Timed Suspension Discussion

Curators
1

This is a discussion area for release of and guidelines for timed suspensions on iNaturalist, you can read more on our blog post and help site.

This is the first time we’ve ever had guidelines for suspension, as well as the ability to easily suspend an account for a specific period of time, so we don’t know what will work and what won’t, what needs to be tweaked, and what gray areas need to be better defined. For example, if user are suspended for a day for, say, copyright violations, do they stop that behavior or do they continue it? What’s working and what’s not?

This is for discussing the functionality and guidelines, and how they’re working in practice. If you find a technical bug, though, please make a new topic in Bug Reports.

A few more rules for this discussion:

  • unless something needs urgent fixing or changing, please give this a week or two before giving feedback about whether suspension guidelines are working or not.
  • please don’t share specific usernames, flags, etc. If you need to discuss a specific situation, tag me and/or other curators on the iNat flag.
2

Thank you iNaturalist team for this good step forward.

3

Thank you so much for establishing this guidance! Hugely appreciate it.

I have one question. I don’t see AI-generated images among the list of infractions — would repeatedly submitting such images fall under “Continued activity that reduces data quality (such as adding many poor IDs)”?

4

That really needs to be made explicit IMO. People posting hundreds of AI generated images should be suspended (I encountered this recently).

5

I think it definitely does. Same as intentionally adding bad DQA votes, annotating intentionally incorrectly, etc. I think the key for something like this is that the user generally needs to have the issue and how to fix it explained to them, and ideally a warning that they may be suspended. If they continue without changing their behavior after that, suspension is warranted.

6

Wonderful to finally have this functionality! Thank you iNaturalist for bringing Curator support to the next level.

A couple of suggestions for the table of Suspension Guidelines in the help topic:

  1. Most, but not all, of the Infraction types specify at what stage any Curator status of the offending user should be revoked. I would like to see this listed consistently for every kind of infraction (even though it’s something only Staff can implement). This is currently missing for Harassment, Copyright infringement, Intentionally adding false content, and Continued activity that reduces data quality. (imho, any of the infractions listed in the table indicate to me that someone is not yet ready to be a Curator on iNaturalist.)
  2. For one of the infringements (Copyright violation), there is guidance to “Add a moderation note that you warned the user.” This should be made general best practice for all moderation actions taken under these guidelines. Even when an action is automatically reflected in Moderation History, Curators should still look at the resulting history and make sure that any other Curator reading that history will have a clear understanding of where that user is in the escalation process, and should add any needed notes to clarify. Otherwise different curators could end up working at cross-purposes. It shouldn’t be an expectation that a Curator taking some kind of moderation action is then obligated to monitor that user indefinitely to detect further violations. Clear moderation notes are necessary to allow the Curator community to share that load in a coordinated way.
7

Small note. The tables are not as mobile friendly as they could be. Marking desktop mode in the browser settings makes it much more readible on mobile.

2689
26891080×1461 173 KB

8

This is helpful, thanks.

Three spontaneous thoughts.

  • First, second, etc. incident for copyright infringement might need a workable definition. Users sometimes upload 10-20 or however many copyrighted images at a time. They may not know any better. That probably shouldn’t count as 10-20 incidents. So how do we count then? Days? Maybe after 1st, 2nd, etc. warning? That would need organized moderator notes, as jdmore notes.
  • AI generated images are missing from the list, are they still too fresh? I could see that, given jenna489’s recent thread about AI flags vs. DQA votes. Might still be worth mentioning though. (edit: well now I see Jenna’s post above.)
  • Lastly, it’s about the “climate” of all this. There was a time, and there were some curators who would at the slightest opportunity threaten unsuspecting users with suspension. Thankfully this has improved quite a bit in recent times. There are some curators now active who make an effort to talk to users as they would to equals, ask questions, offer help, and the threat of suspension is no longer the first thing that comes up. In my experience here and elsewhere on the internet, users generally respond in a more productive way to this.
    Looking at the How to suspend an account page, I’m concerned that the actual moderating is going to decrease again and mechanical application of all those rules and tools will increase. It’s going to feel unwelcoming for a lot of the less experienced users.
9

This is great to have! Thank you!
One small question: What is the difference between an indefinite suspension vs a permanent suspension, and an account ban vs a ban from iNaturalist?

