A month ago, I saw a Narceus in the kitchen and grabbed my camera from the bedroom. The millipede had run under the stove and I couldn’t get it out, so I spent a while trying to draw it. Fortunately, the millipede reappeared the next time I went to the kitchen. I didn’t know what it was, except that it was a millipede of some kind I didn’t know about, and was drawing from the estimated length, thickness, and number of turns when curled.
This level of visibility and the amount of editing needed to bring out the details is pretty normal for underwater photography. A lot of fish would be difficult to identify at only a few metres away in many settings otherwise.
3 Likes
What does it say about me that I assumed it was bad actors 
And yeah, now that you mention it, I follow a lot of wildlife photographers on bluesky who wouldn’t think of posting an image that hadn’t been spruced up a little. The end result is a great image of what was actually there. I haven’t checked, but I hope they’re all on iNat as well. 
So giving people guidance regarding what belongs on iNat and what might be better suited to another site makes perfect sense.
1 Like
Everything here seems fine and well within the bounds of normal photo editing. The swallow is quite a bit more blown out than I would prefer, but you don’t even need editing to really accomplish that, having the camera set to the wrong settings on an extremely sunny day could accomplish that.
I really think you need to take a step back and look at the rules again quote '“that fall outside of general accuracy” is a phrase that is doing a LOT of work and is basically going to cover any mild photo editing that is traditionally done to either make a subject of a photo more visable, or to bring the appearance of it to be truer to life.
No camera is going to take a photo that is 100% true to life, no matter how expensive of a setup you have. Small adjustments to things like brightness/contrast/saturation that fix photo issues aren’t the problem; as Tiwane notes below, things like artistic uses of generative fill are really the issue here.
EDIT: (I love edits)
https://www.inaturalist.org/observations/179742222
Here my observation with probably the MOST editing. I’m not sure where the unedited files even are at this point, but believe me when I tell you that these photos looked like a black fuzzy blob before editing. All edits were done just to make it actually true to human vision and compensate for the limitations of cameras. It’s fine and I bet I would struggle to find anyone that would disagree with me.
5 Likes
This is off-topic here, so please move my request as appropriate. I don’t know how to move it elsewhere myself.
Drag and drop, or select photos and click on the large “Combine” button at the top of the upload page. See https://help.inaturalist.org/en/support/solutions/articles/151000170784-how-to-make-an-observation-with-the-inaturalist-website-video-tutorial
This a helpful video, through which I learned something new (I’ve never really noticed the Combine button, so I’ve only ever combined observations by dragging and dropping). However, its illustration of IDing the lizard begins with that field rather than the location, which remains blank, unlike the location of several other observations that are already geotagged. For the sake of more accurate CV suggestions, could you please update the video to illustrate the recommended sequence of location first.
3 Likes
I’ve always considered that to be giving power to said bad actors. As in the bad actors get to determine what everyone is subjected to.
In the most egregious cases, it punishes everyone but the bad actors – looking at certain public restrooms, closed “because of vandalism”: it punishes people who need to “go,” but not the vandals.
For you, maybe…
Thank you for your explanation. Less about thwarting bad behavior and more about distinguishing expressive art from scientific data. We have had discussions before where, for example, people complained that the taxon photo was beautiful but did not show the distinguishing features. I would expect different kinds of pictures in an art gallery than in a scientific publication.
And add a melanistic morph of the Cougar, with a range map far exceeding that of the regular Cougar. Those would probably outnumber all the Bigfoots, aliens, and Chupacabras put together.
1 Like
That wasn’t my point at all, actually. I’m not suggesting “all pictures lie” (i.e., do not represent reality).
The purpose of these new guidelines, as I understand it, is to prevent falsification of data (adding features that aren’t there, or suggesting a different context such as a sunny sky when it was actually a cloudy day) and secondarily to emphasize the documentary nature of iNat observations – that is, photos should not be portraying an artificially cleaned-up or exaggerated version of reality. The implication of this second part is that manipulation for purely artistic reasons is not suitable for iNat, at least if doing so changes the portrayal of the organism and its natural environment in a way that is misleading or hinders our understanding of them.
While most “artistic” changes are likely to take place during editing, some lenses also produce optical effects that people may use for artistic purposes – for example, a fisheye lens.
It would conceivably be in the spirit of these guidelines if one were to say, iNat isn’t a place for experimental art, please refrain from posting photos taken with weird artsy lenses. But this would also create certain issues: unlike changes introduced during photo editing, if I happen to be using such a lens, I can’t “turn off” this effect, because it happens before the light hits the camera sensor. So if it is not allowed, the photographer needs to make certain choices long before they even take any photos (i.e., which lens to put on their camera body or take with them on their outing).
