How to correct thousands of misidentified observations uploaded by a class?

[jasonhernandez74]

@cyanfox

Are we seriously adding a DQA point “Observer is qualified” or something?

The desire to do just that has been a covert theme through many, many threads.

[neylon]

@cyanfox

But it’s the users, not the project. The project shouldn’t have all its observations set to casual, especially since it’s a collection project. If anything it should be the users who get that penalty.

We’re talking about class projects, not major collection projects and not projects that active users use. I have to assume that you have very little experience with these projects. In the vast majority of cases, their accounts are active long enough to upload the requested observations from the teacher and then never log in again. Recently, I ran into a class where the teacher made the questionable requirement to have all obervations had to include a photo ID (one kid used their driver’s license with address visible!). I added a comment to every single observation saying to alert their teacher that this is dangerous and not allowed. Not a single student responded. They don’t care about their observations.

@cyanfox

Also, what if someone made all observations in the Alive Animals project casual? Consider the disruption and damage. Who will set projects to auto-casual? Curators? Staff? Will collection projects only be eligible if they have a member-only rule? Will this even be a thing for traditional projects, given that you can only create them if you have 50 verifiable observations?

Again, this is about class projects. I don’t see why you keep lumping observations that were made by people of their own volition, and students adding what are often duress observations (like: make 10 observations for class).

@cyanfox

If anything it should be the users who get that penalty…Not to subject the user to a penalty that assumes their observations are not worthy…when a user is subjected to the auto-casual penalty, how will their observations get out of casual? Are we seriously adding a DQA point “Observer is qualified” or something?

Which one is it? Penalize the users who do this or not?

[cyanfox]

Of course I oppose the auto-casual penalty. But if it exists, it should be a penalty against users not projects. It shouldn’t be possible to impose it on a project with non-member observations. Nor should it be imposed on the users in a project who actually make good observations, just because of them and therefore their observations being in the project.

And please understand that penalising a collection project that automatically contains all observations made by members is essentially penalising users.

Also, you shouldn’t assume that no student cares about their observations. Some might. The assignment might require that their observations be verifiable or research grade.

And consider that some students continue to use iNat. No, I don’t want iNat to punish users for being students. What kind of a message does it send it we do that?

[clint_tucson]

please help my newbie friend who wonders what CV means…

[DianaStuder]

Computer Vision - when you click iNat for suggestions - what IS this ??

[DianaStuder]

@neylon

include a photo ID (one kid used their driver’s license with address visible!)

this silver surfer wonders why neither the teacher nor the student have ANY idea about online security?!

[astra_the_dragon]

For the last couple decades, corporations have conditioned people on the Internet to give up all their personal information – to abandon any conception of personal privacy. Google knows everything about you, more than you know yourself. Passwords are leaked on the regular. Every person on the street thinks they have the right to record strangers and post that online.

We live under a de facto surveillance state – the horrifying thing is that many people have forgotten that privacy is even an option.

As a zillenial (gen Z / millenial border), I’m one of the youngest who was still taught internet safety measures. gen Z and alpha have been neglected. It’s up to us to teach them.

[grampianshiker]

@astra_the_dragon

Every person on the street thinks they have the right to record strangers and post that online.

I don’t! I only post the non-humans and plants I pass… :)

[neylon]

@cyanfox

Also, you shouldn’t assume that no student cares about their observations. Some might.

I have given numerous examples that I’ve actually encountered of students and teachers uncaring attitude. You’ve countered with “some might”. Do you have any examples? Or is this a case of the theoretical vs the factual? Further, in cases where the teachers actively monitor their projects, they wouldn’t have their projects set to auto Casual if this discussed idea was implemented. As an aside, I have also run into teachers that do take an active part in their students observations and I appreciate those people. Do not take this as me wanting to ban all class projects, but I would like a way to push the bad ones out of the way of the good ones.

@cyanfox

The assignment might require that their observations be verifiable or research grade.

Also something teachers are asked to not do. https://help.inaturalist.org/en/support/solutions/articles/151000170805-inaturalist-educator-s-guide#Determine-whether-you-want-to-use-iNaturalist

@cyanfox

And consider that some students continue to use iNat.

Yep, and these kids are using iNat because they want to, not because of a class requirement (which is what we are talking about here). These are the people I would much prefer to help with ID’s.

@cyanfox

I don’t want iNat to punish users for being students. What kind of a message does it send it we do that?

