inappropriate pictures. For example, the default picture for a very common North American mammal has been a gross picture of a dead individual for a few hours yesterday (before I reverted it).
I think this is a challenge to the credibility, stability and usefulness of the website to allow that to happen.
Feature request details:
The possibility for anyone to edit default taxon pictures is useful for the vast majority of the taxa, since it allows for a participative improvement of the website. Also, implementation of the history has been useful to call out users who made these kind of changes (e.g. via direct message), I have done that in the past and and I think it does reduce this kind of behaviour.
Now, most of the inappropriate or annoying changes are done on “popular” taxa with lots of observations and - I assume - of traffic.
I suggest locking up editing the default taxon pictures for non-curators, above a certain threshold of observations for a given taxon. Something like 1000 observations or even more. This would increase stability for these popular taxa, by having well-curated pictures. Good quality pictures are readily available for these taxa, since they have lots of observations, so the default picture does not have to change too often once one has been chosen and agreed upon.
This is a fantastic idea. I would personally put the line well above 1000 observations. Maybe just the top 100-1000 observed species, and update the list every few years.
Some kind of locking seems justified in a few situations, but getting the balance right could be tricky. I sympathize with @donalddavesne’s aim to reduce disruption and vandalism, but I would recommend that iNat takes a minimally restrictive approach so as avoid creating excessive workload for staff or curators.
Right now, photo selection is open to most users for almost all taxa. I get the impression that unhelpful updates are mostly made to taxon photos for common and “charismatic” taxa—mostly well known vertebrate species. And the two most common scenarios are inexperienced/young users selecting their own photos as the taxon photo (“Isn’t my puppy cute?”) and inexperienced/young users intentionally posting disruptive photos (“Dead critter. Made you barf! LOL!”) Does that seem right?
If those are the big issues, it seems a lot of the problem could be addressed either by:
Restricting photo selection to users who have made a certain number of observations or IDs
Limiting who can select taxon photos for a small number of charismatic taxa (not necessarily the most-observed)
Some combination of 1 and 2.
There’s a big risk with limiting some taxon photo changes to staff or curators that these people will lack the time or desire to make beneficial updates. I’m a curator and I do try to review photos for the families and genera where I have an interest, but I’m not sure I would spend much time reviewing photos for other frequently observed “locked” taxa.
And having recently experienced an edit war around selecting taxon photos I’ll also note that curators are not always infallible and well-behaved!
This has been an issue for me for quite a while now… Nothing more frustrating than when you want to compare your observation to a certain species and the default picture is a close-up of the center of its flower shot with a macro lens… How anybody thinks that is a useful image is beyond me, additionally, to make matters worse you’ll find some taxon only have that one image instead of the slideshow of 10 that is provisioned for, and often it is some closeup of a flower that is not helpful at all, I have gone through those taxon and added more pictures to the slideshow which then has pictures removed for some reason within a few days…
It would appear that this functionality exists, as it has been implemented for Homo sapiens. So my suggestion would be to flag the taxon and ask staff to lock it in particularly problematic cases. I would be opposed to doing this automatically for a large number (even just a few hundred) taxa, because that would make it very difficult to correct errors or add better images later. The example of taxa that have been locked for curation has made clear the downsides of doing this - there were ungrafted salamanders for a long time that could not be fixed except by one or two people, and changing taxon photos is likely to be even further down the priority list for staff.
I voted for this as I support the concept, but I think the threshold should be far, far, above 1,000 obs
I also disagree with the suggestion some have made in the comments to restrict curators’ ability to edit photos, as I don’t think a need for this has been shown, and I think if we have to fear disruptive editing from curators we have a problem that will not be fixed by just locking taxon photos, but by staff removing disruptive curators.
There is also some non-specific language in this thread about “locking taxa”, so I would like to clarify that I am only voting for locking taxon photos to non-curators, and not voting for any locking of taxonomy
I completely agree with this articulation of the issue. I do in fact think that a more proactive hand needs to be taken with potential disruptive curators; I’ve watched in the past as others dealt with these kinds of issues – but I doubt that community-disapproved choices of taxon photos for semi-locked default image galleries will be the first sign of trouble with any given curator.
I would like to note – though this could be a little subjective in some cases – that I would prefer to see this implemented only for taxa that have relatively recent and well-accepted support for a stable taxonomic treatment. I am a PhD student working all-around on at least one (or two) specific group(s) of plants that does not have recent support for its taxonomy, and will probably see some upheaval in its classification as a result of my work – changes that will actually affect the choice of taxon photos in some cases. I guess there is not often enough area-specific taxonomic expertise to judge cases through community consensus per se, but it still seems like a consideration to make – that is, trying to assess whether semi-locked taxon photos will indeed need to change for reasons beyond aesthetic representation, such as in changes to how a taxon is defined (as I described).
Usually when I come across this, it’s a taxon photo that has been imported e.g. from Flickr at some point in the past and has not had any updates since that original one was chosen years ago, presumably when iNat was a lot smaller and had no or few photos of the taxon in question. I’ve replaced a few of these.
You can check the taxon history for photo changes and see if maybe a reason was provided or at least who made the changes. I’ve occasionally seen a curator follow up on my taxon photo replacements with more edits, but usually they seem to be targeted at improving the selection of pictures. I haven’t seen anything I would call an “edit war” on any of the plants I look at. Seems plants are not as much of a target for this as some of the more charismatic animals.
