This is a response I posted on a new users’ observation when I noticed they “agreed” to my observation:
Just a word of caution, several iNaturalist IDers (like me) are “enthusiasts” rather than “experts,” so it is good practice not to agree with an ID unless you are sure of the species yourself. Once things have 2 IDs they are often taken out of the workflow of a lot of IDers, which means it might be missed by someone who can confirm if the initial ID is correct or not. I’ve seen a lot of ____ around so I recognize it, but it is possible there are ones that look similar I don’t know about.
3 Likes
@charlie, I’m glad you reopened this discussion, although I would make one change to the title: I would eliminate the word “Inexperienced.” As various folks have already noted, there are different notions of “experience,” and the beauty of iNaturalist is that it puts each of those types of experience on a level playing field. It also provides a forum for discussion where we can sort out the observations that come from one another’s experiences and produce new knowledge and theories from this aggregation of information.
To me, the most interesting part of this discussion is the number of different ways people have decided to phrase, “Tell me more about how you reached that decision.” Just last night I could have completely ignored the idiotic suggestion that a plant I observed in Mexico was a Bromeliad, not an Aloe. After all, it was growing in the ground, and it had the texture of a succulent. But by asking a question, I learned that some bromeliads are terrestrial, and a few even have the texture of a succulent. The suggestion, it seems, hadn’t been idiotic at all.
I can’t stress enough how important it is that iNaturalist integrates that forum for discussion into the database of observations itself. My first experience with a crowdsourced database for observations of organisms was with the Invasives of Texas Database. The differences between that experience and my experience here with iNaturalist illustrate what science is all about.
Science isn’t knowledge. It isn’t rules or guidelines or checklists. It’s a process of observing, describing, predicting, reviewing, questioning, and discussing—not necessarily in that order. In fact, not necessarily in any order. Remove the ability to describe, predict, review, question, or discuss, and all you have is the observations. Unquestioned observations have a way of becoming accepted as unquestionable truth. Once we establish unquestionable truth, we can no longer improve our understanding of the world around us.
As a case in point, consider two species—Ligustrum lucidum and Ligustrum japonicum. I don’t know how many observations of each you will find here, but one thing is certain: the two species are often confused. Best I have been able to determine from the information here in iNaturalist, L. japonicum does not escape cultivation in any significant numbers anywhere in North America. L. lucidum, on the other hand, is perniciously invasive—at least that seems to be the case throughout the American South and at least in riparian areas and wetlands across the American Southwest to the Pacific Coast and down into northern Mexico.
Check the Invasives of Texas Database and you will find 243 observations of L. japonicum. Best I can tell from the available photographs, only 24 of those are actually L. japonicum. In other words, only about 10 percent of those observations are correct. But it gets worse: I can’t find a single observation that definitely shows that L. japonicum has even escaped cultivation, let alone established an invasion.
Of the remaining observations submitted as L. japonicum and validated by the person given that duty, 186—75 percent—are actually L. lucidum. A couple of dozen are split between two other privet species that are invasive in Texas—L. sinense (16) and L. quihoui (8). A handful of observations have images from which nothing can be identified, and a couple are plants but probably aren’t any privet at all.
Worse yet, this seems to be a symptom of a larger problem. As I was drafting this post, I discovered these two gems:
-
Bugwood, a resource used by many people who are interested in managing invasive species, has collapsed L. lucidum and L. japonicum into a single entry. They aren’t saying it’s the same species, but the implication is that they are so hard to tell apart that there is no point in trying.
-
The Invasive Species Institute, a new organization I just learned of this week, lists only L. japonicum, not L. lucidum, in its inventory of invasive plants in Texas. The entry begins, “Ligustrum japonicum commonly forms dense thickets in fields or forest understories. It shades and out-competes many native species, and once established is very difficult to remove.”
If you know me, you know that I advocate strongly for using native plants in landscapes whenever possible. Still, I gotta feel some middle-child empathy for L. japonicum here. As a middle child, you catch all the blame for stuff the older kids did because they know how to make you look guilty, but you also catch blame for the stuff the younger kids did because by that time Mom and Dad have figured out how the blame game works (weren’t they ever kids?) and figure that when you say the younger kid did it you’re just shifting the blame. So long as the parents stay in the authoritarian role, the only way for the middle kid to get relief is for an aunt, uncle, grandparent, or other trusted adult to notice, pull the parents aside, and get them to see the real picture.
So here’s L. japonicum, quietly sitting where it’s planted and doing what it was planted to do, but being blamed for everything that three of its siblings are doing throughout Texas and, more broadly, much of North America. With the tools that let people log their observations but not review and discuss the observations of others, that’s where this introduced plant has gotten stuck. In the websites built to serve people who work on managing invasive species in North America, it looks like L. japonicum ruins ecosystems. In fact, it might be the only privet that doesn’t. In North America, anyway.
