I meant coding workaround (though I don’t know why, as I’m not IT-savvy), when adding ids adding a species with note about subspecies will pretty much do the work, then after someone else ids it you can add a ssp id and if it was a species id - mark it “can it be improved - yes”. But just leaving ssp is okay most of the time, there’s no rush in identifications.
While I understand this line of argument, the only reason this is a problem is because the community taxon is (IMO) inappropriately de-emphasized on iNaturalist. It’s misleading to place the RG banner next to the observation taxon, when the “grade” is based on the community taxon. I also believe that when data are exported to GBIF it should be at the community taxon level, not the observation taxon. I realize this change would probably be a major shift in how iNat deals with its data, but I think it would be more honest and more reliable.
And, I note, an analogous situation to what @tiwane describes is still possible when people fiddle with the “Can the Community Taxon still be confirmed or improved?” box in the Data Quality Assessment. The sequence [Genus, Genus species, “No, it’s as good as it can be”] leads to an RG observation at the genus level with an observation taxon at the species level. See the screenshot below, where I temporarily made one of my observations RG in this way. Obviously the subspecies case is likely to come up more frequently (as most people never interact with the DQA), but it’s the same issue in principle.
8 Likes
That’s a good point. It took me months to learn the taxon at the top of the page is not the community taxon.
5 Likes
That top left corner is position position position, prime real estate. Its content needs to match in relevance and importance.
4 Likes
It would be cool if variaties, e.g. domestic pigeon, would be in cv too, instead of just Columba livia. All feral pigeons not only bear signs of domestication, but they’re in fact hybrids that get real wild populations in danger, so diving them means a lot, now cv doesn’t make such suggestion and there’re thousands of “pure” pigeons all over the world.
3 Likes
A lesser-known shortcut is that you don’t have to type all of the words. For example, if I want to add an ID of Eriogonum umbellatum var. dichrocephalum, all I have to type is erio dich to bring up that taxon as the only choice in the list.
If the identifier knows the subspecies, then all they need to do is pick the species from CV, then start typing the first few letters of the subspecies epithet at the end, and it will come up as a list choice.
5 Likes
It’s not working for species with common names.
I just tried with the English common name for that species, bicolor sulphur flower. Typing bic flow brings up the species 2nd in a list of 3. Typing bico sul brings it up first in a list of 3. I haven’t tested to see how well it works with other writing systems like Cyrillic.
1 Like
I mean second part with adding subspecies, don’t know if it works with English though.
Also I’d add that system should recognise when both common name and latin name are entered (both known for it), don’t know why it can recognise them separately, but not together.
Each name of a taxon, common or Latin, is stored and indexed separately, instead of as a hash of all the common and Latin names attached to a taxon. I think doing the latter would probably make the look-up process less efficient, not more, from a technical standpoint, especially for taxa with many names attached.
2 Likes
Got it, thank you!
1 Like
Was this rejected? I read through the recent comments and I’m more confused now.
I think first tiwane answer stands up to this day, it’d cool, but they don’t know how to code it.
Sure, but this one is open?
One simple possibility is to choose the species rank instead of a subspecific rank when there is a taxonomic reference that supports this choice and it appears reasonable in light of the relevent literature as well as consistent with the adopted taxonomy where a given taxon is distributed.
1 Like
They only close requests when they say no future work in this direction will be done, at least in my experience that’s the case, the answer I refered to was done 2 years ago, and topic is still open, so I don’t think it’ll close any soon or ever, as again, this answer meant that it’s how things should work and now they don’t, they need a fix, so let’s hope it’ll happen in a short time!.)
This sounds like a complex software issue. Just to chime in… I have recently noticed that several observations which I only ID’d to Class level and were subsequently ID’d to Subspecies level by another individual are not refined to that higher level. They remain listed at the Class level.
Example:
https://www.inaturalist.org/observations/102392148
It seems like a lot of important opportunities for refinement and accuracy in iNat are being missed by this behavior?
It seems it should work the same as when an observation is further refined to species level as in this case.
https://www.inaturalist.org/observations/102255266
2 Likes
For your second example - the Community taxon is still Dicots.
Needs the over two thirds to tip it. That is how iNat works.
The issue affects the leading ID rather than the community ID.
The main cause seems to be the wish to gamify the experience a bit more by putting a ‘Research Grade’ badge on observations. If I have interpreted things correctly, the complexity in coding this has kept it at bay - the risk of confusing users about what the ‘Research Grade’ badge refers to is prioritised above having the correct leading ID.
4 Likes