Now, in general artsy lenses are probably not going to be a significant factor on iNat, because most such lenses are not particularly well suited for the goal of taking identifiable photos of the organism of interest. But in some cases there may be other reasons than desired artistic effects why someone is using a lens that has certain optical characteristics (hence my example of the mirror lens, which has some possible use cases for the naturalist looking for a compact and lightweight telephoto). So it seems relevant to tease out where camera-side optical effects fit into these new guidelines.
1 Like
I see it the same.
Just, if i would let a current generation cellphone make photos of the different wild geese in my area from the 150m distance they allow, i am sceptical if an AI assisted photo would not favor one species against the other while the raw format would be too blurry to actually judge.
As such that’s not a big deal as long there is no automatism ‘en masse’ behind and as long no new species show up which may be art or may be real.
The tricky thing is if people start with observer statements like “i know what i have seen”. That may be right or perhaps not and gnaws on the need of documented evidence as such. Again not a conflict that should be resolved within the individual ID confirmation process.
The rules about photo adjustments would benefit from some rewording.
Adjustments are not allowed but
Standard adjustments are allowed, both listing the same adjustments.
This is too much open to interpretation.
I’m confused actually. Is this the exact wording anywhere?
The blogpot concerning the changes is here - https://www.inaturalist.org/blog/118284 . This entire post reads less like a strict rule and more like a list of guidelines for how to approach the new flag responsibly, and what sort of photos are appropriate on inaturalist. I’ve cut out the relevant part, as I realized the OP here didn’t even quote the entire list of guidelines.
The complaints about the ‘not acceptables’ were listed while completely ignoring the ‘allowed’ section, which frankly seems to adress most of the issues raised in the first place.
@vbjanos you say that saying that ‘adjustments are not allowed but standard adjutments are allowed, both listing the same adjustments’ is too open to interpretation reads very strange to me. I’ve done a lot of photo editing over the years - I don’t edit that many iNat photos, but I know my way around photo editing software - and I feel like anyone that IS familiar with digital (or even physical) photo editing is going to understand both the spirit of the guidelines and that it needs to be necessarily vague.
It isn’t like there is someone out there nitpicking every edited photo and clearly defining the line of ‘too much editing!’ This isn’t possible, nor could it ever be possible, and with something as subjective as photography there is no way to make it a firm guideline.That said, I feel like it is completely reasonable to expect people to understand that tweaking the white balance because the camera sensor decided something yellow was actually white and now the photo is entirely too cool toned is acceptable, whereas cranking the white balance so hard one direction that a bluejay now looks fuscia is definitely not acceptable.
How would you even word this so it is less up to interpretation? I don’t see it as being possible, and the guidelines as they stand seem very clear to me.
Like, this is a collaborative site. If you think one of your photos has been DQAed unjustifiably, you can explain your justification as to why it is appropriate and disagree with the DQA. If you feel like you are being unjustifiably being targeted by users to the point where it feels like harassment, contact the moderators, that’s what they are there for.
6 Likes
Thank you for the link.
Adjustmets to “contrast, color, saturation, shadows and highlights, white balance, lens distortion and vignetting, sharpness” etc. are listed as both acceptable and not acceptable.
I think the not acceptable adjustments should be prefixed with “Too much” or Excessive for clarity.
Otherwise I am happy with the intent and wording of the change.
I also agree with you with the caveat that identifiers do not need to be familiar with editing software. They don’t even need a camera.
2 Likes
For me, the key to the whole question lies here. For example, if you’re working with a camera that takes RAW files, these can be considered as genuine “digital negatives” and require a certain degree of processing to bring out the image as intended by the photographer. As I read it, this is entirely fine. I certainly hope so as otherwise I would have to cancel ALL the photos I’ve posted on iNaturalist since I joined as ALL my photos are taken in RAW then processed before posting.
I absolutely see the potential problem with photos taken with some advanced smartphones (for example) which automatically adjust a number of parameters without the person taking the photographer even being aware.
But it does seem clear to me that the main aim of the new data quality assessment is to avoid major modifications which result in a photograph significantly different from the original scene. This seems absolutely justified but… the problem is, how do we know? A good manipulation is extremely difficult to detect, even on a high resolution image, never mind the low resolution images uploaded here. It requires pixel by pixel examination and even then, it’s not easy. Are we IDers really going to have the necessary experience and, above all, time to detect a well-done image manipulation?
1 Like
The point is not to make sure that identifiers pick up every problematic photo, but to give them a way to deal with any they do notice.
4 Likes
i would argue that the point should be to give observers some feedback about what is acceptable or not, and i’m not certain that things like flags and DQA metrics do that in the best way.
1 Like
I don’t think I said that IDers need to be familiar with editing software. What I DID say is that anyone who is editing their photos enough to be worried about this guideline, is also going to be familiar enough with the purpose of photo editors to understand the spirit of the guideline. (paraphrasing myself here.)
2 Likes