How do i search for observations that both I and another user have IDed?

By this i mean finding all observations that include IDs contributed by both me and another specific user, NOT observations that we’ve each identified separately

Is there a way to do this?


1 Like

On a related note, is there a thread I can refer to to filter out observations of a certain taxa by identifier?

1 Like

it’s easy to get observations that include IDs by either you or another user. for example, this will give get observations identified by either you or bob:,bob

it’s less easy to get observations that include IDs by both you and another user. however, this set set of observations should be a subset of the observations above. so you could use the either/or set above as a starting point and then parse through those observations, looking for only ones that have IDs from both of you. (this can be automated a bit by using the API, as long as you’re not dealing with a huge set of observations.)

alternatively, you can use the either/or set as a starting point and then:

one last alternative would be to look for identifications by you or another user and then find where identifications by both you and the other user exist for the same observation. (here again, you could use the API to help automate this process.)

here are two pages that will allow you to see identifications by user (specifically, only identifications made for other folks’ observations):

not sure exactly what you mean by this, but this is a good reference for this kind of thing:

"alternatively, you can use the either/or set as a starting point and then:

But in doing so wouldn’t i be filtering out the observations that have both of our IDs? Which means i’m effectively left with observations that only one of us has IDed

What i’m trying to get is [ident_user_id=atronoxychump] + [ident_user_id=bob] -[ident_user_id=atronoxychump,bob]… because this difference is the overlap (intersection) between us i.e. the observations we both gave IDs to, which is what i’m looking for.
(I hope the math makes sense because [ident_user_id=atronoxychump,bob] should not display the ones we’ve both IDed twice and i’ve proven that by calculation)

suppose you have a standard 2-circle Venn diagram. observations you identified are one circle, and identifications bob identified are the other circle.

the intersection/overlap of the 2 circles is what i’m referring to as the both/and set – let’s call this X:

the part of the you circle which does not intersect with the bob circle (aka only you) is what i referred to as you but not bob – let’s call this A:

the part of the bob circle which does not intersect with the you circle (aka only bob) is what i referred to as bob but not you – let’s call this B:

(note: A, B, and X are mutually exclusive. no observation exists in more than one of these areas.)

lastly, what i’m calling the either/or set is A + B + X:

so with that in mind, reread my earlier post, and let me know if you still have questions.


@pisum what i meant is, do we have a search feature that is equivalent to (under the “more filters”) to having an option to search observations by identifier? Hopefully that makes more sense.

the Explore page filter menu does not include a way to filter by identifier, but if you look at the tutorials i referenced in my earlier post, you can see how to filter by identifier using URL parameters. (my posts above also include examples that use such filter parameters.)

that said, if you’re looking for identifications of specific taxa by a specific identifier, then you should search for identifications rather than observations. (you can reference my earlier post for links to pages that can provide identifications.) for example, here are all your identifications of Pale Grass Blue butterflies:

1 Like

Sorry yes, it’s all clear now thanks!

But this means i still have to do a lot of the filtering manually right?

Because I’ve tried combining them like ?ident_user_id=atronox,bob&without_ident_user_id=atronox,bob and, as expected, i get zero results.

I’m not sure if i’m doing something wrong or if there is another way to write it like a formula with parentheses or something, like (A+B+X)-(A+B) but with url parameters

any approach to get the both/and set will involve some step that the iNaturalist platform will not perform for regular users, as far as i know. how manual that step is will depend on how comfortable you are with programming. the observation-set-based approach requires some sort of comparison between 2 sets (finding either the intersection or difference), and the identification based approach requires either an aggregation + aggregated filter or a comparison of 2 sets (intersection).

it’s worth noting that the observation-set-based approach that i described above was some version of X = (A+B+X) - A - B, but you can simplify that as X = (A + X) - A or as X = (B + X) - B. you could also find X by comparing (A + X) and (B + X) and finding the intersection, and this is an alternative i didn’t describe earlier.

it’s possible someone has written a tool to find the both/and set that i’m not aware of, and it’s possible that there may be other ways to find the both/and set in specific situations. that said, without more knowledge about why you’re trying to get this information, it’s not possible to determine whether your use case is a special case that can be handled in a different way.

Thanks for all of that, i really appreciate the help. Unfortunately i’m no expert in programming so i’ll try to find some alternative way to get those observations using filters

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.