Identification Etiquette on iNaturalist - Wiki

Background

The intended emphasis here is on general etiquette and best practices of identification with an eye toward newer users, not on the technical details of specific identification issues. I intentionally stayed away from the weeds of DQAs (Data Quality Assessments), and when they should and shouldn’t be used.

This is a wiki so that we can discuss, add, and refine as much as possible, and perhaps eventually add a topic like this to the site’s Help pages:

Note: most functionality mentioned below is best accessed via inaturalist.org or your network web portal. Currently identification and messaging tools are more limited on the smartphone apps, especially for iOS.

Identification Etiquette on iNaturalist

1. Be Bold. If you think you know what something is, say so! That’s what the iNaturalist community process is all about (among many other things). Note that the tab for IDs is called “Suggest an Identification.” Not sure, but have some ideas? Leaving those ideas in a comment instead can still be helpful. Or you can just come back to the observation later, or mark it as “reviewed” and move on.

2. Disagree agreeably, and with evidence. If your Suggested ID disagrees with an existing ID, it helps to politely note your evidence or other reasons in a comment, so the observer knows where you are coming from with your disagreement.

3. You don’t have to know which species. If you can help get the ID to a finer level, please do. For example, if the current ID is “plant,” and you are sure it is a conifer tree, but don’t know which one, just ID it as a conifer. iNaturalist lets you ID at any level.

  • Adding coarser IDs - if finer-level IDs already exist on an observation, adding a coarser ID in the same lineage will give you the option (when you save it) to explicitly disagree with the finer ID (orange option), or to simply be uncertain about the finer ID (green option). If you intend to disagree, again you should say why in the comments. (Keep in mind that sometimes an observer will have recorded notes about features that are not visible in the photographs, which may be important to the identification.)

    • Special case - as discussed in this topic, be aware that iNaturalist currently treats disagreements with subspecies somewhat differently than with other subordinate ranks.

4. Assume good intentions - Everyone makes mistakes. When disagreements happen, ill intent is almost never involved. There are many different experience levels, ages, personalities, nationalities, and languages on iNaturalist. Keeping it civil and evidence-based helps. Be patient with new users who may be unfamiliar with how iNaturalist works. Even if bad intentions are evident, do not act uncivilly in return. That said, any user who shows a pattern of random, intentionally wrong, or “joke” IDs is at risk of having their account suspended. Seek for others to help resolve a dispute such as by flagging inappropriate comments/IDs or reaching out privately to the staff.

5. Don’t automatically agree with someone else’s ID. The agree button is not the equivalent of a “thank you” or :+1:, and does not provide any “reward” to the identifier. The agree button is a convenient shortcut to add your own matching ID, when you are convinced that ID is correct. Adding your own ID automatically withdraws any previous ID you added to the observation. If you have become uncertain of your previous ID and are not sure what it should be, then consider adding a new, coarser ID of which you are certain. Or alternately, you can use the dropdown next to your old ID and just select Withdraw without adding anything new (not currently available in the iOS app).

  • If you are not sure if you agree with someone’s ID, try doing a little research first. One easy check is to use the Compare tool (or the Suggestions tab in the Identify modal) to see related organisms found in the same area. You can right click on the name or thumbnail of any suggested ID, and visit the page for that taxon. Check especially the Map tab, the About tab, the Similar Species tab, and the existing Research Grade photos. Do these seem consistent with the observation? Could it be confused with a similar species? You could also look at the pages for some of the other taxa that Computer Vision suggests in a blank ID box. Or you can just wait for more community members to weigh in first.

6. Be careful using the automatic Computer Vision suggestions. For some taxonomic groups and areas of the world, the CV suggestions are amazingly good, and getting better every time the model is updated with our new observations (about twice per year). In many other parts of the world there are too few data yet for reliable suggestions, and some species just can’t be identified from photographs alone. But iNaturalist will always give you suggestions, even if the species you observed is not yet part of the CV model. Suggestions may even be from the wrong continent if iNaturalist can’t find appropriate species observed nearby (or if you tell it to include non-nearby suggestions).

7. View all photos, description, and comments before adding an ID. They can help solidify your ID. Or, you might encounter one of these situations:

  • Single photo shows multiple species. If it’s not clear which species is the observer’s focus, did they leave a provisional ID, or say in a comment or description which organism they are focusing on? If not, you can either leave a comment and ask which species they want identified, or leave an ID for one of them, and describe in a comment which one you are identifying. You can change or withdraw your ID later if they respond that they intended a different focus. (See #11 below)
  • Different photos in one observation show different species. This sometimes happens with new users who are still learning how iNaturalist works. For this situation it is best to leave a comment suggesting that they make separate observations for each species. Some boilerplate comment language can be found here, but feel free to make it more friendly and personal.

