I had an instance yesterday where an observer expressed dissatisfaction with my addition of a subspecies ID, indicating that they didn’t want their observations “turned into” subspecies when only one subspecies occurs in the location.
For now, I have excluded their remaining observations of the species (a few dozen) when adding subspecies IDs, but I am adding the IDs so I can analyse climate effects on the distribution of the two subspecies and compare their phenology. Depending on how large an area they mean (city? state?), I also disagree about the minor subspecies not occurring there.
Is this request something I should respect? I don’t see how my addition of a subspecies ID creates a problem for them or why they should want this exclusion to apply (only) for their observations.
In my opinion, this is not a reasonable request and not how iNaturalist works. However, I would respect the observer’s wishes, and add no ID.
It’s likely others will add the most accurate possible ID in the fullness of time.
If the infraspecies is not an autonym, then the observer is potentially shooting themselves in the foot, as there is a non-zero chance that it will eventually be promoted to species, and their observation will not be automatically migrated.
Essentially it seems most people will be forced to use subspecies If they want RG as a single subspecies ID will turn observations with multiple species IDs ‘needs ID’.
Although this is getting off topic some. The point I’m making is, I don’t blame anybody aware of the recent DQA changes requesting no subspecies IDs in the mean time. I still don’t understand the changes and do not understand the taxon identification behavior behind a number of observation examples.
I should add that in most cases, I applied ID at a species level in a previous round of identifying, so the observations are already RG. It only occurred to me later that there was more information to be had if I added the subspecies ID. It isn’t apparent to me that I am “turning” this from a species into a subspecies. The observation is still searchable as a species regardless of whether anyone accepts the subspecies ID to make it searchable as a subspecies.
I came across a similar problem with a bald eagle observation, that has 5 obs. but isn’t RG. I just haven’t gotten around to putting in a bug report for it yet.
I’d be kind of torn. On the one hand, adding a subspecies ID is certainly OK to do. On the other hand, it’s polite to accede to the observer’s wish, even if that wish is unreasonable.
As to the recent change in reaching RG with subspecies . . . . I recently ran into an observation with one ID, a subspecies. I added my one ID at the species level. The observation went to CASUAL. Not Needs ID at either the species or subspecies level. (It wasn’t marked by anything in the DQA that would require that.) Is that result stupid, or what???
A solution that might work for me (assuming the observer permits observation fields) would be to add a custom field with the ID, allowing me to access that data subset without changing the observation’s ID status.
All four of those examples show expected behavior since 2023 (or longer ago, maybe since the current Research/Needs ID/Casual system was originally put into place in 2015 or so, in the case of the fourth); they aren’t related to changes from the past couple months.
(Prior to a February 2023 update, if an observation had a higher-level ID and you added an infraspecific ID, the observation taxon wouldn’t change at all. For example, if an observation started with an ID of “birds” and someone added an ID of a particular bird subspecies, the observation taxon would stay at birds. A change was made in 2023 to allow such an infraspecific ID to at least shift the observation taxon to species level [as it has for your first three example observations]. The observation taxon in such cases [before or after the 2023 change] will only update to subspecies once there’s a second subspecies ID [assuming no disagreements], i.e., when the community taxon is at subspecies. As for the fourth observation [when it had one active species ID followed by a subspecies ID, which i assume was the case when you posted your comment—there’s now a second subspecies ID making it RG at ssp level], its community taxon was at species level, and thus it was eligible for Research grade. As with the other examples, given that it had an ID at a rank higher than subspecies followed by a single subspecies ID, the subspecies ID didn’t affect the observation taxon. Recent changes to how infraspecific IDs interact with quality grade specifically concern cases where a subspecies ID is the first ID on an observation [or at least the first ID of multiple of that species on that observation], which isn’t the case for any of these four.)
Is there a reason why we wouldn’t want the addition of an intraspecific ID to change the observation taxon to that intraspecific level? My personal frustration with subspecies IDs comes from my workflow for IDing “easy” species in my group of interest. To take care of the “low hanging fruit” in an ID project, I go through the observations and click “a” for “agree with current observation taxon” for each observation that’s correctly identified. This keyboard shortcut can double the number of observations I get through in an hour because I don’t have to scroll down to the bottom of the IDs to find the one to agree with, or have the “agree” button run away from me as I’m about to click on it when the line saying “iNaturalist iPhone App” loads a second after the rest of the observation. It’s just a few seconds difference per ID, but it really adds up.
But because the “observation taxon” is only at species level when someone added a subspecies, and because now adding a species-level ID after a subspecies-level ID doesn’t result in RG, I find that clicking “a” on taxa with a species-level observation taxon fails to send these observations to RG. Weirdly, if I click “a” a second time after my first ID was added, this does then send them to RG, because the observation taxon changes to subspecies after my first species-level ID was added, so the second click of “a” adds a subspecific ID.
I know it’s yet another “edge case”, and a minimal amount of effort can get around it, but as someone trying to get through 100s or 1000s of IDs in an afternoon, having to pause every time I get to a “cabbage white” or “monarch” because someone ID’d them to subspecies ultimately cuts down noticeably on how many IDs I get through. If I see that someone has been “subspecies-ing” a taxon in an area I’m IDing, I admit I sometimes just start skipping over that taxon. It seems like an odd choice to set up subspecies in a way that agreeing with the observation taxon when it’s at species level fails to RG the observation.
Since this has come out, I have had noticed one significant benefit. I am person who will jump on subspecies whenever opportunity provides, but I also play it smart by doing my research beforehand. And here is what I noticed with this change. It makes finding subspecies easier, and as such, it helps me locate clearly erroneous observations that were set to RG at species level. This can be a great learning opportunity.
There is no official policy I’m aware of to abide by this sort of request. Personally I think it’s unreasonable for someone to request that you don’t give an accurate ID just because they don’t like it (of course if they think it isn’t an accurate subspecies ID, that’s another story).
This is pretty much what the ability to reject Community ID is for, so you could suggest that they do that if they really want the ID to look a certain way.
The way the site shows subspecies can be confusing and I would definitely agree that it could use improvement somehow, but I’m not seeing that it’s broken right now. It looks like it’s working as intended to me.
While I agree that it is a somewhat unreasonable request based on how iNaturalist typically works, I understand why an observer may want this. This conversation comes up a good bit in the birding community, where identification to subspecies is typically limited to field-identifiable forms. As such, range alone is generally not sufficient for determining a subspecies. However, some folks on iNaturalist are dedicated to adding subspecies IDs to birds, even when the available documentation is insufficient to identify to the subspecies level (relying exclusively on range). This can be incredibly frustrating, and probably adds some inaccuracy to the dataset.