Platform: Browser
Browser, Chrome 152.0.7977.82
URLs (aka web addresses) of any relevant observations or pages: https://www.inaturalist.org/observations?place_id=any&user_id=archville&verifiable=any
Screenshots of what you are seeing:
Description of problem:
Step 1: Go to https://www.inaturalist.org/observations?place_id=any&user_id=archville&verifiable=any
Step 2: Select filters and type Agrotis in Description / Tag section
Step 3: Search is not relevant to entered text.
Hello archville, I’m not sure that’s what that box is for, perhaps it’s more for if you’re trying to isolate caterpillars from adults, for example.
If I wanted to see all the agrostis, I would type it into the main species search box on the explore page? (that you can just see on your photo behind the pop-up filter box)
@archville Nik, you’ve pointed out an interesting quirky feature on the Filters screen. I tried a few different types of searches (of my own observations) using the “Description/Tags” and got some surprising results:
In that field, I typed simply, “green”. The set of observations which were collected in the search results included, (a) the word “green” in the primary common English name of an organism (species, genus, or higher), (b) several observations of taxa for which the word “green” appears among secondary common names listed in the species “Names” lexicon, even when they don’t appear on my display, and (c) observations in which the word “green” appears in my Notes section, even if the word usage is irrelevant to the organism at hand (e.g., “I went to a Green Day concert and I saw this hedgehog in the audience.”)
I attempted to search for One-banded Lichen Moth, so I typed “One-banded” into the Description/Tags field and got an array of organisms with either the word “one” or “banded” in the name, as well as any observation in which I had written the word “one” or “banded” in my Notes section.
So my conclusion is that the “Description/Tags” field is a very inclusive search (looking for any usage of “A” or “B” or “C”, etc.), and it includes all taxon names and any words in the Notes section added to an observation.
Staff or moderators can correct me if I am misstating that function.
but for the scientific name, you would go top right in your screenshot – to the box which says – Species.
Where you need only type
d e i l – and you will get your Genus.
You can also set your iNat to display scientific name first, if you would prefer that.
It’s kind of surprising to me that a filter specifically labeled “Description / Tags” is not confined to the fields for Description (aka Notes) and Tags, but is also looking at the species name.
It also searches for words used in location names!
This does make it rather difficult (sometimes impossible) to search for something that is actually in the description and/or tags.
It is relevant, because this kind of search always includes synonyms and common names in the results. To find an exact taxon, enter it in the “Species” search-box on the observation page, since that will use the taxon-id instead:
Of course, this particular search won’t get any hits, since you haven’t currently made any observations within the genus Agrotis.
The other search methods use a much more general, word-based approach. This looks for any word that’s part of a scientific name, including all of its synonyms, as well as any word found within a common name, in any language.
Thus Xestia c-nigrum will appear in the results when searching for “panewka”, since that’s part of the common name in polish:
This may seem a little unintuitive at first, but the upside is that it greatly increases the chances of finding what you want when you don’t know the full canonical name.
You can use the search_on parameter to refine the query. For example:
This also works with “tags”, “place” and “names”.
I agree this search operation is very inclusive, which can be very useful if you only remember part of a name and are not sure if it comes from a common name (in any language), scientific name, or synonym. When I typed “gaura” in the Description / Tag section, my results from my own observations included:
In the OP’s case, the search term “Agrotis” pulls up Xestia c-nigrum because that moth used to be in genus Agrotis (synonyms include Agrotis degenerata and A. depravata).
Sure, but I guess I just expected a filter named “description” would already correspond to search_on=description. Like how when you set the “place” filter it sets the place_id parameter in the URL.
Yes, the real bug here is the confusing label “Description / Tags”, as well as its slightly odd position in the dialog. It doesn’t look like what it actually does. It’s actually the main search query. By default, it searches everything - where “everything” means not just descriptions and tags, but also places and taxa (including synonyms and common names). All the other options in the dialog are extra filters that refine the main search query (which can be blank).
Personally, I think that the search-box should just be labelled “Search” and have a drop-down next to it containing the items “All, Descriptions, Tags, Places, Taxa” (i.e. corresponding to the search_on parameter). It might also be better placed at the top of the dialog, so that the search is clearly separated from the filters. This would make the layout similar to the one used by this forum, and on many other sites elsewhere.