It’s kind of like the old question about if a tree falling in the forest still makes a sound if no one heard it. Considering that most of these students and teachers are not going to log in again or check their notifications, does it send any message at all? On the other hand, if thousands of class project observations were pushed out of the way, that would make it so that the people who put more effort into their observations (including other class projects) would get their observations checked quicker and might make it so they want to add more good observations. I’d rather help that crowd.

[neylon]

Some teachers heard that stolen photos are an issue that we deal with, and some require that students include some kind of token in their observation to prove that they did indeed take the picture. Occasionally you get some bright spark that tells the students to use an ID card.

Well, you’ve proved that was you’re photo, and I now have your name, address, and license number.

[cyanfox]

Of course it sends a message. People who aren’t students or teachers might see that users are being punished for being students. And some students do continue to use iNat. They don’t all abandon it.

What if this punishment for being a student actually drove a student away from iNat? Like maybe they abandoned iNat after they no longer needed it for school, but if they weren’t punished, they would have continued to use it?

And consider that the punishment prevents the students from interacting with other users on iNat, as casual observations tend to get far less attention than verifiable ones.

And as for your point that teachers are discouraged from requiring their students to make research grade observations, well, they obviously do all kinds of things that iNaturalist discourages, and you know that. Hence why you want their students to be subjected to the auto-casual penalty.

Not to mention, I guess that even if a teacher doesn’t require students to make research grade observations, the student might want their observations to be research grade anyway. Unwarranted species IDs and blind agreements definitely seem to be big issues on iNat.

And why is no one telling me how an observation is supposed to get out of casual if its uploader is subjected to the auto-casual penalty?

I see no reason why we shouldn’t just treat ignorant students the same as any other ignorant users. Why, because they leave after their assignment is done? Maybe they’d stay if they were treated with less hostility. Besides, there might be other users who aren’t on iNat for long.

[DianaStuder]

We have same issues around the City Nature Challenge.

We are here for the weekend … scatter some random (isn’t this fun!) wrong IDs … and disappear.

But some return next year. And few settle in with us and become dedicated (we don’t DO that here) iNatters.

Maybe we need a ‘kiddies playpark’ for new iNatters ? To be promoted to ‘big school’ when they pass the ‘admission test’ ? AKA The Onboarding Tutorial.

[ShotShot]

Assuming the site would care a bit for benevolent identifiers, it could just provide a sandbox for e.g. students and teachers, plus an option checkbox in users’ profile “Do not display sandboxed obs”.

[neylon]

@cyanfox

Of course it sends a message. People who aren’t students or teachers might see that users are being punished for being students.

I don’t think anyone said punish students. The argument is to have a way to get projects out of the way if they are consistently putting up poor observations and the teachers aren’t involved with the project that they created.

@cyanfox

What if this punishment for being a student actually drove a student away from iNat?

Drive away the the ones uploading stolen photos, I-don’t-care photos, and slurs, which is what you get with unsupervised projects like we are discussing? Oh no, what would we possibly do without them?

@cyanfox

casual observations tend to get far less attention than verifiable ones.

Part of the problem is even Needs ID observation are stuck in the bottleneck. Right now I’m working through western bees and running into two, three, four year old observations that never got even a check. These class projects add to the congestion. Which again brings back to my point of preference to ID for people that are here by choice rather than duress.

@cyanfox

for your point that teachers are discouraged from requiring their students to make research grade observations, well, they obviously do all kinds of things that iNaturalist discourages, and you know that. Hence why you want their students to be subjected to the auto-casual penalty.

@cyanfox

And why is no one telling me how an observation is supposed to get out of casual if its uploader is subjected to the auto-casual penalty?

I did, and so did others back when the ability to flag projects was first brought up:

@neylon

It keeps all observations in the same place for ID purposes, it allows the students to upload, but it doesn’t get in the way of the rest of the observations which may have had a little more effort put into the observation. In this scenario, the teacher could get the flag lifted by getting on the site and commenting on the flag.

And if there is a student that is genuinely interested in what they observed, then maybe they will question their teacher and maybe the teacher will look into it.

@cyanfox

Maybe they’d stay if they were treated with less hostility.

Another theoretical vs factual argument? The facts are that most of these students are gone before their observations get any ID’s (seriously, some accounts are active for less than a week). And based on the fact that I get no responses to questions and feedback I put on observations, I have to assume that they aren’t reading my comments either. If memory serves, I’ve had less than six teachers actually respond to questions on strange observations, and can only remember two right now.

