URLs using both ident_user_id and ident_taxon_id

[dan_johnson]

I will frequently use both ident_user_id and ident_taxon_id to search for identifications of a particular species by a particular expert. The problem is that it shows observations that the expert has provided any identification for and that anyone had identified as the requested taxon. Here’s an example: https://www.inaturalist.org/observations/identify?reviewed=any&quality_grade=needs_id%2Cresearch&taxon_id=50190&ident_taxon_id=469268&term_id=1&term_value_id=3&ident_user_id=billreynolds

In this example about a third of the observations listed were not identified as this species by the expert, which is annoying. Is there any way to prevent this behavior? Might this be considered a bug? Is there any reason someone would want it to work this way?

[hanly]

The query is not a bug. It asks for observations that have both been ID’ed by that user and have an ID as that requested taxon – there is no search logic in the query to subset the IDs to those of the chosen IDer.

I’m not sure if this interface is good enough for your purposes, but you could use the identification interface instead to get that user’s IDs for that taxa:
https://www.inaturalist.org/identifications?user_id=billreynolds&current=any&taxon_id=469268%2C50190&order=desc&order_by=created_at

[dan_johnson]

The ident_taxon_id doesn’t seem to work for the URL you give. It lists species that haven’t been identified as the requested taxon.

[hanly]

Did you mean only to include Scissors Grinder Cicada? That would be: https://www.inaturalist.org/identifications?user_id=billreynolds&current=any&taxon_id=469268&order=desc&order_by=created_at

In the above you had taxon_id=50190 as well, which is for the entire superfamily.

[dan_johnson]

The taxon_id is correct. Many of his ids have disagreements and so the community id might be much higher.

[hanly]

The link will include his IDs that are not currently the community ID. E.g., this observation is included in that query: https://www.inaturalist.org/observations/31208411

[dan_johnson]

Again I wonder if there’s any reason why somebody would actually prefer the current behavior as apposed to what I prefer. It not, could it be changed to what I prefer.

[pisum]

i don’t see any official documentation of the ident_ parameters, but i’m guessing they work the way they do so that they can be used independently. i suppose a case could be made that they should work the way you’re describing, but to me, that kind of parameter would be better named identifier_taxon_id as opposed to ident_taxon_id.

besides the page that hanly noted above, the API also provides a way to view identifications (https://api.inaturalist.org/v1/docs/#!/Identifications/get_identifications) with a few more filtering options (though no annotation filters), and i made a page that might help to display the response from the API endpoint in a more human-friendly format (https://jumear.github.io/stirfry/iNatAPIv1_identifications.html).