Yes, all the photos. We just ran this on a small set for speed.
5 Likes
Great to see this being addressed! I agree it is worth trying, perhaps initially with the more conservative script, wait and see if that generates any problems, and then with the other script. The nice thing is that we now have the taxon history, so any unwanted changes can be identified and reversed if necessary.
Update: based on later information in the thread, I don’t support running this script unless the issue raised by @amzamz can be addressed.
5 Likes
Can you make the - this taxon has had wrong photos removed - searchable, with a project or filter? Then taxon specialists could check for problematic changes. (I am not a taxon specialist of any sort)
2 Likes
Just to be clear, by “conservative” I was referring to the first script - the one that removed 157 photos.
3 Likes
… this worries me a bit, @ tiwane. The reason is that I spend a lot of time to add missing moth species pictures from old copyright-free textbooks from http://www.biodiversitylibrary.org to iNaturalist. They are not on Flickr, not on Wikipedia. That means, I will create an “observation” without date or location on my iNat account for it. There seems to be no other way to get those pictures onto an iNat taxon page. After use, I “downgrade” my own “observations” to a more general term, like “Lepidoptera” because I do not want all these species to get into taxon statistics or my own life list. This work has taken up most of my time in the last 2-3 years.
Since August 2021, I have uploaded about 13,400 such pictures! Each of them for a moth taxon page which didn’t have pictures on its cover yet. My activity is ongoing. It was a lot of work to implement the best actual worldwide species checklist, find the pictures, allocate them to modern taxonomy, crop, edit and upload them to the correct taxon page. At the same time, I also had a look at geographical distribution and synonyms, updating it if necessary. You can find all of the pictures here:
https://www.inaturalist.org/observations?order=asc&place_id=any&q=for%20species%20page%20only&subview=table&user_id=amzamz&verifiable=any
(note that the species numbers are incorrect, because I reset batches to family or higher once every few months only)
If I understand it correctly, your new algorithm would annihilate my work within a second!
Please do not have them removed, because they are helpful, particularly for tropcal America, tropical Asia, tropical Africa and for Central Asia. They will show up in country checklists with those pictures, or when you use the “compare” button within an observation together with “checklist” and a geographical area. Thank you for understanding.
4 Likes
This is a very interesting argument against running the script. An alternative would be to run a script to put observations (those that are used for taxon photos but without matching taxon IDs) in a project, which would then make it easier for interested people to work through them and correct any true errors.
1 Like
Or to add the biodiversity library to the list of acceptable sources of external photos. I get that that list should be as small as possible but the work @amzamz is doing sounds worthwhile and significant. Is there something about the source that disqualifies it @tiwane ?
5 Likes
I’m sorry, but this is what Wikimedia Commons is for - put the images into Wikimedia Commons and then use that as the source for the taxon image (it looks like there’s an existing project for this endeavor).
Observations are for recording your encounters with wild organisms, they are not meant to be vehicles for getting these types of images into iNaturalist and we don’t think that use case should be sanctioned as proper by iNaturalist. So if we do run this script, we would not make exceptions for these images.
6 Likes
I think it would be worth waiting to run the script until this was sorted. Regardless of whether it was the best way to do it, it sounds to me like the loss would be far greater than the gain at the moment.
3 Likes
I’d like to reverse my previous support for running the script, if it would wipe out the work that @amzamz has been doing. Even if this was not the correct way to go about adding these images to the platform, they are important additions, and I’d urge iNaturalist to try and find a solution that respects the contributions from this valuable member of the community. Perhaps there is some way to bulk transfer the images to Wikimedia Commons, for example?
As well as these 13,400 images, Andreas has also added over 400,000 identifications on the platform, presumably often with help from these images, and I have benefited from his expertise in my own observations of South American Lepidoptera. For biodiverse parts of the world, these are the only reference images available for many species.
6 Likes
Sure, and in a perfect world that would mean that the rules are enforced without reference to the costs. But this isn’t a perfect world and rules that were put in place to forestall harmful behaviour should be applied with care when unanticipated circumstances arise and the outcomes may include the destruction of useful creations from constructive members of the community who have run afoul of the rules while working in good faith. Emerson’s observation that " A foolish consistency is the hobgoblin of little minds," is not just a witty turn of phrase.
So before the script is run, let’s figure out how to make good on the work @amzamz has been doing and bring it within bounds of the rules rather than trashing it. I will help, if somebody who understands these things is able to put a plan in place and give me a job to do. I expect others would as well.
As an aside, somebody needs to be asking how it is that this much was able to be done without anybody noticing that the rules were being broken and the work, which is obviously worthwhile, redirected. Seems like a problem that needs addressing.
6 Likes
M, we noticed that these happened, I commented on it months ago, but it’s not my place to reinforce the rules upon those photos.
2 Likes
Thanks! It would also have the benefits that iNaturalist gets the “Google hits” when searching for a (tropical, not commonly featured) species, and not some Wikipedia websites … This may attract professionals, which in turn could potentially become acquainted with iNaturalist and be valuable help for ID of exactly these (tropical, not commonly featured) species. Recent example: Michel Laguerre, PhD, Université de Bordeaux, and IECB - Institut Européen de Chimie et Biologie, France; 270 scientifc publications; specialist for Neotropical Arctiinae (Lepidoptera)
2 Likes
I don’t personally feel that optimizing SEO or even adding taxon photos is a justification for adding pics/observations to iNat that don’t represent actual observations. There are clear ways to add taxon photos to iNat that don’t involve creating observations that are not actually observations.
the ones uploaded from Flickr? Which are not supported by an iNat obs and have ‘dubious’ IDs ?
Then I prefer a ‘fake’ obs with a reliable identifier behind it.
1 Like