But not here in iNaturalist. Here we have had a discussion. Alison Northup (@alisonnorthup; that won’t link because she hasn’t joined this discussion app, I think) reviewed as many observations of privet as she could. Asking a question here, making a comment there, she got us to unclutter our data. We straightened out glossy privet (L. lucidum) from wax ligustrum (L. japonicum). We even sorted out Chinese privet (L. sinense) from quihoui privet (L. quihoui) from common privet (L. vulgare)—realizing as we did that L. vulgare probably isn’t even found in Texas. (That might not surprise you, but it should surprise somebody. The Invasives of Texas database has 193 observations to the contrary.)
So I would argue that it isn’t just okay to make a comment or ask a question when someone misidentifies an observation you made or are following. It’s essential. If you don’t, you aren’t participating as a scientist. We each need to participate as scientists if science is going to happen here.
14 Likes
I am not a “experience” scientist nor do I have a degree in any major field that is related to identifying taxon in general. When I click on a picture, I am doing my best to double check facts and other resources before agreeing to something (though I have been guilty of that before).
If I don’t know what something is, I’ll at least try to put it at a section that I know it’s (ie. Aves, or Dicots). I joined this site mainly as a hobby, something to do in my free time. I will make wrong ID’s in the future, because as I said before, I am not an expert. But don’t people also learn better from making mistakes?
I have also read through the forum on how to better take photos of species, so that the actual experts can identify them better (those should really be made like a sticky note somewhere), in which I’ll do my best in the future.
9 Likes
sounds like you’re on the right track! 
2 Likes
Though for what it’s worth experts make mistakes too :)
4 Likes
I’m an expert in some groups, and an enthusiast in others. One of the things I’ve learned is how much I don’t know! (And that sometimes I click the wrong button.) We all make mistakes, even about the groups we’d like to think we’re experts in.
6 Likes
I try to keep an eye on mayfly observations. The most common misidentification (or questionable ID) seems to be Hexagenia limbata. I’m not an expert on these, but I do try to advise caution when I see these IDs offered (without any supporting justification) by copying and pasting the following text that I have saved: “Do you know there are 7 species in genus Hexagenia in North America (https://www.entm.purdue.edu/mayfly/na-species-list.php#FamilyEphemeridae)? As I understand it, some of these cannot be identified without looking at various body parts under magnification – i.e. very difficult from a photograph. To be safe, it is often best to ID to genus level.” People generally seem to appreciate this feedback.
I would like to create more of these for other questionable or incorrect mayfly taxa IDs - and would REALLY like to engage more knowledgeable entomologists to participate in reviewing IDs, and to write up more of these, cautionary templates. It would also be nice if there was an easy way to have such “warnings” stored in iNat and automatically available (e.g. by a clickable link) when someone is about to submit an ID. This of course would take a lot of collaborative work by the iNaturalist community to create and agree upon taxonomically appropriate warnings (or tips) and then modify the data entry system to accommodate them.
- Jerry
6 Likes
That is an interesting idea. i had started using textblaze, recommended by another iNatter, and have found it excellent for managing standard responses. It would be excellent if iNat had that ability built in. I think you intimate that it could be “community maintained” meaning that everyone could add and edit them in a similar way to how we can add places, or to how tutorials work here in the forum and are editable by everyone. I think it would be really useful to have that “shared and collaborative” approach, as well as the ability to create personal ones that you can use. For instance, I have one for Uliodon that I use whenever I see certain IDs given, and it is very similar to yours, stating the background but still indicating to the reader that they can ID as they see fit. I also have one for tagging in the “spider group”, those iNatters that have expertise or interest in anything related to NZ spiders, and where I see the need for second opinions or even just for particularly “wow” observations I use that “\ccspidergroup” snipet, far easier than trying to remember their inat account names and mistyping them etc.
If someone doesn’t beat me to it, I’ll write up a feature request for it.
3 Likes
The idea of warnings is really interesting. I don’t necessarily think too much collaboration would be involved, as there’s probably only a few scenarios. iNat would just need to provide a few examples of boilerplate text, and it’s only the variable content (which will usually be taxon names) that needs to be modified. ie:
“(species1) is commonly misidentified as (species2) - please ensure you have noted the relevant diagnostic markers.”
“This species cannot be reliably identified based on appearance and/or range alone. Please consider revising your ID to (higher-level classification ie Genus or Family)”
In theory, you could possibly even automate this. If a particular taxon is often corrected, or dropped down to a lower taxonomic rank (especially if it’s accompanied by a relatively low Computer Vision confidence score), this information could be used by an algorithm to both select the correct boilerplate, and to fill out the details.
