I think, becasically people need to consider records without photographs are accepted and stay as causal, there is nothing wrong with a photograph staying casual if you cannot ID it to a species level.
In my humble opinion, as a newbie, and having been guilty of said “agree” identification, I don’t think it is usually malicious. The AI guiding suggestions is really impressive, and I have used it to suggest ID’s and to “agree” at times. While birds, butterflies, dragonflies etc are usually quickly identified, other taxa such as wasps, mushrooms, etc are woefully underidentified, whether as a result poor quality images, or not enough experts I don’t know. If the goal is to identify and gather information regarding all species, then we should embrace those who are trying to identify, whether expert scientist or not, while gently correcting the excesses.
2 Likes
One solution would be to post the details you used to field identify the species in the comments section. That way other identifiers know that you are very familiar with the group and can use the information to decide if they want to agree with your ID or not. If someone says a moth identification is based on dissection, I won’t question the ID. Without the ID I may be tempted to change the id to genus if I know it needs dissection.
5 Likes
I think the “genus-level ID” discussion needs to be clarified a little. A genus-level ID can come from two directions: It can result as a lower level ID for an observation initially placed at family or higher level (or in an erroneous genus). That is obviously an improvement that few would argue with.
The other use is placing genus-level IDs on observations already at species level–a “downgrading” ID, if you will. As already much discussed, this can come from either a discussion/disagreement on the species ID, or from the nebulous, “I don’t know but I’m sure that it is in this genus” basket.
That “I don’t know” category of genus-level IDs itself is an amalgam of opinions that run the gamut from a well-informed, “I’m knowledgable about the species in this genus and I still can’t decide what it is,” to the less-informed, “I don’t know species IDs in this genus but I’m sure it’s in the genus.” I have long questioned the utility of the latter use of genus-level IDs and I still don’t understand how iNat’s algorithms take these into account. (No need to try to explain it to me! I’m just too dense!)
The assumption behind this comment seems to be that the observation wasn’t already placed in the correct genus (whether right or wrong at the genus or species level). Certainly adding the proper genus for something placed in the wrong genus or only at family or higher level is useful for this purpose. From the opposite direction, and for those of us who are subscribed to certain genera in which we have an interest, we will automatically see any/all species-level IDs that come up along with those at genus level. Even without subscriptions to a taxon, there is no need to place a genus-level ID to “notify interested taxon experts” unless the observation is totally misplaced (like a mayfly I recently reviewed which was IDed by iNat’s AI as one of my favorite moth genera!). The iNat Explore search function is perfectly capable of collecting all relevant observations for someone interested in a given taxon. Only in the grossest of misidentifications will the latter search effort fail to find an observation of interest.
2 Likes
It must be frustrating when your species ID’s are correct but get backed up to genus. There are things you can do to help avoid that.
Realize that we identifiers can only base our ID on your photos plus any comments you write. So write comments. Explain what you observed that is important but can’t be seen in the photo. (Behavior? Dissection?) If you know there are two similar species but only one occurs at this location, write that down. If you realize you need to write this only after your photos get ID’d to genus, write a comment explaining why you’re sure it’s that species. In a case like that, I would withdraw my genus ID or change it. Even if the identifier doesn’t come back and change the ID as he should, someone else coming by will understand why the species is what you say it is.
There are two ways to take the ID to genus. In one, I say I disagree with your species identification. In the other, though, I confirm the genus while indicating that I don’t know what species this is. Theoretically, if I don’t know I should choose the second way, though I’m sure I don’t always do that.
12 Likes
Nobody should be bumping back to genus unless they see evidence it is NOT that species, OR they know the wider community would agree with them in lieu of an absent or non responsive identifier (ie it should not be “irritating” anybody in doing so!)
“Agree” identifications are not problematic per se, it is when they are made and the identifier becomes inactive or non-responsive. Someone who identifies something “wrong” but “corrects” their ID when it’s shown to be wrong are very welcome participants!
3 Likes
Like your comments. We seem to think alike on many things… including spending time on iNat rather than computer solitaire!
2 Likes
iNat uses a reputation system without calling it one, but it is poorly quantified and maintained. Firstly, one’s profile remains hidden until they submit “several research-grade observations.” As I joined here mainly as an identifier and not an observer, I think I passed 6000 arachnid IDs before my profile could say “I like spiders.” Secondly, it seems to become a curator, you email them and ask for it, and they they decide if you have used the site enough.
And that is precisely how reputation on StackExchange works … reputation is build upon activity on the site, and the activity includes things like creating posts, correcting posts, answering questions, fixing mistakes, getting ‘likes’ for your posts, getting ‘likes’ for your comments, getting ‘likes’ for your answers. It does not use external qualifications of any kind. The system works quite well and those individuals who grow in reputation on StackOverflow are people who make legitimate and devoted contributions to the site and it is quite difficult to ‘game’ that system because everyone is effectively watching and up-voting or down-voting your activity. As one’s StackExchange reputation grows, they are able to do more things like edit people’s posts (which, again, are scrutinized by moderators) or close the comments on certain posts if needed.
On iNat I can change the top-page photo of a taxon without anyone saying a thing. But it seems to make a correction to the taxonomy, one has to be labeled a ‘curator.’ And as other people have mentioned in forums, about half of the curators have 0 taxa curated.
