Reword DQA “Evidence related to a single subject” to "A single taxon is consistent in all evidence"

Platform(s), such as mobile, website, API, other: All

URLs (aka web addresses) of any pages, if relevant: Observation pages with DQA voting, including Identify

Description of need:
The DQA “Evidence related to a single subject” is ambiguous in multiple ways. Because it says “single subject”, there has been confusion on whether it means more that one individual of the same taxon (see here). Likewise, this DQA is only supposed to be used for multiple photo/recording situations. The help file says “Please do not vote “No” to the “Evidence related to a single subject” DQA condition in these scenarios: When there are multiple species in a photo and it’s unclear what the subject is (e.g. a single photo with both a duck and a pigeon in it).” I found out about this exception to “Evidence related to a single subject” here. The current wording and use of “Evidence related to a single subject” do not align. Most people will likely not consult the help file and will tag this DQA when the evidence is not related to a single subject rather than just the subset of this the staff intend.

Feature request details:
Reword DQA “Evidence related to a single subject” to accurately reflect what it is supposed to mean. I propose the following rewording: “A single taxon is consistent in all evidence”. “A single taxon” fixes the ambiguity of subject/individual vs. taxon. “Consistent in all evidence” allows for multiple taxa when there is a single photo/recording but does not allow for different photos/recordings to have different taxa.

I think this wording is much clearer and should cause less confusion. The word “taxon” could arguably be replaced with the word “species” in my suggestion. That would perhaps be less accurate but more beginner-friendly and there would rarely be an observation with one species and multiple subtaxa. Feel free to offer alternative rewordings and note what people may find confusing/misleading in the original or any suggested alternatives.

I personally find using the term “taxon” to be vague/inaccurate for what this DQA means. Taxa can be at any level. For instance, an observation with separate pictures of an ostrich, a tiger, and a skipjack doeshave evidence related to a single taxon: “Vertebrata”. The lowest ranked common ancestor for any set of photos could always be considered a “single taxon”. As mentioned in the request, I’m also not sure whether beginners would be familiar enough with “taxon”, though I don’t think most iNat beginners use DQA extensively, so this might not be a major issue. I’m also not sure how well “taxon” translates to other languages which I think could be considered as well.

I do think that the “on page” info text helps clarify quite well and is easily accessible:
" the observation doesn’t present evidence related to one subject (e.g. it has four photos of unrelated organisms. A photo showing habitat of the observation’s subject is OK.)"

I don’t think that any one term/wording will be perfect - all options will have tradeoffs in understanding because they need to be very short/direct to be on observation pages/the app. Because of that, It might also help for some of the more confusing/contentious DQAs to have links to the detailed explanations of how to use them (single subject, wild, etc.).

One other consideration might be: This DQA and associated wording are quite new - there’s always going to be some confusion when things are new (and even well established as with the captive/wild/cultivated DQA). It might be good to give the situation a few months, see how users work with it, and assess whether the DQA is being used accurately. If the DQA is used accurately in a high proportion of cases, there isn’t a major issue. If the error rate is too high (whatever that is determined to be), making a change might make good sense.

11 Likes

I agree that this wording is very confusing, but I think the wording proposed by the OP is also confusing, so I thing it should say “all photos depict the same species” although even that is not totally clear

7 Likes

As mentioned in the last but one comment there might be useful photos of the habitat without the subject. Or the subject might be the pupa of a parasitoid fly and an additional photo could show the dead katydid from which it emerged. So these photos would be related to the subject although it isn’t shown.

4 Likes

I agree. And if anything, all evidence relating to a single “taxon” or even a single “species” isn’t even what the DQA is meant to address. If I post one observation with 20 close-up photos of 20 different blue jays, all my evidence relates to one single species, but I’ve lumped 20 different subjects into one observation. That is exactly the sort of observation that should be marked as “evidence not relating to a single subject”, because each photo has an entirely different organism in it, but the proposed re-wording would explicitly discourage such a designation, as the photos do consistently show the same taxon.

If my understanding of this DQA is correct, it’s basically saying “all the photos show the same individual organism”, a few people are misunderstanding this to mean “only one organism is shown in the photo” (which doesn’t make sense to me as an interpretation… virtually every nature photo shows multiple species, unless it’s a bird flying against an empty sky or something), and this proposed re-wording seems to alter the meaning completely to “all the photos show the same species anyway…”

I guess I just don’t see the connection between this DQA and an observation where the user was simply unclear about what they observed. Five photos of a bee on a flower show the same organisms in each piece of evidence; all evidence relates to the same subject. DQA applied. Any ambiguity about whether the bee or the flower was the intended subject of the observation has no bearing on the fact that the evidence all shows the same creatures. It seems pretty clear to me that this DQA is intended for observations where the different photos literally show entirely different organisms. Maybe re-word it to “are there multiple unrelated pictures showing literally entirely different organisms in each one?” with a yes/no option, and add a separate DQA field for “are you personally confused about what this is supposed to be a picture of”? lol

3 Likes

How about a simple non-technical “Same species present in all images”?

3 Likes

This is intended to take - 3 pictures - of a beaver, then a daisy, then a mushroom - out of Needs ID to Casual - since we cannot ID the beaver - what about Daisy and mushroom?
We need an unambiguous way to say - do all 3 pictures show the beaver ? Terminology suited to both newbies and ivory tower scientists, and straight forward for translating.

Is the first subject in all photos / sound recordings ?

If you are battling with a photo with no clear subject - do you want to move that from Needs ID to Casual? But then with a new DQA, perhaps for Habitat photo. That is a separate issue which needs its own solution.

