Did it mention color, size, surroundings of where found?
If so, for some things these details might be enough to be differentiated down to a pretty good level, even if the drawing itself was fairly simple.
Photos and recordings are fine, but at a certain point we do have to rely on trust, our ability to assume the dates and places are correct, that questions asked are answered in good faith, that information in Notes is accurate to the best of the Observer’s ability.
Please let us remember that a drawing’s perceived quality can also be linked to ability, difference and challenges, that a camera is not always at hand, and we are here to lift each person’s enthusiasm for nature up rather than trample it in the name of scientific accuracy.
Finally,
Someone please point me to this being the reason for this requirement in the TOS, because I read them extremely closely the day I joined, and I do not recall this. If this is really in there I would need to rethink my presence, as I am not a scientist, merely a nature person.
There are plenty of observations on iNat that consist of really bad, blurry photos lacking clear details, which the observer has IDed to species, and other IDers have agreed, bringing them to RG. No doubt some portion of them have been IDed incorrectly. But there is no rule that only clear, in-focus, detailed photographs are permitted, or allowed to become RG. Why should drawings, good or bad, be any different?
“Research Grade” does not mean indisputable, certifiable, 100% verifiable beyond a shadow of a doubt scientific data. It simply means that 2/3 of the people who looked at this observation believe that it represents this particular organism. They’re called “observations,” not “proofs,” because the primary goal of the platform is encourage people to connect with the natural world, and through it, connect with each other.
Which perhaps explains why more than 1.3 million people interacted with iNaturalist in the past year.
That amphipod field sketch was by me. I don’t mind disclosing my identity and as the owner of the observation that elicited this discussion, seen as many discussions on here would prefer to keep the observers being pointed out as anonymous.
Here is the observation, to put more context to the discussion at hand. Here is the sketch op is referring to:
Recently, I uploaded a handful of field sketches from before I joined iNaturalist. This was one of them. Admittedly, I feel it was the worst sketch of them all, mostly because I really had no idea what I was looking at the time, and just drew what I saw, trying my best to put down everything as I saw it. Coming back to it at a later date, the description and overall shape reminded me of Gammarus, seen as I had seen some a couple months back. I purposely put it as general of an ID as I could, as it definitely could have been another species.
I read a while ago, the general rules of posting notes and sketches of organisms, and everything seemed to line up that this is fine. In fact, I have seen others post similar observations, perhaps with even vaguer details!
Me when IDing, I trust that people posting low quality observations not showing field marks and then describe certain field marks (eg. posting a blurry pic of a hawk and describing it as having stripes on the breast), are doing it in good hope. I was in that mindset when posting this.
Interested to see what the consensus will be here…
It makes me happy to think that during the pandemic, you were across an ocean making field sketches.
When I scrolled down on your link, I could see the others and together they remind me of a book my late father bought me some forty years ago, Dutch Treat: The Artist’s Life, Written and Painted by Himself by Rien Poortvliet. He did wonderful illustrations (which he later watercolored) labeled similarly of species in Holland.
(There is a marvelous bit about the gift of the rotting carcass of a wild boar, which his wife does not appreciate, but from which he can learn much about anatomy.)
While I can understand the sentiment of allowing drawings on iNaturalist when it comes to stuff like that observation specifically, I just don’t think it should ever reach research grade, which is why I ticked no for providing evidence on the data quality assessment. I know for a fact if there were more people drawing organisms than those actually photographing them the websites credibility would be impacted by this because it’s harmful at a data accuracy standpoint. While if a photograph is poor quality and cannot reach a species level ID it can be marked as “no” on the data quality assessment for “Based on the evidence, can the Community Taxon still be confirmed or improved?”, but for drawings you cannot tick this unless you confirm it’s what the observer said it is. In the amphipod drawing example I don’t think there is enough details to agree with an amphipod ID, so it’s like it has failed the Data Quality Assessment, just not in the way you’re allowed to mark it as.
A big issue I have with allowing drawings is if drawings are considered evidence then so should photographs of paragraphs of text describing an organism they saw.
Also if someone was to upload an audio file of them mimicking a sound an animal makes such as bird calls or cricket chirping. It just opens up the door to subjective and inaccurate interpretation, which undermines the reliability of iNaturalist as a whole.
It seems like everyone has their own bar of what is acceptable evidence or not. I think from a data quality standpoint drawings represent a problem as they are not as credible or accurate. If someone was to upload a drawing of a fictional organism for instance it won’t be as easily proven to be a fake observation compared to if it were an observation of a toy or something.
That is essentially a misuse, a downvote based on your own opinion, not on the site’s design.
This is a community venture so we agree to the community terms. Are there things we each would like to tweak? Undoubtedly. But we agreed to abide by these terms.
The proportion of those submitting field drawings is quite low and the CV is here to help us, we are not here to help the CV.
