When I post an identification with my observation I only post identifications that have been supplied by specialist. Yet curators, some who no specialized knowledge of the animals, are able to over ride my identification and there is no way to correct their mistakes. How do I realize this problem?
iNaturalist site Curators are a small group of users who have the ability to edit taxonomy, but they do not have any way to make their identifications count more than anyone elseâs. No one does.
I think youâre just seeing the iNaturalist Community Taxon at work, but without specific examples I canât say for sure.
12 Likes
You currently have three observations with conflicting IDs, of which one was made by a curator:
Gattung Gyratrix von Lorain County, OH, USA am 21. Mai 2020 von johnnlavelle ¡ iNaturalist
Neither of you provided reasoning for the ID made, so I cannot say who is right or wrong.
You can opt-out of the community ID, if you donât want identifications made by others.
6 Likes
As IDfanatic already pointed out, three of your observations have a disagreeing ID.
In one, the IDer has helpfully pointed out the feature leading to their ID. If you disagree with this, it is best to post a comment with the reasoning behind your ID (ideally pointing to a specific feature or something like habitat rather than âxyz said soâ).
In another, the IDer has kicked it back to genus. This often happens when an organism is unidentifiable to species from photos. In this case, it would help to state if it has been IDed by an expert in person.
In the third, no reason has been given for the disagreement. In this Iâd recommend just asking the identifier what the reasoning for the ID is. Many are happy to reply.
13 Likes
iNat is based on the principle that observations should be confirmed based on the evidence provided (i.e., the photo or audio that you upload with the observation). It is also based on the principle that anyone can add an ID. Most users do so responsibly, particularly with the sort of small aquatic invertebrates that you observe.
If the photo evidence is definitely not sufficient to determine the species, IDers may add a higher-level ID at whatever level they feel is appropriate. Some people will disagree (there will be a note under the ID indicating that they disagreed) if they feel that the ID is not justified at all, others will add a non-disagreeing broader ID. Such a non-disagreeing ID does not count against the finer ID or degrade it in any way.
Agreeing to IDs or demanding that others do so simply because the specimen was IDâd by an expert is strongly discouraged â even experts make mistakes or may interpret the evidence differently. Nevertheless, it can be useful to indicate who IDâd your specimen so that people have an idea where the ID came from. Most users will want to evaluate it themselves, but they are less likely to disagree or they are more likely to explain their ID or ask for additional photos if they know that your ID was made by someone with some expertise and not randomly chosen.
You have a number of observations that got uploaded as âunknownâ (without an ID) because you mistyped the taxon you wanted or it was not found in iNatâs database. When this happens, users with some general knowledge will add broad IDs to help sort the observations so that they get seen by specialists. These broad IDs are not disagreements with your ID. If the taxon you want is missing from iNatâs database you can request that it be added by going to a parent taxon and creating a curation flag.
18 Likes
Thanks for improving the quality of iNat data by adding expert verified IDs. As others have said, you can opt out of the community ID when necessary. Community identification works well for some species such as birds, but for more obscure species it can be a tiresome and sometimes fruitless exercise, and reliance on experts makes much more sense than just tabulating votes of random people who are sometimes just following computer vision suggestions. Communities get things wrong all the time (just look at election results). Brief discussion of characteristics isnât going to provide what experts have learned from years of experience. But I would at least add a comment when you submit your sighting that it has been identified by an expertâthat should guide any subsequent discussion in a more constructive direction.
A word of caution about opting out in this particular case.
The user in question has a number of observations with a placeholder and no ID because the taxon name was mistyped or not in the database. One source of concern about inexpert users interfering with their observations seems to have been a reaction to other users adding broad IDs to these observations, which was interpreted as downgrading the ID.
Opting out would not solve anything for observations like this, and in fact only create an obstacle to getting the right ID. Opting out forces the system to only display the observerâs ID; where the observer has not entered a valid ID, this means that the observation will permanently be listed under âunknownâ, with no way to fix it unless the observer figures out that the observation does not have an ID. So I think the first step is becoming more acquainted with how iNat is set up and making sure that all observations have been entered correctly and with the intended information.
