Only 526,292 if you only include observations where the CID is Dicots
https://www.inaturalist.org/observations?disagreements=true&lrank=class&taxon_id=47124
Only 5,284 of those in Southeast Texas (where I do most of my IDs)
2 Likes
Before the DQA for “evidence refers to a single subject” was introduced, people were using the DQA “ID cannot be improved” to take observations out of Needs ID if they included photos of different organisms.
3 Likes
11,777 observations without ID are marked as “No, it’s as good as it can be”:
https://www.inaturalist.org/observations?fails_dqa_needs_id=true&identified=false&verifiable=any
(The project for “The Community Taxon is as good as it can be” does not contain them all. Created for technical reasons, it is one of the inputs for the Exclusion list for Phylogenetic Projects. Now that we have an URL filter for this DQA, I could delete this project and update the software bot that populates the Phylogenetic Projects for ‘unknown’ observations).
Surprisingly, adding a first ID does not clear the DQA “No, it’s as good as it can be”:
https://www.inaturalist.org/observations/326581795
3 Likes
In the past couple of days, I realized that if I go through plants at the Kingdom level in my region, sometimes I can get observations down to the genus or even species level. Great! But I can also ID observations to Mosses or Vascular Plants or even the dreaded Dicots, when that’s the best I can do. Then, when another plant identifier comes along, they can agree with me (if that’s appropriate) and maybe mark the DQA “as good as it can be.” Or maybe they leave that for a third identifier, who can then clearly see that at least two active plant identifiers couldn’t improve the ID past, say, Dicot, and that third person will check the DQA “as good as it can be,” finally sending the observation to Casual ten years after it was uploaded. (Ah, the lightning speed of science is so impressive.)
In other words, I’ve realized it’s a good idea to ID things as Dicot if we’re ever to have a hope of clearing some observations out of Needs ID.
ETA: In other news, apparently I’ve made almost 11,000 IDs just in the last month.
12 Likes
I don’t feel comfortable with that. Assuming that 3 identifiers have identified these observations as dicots, would you/they mark it as “No, it’s as good as it can be”? On the contrary, I expect that someday these observations can be identified at rank species:
https://www.inaturalist.org/observations/332760023
https://www.inaturalist.org/observations/331931159
https://www.inaturalist.org/observations?field:Similar%20observation%20set=333028344
https://www.inaturalist.org/observations?field:Similar%20observation%20set=332559666
https://www.inaturalist.org/observations?field:Similar%20observation%20set=332986163
I agree with you that sometimes such observations can indeed be identified to genus or species. But sometimes they can’t be. I simply want to argue that, sometimes, identifiers should have as much confidence in sending an observation to Casual as they do sending others to Research Grade.
One of the things I am acutely aware of is exactly how much I don’t know, even about groups I often identify, like plants in eastern North America. I try very hard to be cautious in my identifying, whether my IDs send an observation to Research Grade or Casual or even just stuck at, say, Family. But I think we all have to recognize that if we, the identifiers, want to reduce the numbers of Needs ID observations (and I recognize that may not be the goal for every identifier), we need to use every tool iNat allows us, including the careful and judicious use of “as good as it can be.”
8 Likes
I was dragged back to Asia by an IDer IDing some of my observations. I wanted to return the favour as a thank you, but unfortunately they are no spider observer, so instead I went to the country the IDer seems to live in and IDed some spiders there.
I also started IDing a new genus in Asia and tried IDing some newly described species there. However, there seem to be few curators active willing to use their time to curate taxon flags and add new taxa, so I am slowed down on that.
So for now I am back in the US and do what I do there.
3 Likes
I noticed that too
Trundling through Plants Kingdom → Order has been what I’ve been working on as well recently. There is still a lot of low hanging fruit in taxon I’m familiar with, but in advance of the NE+NY IDathon I’m trying to bump as much as I can to genus or species in that area. So many very IDable things trapped in that zone, a lot of generalized IDers who put in the work to get the unknowns slightly more known.
6 Likes
Ah, I see the problem. This DQA is related to the Community Taxon. Apparently, it can checked at Unknown (no Community Taxon), but when the first ID is added, there’s still no Community Taxon. Adding a second ID can establish one, at which point the DQA vote is removed.
This may be a legacy issue related to DQA votes that were added before the recent change that linked it to Community Taxon. I haven’t tested this, but it would make sense that now you should not be able to vote on this before a CT is established. [Edit: yes, that’s how it works now.]
But these observations are back in the Needs ID pile now. Is that related to another recent change to how “ID cannot be improved” is handled? Observations at a broader ID level than family is Needs ID instead of Casual?
At least this is easy to fix. And since the earlier identifiers have usually already left a comment about multiple subjects (and the observers are often long gone), I have no hesitation in checking the new DQA, again banishing these observations to the Casual pile, without leaving another comment.
Using another account (for test purpose), I was able to add a 2nd DQA vote (then I removed it):
to this observation that has only one ID:
https://www.inaturalist.org/observations/326581795
DQA votes that predate the recent changes are not automatically removed (unless the community ID changes) and the field remains active even if the DQA would be deactivated in the current system, so I don’t know what you think this test will demonstrate. If you remove the DQA from an observation that does not meet the criteria for this field to be active, the field will be deactivated.
I cannot remove the vote “No” (it’s not mine) but I could add a “Yes, it can be improved” (then I removed it). Then, using another account, I could still add another “No”. So, the votes for this DQA are still active:
I just wanted to check what we said. I think it shows that things are not as simple as we said.
The DQA “No, it’s as good as it can be” is in such a bad condition that, from now on, I don’t take it into account anymore for populating/cleaning my traditional projects. Still possible to filter out observations with the URL parameter &fails_dqa_needs_id=false if one wants.
2 Likes
Yes, because there is still a vote. As long as there is an active vote, the field is not deactivated. So you cannot test anything because you cannot remove the vote. The “erasing” of DQA votes and activation/deactivation of this field is related to changes to the observation (in this case, IDs), not other DQA votes. Adding another vote demonstrates nothing.
I also fail to see why you need a test account to add and remove DQA votes for an observation where you have not previously entered any DQA votes.
1 Like
Not as simple as this, because adding my ID (the 1st ID for this observation) didn’t erase the DQA votes (as I said).
I do have entered a “Yes” vote (as I said), in order to cancel the unexpected “No” vote that I could not remove because it is not mine.
1 Like
On the other hand, if those observations languish in “Unknown” I’ll never see them and neither will most people who ID plants.
Thanks for pointing out:
Sounds like a valuable project to look over!
2 Likes
This project will be deleted in a few weeks. To explore/identify observations marked as “as good as it can be”, consider using these URLs:
https://www.inaturalist.org/observations?fails_dqa_needs_id=true&verifiable=any
https://www.inaturalist.org/observations/identify?fails_dqa_needs_id=true&quality_grade=any
5 Likes
We have already explained this: the DQA is linked to the community taxon. According to the staff blog posts, the resetting of the DQA only happens when the community taxon changes. That observation does not have a community taxon because it only has a single ID.
The DQA’s behavior changes in response to IDs but this does not mean it changes in response to any ID, nor did I claim this.
Adding additional DQA votes to an observation where the DQA is active will not cause the DQA to become deactivated. It may change the status of the observation (casual/needs ID/RG), if the addition of a vote changes the balance of yes and no for that DQA item.
1 Like
(Chuckling here! Of course it will!) Thanks for the information and the work-around.