It’s an odd spot to be at as in one side there is military land, the other is private property and a busy road where you can’t really park. So very few obs there from locals. As a local I’m wondering how he accessed this area. Possible but odd.
”Yes, for ease of submission, I take a screen shot of photo I’ve downloaded to my laptop. Not sure how these affects the submission”
What’s the best way to deal with these kind of submissions?
noting I’ve removed the link to the record, please don’t post links to individual observations like this in forum posts where a user’s actions are being called out or questioned
So the original photo is an observation of sorts, though perhaps not an iNat observation. The screenshot, then, is an observation of an observation, rather than its being a direct observation of actual current or former life. Shall we term it a metaobservation?
If I recall correctly, the posted screenshots were not even accompanied by information about their sources. It seems a good policy to discourage the posting of such material on iNat.
Sometimes users will take a photo of a screen instead of uploading the original file because this fits their workflow:
photo of the back of the camera taken in the field to capture location information (ideally with the original being uploaded later, but users don’t always seem to remember to do this)
photo of their computer screen or a screenshot from their cell phone because they are only familiar with the app and don’t know any other way to get photos onto iNat
In both cases, I encourage the users to upload the original photos. Usually the photo-of-a-screen method results in a loss of image quality which can substantially affect the feasibility of IDing many taxa. In the second case, it also often results in wrong metadata (location and time) associated with the observations because the users often don’t know they need to edit it to reflect when/where they saw the organism, or they forget to do so.
While these are legitimate (if not ideal) reasons for uploading a photo of a screen, other times it may be a sign that there are problems with the observation – e.g., that the photo was not taken by the observer.
So unless the observer provides additional context about the source of the image (and sometimes even if they do), it is often a good idea to treat such observations with caution.
I marked them casual just in case. Didn’t want them on ALA.
The strange thing was that I have notifications for my postcode and I saw a camera screen pic pop up on that. And a casual glance through their observations show a few back of camera shots. But this is a person who is a frequent accurate herp IDer as well as observer and given that we have just had a cyclone hit this city and there are still many places without power this might be the only way to upload it at the moment if they normally hook their cameras to their computer. There are some that are taken in remote areas so this might be a way to get an accurate location especially for species that are only just establishing themselves in the area.
So I don’t have issues about that person’s observations as the location is very specific whereas the original observer gave some very imprecise locations which seem much more random. Two extremes of the same thing.
It could be that it was necessary for the observer to record the observation in an unusual manner. When that occurs, it seems best for the observer to clearly describe the circumstances and method of recording of the observation in the notes. Without that, it leaves the reader (us, for example in this case) confused.
Please don’t do this - observations should not be downvoted in the DQA solely because another user considers that they aren’t wanted on ALA or GBIF, etc. Doing this could be considered DQA abuse depending on the scenario. Observations should only have the relevant DQA fields downvoted because they meet the criteria for those fields (ie, there is clear evidence or strong reason to suspect that they are not correct). Just being a picture from a laptop would not rise to that level. As @graysquirrel noted, some users do regularly take pics of their own pics (or screenshot them) on other devices. It’s not ideal, but not forbidden. A screenshot or pics of a laptop might be grounds for reasoning that a user is not using their own photos, but it is not inherently disallowed. The best approach would be to leave a polite comment for the user or message them asking about whether the photos are theirs/their workflow.
If the user actually responded
then there’s not a problem with that and I would say there is not strong evidence that the observations should be addressed with a DQA.
Still curious and confused about the workflow with which these observations were submitted, I managed to find them again in order to take another look. Out of consideration for the observer, no link to the observation described below will be provided here.
There is a submission that is about a day old as of this writing that includes a screenshot of a computer screen. To the left of the frame of the computer screen are a battery and a tissue box that are evidently sitting on a table. What confuses me it why it would be easier to take a photo of the computer screen and to submit that rather than to submit the image directly from the computer. Would someone be able to explain why the original image, already of the computer, would be needed to be uploaded indirectly?
