ID based on observer notes

General
21

I was trying to find a suitable thread for this rather than starting a new one. Some of you will remember my comments in the past about lack of confidence, a.k.a. Fear of Being Wrong as one of the bottlenecks on iNaturalist. Well last night was a frustrating illustration of the problem with it.

Thank you for saying it, but the phrasing “their expertise” can still put people off who might have helped. Last night I was identifying and came upon an observation submitted in 2016, which recieved its first suggested ID (“Animals”) in 2019, and then nothing else. It was the observer’s sole upload. I don’t expect they will see the tentative genus-level ID I added last night, because they’re probably long gone.

Now, if they had uploaded some ubiquitous lawn weed, their quick departure might not seem like such a loss. But then, if they had uploaded some ubiquitous lawn weed, they probably would have gotten engagement and been more likely to stay. But this was a seldom-observed organism. What if they had gotten the engagement needed to get them to stay? Maybe they would have been one of those observers who habitually upload less-often observed taxa, filling in gaps in the data. But the same reason that made their departure such a loss also made it more likely: they didn’t get any engagement because they uploaded a seldom-observed organism.

Fear of Being Wrong means that the very observers who upload the unusual data are exactly the ones most likely to slam into the discouraging wall of silence.

The thing is, they left clues, so the “no idea” line doesn’t even apply. The observers notes referred to the organisms as “Worms living in casts?” and mentioned that they were found in shallow freshwater. So right away, if you read the notes, you can’t say that you have no idea; you have an idea that it might be a worm.

Golden Guides were a series of nature books meant for primary and secondary school readers. They’re not sources you’d cite in a manuscript submitted for peer review. But that doesn’t mean that they are inaccurate; just not comprehensive. As it happens, I still have my Golden Guide to Pond Life from when I was a kid. It has a section on worms, so I figured it was worth a look. As of last night, that observation, unidentified for nine years, has a suggestion as to genus. And even if there are other genera that could fit, that weren’t in the Golden Guide, still, ruling out roundworms and flatworms still gets us as far as Phylum Annelida, which is a start.

Aquatic worms are not “my expertise,” but I doesn’t mean I throw up my hands and claim to have “no idea.” Too bad nobody who saw that sole observation when it was first uploaded had the courage to use what sources they had. Even if the source is a nature guide meant for kids.

6 Likes
22

That is why I ID for newbies. Try to catch and encourage them. And ask the questions while they might still be coming back to answer them.

4 Likes
23

Several times now, I have encountered a different phenomenon: The observation notes give a description that is supposed to substantiate the given ID (in these cases an incorrect one), yet it does not match the photos at all. Presumably, the observer first accepted the CV suggestion and subsequently added a some characters of that species based on literature review, without evaluating what they actually saw. This can often give the impression that they are quite sure how to identify a species when really the opposite is true. Thus, I am usually somewhat critical of IDing based on notes.

2 Likes
24

I don’t explain much, tactfully or otherwise. If the evidence isn’t there for me I don’t try to ID and move on. Others might see something I don’t or be influenced by the notes and add an ID and that’s fine.

25

This is not a case of IDing based on observer notes (i.e., using evidence only included in the notes but not in the media).

It also probably not a case of “fear of being wrong”, but of a new observer who doesn’t understand how to add IDs or didn’t know what ID to add. One can only be afraid of being wrong if one knows what to enter in the first place. If someone doesn’t have experience with nature or taxonomy or using field guides, they may not know where to start to research an ID and they likely do not realize why iNat needs them to enter a broad ID to get the process started. Many observations entered as “unknown” aren’t entered as unknown because the observer has no idea what they saw, but because they don’t know what label to input. They may think that if they don’t know the species their suggestions are irrelevant.

Users who provide broad IDs are likewise often people without advanced knowledge. Reading observer notes may be of limited use to help them classify the observation: there are worm-like organisms in many branches of animalia and observers, particularly new observers are not always reliable in their guesses about what things are. The person IDing the unknown observation may not have known where to begin to research something far outside their area of knowledge.

People who ID unknowns may pursue various strategies – some may spend time researching individual observations. Others may decide that they can be the most effective if they quickly add an ID, however broad, to as many observations as possible, according to the reasoning that this filters the observation to the next level of IDers who might have a better idea what it is. There are ongoing disagreements about whether this is useful or not, but I don’t think that users who add such really broad IDs are actively choosing not to add finer IDs because they are afraid of being wrong.