  1. Ask the observer - or
  2. Choose your favourite taxon - or
  3. Mark as Reviewed (and move on) - or
  4. New DQA for habitat ? Which someone else can tackle.
1 Like

I think “single subject” in the current wording is a cause of the lot of the current confusion and uncertainty. Even if current users learn what the DQA is intended to mean over time, the phrasing should still be something reasonably intuitive and not something likely to lead new users to wrong conclusions.

Photos with multiple organisms and no single obvious subject are a challenge for IDers and I know that there are people who feel that such observations should not or cannot be ID’d. So a DQA that sounds like it applies to such situations but is intended for something else is a recipe for misinterpretation.

However, I don’t find the formulation proposed in the feature request to be the right solution either – “A single taxon is consistent” in all evidence" is not grammatical for me.

It was decided in one of the the threads on the new feature that this DQA should not be used for different individuals of the same species. While it might be (reasonably) feasible for large vertebrates, if one is photographing fungi, or plants, or groups of small, fast-moving insects like an ant colony, it is often difficult to be sure that the same individual is in fact present in all photos.

See the FAQ on the help page: https://help.inaturalist.org/en/support/solutions/articles/151000171680-what-do-i-do-if-the-observation-has-multiple-photos-depicting-different-species-

I like the simplicity, but I see two issues with this: observations can also include audio files (even if in practice the issue of multiple audio files of completely different organisms doesn’t seem to be as common as multiple photos of completely different organisms), and observations may contain indirect evidence where the organism itself is not present (e.g. a nest or track).

Maybe the formulation I mentioned in one of the other threads: “All evidence relates to the same organism/species”.

2 Likes

Rewording might help to some extent, if kept simple and jargon-free enough.
What about “The subject(s) of the observation can be found in all the photos/recordings”? Covering both the ant portrait and the anthill cases.

(Still, the explicit tolerance for habitat photos seems self-defeating of this new DQA thingy. In practice, various cases of “1 closeup of snake + 1 photo of alder bush” would have to remain unresolved as either DQA -upload mistake- or QA -habitat view- in the prolonged absence of a proper statement by unresponsive observers.)

1 Like

is not a snake. Use the new DQA.

…but that’s its habitat, ergo it’s tolerated, ergo “please do NOT use the new DQA for” yaddi-yadda…

From the iNaturalist Help:

Please do not vote “No” to the “Evidence related to a single subject” DQA condition in these scenarios:

  • When there are photos related to the subject that don’t show the subject (e.g. a picture of a bat sonogram, or a picture of the habitat the subject is in, or a picture of a drawing of the subject).
4 Likes

That is - it’s complicated. The habitat photo is supposed to show the snake tho, not just a picture of an alder bush - where snakes might live but we can’t actually see one.
A judgement call on each obs.

2 Likes

And yet, it was a direct copy-paste from iNaturalist’s very help page! nothing overly complicated:

(Emphasis mine)

1 Like

This might be possible with a snake.
But this is not possible for a 5mm beetle.

I consistently add habitat shots showing if insects I find are in woodland, etc.
Also to show the precise location for myself or others to check in future.
The habitat shots cannot show both broader environment as well as a small insect.

4 Likes

And even with this, I was flagged for “deliberate misuse” of the DQA, for making an argument for why an observation with an additional image that functioned as a habitat shot, but that lacked the subject of the observation, should not be marked “no” for “Evidence related to a single subject”. A staff member also commented on that observation that it should be marked “no”.

I am still left not understanding if there even is a clear answer to how to handle an instance like that because it doesn’t seem like all of staff agrees with the wording in the guidelines / help sections.

3 Likes

Perhaps this particular new DQA item could be moved to the very end of the list, reworded to something like “photos belonging to different observations have erroneously been added”. With the “yes” and “no” ckeckboxes working like whether the community taxon can be improved (“yes” preventing RG) or not (“no” allowing RG).

The checkbox currently makes observations casual. Are you suggesting that observations with photos of different subjects should not be made casual, but only prevented from becoming RG? I would be against that.

The value of the current option is that it takes observations that have specific issues out of the “needs ID” pool – and as long as they stay there, they will continue to be seen by users who may be tempted to try to ID the first photo, thus wasting people’s time and creating all the problems and potential conflicts we have been dealing with up until now.

2 Likes

You need to push for an iNat guideline for habitat shots. It is a broad issue.
We have a (clunky) workaround with a project -
https://www.inaturalist.org/projects/vegmaphoto-s-afr

  • for landscape photos deliberately intended to show the vegetation type, not what iNat wants. But. WHICH plant?!

Sorry, I can’t find it now, but - as someone said - I don’t get cross with my house key because it doesn’t open the car door.

iNat clearly has to work on - DQA for - is that badger in ALL the photos?
I would like to annotate individual photos

  • flower
  • fruit
  • leaves
  • bark
  • wide view
  • habitat intended, thank you
  • oops - pollinator - needs its own obs

https://forum.inaturalist.org/t/suggestion-on-how-to-include-habitat-images/3553/32
https://forum.inaturalist.org/t/policy-recommendation-about-habitat-photos/10866/12
https://forum.inaturalist.org/t/habitat-shots-acceptable-with-audio-observations/26630/74

1 Like

I agree with you. Subject plus habitat is not an obs of multiple sp.
Did you @mention tiwane on that obs, so he can see a problem case in action?

1 Like

This is a better wording than the current version or one posted by OP in my opinion.

This issue should be about judging user error, and only be used when that’s clearly the case.

Maybe this feature request could be bettered by logging different options and polling users rather than suggesting a single alternative.

2 Likes