alyaza is unbanned
A couple of days ago, I loudly banned alyaza. After investigating it more, I no longer believe that DearDeer was their alt account, so everything I accused them of doing in that post was not true, and they've been unbanned.
There are a lot of justifications and excuses I could give for why I got it wrong, but in the end it doesn't really matter. I made a somewhat-rushed decision, but I was confident about it at the time. Yesterday I spent more time looking into it, including following the invite chain and managing to get in contact with the person that sent the invite that DearDeer used to register. Between talking with that person (who was remarkably helpful) and some other info, I found more evidence that DearDeer wasn't alyaza than I had used to originally decide that it was, and realized that I was wrong.
This is a good example of why I don't like publicizing bans. Without me making that post about it, I'm sure this still would have been noticed by some people, but it could have been a relatively quiet temporary ban that lasted for about a day while it got sorted out. Instead, it ends up as a multi-day unnecessary spectacle. I'm not bothered by the effect on me because of that—I screwed up and deserve the embarrassment and criticism that comes from it, and I fully accept that. But it was unfair and cruel to alyaza to be falsely accused of things publicly, and that can't be reversed.
Decisions like this (and moderation in general) are often judgment calls that have to be made quickly and with incomplete information. Sometimes, like in this case, you make the wrong call, and more time, information, or an appeal leads you to reverse it. There can be value in having that happen in public, but there can also be harm, and I think this case absolutely leaned more towards the harmful end.
Anyway, I'll leave the comments open this time so that you can berate me appropriately. Please avoid commenting on alyaza personally though—I've already done enough damage and we don't need to continue that.
Let's not. You made a mistake and you owned up & apologised about it. You're only human. As far as I can see, no berating or criticism is needed. The internet has enough miserable commentary as it is, I see no need for myself or anyone here to pile on.
Everything you said is true, however, these 2 lines don't sit well with me:
and
I disagree with these two lines. If one is spending time on a given site, helping it grow, adding content to it, etc, I think one deserves to know how it is being run. But I am on the team of "total transparency" and I understand that there are others who are not.
I think you posted at the wrong place, this doesn't seem to be a response to @emdash ;)
Well, it looks like I need to publicly apologize again myself too, since I was the one that first brought the accusation against @alyaza and @DearDeer to @Deimos. I thought I saw some similar patterns in behavior and writing style between the two accounts that lead me to believe they may be being operated by the same person, but I was clearly very wrong. And I was also the one that encouraged @Deimos to make a public post about the situation, which only exacerbated the problem before it could be properly resolved.
So for all that I am truly, deeply sorry. I am sorry to @alyaza and @DearDeer for my (now obvious) paranoia and accusing them of being the same user, and I am sorry to @Deimos for starting this whole mess and potentially making it worse. And I would also like to apologize to @Whom, but also thank her as well, for standing up for @Alyaza and pushing back against this accusation.
I obviously still have a lot to learn and improve on... I keep harping on about the Principle of Charity, but I clearly didn't follow my own advice in this situation and assumed the worst of two users who didn't deserve it, which put my friend, the admin of this site in a compromising position as a result. So I genuinely am sorry for everything, and while I can't promise I won't make any further mistakes in the future, I can at least promise to try my best to take this lesson to heart and do better from now on.
OK, I'm going to essentially restate my prior comment in the other thread which is that I firmly believe Tildes should look at SomethingAwful as an example of a moderation system. Which is that every user has a public history of [probations, bans, permabans] as well as the moderator that banned the user and a note from them. Having a public system like this is good for both the moderating team as well as the users: It allows moderators to establish a paper trail for why a user might see escalating punishments and allows a user to similarly understand why someone was banned and how.