It all very much depends on the quality of the drawing though. From the information and sketching above it seems fairly clear to me that this is a crustacean at least. I’m not much good with crustaceans, so I don’t know if it is possible to confirm it as an Amphipod, but I kind of doubt it would be possible to get it below family, so it wouldn’t get research grade anyway.
I once came across a superb field sketch of Volucella pellucens - so good I could even tell it was female (and that’s a species where the sex can catch people out!). Person said they didn’t have a camera, but they did have a pencil and paper, so they outlined it while the animal was there and remembered the colours to fill in later. Anyway, if it’s identifiable we should identify it to the level to which it is identifiable. If it’s such a good sketch that you can get it to subfamily or better it will become RG, if not then not.
I’ve done that. Crude drawings of fishes seen on a dive (perhaps the diver didn’t have an underwater camera?). One was clearly of a juvenile Pomacanthus, but didn’t show the caudal area accurately enough to determine whether it was a Gray Angelfish or a French Angelfish. I, in fact, commented to that effect when identifying it to the genus level.
Clearly a crustacean, at least. My main difficulty would be in distinguishing whether that is a tail curled under (amphipod) or an oval body shape (ostracod).
I have not found that so. I can add a disagreeing, broader ID and then tick the box. This one, I would say “crustaceans,” or, if I was convinced about that curled tail, “amphipods,” and mark it as good as can be, as I don’t think there is enough evidence to go finer than that.
What is the consensus when a drawing is purposely stylized in a way that the details are either removed or fabricated? For example corporate flat style , stylized like an emoji, abstract, etc… (There’s 1000s of art styles). I think there is a difference between making a drawing of an organism to the best of your ability, and creating an art piece by creating an image in a very purposeful style.
These two heavily styled images above could still be theoretically IDed.
I am asking these difficult questions now becuase I have encountered the first drawing I’ve seen on INaturalist that was done in a purposeful art style (rather than drawing a realistic image to one’s best ability). If theoretical observations like the images I posted above are allowed to be RG. Is there no limits to what one can draw as evidence if it is based off a real encounter?
In that case I will cease marking drawings as a lack of evidence provided. I guess I misunderstood the use of this feature and believed it was a tool to mark observations as casual because they lacked scientific evidence. I understand now that it’s not meant to be used that way.
I would also like to bring up the incorporation of AI images / AI art into iNaturalist. People seem to dismiss the damage drawings can cause, despite the human error element involved, yet when it comes to AI images they seem to take a much stronger stance against it and bring up the errors associated with the AI’s artistic capabilities.
If someone saw an animal, but failed to take a photograph, surely using an AI to generate an image of what they saw would still count as evidence as it’s being used as an interpretation of what they had seen? Given the argument in favour of drawings it seems almost as if AI generated images and art should also be allowed? I saw another post where people worried the AI art being allowed would be bad because the images would pollute the iNaturalist training model, yet I see very little difference between the issues associated with both AI images or human drawn images.
But. You will NEVER. Have more drawings than photos!
AI versus a (human) field sketch problem - is that AI may add or delete field marks regardless of what it started from. We presume the drawing is done in good faith as a ‘record of what I am looking at’.
@zoology123 I would ID your examples as human - art - inspired by nature. But not as the biodiversity as such.
‘Evidence’ just means something others can see or hear to confirm that you did in fact see or hear something. So a photo of a species is evidence - not evidence that you can provide a correct ID, but that you’re not making up the fact that you saw a polar bear in Mexico.
Of course, this assumes a certain amount of good faith in terms of location, and if it’s too way out, things may be marked as incorrect location or not wild - but that’s a different matter.
All that is required for observations to be eligible for research grade status (correct me if I’m wrong…) is that they have some form of media attached ‘showing’ the organism, include correct metadata (date, position), and not be cultivated (well, also a few other minor points you can see in the DQA, but I’d say those are the main things).
Scientific drawings are a far older way of documenting wildlife encounters than photographs which at least goes to show that drawings aren’t inherently untrustworthy. In fact, with photo-manipulation being so easy nowadays, it requires the same amount of trust, IMO, as photos do.
For the reasons already mentioned by others, I think the advantages of allowing drawings outweigh the disadvantages.
All business, including not for profits like iNaturalist are based on trust.
It just means that submissions should be made in good faith and evaluated as such until it is evident that it’s not the case. In a similar manner, errors and omissions should be just honest mistakes and I’ve done my fair share of them.
Once I described the call of a corvid as an argument for my ID and it was rejected because the other species makes the same call, not because I did not include a voice recording.
A sketch is not worse than a low quality photo and I think both could reach RG if accompanied with detailed and relevant description.
Would a photo, of the environment only, count as evidence, at least of the observer visiting the place?
Completely agree. I wasn’t trying to argue against any of that, just trying to clarify what seemed to me to be Jarronevsbaru’s misconception that ‘evidence’ had to be evidence of it being the species claimed. Thanks for the further clarification, though.
Good point.
An empty branch is “no evidence” for a bird observation while a blurry beak or wingtip or toe is evidence, just “not enough evidence” for anything specific.