10

These tools and refined guidelines are great to have, especially for us newer curators. Thanks for putting in the work, and to many long-time curators for asking for these changes over the years.

I like the new norms for copyright violations.

I also think this should be part of the reading material for new curator applicants. I’m thinking even an email (if that’s possible) reminder to all current site curators would be a good idea?

11

I like the concept of the timed suspensions but have concerns with some of the specific guidelines, specifically that they will result in harsher treatment of minor offenses and more lenient treatment of major offenses, especially threats.

In the case of minor insults, a 1 day suspension is recommended for a second offense. In the past, arguments that turned personal could often be deescalated by reminding people to attack the argument not the person, and hiding offending comments. Sometimes this would cause people to get upset that they were in trouble, and telling them that I am just reminding them to be civil and not doing anything to their account would in many cases be taken positively and lead to deescalation of the conflict, I have found that people are more open to recognizing that their comments were uncivil if they are not defensive about a possible suspension. Now, if the user in question has any past record of removed insults, even minor ones, I cannot honestly tell them I’m not suspending them without waiting for another curator to agree on deviation from suspension guidelines, which delays deescalation, allowing things to escalate further. I think wherever possible we should prioritize deescalation over punishment, and would prefer to see the suspensions for minor insults be treated more as maximums than defaults.

My biggest concern is the conflation of major insults and threats. A threat is defined here as “any content that indicates an intent to harm another person”. This is a fundamentally different thing from an insult, I have a really hard time seeing how a 1 day suspension is appropriate in any case where someone has stated intent to IRL harm another user, and I don’t think waiting for another curator to agree to deviate from guidelines should be required here. A non-joking statement of intent to do IRL harm is not just an insult but in many cases a crime, and should be an indefinite suspension pending a staff decision, similar to sexual content. Under the current guidelines a user is allowed to come back to iNat 2 weeks after the 3rd threat they have made. A serious threat should also lead to loss of curator status on the first offense.

Last but not least, I am concerned about the revocation of curator status for second offense minor insults and non-hate discrimination. I’m glad to see that curators are no longer allowed to develop a pages long moderation history of antagonistic behavior, but I fear an overcorrection where people who are fine curators 99.99% of the time, but who very rarely minorly lose their temper or talk down to someone, end up removed after enough time passes that 2 incidents accumulate on their record, leading to attrition of longtime curators.

Also, should there be a limit to how long ago an incident counts toward the total? For example, a 30 year old makes a single insult. Is this a second offense if they also made an insult when they were 20? 15?

And then there is the question of how to make sure users have received a warning, it is very easy to miss comments and tags, but then there are user who don’t receive messages

The selection of suspension reasons from the drop down menu also creates a possibility that someone will be suspended with the wrong reason given, similar to the misclick IDs where Black Carpenter Ants are IDed as Black Carp (a fish). The ability to edit suspension reasons largely addresses this, I just hope that users are notified when their suspension reason changes, I imagine a sockpuppeteer being very confused by a suspension for sexual content, for example.

One more thing, this was not widely disseminated and I found it out from another curator’s comment on a flag, some sort of notice needs to be sent out to curators.

12

I also have concerns of this and planned to bring it up at a later date. I think overall this whole release is good. But i can easily see a ton of people getting suspended over little things like “minor insults” and I can easily see many active Curators getting their curator status removed.

I think part of being human is having emotions, and occasionally emotions can run high including to the best of us.

Time will tell how these new guidelines preform.

13

I think part of being human is having emotions, and occasionally emotions can run high including to the best of us.

Well said, revoking curator status based on the accumulation of 2 incidents of behavior, even if that behavior is neither serious nor the person’s typical behavior, could end up functioning as a limit on the duration of curatorships, leading to less experienced curators (and less curators in general).