@rupertclayton While the cases you mention definitely exist (especially for mammals), the issue can be more diverse. As a curator focusing mostly on ray-finned fishes, I often run into two issues:
high-quality photographs, usually taken by skilled underwater photographers, but that are not ideal for the purpose of the picture gallery (i.e., IDing the taxon) by being very zoomed out, in awkward angles, showing several species etc… This is quite common, and I noticed that sometimes it’s the user themselves that put it there - a poor behaviour in my opinion.
photos of dead fishes, usually taken by sport fishermen. In many cases, they are the best available picture so I have no issue, but you also often see cases where they replace perfectly good pictures of the animal taken in situ by e.g. divers. Maybe it’s just my personal opinion, but as a naturalist website, I think pictures of the animal alive and in its natural environment should be favoured if available.
To go further, maybe locking down is not ideal as some of you said, and the 1000 threshold definitely feels too low (10 000? maybe?).
Another option would be to have community-driven guidelines for appropriate taxon pictures, and make them pop every time someone wants to modify them. The advantage of this would be that they could be taxon-specific, for example:
for plants they would ideally include both flowers/fruits and well-visible leaves;
for insects the larval form should be visible as the second or third picture for easy comparison;
for birds and other sexually dimorphic animals the male and female should be the top two pictures, plus juveniles later on…
This would be less curator-intensive and allow more user agency, but still show a clear guideline.
All in all, I feel this is an issue that would be worth discussing, since default taxon photos are such an important part of the website/app.
I agree, (although I don’t do this to popular species) I will occasionally try to do this for fish species I come across so that at least two of the pictures are in situ. I’d really prefer a fish IDer do it but the species I mess with usually lack a photo history and only have a single picture so I don’t think I was hurting anything.
It depends, at least yes, but as I mentioned it could be worth looking into having at least always of picture of the larva / fruit / female etc… in addition to the “default” one if they are not already visible.
For plants - I often remove the first default picture - which is beautifully focused on the insect. No wonder CV sometimes offers weird choices. If it is a taxon picture for the plant the insect should not be there.
The stable taxonomy point is a good one, but I don’t really see how this can be anticipated (except if a publication is already out and is not yet implemented in the reference databases).
IMO it should be the responsibility of the person that creates/validates the taxonomic change. I did it recently for e.g. the American goshawk. It’s not too hard once the split is done since each obs will be marked as the correct taxon.
It would be useful to lock at least the main image but I’m not sure how we determine the best image. Maybe a taxon expert can be called in to set the images and then iNat lock the lead image. We can leave the remaining image slots open for new/better/additional images that can be added the old fashioned way.
My main struggle with this is people trying to make caterpillars the lead image for moths. i know the caterpillar is very distinctive and charismatic (think Woolly Bear) but I think for species page formatting the adult image should lead and the caterpillar be the second image. I also have formatted Catocala images to lead with an image showing both forewing AND hindwing as this is important for ID. I often run across Catocala that have been changed to a nice image of them on a tree folded up. These are good for the image collection but not the lead image.
Removing problems for some usually means increasing problems for others.
Will an expert in that taxon be the one determining which photos are significant?
If there are no really good photos available now, who will monitor them to update when a better photo becomes available.
We can’t assume that there already exists a photo that accurately shows all of the criteria needed to make an identification, and that the person locking the photos is aware of them. If you don’t like the main photos, just click on “more photos” and search through them.
My philosophy in this matter is “More education, Less legislation”
One option for this ( I think mentioned by someone before? )…would be to have a voting system for the taxon images.
If anyone who wanted to change an image was presented with the chance to add their vote to the different images, rather than simply swap them out, it would be a bit more slow-moving, less chaotic and in most cases the main image would probably rise to the top from a consensus of taxon experts.
I’ve also stepped into a taxon gallery war with someone inadvertently since the history came in.
I just left them to it as they were a new user and it seemed churlish…but they were consistently switching the photos back to suboptimal images …many of which weren’t even RG and looked likely incorrect in the first place.
I think you and I (and most of the iNat community) agree on a lot of the criteria for selecting good taxon photos.
Can I ask a little more about the two scenarios you mention?
high-quality photographs, usually taken by skilled underwater photographers, but that are not ideal for the purpose of the picture gallery
Are you finding it a common problem that these skilled photographers replace existing good photos with their own, less-suitable ones? Or is it more just that the these photographers have selected photos that have these problems and we could choose better ones?
BTW, I don’t think people selecting their own photos as taxon photos is necessarily a problem, so long as they meet all the criteria and don’t needlessly displace existing photos of similar or better value.
photos of dead fishes, usually taken by sport fishermen. … you often see cases where they replace perfectly good pictures of the animal taken in situ
This does seem like a problem. I guess it’s a self-promotional thing a little similar to people uploading photos of their pets. Is it very frequent? Does it seem there could be an easy way to exclude the people responsible (e.g. by requiring a certain minimum number of improving IDs)?
I do think that if iNat were to restrict adding taxon photos above a certain number of observations, that would have to be 10,000 or higher. Even so, that would likely mean locking taxon photos for many thousands of species.
As someone who works on plants, I don’t think I’ve yet seen this problematic behavior with taxon photos for plants, and so I would much prefer a fix that targets just those the areas where the problems occur (mostly vertebrates?)
I would love to have integrated, community-created guidance on selecting taxon photos (and maybe integrated, community-created identification guidance, too). But I think those are topics for other feature requests.