Greetings. Some users post multiple observations of the same encounter. It’s obvious from the photos and also the time stamps on the photos. Unfortunately, it’s like spam in terms of drowning out relevant observations that other users want to look at and contribute identifications to. It doesn’t seem like this practice helps iNat at all. Is there any way to report users for this in the hopes iNat can get the user to “fix” this by removing redundancies?
6 Likes
There’s no special way to deal with duplicates on iNaturalist so the only really simple thing is to ignore them. That’s my preference. Duplicate don’t seem to be all that common overall (though you may have hit a nest of them). Theoretically, they could mislead about abundance, but iNaturalist is really bad at assessing abundance anyway. Duplicates waste identifiers time, but not much time if we’ve figured out how to ID the first copy. Researchers may want to remove them, a waste of their time. Duplicates generally don’t much harm, though.
Also, some “duplicates” aren’t duplicates. Sometimes people will post very similar photos, sometimes clips from a film, as separate observations. These should be combined into one and we can recommend that. Sometimes a class will post observations from a field trip and that will result in 20 similar or identical views of the same tree, all from different students. Sigh.
The way to “fix” duplicates is to leave a comment saying that they are duplicates and ask the observer to remove one of them. To help the observer find the duplicate, it’s best post the URL of the duplicate in a comment, with the request. Doing all that is frustrating. Of course, the observer may or may not fix this but we’ve done our part.
There has been a request for a way to flag duplicates for removal by iNaturalist. I don’t think it’s gone anywhere.
10 Likes
I would like 2 new DQAs. For which I currently use text expanders. I will ID one obs.
Duplicates
See also (include obs nos)
please delete one
Multiple obs but all of the one subject
See also (include obs nos)
Please combine multiple pictures of the same individual of a species
https://inaturalist.freshdesk.com/en/support/solutions/articles/151000170775-how-to-turn-multiple-observations-into-one-observation
7 Likes
As a IDer, if you can ID the duplicate to a point that will drive it out of the needs ID pool eventually (so usually if it can reach RG), please do it. I think that is the best approach to help that the duplicate does not waste IDers time. Leave a comment like suggested above and then just move on.
7 Likes
There is no means for downgrading a duplicate observation to “casual.” It will remain as “Needs ID.” There are a bunch of the “unknowns” that are stuck in this limbo bc the issue was never addressed.
1 Like
I would very much appreciate this
2 Likes
When I find observations by new users that are duplicated, I usually ID both, and comment that they have created 2 observations with the same photo, so one can safely be deleted.
At times when identifying CNC observations from certain places, I have found people who post the same observations over and over again (because it has been all about numbers of observations). I don’t have the patience to list all the copies and ask them to delete the extras. My way of dealing with it has been to comment “Duplicate” and move on, although another prolific identifier has pointed out to me that what we are meant to do is identify the duplicates. So in future I will not add the comment, I will just step away completely.
1 Like
Yeah, this is a major problem. My hope is that the rules changes this year fix this, but this is why I am not a fan of competition when it comes to observations. We need quality, not awards for spotting the most things - especially when in an urban environment where more plants are cultivated. If you look at the data for “unknowns”, there is a big jump around the time of the CNCs (I think) or some other event, and that is a problem.
1 Like
I usually leave a note in the comments with a link to the other observation.
If you have spent time and effort and you know ‘that is a duplicate’ do comment. If others want to ID all the multiples … over to them.
How do you find the URL of the other ob when you’re identifying and the ob you’re identifying is in a dialog box on top of the browser window? I write “duplicate”, but then the observer has to click on “edit observations” to find the other (which I probably have identified).
Open each obs in a new window - to make sure they are duplicates. Same observer (not one ‘borrowing’ a photo). Same place and time. Really the same photo - not 3 people taking slightly different angles of That Same Plant.
2 Likes
I move back and forth between the two observations, in the same window. Obviously, this only works if both were posted very close in time or in some other way that makes me run into them together. And sometimes as I work back and forth between the two, I add the URL of photo #1 to – photo #1. Life is generally better when I ignore duplicates, really.
This is the best approach in terms of time saving. I come across so many observations where if the first identifier to notice it was a duplicate had just ID’d it, it would be RG and out of the Needs ID pool months ago. Posting a duplicate is usually just an honest mistake on the observer’s end. An identifier commenting with a series of URLs and copypasta but not adding an ID, so that a later identifier has to spend more time on it, is a choice. By all means, comment that it’s a duplicate if you want to, but don’t add an ID to only one of the copies; that’s the practice that I get frustrated with.
4 Likes
I would vote for a feature request for this
2 Likes
I feel like pushing an observation to RG gives the observer no incentive to remove duplicates, so if they’re still a relatively new user (based on number of observations) I often prefer not to ID, just add a comment requesting deletion. Yes, that is less than ideal in some ways, but I’d like to think it might help encourage better behaviour? (Wishful thinking? Probably. Ah well.) If someone has thousands of observations I treat it differently, depending on the observer - thankfully I rarely come across it in more established users.
2 Likes
How do you open it in a new window, when you have thirty in a page and clicking on one puts it in a dialog box on top of the page of thirty? When I happen to find duplicates, I’m on the identify page, not looking at a list of one observer’s observations, in which clicking on one brings up the ob in its own page.
Also in Identify - click the word View - and it opens in a new page. I often do that for the ‘more info’ Location Accuracy for example - is it really at the corner of that building? Sigh. I tweak my copypasta for each obs, as needed.
Location Accuracy Not Recorded. Garden plant ?
Why does it matter? Data cannot be used for small nature reserves or for Red Lists.
1 Like
Say the picture was accidentally uploaded 3 times. I notice because my workflow picks up the problem obs. So … all 3 are sitting in Needs ID, but in 3 random taxa - depending on who saw That One first - is A Plant 
a big sturdy one, a tracheophyte, not it’s a dicot 
, ?? a garden rose 
- that is the waste of identifier time and effort. Take it out of Unknown to make a different and bigger problem ? I don’t get the appeal of bashing down Unknown any way we can, so Needs ID explodes into a bigger mess.
With a DQA we could find and sort the obs. And eventually apply a solution.
When there are multiple observations of the same individual, whether duplicates or multiple observations by different users, I appreciate when people leave a comment and a link to the other observation or observations even if they don’t add an ID – it saves time for anyone who comes along subsequently and does want to ID it. This is particularly the case if the other observation has a slightly different image that is helpful for ID, or sometimes all of the observations will have wrong IDs that need to be corrected, so this allows people to address them as a set.
If the duplicate is incorrectly ID’d, I also always add a correct ID so that it does not continue to be listed under the wrong taxon in iNat’s database. If it is correctly ID’d (or has a broad ID because it has been corrected by someone else), I generally do not add an ID – I tend to follow the philosophy mentioned above that users are less likely to notice the comment and less likely to be motivated to delete the duplicate if I have also ID’d it.
1 Like