Now there was a problem with curators whose typical behavior was antagonistic not getting removed, but that can be fixed without automatic removal for 2 minor incidents ever

14

First off, I think this is a very good change. I am sure that there are kinks to be worked out, but I am really hopeful for this. My initial thoughts and questions (some of which echo others):

Why is “Add a moderation note that you warned the user. “ only for copyright? Other infractions have instructions to warn, so I think this should be consistent.

“Human nudity” may need a definition. Some relevant considerations - is topless nude? Front vs. back? Male vs. female? Does it matter the age? There are varying cultural norms around what is considered “nudity” worldwide, and I’ve seen quite a variety of what users will flag as nudity. The guidelines currently equate human nudity with sexually explicit human content and pornography in terms of how they are treated which are very serious offenses. One real life example of a difficult area was a parent posting a photo from their yard/garden that had one of their young children without a top on. The parent saw nothing wrong with this (and I think seemed a bit offended by the idea that they might have posted sexually inappropriate content), but the flagger was sure that the photo was a serious issue. Having a clear definition here will help avoid confusion.

The guidelines recommend that a warning be “a civil comment or a message to an account with an explanation of why their action(s) violated the Community Guidelines/Terms of Use and what consequence will follow if they continue to violate guidelines or Terms of Use.” I think this is good, but I would suggest that the warning be a comment whenever practicable. The content of a message can’t be verified/a second opinion given by anyone other than staff and the two people involved. A comment is much more transparent. That said, I know that there may be times when a message might be preferable (some users get really defensive arguing in public, but are more receptive to correction when in “private”, etc.)

I agree that revoking curator status for second minor insult seems like it might end up catching a lot of people.

Can we clarify if there is/isn’t crossover between the infraction categories? For instance, if a user gets two minor insults (and thus a one day suspension) and then gets a major insult would this be a one day suspension (first major insult) or a one week suspension (which is what a third minor insult would get)?

How is indefinite suspension different from a ban? An indefinite suspension can still be appealed but a ban is a final decision?

I appreciate the desire to make everything more transparent, but I also worry about the amount of “paperwork” that some of this will involve. Curators will need to check moderation histories more now. This was of course good practice already, but now it will essentially be mandatory for any flag to see if the user has any previous infractions that would lead to suspensions/other consequences. Curators will also need to make notes whenever they flag for copyright infringement and warn users. This can really add up (I am thinking of currently how long it takes to do flagging and then hiding, entering reasons for each, then resolving flag for content) which really slows things down. As a one off, not a big deal, but during CNC or similar it really adds to the mental wear of moderating. I wonder if there could be some display/functional additions to the moderation history that could make this easier.

Example current moderation history for reference:

Moderator History
Moderator History2638×706 45.3 KB

There’s room for a few more relevant columns in there. Here are my suggestions:

The “Actions” entry “View” currently leads to the flag itself. Just make the “Flag” text under “Type” the link to the flag to save an entry.

Add a “+” button or similar to allow curators to add context/columns to an entry. Specifically the type of infraction and whether a warning was given. Sometimes the type of infraction is obvious from the flag (copyright), but other times it is not (as in the case of the “inappropriate” above which could be almost anything. A dropdown for “Infraction” that allows a moderator to choose some of the same things as the suspension dropdowns would really help keep this organized and clear. A similar dropdown (or maybe a button and popup?) for whether a warning was given would help make entering this info quicker and standardize it. It would also make it readily apparent what specific infraction/flag the warning was for. If there are users with multiple issues, this could be very helpful. There would probably be a need to have some type of history at least for who entered/last edited like a taxon status.

At some point, it might help to have a definition of “major” insult. There is currently one for minor. Personally, I tend to consider the default for an insult to be minor until my brain thinks “whoa, that’s uncalled for/over the line”. But people (users and curators) will have vastly different opinions of what major and minor are in this context based on all kinds of personal factors. I personally find the specific examples in some places in the guidelines to be quite helpful at adding some consistency. For instance, the Hiding Comments guidance:

"Examples of comments that should be hidden:

  • Hate speech
  • Insults, including belittling someone about their knowledge or experience level
  • Curse words used as an insult (eg “You little shit.”)
  • Continue to have the last word of an argument
  • False information
  • Libel

Examples of comments that are OK:

  • Curt comments
  • A curse word not used in an insulting or obscene manner (eg “Holy shit, that’s cool!”)

really makes it clear with two examples how to moderate many instances of profanity.