A comment I see oft-repeated is that the public will second guess every ban you do. In my experience of 10+ years posting on SA, I've rarely seen this to be an issue. If anything, I'd argue making bans non-public make this more of a problem than you might think, considering Tildes has had this issue pop up twice now. The first time being due to Alyaza, and the second being due to a disingenuous user (whose name escapes me) and resulted in another poster going to bat for them. At scale, eventually what will happen is that higher profile users will be banned, resulting in friends of that user asking for reasons why they were banned, and causing more issues because now there's no paper trail showing that so-and-so called transsexuals 'mentally ill' or other toxic nonsense.
With SA's model, you can point directly to the post where so-and-so was a Gigantic Asshole and close the case right there. In this case, the only way to avoid the problem is to essentially close down or ban discussions of bans, which I believe would lead to conspiracy, resentment of the moderation team etc. As an aside, SA also uses bans for humorous purposes (like 'toxx clauses') which also helps to reduce the sting someone might get when they get a short term ban or probation, but I don't think that element would apply as well here due to the fundamental nature of SA being different from Tildes.
That's where I was going when I talked about having 'user report cards' and muting and strikes. That's a fairly simple system, still keeps the human in control (rather than bad bots), and it's transparent. I'm totally in favor of moving in this direction for the short-to-mid term.
I believe the user you are referring to was @go1dfish who went by a few other names on other places, and the post you remember with someone else discussing the topic was this one: https://tildes.net/~tildes/bx1/banning_of_users_and_the_discussion_of_controversial_subjects
As someone who does not post often but lurks, this being only the second relatively visible "spectacle" that I have noticed means Tildes does well with dealing with potentially dramatic situations.
Glad to see this reversed!
In regard to publicizing bans, I wouldn't say the negatives of what happened here are inherent to doing so. Being open about events like this doesn't mean an announcement has to be immediate. Rather than doing so at the point of the decision or not doing it at all, would waiting until you've investigated every aspect of it that you were planning to and letting things play out first have worked?
Or are other options like displaying brief ban reasons / terms / length on a user page worth looking at? That sounds like it would still be open, but not an event in the same way. I imagine the ban volume is low enough that this wouldn't be a big deal, but maybe that's not quite right.
I think displaying more information on a banned user's page could be reasonable. I'll have to think about it more, I'm not sure if there could be some cases where it would cause issues or need to be omitted (which would make people think there was no reason, if everyone else has one).
It could be just a vague description, rather than a specific list of charges: "trolling", "spamming", "personal abuse", and so on...
I think a moderation log like that would do better. I must be out of the loop as I was not aware of this scenario, but as Tildes continues to grow I don't think any publicity towards bans need to made outside a log.
Mod logs have their own problems which I've documented on Tildes before, and often are actually a useful outlet for spammers to determine what techniques work and which ones don't. Also, you fall into the "why was this person banned?!11" trap where every action becomes some potential moderator-flogging event.
Add me a countervoice for "let's not have a modlog".
Good on you for owning up to this. It just reinforces why you're the right person for this job.
(And you can call @alyaza "she" - it was in her user bio!)
Oh God, the emojis. Can I label this noise?
If it deserves a label, the label would be "Joke". It's evident @Bishop is exaggerating the message for humorous effect.
Nothing wrong with not liking the overwhelming amount of emojis, but – context is important. Here it is entirely appropriate.
Oh I was just kidding around.
First of all, I applaud you for owning up to the mistake and taking some steps to fix it.
However, I do want to ask: Are there any ideas in the works to prevent this type of thing from happening in the future? I mean things like a public set of guidelines for what will constitute a ban, and guidelines for how those bans will be carried out and how they can be appealed? I agree that the initial ban shouldn't have been publicized, and your justification in that post for publicizing it was that "users would have questions". Having a common set of guidelines for bans like this might help in such situations, so users know exactly what to expect from bans.
Overall, it doesn't sit very well with me. The entire process seemed somewhat arbitrary, from the original ban to now. What counts as 'too hostile', how many infractions are allowed before a ban, some type of systematized way of determining duration of bans, a public way to see why someone was banned (and I would argue comment on it), and an appeals system are all things that might be considered and formalized.