Once a boilerplate response becomes active for a taxon, I would envisage the following things happening:
-
The first time someone adds that ID, a auto-generated comment is created on the observation. The original identifier will get a notification, and other people will be able to see the comment inline
-
The identification (ie, wherever the “agree” link is) is flagged with an icon indicating care is required before agreeing - With maybe a tooltip summarising the auto-generated comment.
None of these things should it more difficult for people to ID observations (like an agree confirmation would), so the only downside I can see is that there will be a bit more notification spam for those who are ID’ing these species.
I agree it is a real problem when people simply confirm IDs of their observations… I don’t think it’s a “thanks” issue; I think it’s a “they must know what they’re doing” situation. It’s not just newbies, I see plenty of people with many observations (but typically having done few IDs) doing just that - after all, don’t we all love having our observations at RG level?
The end result, as mentioned above, is plenty of observations reaching RG based on a single informed opinion. I find this particularly problematic because I am focusing on a group with few experts (land snails) in an area of relatively high endemism (SW Europe), and many of my IDs seem to be neither questioned not confirmed by people with snail expertise. So I am torn between wanting to add to the pool of observations with my imperfect knowledge and being worried that my mistakes will become deeply engrained and propagated in iNaturalist.
For what it is worth, this is what I have been telling people who seem to be taking my word as Gospel:
I see you have now agreed with my ID - is it because you have checked the species out and reached the same conclusion; or because you trust my ID?
If it’s the first: YAY!
If it’s the second: keep in mind I could be wrong, so it would be good to have a second opinion. That is the logic of iNaturalist: it takes two people to agree on an identification to make it Research Grade. If you simply agree with me, the observation has been in practice identified to Research Grade by a single person (me). It will fall out of the radar of other people doing identifications (who mostly focus on observations that Need ID), so if I made an error it won’t be easily corrected.
As general advice: do not “simply agree” when someone identifies your observations. Only agree when you recognise the species yourself, from your knowledge/experience, so that your identification is adding new value.
This is really cumbersome. I propose that instead whenever where are in this situation, a pop-out message appears that makes the same point.
To clarify, the situation is: Person A enters an observation (with or without proposing an ID). Person B proposes a new identification (that either adds, changes or refines the previous ID). Person A presses the “Agree” button. [these being the only IDs for the record]
They would then get a message saying
Wait, why are you agreeing?
Is it because you checked your record again and reached the same conclusion as personB? Great! Click “Agree” again.
Or is it because you assume personB’s ID must be right? If so, click “Cancel”. Please agree only if you recognise the species yourself (from your knowledge/experience), so that your identification adds new value!.
[with an Agree and a Cancel button]
I realise this adds extra annoying clicks to people in the first situation; while people in the second situation who just want to bump their observations to RG will just learn to click on agree twice. But a number of people who were clicking “agree” mindlessly will hopefully stop doing it…
5 Likes
I’ve been proposing exactly the same on a different threat recently. :-)
I think this could be helpful
- for new users
- for those prolific agreers who buzz through 1000s of records
- for anyone who thinks that the identifier is probably right and/or that its polite to agree to acknowledge their effort.
3 Likes
I agree, especialliyt if you’re a person iDing something late in the night…;).
I don’t think it would be annoying, also if someone who has more knowledge of said species could be maybe given a little leeway? Like not as many make sure pop-ups? If that makes sense.
I you are referring to the suggestion of anasacuta someone with ‘more knowledge’ would (as good as) never see that popup, because it would only come up if the user suddenly changes his/her mind after seeing a new ID from someone else on the users own observation. This popup suggestion has basically zero potential to annoy high volume IDers.
1 Like
I generally try to start a brief conversation about the subject of the observation and to say welcome before I bring up problems. It helps a lot.
A few weeks ago, with this one guy in (Turkey? Qatar? Don’t remember) who Agree-d with some 40-50 of my coarse IDs of his stuff, it took my gentle nudging and friendly asking a number of times before he “got it” and changed his behaviour. Thankfully he was grateful to learn how to properly use iNat and pretty patient with me. Nice guy. I’ve had a few experiences like that.
Except other times I’m lazy (or busy) and let it go, provided the Agree didn’t / wouldn’t bring “unqualified” RG.
5 Likes
Last week I tried to tell schoolboys agreeing with my ids of my own observations that it’s wrong to just agree and wrong to “assume I kow what it is”, they tell me they did research and then agree with my moss genus id, while even I am far from sure it is one (key is hard to follow). I’m quite tired of that actually, I understand if (why) people agree with their observations ids, but that situation is new to me.
3 Likes
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.