Edit: I also tend to discourage opting out as the first reaction to dealing with ID disagreements, because I think that sometimes users opt out too hastily without fully understanding the implications. Zooplankton is obscure enough that it does not tend to attract well-intentioned wrong IDs, so most people adding IDs are likely to have some expertise and a good knowledge base. I also find that most users are happy to discuss IDs if asked and are open to changing their IDs if they can be convinced that the evidence supports this. Opting out tends to have an effect of shutting down this discussion because it creates a power imbalance and suggests the observer is not interested in participating in a shared, mutual process of coming to a consensus about how to identify the organism â and this, it seems to me, is precisely one of the strengths of iNat, that it opens up space for people of all backgrounds to interact and learn from each other.
14 Likes
Agreed; if itâs a taxonomic issue, then opting out is not the solution. The OP should research alternative taxonomies to understand how things fit in the iNat taxonomy. I also agree that opting out of community ID should be a last resort. Iâve never found the need myself.
3 Likes
John, are you referring to when your placeholder ID is not selected properly? You messaged me about this, and I did explain.
https://www.inaturalist.org/observations/293150083
When you create an observation but donât fully select a taxonomic species/family/etc, sometimes your text is saved as a Placeholder, but the observation itself still shows as âUnknownâ. You have multiple instances of this.
https://www.inaturalist.org/observations?iconic_taxa=unknown&user_id=johnnlavelle&verifiable=any
For the above observation, I did not know enough about water fleas to confirm your Kurzia ID (though I donât doubt you are correct), but I did add the ID of Anomopod, hoping more educated users would be able to move the identification process further along. When I added the Anomopod ID, it didnât override your ID, your ID was never properly selected.
Whenever there is a Placeholder ID for an âUnknownâ observation, I copy and paste that text in my identification notes so that it is not lost. It is never intended as a disagreement.
9 Likes
Another issue is that the observer may think the person they have consulted is an expert because they have a degree or something, but they may have far less expertise in a particular taxon than âamateursâ. Or they may not be up on recent research.
I run into this from time to time when making an ID correction. The observer will respond that they consulted with a particular expert and this is the source of their ID. I usually recognize the name of said expert and know that while they may be a professional entomologist, they arenât necessarily experts on lepidoptera, and they are unlikely to be up on the most recent research around the taxon in question.
Observers participating in bioblitzes may not realize that the âexpertsâ who are helping out will necessarily be doing their best to cover taxa with which they only have a passing familiarity.
4 Likes
True. In most cases where someone adds a disagreeing ID, itâs because they have expertise to contribute and are right. I have seen few cases (though I have seen some) where someone will add a disagreeing ID out of ignorance.
I think this probably varies a great deal. With butterflies, I frequently see incorrect IDs applied by folks from outside our region. These folks arenât usually âregularsâ, and I suspect they are just folks who are trying to cut down on the âneeds IDâ backlog - though they are often doing this with brand new observations (I usually get to reviewing observations within a day or two of them being submitted).
We could probably use a filter based on how long an observation has been in the Needs ID state (to allow folks who are trying to tackle the backlog to see only those observations that really are part of the backlog)
Iâm not sure weâre talking about the same thing⌠I frequently see incorrect IDs too, but not usually as explicit disagreements. When someone goes to the trouble to disagree, at least in my experience, they usually have a good reason for it. Explicit disagreements wonât cut down on the Needs ID backlog because it then takes even more identifiers to overturn the initial ID.
1 Like
I review several hundred needs-id observations every day, so itâs hard to remember how many cases I see for the different scenarios. You may be right, it may be more a case of folks incorrectly refining higher level IDs than explicitly disagreeing with species level IDs, but Iâm pretty sure I see both. And yes, itâs frustrating because instead of being able to push the observation to RG by simply clicking agree, I have to enter an explanation as to why the original ID was correct, and (often) enlist the help of other experts to push the ID back to where it belongs.
2 Likes