I don’t know that it really matters, as staff have noted that uploading screenshots/photos of screens isn’t inherently violating iNat’s guidelines. But I could think of lots of reasons, some of which are referenced in the previous threads linked above:
The observer doesn’t know about the website or how to use it, just the app.
The observer learned about iNat using the app and just prefers to continue using that.
The observer doesn’t have internet on their computer but does have it on their phone.
Since we’re asked to assume that users mean well, if the user responds that they are doing this for some reason they feel is valid (which seems to be the case based on the quote above), that seems fair enough.
I encounter a surprising number of people during iNat workshops that I run who have no clue the website exists, and thought iNat was a mobile app only. IMO a large contributing factor to this perception is that online media articles about iNat invariably only refer to it as an app without mentioning the desktop version
Ah, yes, my perspective centers around the website, with my having learned of it long before my knowing about the app. That led to my thinking of the workflow as either 1) getting the photos onto a computer and uploading directly from there, or 2) uploading directly from a phone via the app when there is a good signal in the field.
Now I see that others may have a different perspective with good reason.
I don’t mind photos of screens, as long as the location and date/time are accurate to when the organism was observed. (Obviously uploading the original photo is better, but as long as the resulting photo is IDable, I don’t think it matters too much.) When I see a photo of a screen, I check the location (is it in a house/building even though the photo looks to be from outside) and the observation time vs. the created at time. I might also go to the photo info page and see if there’s a timestamp there. If the photo timestamp/observation time is too close to the created at time, or if I feel the location is inaccurate, I leave a comment asking the observer to verify the location and date/time. I’ll also leave a similar comment for specimen photos where the location is inside a building or when the labels are in frame and disagree with the observation date/time/location.
So a photo of a picture on another device is an evidence of observation? There’s a difference between benefit of the doubt and naivety. Aren’t we crossing it?
I’ve recently seen another thread here in the forum complaining about people in academia snubbing iNaturalist data, but also seeing this thread…it’s seems contradictory.
Birders have a culture of often sharing “BOC” (back of camera) shots instead of the real deal, and then sometimes when they start photographing other taxa they’ll continue this behavior. (I never knew about this being an accepted norm until I got into birding myself.) But I agree with the concerns. I’d say ~40% of the computer screen shots I see in insect iNat have obviously wrong metadata (date and locality are not realistic) and 50% are iffy (i.e., they could be plausible date/locality but probably not given the pattern of behavior of the user uploading a bunch of observations en masse). Is it really worth keeping this loophole for the 10% who actually do it right…? Always assuming people mean good does mean a lot of these iffy records are by default accepted, since we’re waiting for the observer to come back and give a yea or nay on whether the metadata are correct, and many observers never respond to comments / return to iNat. Since the metadata in the iffy ones are sorta plausible I guess it will all wash out in the end with enough data, but it still rankles me sometimes, and it might take years for taxa that don’t have as much coverage as birds.
If you have reason to believe that the metadata is likely incorrect in a particular case, you can use the DQA to mark it as “date inaccurate” and/or “location inaccurate”, regardless of whether or not the user responds. The user is free to later countervote “yes” on the DQA items if it turns out that they did upload the observations with correct metadata (or if they later edited it to the correct information). Using the DQA does not preclude writing a comment; I think it is still considerate to do so, to let the user know that there is a problem (they won’t get notified that their observation has become casual), what they need to do to fix it, and encourage them to use a different workflow. In at least a small portion of cases it seems that observers do come back and edit, though they don’t always respond that they have done so. Assuming that people mean well does not require assuming that they know how to use iNat correctly. Most times there does not seem to be a deliberate attempt to falsify information, merely a lack of understanding about how to upload files from one’s computer and/or how iNat assigns place and date information to observations or that this matters to people using the data.
But just because an observer took a photo of a screen should not be a reason to automatically assume that the data is incorrect – it is a reason to check for plausibility and decide whether it in fact seems iffy. (The problem of needing to assess the metadata comes up with photos of pinned specimens as well; it is not exclusive to photos of screens.)
