Help with curating big influx of low quality observations

[bouteloua]

As Janet mentioned in a different thread, there was a group of about 100 students instructed each to upload I think 200 observations as part of a class project. I think the deadline snuck up on a bunch of them and unfortunately most of the observations are garden plants, trees in an arboretum, pet store/zoo/aquarium animals, and even grocery store fruits and vegetables. There were also a few hundred copyrighted images, most of which have hopefully been flagged as such by now. Also a few strange things like the dates clearly being wrong (spring flowers observed in December)--sometimes from the students uploading old photos but not realizing the date needed to be adjusted... but sometimes because the photos were lifted from elsewhere.

It would be a great help, if you have a few minutes, mostly to tag things as captive/cultivated/not wild when they clearly are.

But also of course helping identify if you can or other curation issues mentioned above. Here are the main links:

https://www.inaturalist.org/observations/identify?projects=14177

https://www.inaturalist.org/observations/identify?place_id=26

-mark as captive/cultivated: keyboard shortcut is x

-date is clearly wrong: leave a comment and mark "Date is accurate?" "No" in Data Quality tab

-location is clearly wrong: leave a comment and mark "Location is accurate?" No" in Data Quality tab

-observation is of inanimate object, like a silk flower: mark "Evidence of organism?" "No" in Data Quality tab

-photo stolen/copyrighted: open the observation page, flag the photo directly by clicking the "i" (white circle) below the photo and clicking "Flag this photo" in the very bottom righthand corner of that page. Then choose "copyright infringement" in the pop-up and save. This replaces it with a big "COPYRIGHTED IMAGE REMOVED" image and makes it casual grade. 

(I've already messaged with the professor of the class pretty extensively so no need to reach out to them.)

thank you!

cassi

[Chris Cheatle]

I'm housebound due to having a foster rescue dog here who cant be left alone, so I will try and do some, at least stuff I can help with.

[Charlie Hohn]

ugh. i wish people would do their research before assigning massive projects not after. 

[Charlie Hohn]

community ID isn't working for this one
https://www.inaturalist.org/observations/9093088

maybe because the user is flagged as a spammer? But if so it shoudn't show up on the ID please page.

[Chris Cheatle]

Just a heads up, not wanting to discourage helpers, but a very high percentage of the observations are plants, so if like me that is not an area of strength, you may not be able to help much. I'm looking at as many birds and insects as I can.

As a note to the admins, perhaps if a user is suspended, their records should not appear in the identity pool, as we cant open them to review.

On Saturday, December 9, 2017 at 11:12:41 AM UTC-5, Charlie Hohn wrote:

[Charlie Hohn]

yeah i am coming up with a lot of suspended users in the ID pool and for some reason the community ID doesn't work so they are going to keep coming up

[Whitney Mattila]

A tip for anyone that finds photos that may be copyrighted: try searching those photos in reverse image searches like Google, Bing, and Tineye. Tineye's more able to find an image that's been modified, but only if the image is old enough.

[Team W700]

Working on it now. Just glad their project is over! 

On Saturday, December 9, 2017 at 10:30:09 AM UTC-5, bouteloua wrote:

[Ian Toal]

Does a pile of dead geese count as "captured"? BTW, I now have an understanding about why hunting images are controversial - first one I've seen. https://www.inaturalist.org/observations/identify?page=6&projects=14177

[Team W700]

Ian,

I can't answer for certain, but it's my guess that because the animal was killed in the identification process, it technically isn't there anymore. I think hunting images like Deer, Geese, Etc... Would be alright. But if someone shoots an animal not in seasons, it makes it difficult to track species progression in areas. It would be like me taking a picture of a Kentucky Coffee Tree, G. Dioicus, and uprooting it but still tagging that I found it. Makes no sense in that circumstance!

If someone shots a goose, and sees 20 more that day, then it should be fine. But if it's the only one you in a day it should probably be excluded.

Not sure if I'm going in the right direction with this?

[Charlie Hohn]

in terms of data quality, it's perfectly fine to add them, it doesn't have to still be alive to count, after all we add roadkill and such. It just needs to be entered in the time and place it was killed not wherever it is taken to. In terms of what is appropriate for the site, I don't think hunting stuff needs to all be banned but people should consider the community and not post things like piles of dead geese or weird gloaty trophy pictures. But that's just my take. There was a discussion about this recently that was rather contentious, but no change in policy was made.

[megatherium]

I think it's fine if, as Charlie said, they're entered at the place and time they were found/killed. Unless they were the very last of that species before it was extirpated it's still a record/indication of species presence for the area.  I don't hunt, but I've put into iNat roadkill, found skulls, fungus and plants before I collected and ate them, introduced pests before/after I trapped them, fish I watched people or animals catch, etc.  

--
You received this message because you are subscribed to the Google Groups "iNaturalist" group.
To unsubscribe from this group and stop receiving emails from it, send an email to inaturalist+unsubscribe@googlegroups.com.
To post to this group, send email to inatu...@googlegroups.com.
Visit this group at https://groups.google.com/group/inaturalist.
For more options, visit https://groups.google.com/d/optout.

[Ian Toal]

I followed a bit of that discussion on hunting. I don't think it should be banned, and don't really have an 'abstract' problem with it, but it was a bit of a shock to see about a dozen dead geese piled up. I do know that Ducks Unlimited has worked hard to preserve wetlands, with the aim of increasing and stabilizing waterfowl populations. It also has the added benefit of increasing wetland environments for other life forms. I did a bit of work with them, and have a lot of respect for what they do.

Ian 

[Charlie Hohn]

yeah, while there's no real way to put it in the rules, i don't really appreciate boastful piles of prey or whatever. I am not anti hunting and in fact am trying to learn to deer hunt (it's hard in Vermont, no deer yet) but i think there's a big difference between posting animals killed when hunting to add to the pool of knowledge vs just posting piles of dead animals as some sort of weird display of supposed hunting prowess.  I almost never see it though so am not too worried. 

[Whitney Mattila]

That gives me an idea, though it's definitely not meant as an answer for any hunting debate. I can move this to its own thread if things are getting too off-track.

Maybe in the future there could be a tag for photos that may be gorey, show visually noticable animal diseases, etc. This way those photos, no matter the context, can be counted and analyzed, but these photos can be grayed out for people who don't opt in.

Nature is messy (regardless of the cause), but this option could give people a way to view the site if they're with kids too young to understand, a mixed audience, etc. Not sure if would instead open a can of worms, though.

[Ian Toal]

That's a good idea, but I really hadn't meant to bring up the issue of hunting. I was just kind of surprised by the image. If you want to start a new thread I'm sure you will get lots of opinions!

Ian

[James Bailey]

Shot/hunted birds (I think without doubt) qualify as a data point, just as pinned insects or botanical specimens do (as long as location of the organism where it was found is provided).