4 Likes
26

I was referring to however many people saw that observation and didn’t add an ID during the three years between 2016 and 2019.

I was trying to say that we have a big problem with potential identifiers also thinking this way.

I was referring to the ones who didn’t even try to add an ID at all. We’ll never know how many of them there were between 2016 and 2019. More importantly, we’ll never know how many of them there were in the days when the observation was fresh and the observer still hopeful.

2 Likes
27

Again, people looking at unknowns or broadly ID’d observations are often not greatly experienced themselves, and worm-like creatures are outside of many people’s area of familiarity, so not IDing such an observation may well not be “fear of being wrong” but not knowing where to start to research an ID. Taking something completely unfamiliar and making sense of it is also a skill that takes a certain amount of knowledge and experience to develop.

Sure, it would be great if more people would spend time to figure out puzzling observations, but there are lots of legitimate reasons why people might choose to focus on something else.

I’ve also seen plenty of observations at all taxonomic levels which just seem to get overlooked – perfectly identifiable, easy species that sit for years without an ID. I don’t have an explanation except that presumably nobody with relevant knowledge happened to look at that particular observation.

Don’t negatively judge “the ones who didn’t even try to add an ID at all” – you don’t know who saw the observation or why they may have decided that they didn’t have anything meaningful to contribute. We are asked to assume that people mean well. It seems to me that this includes the assumption that people helping ID observations are doing their best. Accusing them of failing to ID something that would have encouraged a new user because they were “afraid of being wrong” won’t motivate anyone to ID or solve the lack of IDers. Empowering people to feel that they can help with IDing also necessarily means accepting that it is OK to not ID an observation if one does not feel comfortable doing so for any reason.

5 Likes
28

Definitely true. If I was forced to ID everything I ever opened, I’d quit mighty quick, because while I might be happy to recognise an insect as an insect and ID it as such, worms are so far outside my comfort zone that I wouldn’t know where to start.

And as far as observations which get overlooked, I’ve seen the same thing: easily identifiable observations that seem to have languished for years, somehow. At the same time, I sometimes wonder whether they did in fact have IDs, possibly multiple, but the identifiers have since left. The occasions where orphan comments indicate missing IDs/comments show it definitely happens - for all we know, the worm mentioned earlier may have been at RG for years before people left and took their IDs with them. Maybe the problem that needs tackling is how to prevent such loss just as much as how to encourage more to ID in the first place… (Yes, I know this is far from a new idea. It just doesn’t seem to ever get anywhere.)

7 Likes
29

Reviewed without my ID (replace my name with yours)

157K (and there I could have, Dumped 90% into Damned Difficult Dicots)

to set against 161K IDs

There was one day during CNC with one MILLION obs. No wonder identifiers never do see some obs.

4 Likes
30

I think the level of the observer and his specialization (if suddenly such is indicated in the profile) are a good sign of trust, yes, you should not believe 100%, but on the other hand it will serve as a signal for you that this is really what you thought about, or what you could not understand, fortunately on the site there is a rating of people who are the best in their definitions, but of course this is not always decisive. It is also worth noting that accurate identification requires a comprehensive approach, it can even be an autopsy and sometimes one photo is not enough

1 Like
31

Just today I got the first agreeing ID on a flower I photo’d, IDed, and submitted 11 years ago. It was a good clear photo of the whole plant and not that difficult to ID. I’m assuming it was simply overlooked for more than a decade. I wasn’t annoyed, more amused it took that long.

6 Likes
32

This is very true. For example, there are certain differential diagnostic features that may be difficult to capture in a photo, but which may have been clearly noted by the observer. The fact that the observer mentions these features specifically is a pretty good indication that they know their stuff and are mentioning that feature precisely because they are aware of its importance as a differential characteristic. I’m thinking for example of two very similar grasshopper species in the Oedipoda genus, where the main differentiating characteristic is the colour of the underwing, almost impossible to capture in an image unless you’re very lucky, but often clearly visible if the grasshopper takes even a short flight while you’re observing. Mentioning this trait in a comment is clear evidence that the observer already has a pretty good idea of what they’re seeing, making (in my mind) the comment reliable enough to be considered as evidence.