Small thing, but I would put the infraction definitions in the same order as in the chart above to make navigating easier.

Sorry for such a long post, but really looking forward to seeing how this works!

16

Neither a parent (nor a grandparent) But perhaps iNat needs to spell out guidelines for child privacy. iNat is not ‘social media for friends and family’. Crop the photo to biodiversity, and only show us the little hand holding a fishie. If that is an actual guideline, it can be enforced politely.

I have seen sad posts from young parents – I have asked granny NOT to post pictures of our kids, but … granny is old, doesn’t care, doesn’t understand. My grandkids are cute and I am proud of them.

17

I was wondering the same thing, I don’t think it makes sense to give a lesser penalty because the insult is worse, so I think the week suspension is appropriate.

In the absence of further staff guidance, I would view minor insult, major insult, and hate speech as types of insult, in order of increasing severity, so if I see one of these infractions I would give whatever penalty is higher, the penalty of that infraction or the penalty that would result from treating it as a lower infraction.

So if someone has 1 minor, 1 major, and 1 hate, and then they do another hate, I would look at the penalty for 2nd hate, 3rd major, and 4th minor, and choose the largest (in this case the month suspension for 2nd hate, or consider the “possible indefinite” for 4th minor)

If this same person did a major insult, I would look at 3rd major and 4th minor (both 2 week suspensions, but with consideration for indefinite suspension for 4th minor)

If they did a minor insult I would treat it as 4th minor, and consider the indefinite suspension more strongly since some of the prior offenses are hate. (I usually would just do the 2 week suspension for 4th minor, if the priors were actually minor)

I think discrimination probably also counts as insult, probably between minor and major

@tiwane is this the correct interpretation?

I agree with this, I would treat most simple uses of profanity in an argument or comments calling another user stupid as minor insult. Major insults make me think of comments that go out of their way to be extra mean, wish harm, or use discriminatory terms in ways that are not intentional hate speech (eg using the R-slur to mean “stupid” in an argument)

This I disagree with, it is very easy to miss a comment notification, and this will lead to suspensions for second infractions in cases where someone never saw the warning for the first. However, messages create problems too because IIRC some app users cannot see messages.

I also think publicly visible comments can continue the argument

18

While we’re writing down definitions and guidelines, how should we treat cases where a user deletes their observations in order to

  • avoid flags (OK in my opinion, e.g. for copyright or AI flags)
  • get rid of other’s comments (probably OK, but would like to hear opinions)
  • get rid of other’s ID’s (not OK)

I’ve always held that an observer has all the rights to deal with their own content as they see fit, including removing it. It’s just the intention and context that matters. I seem to recall prior conversations about deleting observations to remove disagreeing IDs, though I’m having a hard time finding that again.

19

In regards to deleting for flags, I think this is fine. At one level, it can show that the user realizes what they did was wrong and is removing it (good in the case of AI/copyright imo). The flag will still show up in moderation history and may note the content is deleted. So it isn’t super-worrisome to me.

To get rid of other’s comments, I think it is not ideal, but could be warranted in some cases if the comments have been flagged and the observer was targeted or something.

I think deleting if you don’t agree with IDs isn’t ideal, but it really only crosses the line if someone deletes the observation to remove IDs and then reuploads it. This is flaggable/not ok in my opinion.

20

Hi @jenna489 and @Myelaphus , thanks for the question! The category “Machine Generated Content” (last item in the table on the help site link) is intended to capture any AI generated images, as well as any other posted content that was generated with AI. This category is a little broader than “generative AI” to capture any observations that do not represent something actually being observed, like a photoshopped image.

21

Hey @cthawley , thanks for the feedback! Just to summarize and confirm the UI changes you’re suggesting:

  • Move View action link to the Type column.
  • Add a field to track curator’s comments on a Moderation Action, which is displayed to other curators, not to users.
  • Add a field to suspensions indicating if the user was warned BEFORE they were suspended.