@cyanfox

Besides, there might be other users who aren’t on iNat for long.

And maybe those users would be on longer if their observations weren’t stuck in the bottleneck.

@DianaStuder

We have same issues around the City Nature Challenge.

That is one I think every identifier has a love/hate for.

[DianaStuder]

Actually that is the intended purpose of Seek.
That should be promoted as a better alternative to use for ‘this year’s assignment and done’.

[lynnharper]

I know we’ve all discussed this before, but we really do need to figure out a way to attract and keep more identifiers - and figure out how to teach iNatters how to take photos that increase the possibility of a confirming ID. I don’t have any new ideas on how to attract identifiers, but when I think about what’s involved in teaching people to use iNat (the on-boarding process), there are three parts that come to mind:

• The mechanics of how iNat works
• The mechanics of taking good enough photos (or recordings)
• Enough about all the many possible organisms out there such that an observer’s photos have a chance of being IDed to Research Grade.

Now, the first two parts are easy enough to teach, if tedious and often frustrating for new people to learn… The third part, though - how do we teach people around the globe enough about, say, globemallows in Arizona that they know what parts to photograph? I chose globemallows as an example because while I’m a decent (but not professional) botanist in the northeastern US, when I’ve visited Arizona in the past couple of years, I made photos of globemallows that weren’t, apparently, good enough to reach Research Grade. And that was with my obsessively reading and re-reading field guides ahead of each trip. Now, I’m fine with not having my observations reach Research Grade precisely because I know how hard it is to learn how to photograph every single kind of plant for ID purposes, but new iNatters, whether students or not, often have no real way to know that. So, how do we overcome that obstacle to effective use of iNat?

[neylon]

@lynnharper

Now, the first two parts are easy enough to teach, if tedious and often frustrating for new people to learn

This is unfortunately an issue. There are aspects of the site that take time to learn, and are sometimes difficult to explain to people.

@lynnharper

The third part, though - how do we teach people around the globe enough about, say, globemallows in Arizona that they know what parts to photograph?

I think this comes naturally with interest and having dedicated identifiers. We’ve had this a few times with bees. An example is few of us were working Bombus vagans/sandersoni which can be unidentifiable from most of the photos we get on here, but we got enough engagement on those observations that we are seeing a lot more observations where the observer took the time to get the identifiable marks.

This was kind of a cool moment where the identifiers showed the observers what to get, the observers gave more to work with, and now with better practice available, a couple of the specialists are even identifying observations that a couple of years ago would have been stuck at Subgenus.

@lynnharper

we really do need to figure out a way to attract and keep more identifiers

Yes, the issue of class projects and student accounts would be much less of an issue if there were more identifiers.

[danly]

@lynnharper

I know we’ve all discussed this before, but we really do need to figure out a way to attract and keep more identifiers

And this needs to happen outside of the forum. I was making observations for a couple of years before I joined the forum, and it was only then that I realized that I could (and should) be making identifications.
What if for new users, a certain number of observations or a certain amount of time triggered a notification to the effect of “It seems like you are enjoying contributing observations to iNaturalist. Here’s how you can become an identifier,” with links to appropriate tutorials?
Right now the onus is on the user to seek out that information, the existence of which most are probably unaware.

[lynnharper]

I’ve thought about a project for new iNatters in my region, to which I could invite them one by one, but I haven’t had the time to implement such an idea. I think it’s do-able, but would need a team of dedicated people to make it work.

@danly, I first got involved with iNat because I was directed to for my job, so I had reason to read all the help documents and poke around a lot on the website. That helped me learn about most of the many functions iNat has. I still don’t know much about curation (and I’m not the best person to get involved there) and I can’t always make the API do what I want. However, I do agree that learning the mechanics of iNat has a bit of a learning curve and, unfortunately, unless staff have the time and motivation to implement an automatic technique like you suggest, we active iNatters are just going to have teach people one by one.

[DianaStuder]

It shouldn’t be possible for iNatters to upload
hundreds
then thousands of obs
without being made aware that they ‘need’ to ID or annotate at least (if that better fits their skill set)

IDing at your level, definitely focuses your mind on better photos - which adequately show relevant and useful field marks. You want us to ID a bird in the sky - insect next to your foot - bit of dirt on your lens? You try it …

Otherwise the balance tips to Unknown plus Needs ID plus problem IDs that need resolving - will swamp the good / correct IDs.