If it is clearly a computer screen rather than a back of camera field photo, I look at things like whether the “observed” timestamp is within a few minutes of the “upload” timestamp, as this is usually a sign that the observer did not manually change the metadata. I may also check other observations posted on the same day for additional clues about whether the observer was diligent about entering their information correctly.
If they upload multiple observations with photos of screens that are wildly different in style, it may be a sign that the observations need to be investigated for possible copyright violation. Unfortunately the reverse image search engines seem to have gotten worse at finding matches for this sort of image theft.
As shared, there may be a number of reasons why people would take a pic of their(presumably) computer screen, and of course the info should be of the original observation, not the date/location the photo of the computer screen was taken. I may have missed it, but another reason this may be done is that for some people it is the only way to submit observations from the past, or present, if a person uses an actual camera. There is a big difference between phone service and internet service. I live in a very rural area, where many can’t get internet service as there is no affordable option, if any. They may however have cell phones. Their computers are basically word processors for typing/printing, and storing pictures (old or new) from their camera. As always, assuming good intentions, while being willing to ask if in doubt (not being too naive) should be the way to go, I think.
Back-of-camera shots are honestly more trustworthy in terms of location data than later uploads from a computer. If I snap a photo of the back of my camera and post it (which I’ve done plenty of times), it means I want to ensure I’ve got the location right on the observation. I don’t have one of these fancy cameras with a built in GPS that tags the location of every photo it takes. But I do have a phone, and the phone has a GPS, so if I upload a back-of-camera shot in the field, I know I’ve got the location right. Later (if I remember to) I can replace that image with the actual photo from the camera, but if the back-of-camera shot was perfectly identifiable, then it’s no big deal when or if I get around to posting the higher quality image.
I honestly have never understood the resistance to “image of another image” as a form of upload. The most common complaints I hear are:
“How do we know it’s their photo?!?” How do we know any photo on inat is “their photo”? If you want to cheat and post an image that’s not yours, you can easily screenshot it, copy it, etc.; most of the things I’ve seen that qualify as copyright infringement are photos someone found on the internet and uploaded as their own- no photograph-of-a-photograph required.
“If they took the photo of their laptop at home, the location of the observation will be off!” True, and the same applies for directly-uploaded photos with no GPS tag, photos of specimens that were collected and photographed later, etc. Forgetting to change the location to the spot where the organism was found is a mistake everyone makes occasionally, and it has nothing to do with the “photo of a photo” practice.
“Why don’t they just upload the original photo??” I don’t know, ask them. I’ve heard loads of different reasons given, some of which I never would have guessed until I asked. It’s not against policy, and everyone has their reasons for doing what they do.
The way I see it, a standard observation involves the light bouncing off an organism being bent by a lens, passing through mosaic filters and being detected by some photosites that transform the light intensity into electrical signals that get copied onto an SD card in the form of a string of 1s and 0s that later get copied again onto a hard drive, copied again to some server which runs the 1s and 0s through a compression algorithm that spits out a lower-resolution version of the image and hosts it for all to see. That process we call posting the “original” image- but if I take a photo of my photo with a second camera, that becomes “not the original” and is subjected to extra scrutiny? There have been whole publications dedicated to philosophizing over what “original” even means in the digital age. But without digressing too much, my back-of-camera upload and your SD card image upload are just different versions of indirect evidence of an organism; the only way to have the “original” is to collect the organism itself, which for obvious reasons is often not feasible, desirable, or legal to do. The debates about whether photo-of-photo evidence is acceptable ultimately comes down to an arbitrary fight over how many steps removed from the organism we can get before the evidence doesn’t “count” any more. I say if it’s identifiable, it’s fine. You want to photograph a lion in your rearview mirror on an analog camera, scan the film negative to a computer, photograph the image on the computer with your phone, screenshot that image, and post it? Go right ahead; it’s just as much proof that you saw a lion as any other lion photo would be.