8. Getting more ID help

  • See the How to Get Identifications tutorial.
  • Did you give it your best shot first? Even if you can get it into a broad category like birds, plants, butterflies, etc., that will make it much easier for knowledgeable identifiers to find. If you are not sure, this is still better than leaving it as “Unknown.”
  • Please do not use this iNaturalist Forum (where you are reading now) to request ID help. That is what iNaturalist itself is for. See this topic for more guidance on using this Forum.

9. Some observations are not identifiable. Different kinds of life have different characteristics that distinguish species, and these are not always visible in the photographs (especially blurry ones) or other evidence.

  • See the How to Get Identifications tutorial and links there for more advice.
  • Don’t be discouraged—every observation still helps! At least a researcher will know where and what time of year to go find what was seen, and who to message if they have questions.
  • A good identifier may leave a comment about the missing information they need before they can make an ID. Don’t take this as personal criticism, they are just documenting their assessment, and might help you know what to focus on next time you see that organism, or ones like it. If you have more photographs of the same organism at the same place and time, definitely add those to the same observation (don’t make new observations for these), and @ tag the identifier to let them know. The additional views may help. Or if you have separate observations from different dates, include the links to those in the description or in a comment.
  • If an observation can’t be identified to species, and you are convinced that the current Community ID is “as good as it can be,” there is a question at the bottom of the observation detail page, “Based on the evidence, can the Community ID still be confirmed or improved?” You can vote “No” on this question. Unless others disagree, if the current Community ID is finer than Family level, the observation with then become Research Grade at its current level.

10. What if I can’t find the right name for my ID? Not all published names have yet been added to iNaturalist. Or you may only be able to find a synonym of the name you want to use.

  • If you found a synonym and can accept it for your ID, then please use it. Otherwise…
  • Look up the next higher taxon (genus, family, etc.), or the synonym in the header search bar and go to the taxon page for that name.
  • Under Curation, select “Flag for Curation” and leave a short reason like “GenusA speciesB missing” or “synonym of GenusX speciesY” and click “Flag It!”
  • If you have more detail to add (strongly encouraged), you will see a link to the flag you just created near the top of your screen. Open it back up and leave a comment in the comment box with the details. Adding links to your sources is very helpful for busy curators.
    • NOTE: with some exceptions, out of practical necessity iNaturalist only follows a single “Taxonomic Framework” for each major group of organisms worldwide. If you are wanting to use a different synonym, and it conflicts with the adopted framework, and you believe the name currently used in the framework should be rejected, you will need to provide whatever documentation and justification you can to support “deviating” from the framework. Deviations are sometimes accepted, but only with sufficient justification. See this help page for a list of external taxonomic authorities currently followed by iNaturalist.

11. Honour the observer’s intention. If the observation shows a photo of an elephant with its toe touching a daisy, and the observer indicates they are interested in the daisy, then don’t add identifications for the elephant. Or if the observer doesn’t add an initial identification, but chimes in later when they see the wrong thing being identified, follow their lead and identify whatever they are interested in.

111 Likes

Good idea to have this thread. The only thing I can think of right now is that the pop-up box that gives a choice to say that a coarser ID is not a disagreement would need to be changed to be consistent with #3 which makes all coarser IDs into disagreements.

2 Likes

Thanks @sgene. We’ll see what others think too, but I’m not sure there is an inconsistency in #3 with the popup box. Both popup choices are disagreements in a sense, just about different things. For everyone’s reference, the pop-up when adding a coarser ID (that is still a parent of a finer ID) says,

Is the evidence provided enough to confirm this is GenusX speciesY?

  • I don’t know but I am sure this is GenusX
  • No, but it is a member of GenusX

The second choice is saying that the evidence definitely doesn’t support speciesY – clear disagreement.

The first choice says that the sufficiency of the evidence is uncertain. While the finer ID was saying that the sufficiency of the evidence is certain. Finer distinction, but still a disagreement to me.

2 Likes

I think your wording is much better than the current wording saying that “I” don’t know whether the evidence provided is enough to confirm the species. Your wording is a definite statement of the sufficiency of the evidence being uncertain, rather than a statement of my personal knowledge about it. Now I see where some misunderstandings have arisen . . . Thanks!

I think it would be best to have this [or whatever the final language of #3 becomes] displayed somehow whenever a coarser ID is added, as a reminder. (For what it’s worth, I see people presumably using apps constantly making coarser IDs. I have assumed this is because they are responding to IDs in reverse order.)

4 Likes

Yeah, the current wording of the first choice in that popup box comes across in a way that inspires a thought process like “even though I don’t personally know enough to confirm the species others have identified, I can still contribute by confirming the genus.” That there are a green button and a yellow button also sort of implies it’s a “soft agree”/disagree dichotomy being presented. I have definitely interacted with newet users who misunderstood it in exactly that way.