6 Likes
33

I do see your point: there are three phyla of worms, and therefore no taxon for “Worms” as a whole. You would have to underdstand about roundworms (Nematoda), flatworms (Platyhelminthes), and segmented worms (Annelida). Mainly I was thinking in terms of the fact that even a book meant for kids – a Golden Guide - had enough information for me to feel that I could say something. Even if the genus turns out to be wrong, I doubt that the phylum will be; and considering that synapomorphies are a real thing, the family is likely correct, too.

In contrast, much of the discourse, at least here on the Forums, seems to suggest that people think they need specialist-level knowledge to be able to contribute meaningfully. I disagree. Put a Song Sparrow and a Swamp Sparrow (both in Genus Melospiza) side-by-side, and they look very similar; but these routinely get identified correctly by people who do not have the level of knowledge to have written the Sibley Guide.

34

I don’t think people need to have specialist knowledge to contribute meaningfully.

However, many of us in the forums ID taxa that – unlike your sparrows – routinely get misidentified both by the CV and by users. So I do consider that learning to recognize the limits of one’s knowledge is an important skill; it isn’t the same thing as thinking that one doesn’t know anything, and it also isn’t inherently separate from learning to recognize what one does know (maybe all ants look the same to me and I therefore shouldn’t be suggesting species or genus, but I can still confidently say that this is an ant, and that is a useful starting point).

6 Likes
35

Oops. I took time off from the forum, and am late picking up the question.

Data quality is important to me, so I work hard to submit the most correct information I am able in identifying my own observations. I work even harder to offer only a correct ID for the observations of others, since: a) I have only their submission to work from — not the 1st person experience or local knowledge; b) a “disagree” undercuts the observer’s effort and takes a lot of work for me to both assert with confidence and to support with competence; and c) a “me too” “looks good” quick ID may really be not much more than a :heart: and does not really improve the dataset — seems likely to delay an accurate ID if in error.

I do try to follow my submissions, and there is nothing which discourages me from contributing time and effort to iNat more than a “disagree” with no explanation/rationale.

My “expertise” over a wide range of taxa is pretty much limited to the spp of my hyperlocal region. I don’t like to judge too far afield, and don’t see very many other observers in my physical space. I’m happy to engage with anyone who wants my opinion, but generally think my time/effort is more productive by adding quality observations — both current and the back catalog.

2 Likes
36

If only more IDers showed signs of thinking this way.

2 Likes
37

Once again, IDers do not disagree because it gives them pleasure. Most of us find it stressful to do so. But it is often necessary to disagree with IDs, particularly if one is trying to clean up a difficult taxon where people often choose unjustified species IDs apparently at random. It would not be doing any favors to anyone, including observers, to simply leave such IDs if the provided evidence does not allow for excluding other species that are also possible in the area.

Experience teaches us that most users do not put a great deal of thought into their IDs. Obviously there are exceptions, but if they do not leave comments we have no way of knowing this. If they do leave comments about how they arrived at their ID, most IDers are willing to take this into consideration and they may include more details themselves if they still find it necessary to disagree. But we are also human; sometimes we overlook observer notes.

10 Likes
38

We each make our choices - but think about the identifiers who will have even more work (and even less time to spell out Why I Disagree Here … I am sure you would be able to help ID a slice which suits your taxon / location knowledge. It seems ungracious to both ‘refuse’ to ID, and also complain about a lack of explanations for IDs that are given.

PS this is your most observed sp, which has almost 1K waiting in Needs ID. If you leave the doubtful ones for someone else - there are surely some obs you could help to RG? No?

You say on your profile, that you are more interested in animals - so this has over 2K waiting

5 Likes
39

I’ve observed two sedges and noted that one has a round stem and the other has a stem that feels like a Reuleaux triangle. Hopefully that’ll give someone a clue about what species they are.

1 Like
40

I understand that it would be so, and am sad that it is.

Again, understandable that would be so. It may be surprising to hear that I very much appreciate a “disagree” ID on my observation, even if I have put hours into coming to the wrong conclusion. If I am able to learn something from my error, or teach an IDer something they didn’t know. But a silent “nope” does neither of those things.

I might figure out the explanation based upon the alternative proposed — immediately or with more hours of effort. Unfortunately, at least as often I will learn nothing useful and wonder whether contributing observations to iNat is worth the effort. There is a trade off between quality and quantity in both submissions and reviews. It might feel important to make 100K IDs, but if it comes at the cost of losing the most careful observers, is quantity the best measure? I don’t know the answer, but it is a question which I think matters.

2 Likes