Dealing with Account Deletion

As was noted in the original post, " iNaturalist will always allow users to entirely delete their content from the site , both as a matter of principle…" meaning while the laws are obviously important, we also believe you should be able to delete your content from iNat, and this topic is supposed to be focused on what alternatives should be provided if someone wants to delete their account and how that might look on iNat. Of which anonymization is one, and one that I would endorse.

The user’s data is on our back-ups, so one idea we’ve kicked around is that if they want their account restored they need to start a new account, make a certain significant number of observations or IDs to prove they are committed, then we can bring back the old account and merge them. I like this because it takes the user at their word (they want to delete their account) and also gets across that rash decisions have consequences.

I like the ideas about exit surveys and advice to reach out to help@inat, I’ll make those recommendations.

I’m also going to clean up some of these off-topic conversations and move them to different topics. Let’s really try to keep conversations as un-muddled as possible here.


it’s too bad iNat admin is taking such a hard line on this. I don’t think it’s such a straightforward issue especially when it comes to identifications and i think it will end up doing a lot of harm if it becomes more common than what it is now. It makes me less enthusiastic being a part of the ID community if i know the ID data may just vanish, and i never really did get a response as to the issue of IDs disappearing from the community ID if a user blanket deletes them so i’m still not sure what to do about that. But… as it seems i have no control over it, i won’t say any more than that.


Is it the observations that are submitted with the copyright designation, or the photos/sounds? I was under the impression that it was the latter, but I may well not have been paying enough attention :-)

It is both, but the observation has a separate copyright designation to the media, so you can have tight restrictions on the media but quite open rights on the observation itself, etc…

As kiwifergus said, there are separate licenses for the observation, each photo and each sound recording. I have not seen any license for identifications though. Personally I am happy to consider any IDs I have made as a donation to iNat. Especially where they were correct :wink:


The new interstitial is well-written and is a big step forward. For those people who choose to delete their IDs even after reading it:

Anyone who acknowledges that something “personal” affected how they ID’d others’ content should certainly have all their IDs deleted. Ditto if the user asks to have all IDs deleted because some of his/her opinions on ID have changed.

Otherwise, the user’s departure should not change the iNat ID or its “Research Grade” status. Just add an an asterisk after the ID or RG tag saying:

“This ID was shaped in part by a user who subsequently voluntarily left iNaturalist and asked us to remove their contributions. Additional ID input is especially welcome here.”


some people have 1000s of IDs, i don’t think we should blanket requre deletion of all because they took to abusing IDs in a few recent ones.

the admins more or less said they are going to let people always delete their IDs anytime they want, basically. Which is not really what i personally prefer, but there isn’t really much we can do about it, so there’s probably no point in repeatedly asking them not to. It didn’t seem like something a user groundswell would cause them to rethink. What i hope they would consider is making the community ID ‘sticky’ so if someone deletes their ID and it was a ‘leading’ ID, the community ID remains with a disclaimer that the original ID was deleted, untill someone else adds an ID. Otherwise we have to all ‘agree’ with experts even if we are unfamiliar with the ID, in case it gets deleted.


I think we should agree with the ID, but if we don’t really know what it is also check the box in the DQA that it needs further ID. That way the ID can’t be lost, but it’s not Research Grade yet. Just having Community ID stick risks losing the ID (which may be right) if Community ID changes (in the wrong direction).

1 Like

yeah, but the problem is people then forget to uncheck it, because it doesn’t uncheck after getting another ID. No good way to deal with it really.


Key point here! Pretty sure that some kinds of user actions are not subject to copyright law. To me, that makes a convenient distinction for when content should only be anonymized, versus being subject to outright deletion.


I’m pleased to see the progress on this topic. I started the google thread a long time ago, so thanks for bringing the topic over and providing the summary in the original post. The proposed language looks like a big improvement, but I am disappointed that the iNat team is adopting this stance. I laid out my issues in the google group posts, so won’t revisit them here.

In addition to Cassi’s interesting point on wikipedia, is another place that data can hang around and there is no obvious way to delete a record there.

Andrew’s note is intriguing. I am also not a lawyer but the iNat approach seems like a very expansive reading of GPDR. Take the example of someone who opens an iNat account with a pseudonym and never provides their real name, what ‘personal’ information are you protecting by allowing them to engage in a straight account deletion?

I support anonymization as the best of the current options so that there is some data integrity and to avoid the incoherent discussions that would result from deleted comments.


I’d like to double or triple heart @dkaposi 's post.


Me three. Anonymization is the way to go with IDs.

The GDPR is a red herring. iNat does not keep personal data on people. The exception might be a journal post that included personal information, such as someone’s battle with a chronic illness.

The personal copyrights on observations (iNat no longer copyrights just the photo) give anyone who chose to retain full copyright the right to take back their observations. It would be different if iNat had included in the user agreement a perpetual license to display the observation. But that is not iNat’s philosophy.

Can you copyright an ID? No.


So, if I understand correctly, it would be OK for users to remove their observations but not their IDs? Where would comments fall? I could see an argument for comments being withdrawable, but (as someone previously noted) that could render conversations unintelligible.

1 Like

for what it’s worth i am pretty sure iNat admin’s insistence on making ‘anything’ deletable extends to observations, so there probably isn’t much we can do about it aside from archiving our observations in CSVs or whatever.

I would also prefer if IDs were never removed. And with comments (including comments in IDs), at the very least have the comments remain but delete the text and replace it with something saying the user left. Then people can tell that there used to be a comment there.
It should definitely be possible to remove all your observations, but I don’t think it should be the default.

I just read through the Google Group discussion and I really like the suggestions by Ben Phalan, particularly the ability to remove media and username from all observations but leave the taxon records and discussions (meaning they will be casual). @jonathan142’s suggestions were similar.


A legal clarification:

notyouraveragecatlady, once iNat publishes an observation with a full copyright line, like the one below, iNat cannot refuse the observer’s request to delete the observation.

Observation © nancyasquith · all rights reserved

iNat could probably modify its contract (or lack of written contract?) with members to give it the right to keep future observations on its website forever. But even if iNat changed its whole philosophy, contract law will not allow iNat to retain existing “all rights reserved” observations if the copyright holder wants to remove them.

IDs cannot be copyrighted. Comments can be copyrighted, but usually are not copyrighted. You could have written “Copyright 2019 Kit Howard” after your comment above and it would be copyrighted.

BTW, even if you copyright a comment, brief quotes from copyrighted written works are permitted.


I think comments and IDs on other peoples’ observations should remain deletable individually, but that the account deletion tool shouldn’t provide that feature, and only allow anonymization en masse instead.


I don’t know if this is the same in other countries, but in our copyright law something also has to be “sufficiently original” to copyright it. For instance if you post a comment saying “me too” you then try to claim copyright that would be rejected.


Is there any method to opt in for delayed account deletion as to prevent the instantaneous process of deletion?

Perhaps the users personal information such as journals and profile may be removed from the public during the process to conform with necessary legal guidelines?

A good example of the benefits of delayed deletion may include, if someone deletes their account by:

  • If someone’s account gets hacked, then the user may recover their account within the 30-day period without the risk of losing their data.
  • If someone is off their face with drugs or alcohol and can no longer make a rational decision.
  • A user accidently leaves their account open on another computer.