9 Likes

in terms of the explicit disagreements, i still think inat has a grey area under dispute in terms of how to manage that. For instance, you can’t tell if a sugar maple is red or sugar maple and you are skeptical the observer can, but you aren’t sure. ‘knocking back to genus’ is something i think one should only do if they either believe the ID is wrong or are very certain the original observer couldn’t tell the difference (a high bar!) I personally put the bar a lot lower when a user hasn’t used the site in several months. If it’s an actuve user, i think it would be best to add a comment like ‘it’s hard to tell sugar from red maple just by bark, did you see the leaves?’ instead of disagreeing right off the bat.

Also on that note… whether or not a user is active is really important. For my part, i don’t post ID help along with disagreements if the user has definitely been gone a long time (~6 months or so) but for an active user i pretty much always do unless i already know them and have talked to them before. I think it’s important to help cushion the impact of being ‘corrected’ since some people don’t respond to that (it is hard to!)

11 Likes

Totally agree. I’m into bryophytes now and I don’t have the tools to take photos of microscopic features to include in my observations. So I submit observations which I’m fully aware can’t be confirmed because key features aren’t visible. Certain bryophyte experts who shall not be named keep knocking my observations back to genus because they “cannot be confirmed”!

I try to only knock observations back to coarser taxa if I’m defensibly certain that it isn’t the species the user originally suggested. Not being able to confirm an ID suggestion because certain features aren’t visible is not a valid reason! /end rant/

On another note, I find myself struggling sometimes to use language that’s instructive without sounding rude. Intonations don’t come through on the internet, and factual statements can sometimes come across as rude or arrogant. It’s hard to find a balance between instructive language and overuse of “lol” and “:)”…lol

20 Likes

@wdvanhem To minimize this problem, on insects for example, which must be dissected or studied under microscope, I include verbiage in the description or comments that specimen has been dissected, or specimen has been analyzed under microscope.

12 Likes

people need to calm down on that business. I feel like we need a more defined policy on that. I originally only advocated for explicit disagreement if you are confident it is wrong but that seems a lost cause at this point and there are grey areas of course… like if it seems wrong and is also out of range…

8 Likes

I think this tutorial is very well done, great work!

3 Likes

Way out of range can be a good reason for disagreement (unless it’s something that conceivably could be introduced).

3 Likes

Yes, that’s a very good point and I should probably do that more often. I still think the knee-jerk “I can’t confirm this so it must be wrong” approach is a little unnecessary, though.

3 Likes

yeah i agree

@jdmore I think this is a great post! I often struggled when I started using iNat with the general idea that it was okay to be bold and take an educated guess at an ID… and to be wrong. I have learned a great deal from iNat, not only about identifications, but about how to be nonjudgmental and cooperative in seeking understanding, which has been incredibly rewarding!

I do have one question about ID etiquette that I don’t see addressed in your post: when is it appropriate to withdraw an ID? I have developed a habit over time to scan my previous IDs to see who has agreed/disagreed with me on taxa I was not particularly confident about. When I see that someone who is a curator or has in-depth knowledge has disagreed with my ID, I withdraw it in the hopes that this will accelerate arriving at an accurate ID. Is this strategy/practice a good thing, or should I not withdraw IDs in this situation?

Thanks for your guidance! My goal is to use iNat in away that helps move toward accuracy.

ocean_beach_goth

11 Likes

i withdraw an ID if someone else offers an ID that seems like it might be right. It keeps a record of your ID so it’s still there for learning and reference purposese. If it’s a coarser ID and I get a finer ID within the same taxonomic unit, i leave my ID alone if I am not sure of the finer ID (like if it is a taxa i know very little about). If someone offers an ID and i am reasonably confident that they are correct and know the taxa pretty well (plants), i will agree with it.

In terms of IDs, if it’s your own observation, i think making an educated guess is fine, maybe along with a ‘not certain’ type comment. if you are helping others with ID, if you aren’t certain, i think it’s better to use a comment to suggest a possible ID such as ‘I think it’s Solidago canadensis but i am not sure’ or ‘Perhaps solidago gigantea’.

Does that make sense?

6 Likes

@charlie yes, this makes sense and is in line with what I generally do already. Thanks for the quick response!

5 Likes

The other aspect that contributes to the thinking that one choice expresses disagreement and the other does not is that one choice ends up stating on the observation that there is disagreement and the other choice does not.

I think I can bring this up without talking about the DQAs, but I would like to see a statement of etiquette for this situation: Observer posts multiple photos, each of a different organism. Someone adds an ID for the first photo. One or more comments are added asking the observer to separate into multiple observations. Then with all the photos still there, another person (not a newbie) comes along and confirms the ID, leading to Research Grade. I think a statement of etiquette about continuing to add IDs in this situation might be helpful.

6 Likes

Point well taken. Either the Wiki or the ID dialog (or both) probably need to be clarified and better aligned on this.

4 Likes

Great thought, we should definitely cover that in the Wiki. Let’s see where the discussion goes on that for a while, and then come up with some wording to add. Suggestions all?

And welcome to the forums @ocean_beach_goth! Glad you have joined us.

3 Likes