Dealing with Account Deletion

Was there a decision on the idea of an anonymizing option?

1 Like

Sorry, I should have addressed that. No decision, I still have to write up a proposal for the team to discuss, as this is a much more complex idea.

4 Likes

Does the first step have a requirement to enter a password? I use my account on a number of computers, including when visiting people, and if I use it there often I will leave the account with the “remember me” option. It is possible someone might access my account (hijack), and while I don’t think it is likely, I can accept having to mop up a mess that they create by making inappropriate comments or deleting a few observations, but if they could delete my account wholesale without a password entered… ouch! Sorry, I catastrophise too much :)

I won’t start a new thread, but I was more thinking along the lines of Friday night and a few too many beers, take something the wrong way, or let a fellow drunk browse from your account and they “jokingly” delete your account… maybe I hang out with the wrong people :thinking:

5 Likes

I hope so.

for these and every other situation

2 Likes

Sorry, here’s the example:

-i found a weird bug in Costa Rica
-Someone identifies it
-Since i am not a bug expert (or even amateur really) and had only been to Costa Rica once. I don’t even have a clue about the profound insect diversity of Costa Rica. Thus i do not ‘agree’ with the ID.
-That person later deletes their account over some sort of dispute over pheasant nomenclature
-The ID is taken away, and so my ID goes back to ‘bug’ and I lose that ID.

Unless i’m really confused, which is possible, that recently happened with a bunch of IDs of a different taxa and i don’t think that’s how the community ID should work. I think it should retain the bug ID even if the person IDing it is gone. Either through an ‘anonymous’ ID or it just keeping the old community ID until someone says otherwise.

The only alternatives are, i risk losing a bunch of data, or i agree willy nilly with every bug ID in Costa Rica and create a bunch of fake RG observations that shouldn’t be, or i agree but hit ‘more ID needed’ then forget to uncheck it because i have literally 35,000 observations and so it sits there and every person in the country who looks at bug ID has to see the same already identified observation.

6 Likes

This seems like a sensible interim step, although I agree with Charlie that it’s sad to be reminded that people can delete so much content. I would suggest to add two things here: first, an additional sentence to remind people that if they are having problems with other users or the site they can contact the help email address first to try and resolve it. Second, a password confirmation step like others have alluded to. This is a standard step in doing drastic things with an online account, and could help avert the “drunk friend” scenario mentioned by kiwifergus!

10 Likes

any drunk person who deletes my account is an enemy 4 life now

4 Likes

Sorry I didn’t make this more clear in my post, but this interstitial is just a first step that seemed pretty easy to implement. Adding anonymization would entail more coding, more thought about exactly how it would look, design work, etc etc so it will take more time. But we will discuss it.

4 Likes

The choices as written so far are (1) editing one’s settings, (2) muting, (3) deleting, or (4) doing nothing. How about a fifth option with a link to send the reason for deletion? Sometimes a chance to vent, or just the act of typing the reason, relieves the pressure to act right now.

8 Likes

If not at the front-end, I think a “back-end” cooling off period would at least be advisable. That is, make it look as if the account is completely gone to the front-end user (and to search engines, etc.), but retain the data internally for a period of time, so that users can still change their mind (or, maybe, recover from a malicious deletion if someone else had access to their account).

And yes, glad to hear that anonymizing options are still under discussion. And hopefully also selective content (like own observations, but not IDs or comments on others, etc.)

10 Likes

besides being useful in itself, this would be another way to slow down the decision

2 Likes

This wouldn’t be terribly different from how Discourse does things. There’s a period of time before content is permanently deleted, just in case something gets processed accidentally. It could also help regarding “rage quit” responses.

Glad to hear there are further discussions as well. I look forward to hearing the results.

4 Likes

i think this is a reasonable interim step, but i think you should also add a mandatory exit survey that captures reasons why people choose to delete their accounts. you should capture those reasons, along with general metadata about the deleted users (sign up and delete date, # of observations created, # of identifications, # of comments, # of messages, etc.).

the standard list of reasons to choose from should include things like “i spend too much time on iNaturalist”, “i argued with too many people on iNaturalist”, “i signed up for a class, and i don’t need it anymore”, “too many notifications”, “prank (this might make someone really mad)”, “stalking”, “i died”, etc., in addition to an “other” option. you should also have a comments box that allows them to offer a more detailed explanation and maybe a place to capture contact info so that you can contact them for more info on more complex situations.

you can analyze this data and get a better path for how to proceed, i think. maybe the solution isn’t to change the system per se. maybe there’s something about the user experience that can be improved to better promote retention.

9 Likes

I’m pretty sure I know exactly which knowledgeable curator you’re referring to, as several of my own contentious comment threads on nonnative conifer observations are now completely lacking in context. It is unfortunate that so many ranges are now of poorer quality, but having experienced and observed others experiencing that user’s cavalier attitude to marking organisms “not wild” in even very ambiguous cases, never seeking additional information from the observer, and resistance to considering additional information from the observer, I have to say I don’t miss their behavior. As someone who often seeks out clear non-planted examples seeding in from plantings, coming back to find all that work lumped in with those arboretum plantings and rendered “casual” was always very frustrating. They always seemed to justify themself by the size of the curation task at hand, but perhaps if they had been willing to work more respectfully with others they would have found more allies in the task and they’d still be here. Some of their DQA assessments on my observations when I was new to the site were obviously valid and educational, it’s not as if I don’t appreciate that- but my other interactions with them left me unsurprised when I noticed that they had apparently departed, though I don’t know the exact circumstances that prompted them to do so.

I agree wholeheartedly with your ideal data deletion format. I still have comments I made with an @ to their original username, which now just redirect to a 404, and that seems to be a bit of an oversight in regards to GRPD compliance.

3 Likes

Just going to take a stand here and say sure, maybe the affected observations are small. But that doesn’t mean something shouldn’t be done. For more obscure groups expertise is critical and not everyone likes to agree with expert IDs, especially if they don’t know on their own. The small number of affected obs may well be mostly cases where the user doesn’t respond/agree, and then that ID and comment source is lost.

I don’t see why we can’t have stand-in IDs and comments – i.e. the IDs are kept, but the username associated with them is removed or made anonymous. Is that still against the laws?

11 Likes

4 posts were split to a new topic: How are iNaturalist Forum and iNaturalist accounts linked?

4 posts were split to a new topic: Advice for using iNaturalist Data

I am not a lawyer, but from what I have read, the GDPR is about the individual person and the fact that data about you is valuable and should belong to you, not a third party. But an ID added to an observation is not information about you, it is information about the organism depicted. The only issue is that your account is associated with the id. Thus deleting the account association is fair. The part of the record that is not about the user should belong to iNat.

That is for IDs. Observations are more tricky. The problem is that if it was submitted under one of the CC licenses that requires attribution, you cannot attribute it. No idea how to solve that.

5 Likes

I’m liking the ideas here, since I think people should have the right to disappear, especially since there’s tons of reasons a person may have to do this, especially for reasons of personal safety.

I do like the idea of a ‘cool-down’ period too, however. I remember on the Google forums someone was trying to get their account back up, due to the shared account being deleted by a student. People get hacked, computers and accounts get shared, and if someone thinks this is a way to do damage, they won’t stop at the warning.

5 Likes

I don’t think I saw this already but apologies if I missed it, I’ll delete if I did: Can there be an email verification required before the account deletion goes through? That would be in addition to a short “cool-down” (front or back end) but the requirement of having to verify deletion through an email would possibly exclude or slow a malicious deletion.

6 Likes