Within the Filter window I’d like to suggest filters for:
obscured location - any/yes/no
location accuracy - < Xm
observation licensing
It was pointed out to me I can set these via the URL, which is great, but exposing these via the user interface would be a huge benefit to making this functionality more discoverable and easy to use.
My specific use case is to apply these filters in the Identify page so that I can focus my efforts on identifications have accurate unobscured locations and licensed under an open license.
I would like Accuracy as a filter, to find those I need to reducd to get them collected by Projects. I just spent several hours bulk editing several thousand of my obs to change the Accuracy of those few that I will never find otherwise.
Like @tonyrebelo, I use the full-screen map view all the time. To see individual observations without closing the map:
Click on the pinpoint for the observation.
In the popup window showing the observation, right-click on either the image or “View.”
The observation appears in a new tab.
I do encounter a problem in Chrome: I still have to close the map view to go to the tab for the observation. I don’t think I have to do that in Safari. Even if I do, I can still open a whole bunch of tabs from the map before having to switch out of the full-screen view.
Haven’t read through all the suggestions, so maybe this has already been mentioned…
Whether in the Filters interface, or just in the URL, I would like to be able to specify multiple values as “all are true” (AND operation) instead of just “any one is true” (OR operation).
For example, &place_id=50,53170 currently will show observations located in Nevada or located in the Mojave Desert.
I would like an option like &place_id=50+53170 that would show observations located both in Nevada and in the Mojave Desert – that is, in the area of overlap between the two regions.
I don’t think this was requested yet (or maybe I just used the wrong search term), but if the ID count is going to be fixed, I’d like to be able to search/filter based number of IDs. In my experience, an observation with 9 or 10 IDs is less likely to need correction than one with only 2 or 3.
Edit: maybe instead? or in addition? Can we have the ability to search/filter by score?
But once the page above is loaded, a click on the observation opens it quickly. So, there is really a performance issue specific to this intermediate page, useless if only 1 observation found.
Save user time perhaps which is important but in terms of server resources it may actually be more taxing on the server to have to look up and dynamically add to the page all the specific observation ID numbers versus a standard slightly altered search url as is done now.
My remark is only about the case with only 1 observation found as the result of a search.
For sure, generating a direct link to this observation is less ressource consuming than generating a page that contains a link to the same observation. (Anyway the server has to generate at least a link, and I say that it is enough, the server should not do more, as it does now unfortunately).
Moreover, this useless intermediate page has a performance issue.
It is impossible that the current behavior is less ressource consuming for the server.
Because what I suggest is stricly less than what is presently done.
The current search wizard allows to sort the obtained results after date added, date observed or favorites.
Would it be possible to sort according to automatic IDs ? I am pretty sure a covariance matrix (or some set of parameters) is stored with each observation, isn’t it?
Alternatively, it would be cool to search (instead of headline taxon) for the same taxon, but in the automatic IDs of a photo. That is, I search for “Noctua pronuba” and get all photos, where the automatic ID has a high probability for N.pronuba, regardless of what the headline says.
Such a thing would be very helpful for us identifyers, because many nice observations are “unknown”, at a very generic taxon or wrongly identified. Sometimes, I would like to search everything which looks similar to a certain genus or species. As it is now, I can only do this for the correctly identified observations…
Copied from original post since it seems to belong here, “There is a filter for “introduced to (geographical region)” but not native species. For example, say the state of California has been selected, and you have searched for “grasses” as a taxon. Since there are a lot of introduced species as well as native species, it would be nice to be able to search for exclusively native species, as well as introduced species. Not sure how plausible this feature would be but I believe this could be a useful feature for the site.”
What about feature we have on id tab, when you click on number of photos of obs, and they are opened inside the box of the first photo instead of opening the opservation?
In this case, given that it’s a request for a filter option, it would be observations of endemic species.
That said, lists would be good too.
Being able to filter observations this was would also help us update iNat info to be more accurate. In my area (limestone island ecosystem in SE Asia), there are a lot of endemic species, but many of them are not marked as such. It would help us fill in gaps in both labeling and observations, and having averified list of observed endemics also helps us push the National Park and provincial politicians we work with for stronger and more effective protective and conservation efforts.
I’ve noticed the omission of endemic in the project filter options as well… We have an NZ Endemic Spiders project that they have to build by manually adding taxa to the criteria for the project, and when I am creating a project I note that I can choose native or introduced as a criteria, and wouldn’t it be cool to be able to do that with endemic as well!
From the New Projects page, having chosen collection:
To make matters worse, Bulk editing the accuracy doesn’t always work, for some reason, despite trying various methods including entering 0 and applying,so I sometimes have to do them individually.