There are taxonomy-tree changes that should be made but curators do nothing about my submissions. I’d like to clean up some geographic regions… region 83275 and 122168 have the exact same name with slightly different shapes. And 64971 and 64972. And probably tonnes of other examples.
There are plenty of 3-4 year old observation submitted by users who logged in once or two and an observation might have multiple photos of different species and commenters will ask that they separate their photos into different observations, but they never will because they abandoned their account years ago. It would be nice if a person who knows what they are doing (i.e. someone of high reputation) could make that change to fix things up.
With an automatic reputation system with increasing moderation power, these things (taxa curation, region corrections, improper observation posts) would get resolved.
Not all curators work on doing taxonomy management. Some do, others work on managing copyrights, user disputes etc. Having no flags on taxa resolved does not indicate a lack of contribution as a curator.
You have a single unresolved flag. As the curator responsible for the area that it sits in, I guess if you want to call me out in public for not acting on it, or take your complaint to the site management, then I guess feel free to do so. The flag itself is unresolved (you can see I have commented on it) because it deals with a larger question than just the specific topic of the flag. Specifically how the site should try and deal with taxonomic levels not contained in the references we link to.
I have had discussions with both other taxa curators, and the other spider curator about how to deal with this issue, including some technical back end issues, and think it is best to try and get a consensus and plan rather than each taxa curator just working ad hoc to their own views. Yes,the superfamily you flag is likely invalid, but before moving hundreds of things, I’d like to have that consensus reached.
I can’t see a workable solution where every user can change taxonomy. I’m assuming your plan would also mean turning off all the locked taxa (birds, mammals, reptiles, fish, odonates, spiders etc). Having ‘open access’ to changes created a big enough problem just with regards to edit wars on common names that it had to be turned off. I can’t even imagine the level of disruption of allowing ‘the US bird people’ and the ‘European bird people’ to fight out changes to taxonomy or many other examples.
6 Likes
It isn’t other users who control the effects of gaming on StackExchange sites. In fact, many of the most common abuses stem from up-voting and down-voting (e.g. voting rings, sock-puppet accounts, serial voting, etc). It actually requires a huge amount of code operating behind the scenes to police the system effectively. The moderators would quickly become completely overwhelmed without all the automated checks. It could be argued that the reputation system is actually the ultimate cause of most of the gaming on the site - because It creates an artificial currency for “buying” privileges that’s hard to accumulate legitimately, and this then sets up the conditions for cheating to become profitable.
4 Likes
I really want to know whether or not posting un-needed “Agree” IDs, multiple times after the ID is already Research Grade, is considered bad user behavior or not.
I can give a specific example. There is an invasive insect that is now ubiquitous in my area. It is extremely distinctive-looking and there is nothing else within thousands of miles that looks like it. So it is very easy to identify. For this species, many users will pile on with unneeded identifications even if the same observation has already been IDed a dozen times or more.
I don’t understand why they do this. Are they trying to create these IDs as a substitute for a “favorites” list, and if so, why? Does it offer some advantage over the existing “favorites” function? Or are they really doing it just to get on the “leaderboard”? Unquestionably, the users who are at the top of the “leaderboard” for this species are people who have done thousands of these these 'superfluous" IDs. At least one of them appears to also be a curator.
Is this a behavior that iNat wants to discourage, or do they not care, or does it have a valid purpose that I’m not seeing?
2 Likes
It depends on how you define bad behaviour. There is absolutely nothing in the formal rules or terms of use of the site etc that prohibit it. Doing it does not break any rules in any way.
There are arguments to be made on both sides of the question if it is necessary. I’m not going to weigh in on those arguments, because frankly it wont impact or change anything.
3 Likes
As made clear in the comments on a recent blog post and companion thread here, people’s motivations are varied and often well-reasoned.
5 Likes
I do this sometimes with the Spotted Lanternfly (Lycorma delicatula). Since this behavior has been brought up & frowned upon multiple times before, I’m trying to get away from adding an ID to every observation. I used to do this w/ this species because (for some reason) I’m super fascinated by it. As you know, this insect is particularly dangerous and is arguably the worst invasive insect in the last few hundred years.
I like to view each observation to see how the species has been spreading and to see any phenotypic variation that may be present. Additionally, I like to monitor any large range jumps where an individual was observed far away from the quarantine zones. In these scenarios, time is of the essence, so having extra pairs of eyes to ID, comment or message the user could mean stomping out a potential range expansion.
I also took this as an opportunity to annotate each observation because I personally think it’s pretty cool to have every observation of a species annotated! The life stage graphs look very cool when all are annotated. These days, I stick to sorting by unannotated RG observations, adding my ID and adding an annotation to ensure all IDs have one.
Hopefully this makes sense! My intentions are not to run up my stats. I don’t see the point in that anyway. I just really like this insect and like monitoring it. In my opinion, it’s always good to have an extra person look over an observation - they may notice something that went unnoticed before!
13 Likes
I’m going to close this topic as it is quite broad, and some of the issues brought up recently are probably deserving of their own more specific threads. Thanks everyone!
2 Likes