Ultimately a mistake is a mistake and there's no sense beating that dead horse when it has already been acknowledged. The only thing that makes sense to do is see what flaws in the system this whole episode has highlighted, and how they might be fixed.
This is a problem with so many moderation decisions. Almost all of it displays some level of subjectivity. It's hard to arrive at "concrete" rationales for bans. Or where exactly certain lines are drawn. What is arbitrary defined as? How about "objective"? What is an "objective measure" for banning someone behind a series of text posts? Do you weight bad words and numerically compute a ban value? How about hostility? How do you determine how hostile something is?
From 4 years of Reddit moderation experience, I can conclusively say there's no objective metric you can use for banning people, and everything is subjective to some extent. This sort of criticism is the issue I have with the users behind communities who complain towards mods (not saying you are); they're asking for something which is impossible to achieve, effectively. The only solution which reliably works in practice is that moderators have to do the best they can to meet as narrow of a line as realistically achievable, and users need to accept the world is a subjective place. It's a give and take situation; requiring compromises from both sides. And unless both sides are willing to assume good faith and make such compromises, you will always have warring factions.
It becomes even more complicated when you factor in people having differing opinions on political topics. One person's hostility is another's "beautifully worded forceful argument".
Ultimately this is something I hope Tildes can solve via a wide, broad group of people having a fairly decent number of powers. Instead of moderators numbering 1 to 1000 people, have 1 in 5 people be quasi-moderators. After all, "trust users, punish abusers". Let consensus decide. Spread the pain of moderation amongst many. It won't be perfect, but it sure isn't going to be difficult to improve on the clusterfuck that is reddit.
I largely agree with what you're saying, also based on some years of moderating medium sized subreddits. I completely understand that there's no mythical objective standard of moderation and that ultimately subjectivity enters in - and that's not even a bad thing. That being said, I don't think it's unreasonable to suggest that things can be too arbitrary, and that this is a case that highlights that. Tildes code of conduct currently reads:
As far as I am aware, the only rule that the user in question came close to violating is the first one, which in my opinion is sufficiently broad as to allow banning a user for relatively minor things. I don't think it's unreasonable to suggest that "Don't be an asshole" is insufficient as a policy when moderating 10,000+ (and growing!) people, for precisely the reason you mention: One person's hostility is another's "beautifully worded forceful argument". For example, while alyaza's comments often took a tone that I myself would not take, I was pretty surprised to find that she had been warned about them as they were often not directed at the users but at their arguments (at least from what I saw, which again is limited). I personally don't think 'good faith' discussion inherently precludes calling someones point 'dumb' if you genuinely believe it to be so. Furthermore, there have certainly been users here that have negatively impacted my usage of the site through a perceived overly-contrarian attitude. It's not immediately clear to me what distinguishes the two, and based on the discussions that followed this banning I don't think I'm alone, although perhaps in a minority.
Anyway, I'm under no impression that there is some objective standard of moderation out there. That being said, I don't believe that "Don't be an asshole" on its own is sufficient when moderating a group of this size. At the very least, having some universally agreed upon standards makes it easier for people to comply to those standards and to understand when there have been violations.
I hope this doesn't come across as just complaining about moderation, since all things considered I think @Deimos did an okay job here. I do think that this can be an illustrative episode, especially when considering how to implement moderation in the future. I personally am a big fan of the stack exchange approach to moderation as it is relatively democratic, and distributes power across a good range of the user base. In any case, my main point is that the inherent subjectivity of the moderation process doesn't mean that some standards shouldn't be written down.
EDIT: As pointed out by @Whom, the actual reasons for the ban were an apparent attempt to use an alt account. I initially thought alyaza was banned for hostility, and then apparently used an alt to circumvent this. This was my misunderstanding. That being said, I still stand by the above points regarding arbitrariness so I leave this comment up. I also want to stress that I am not advocating for explicitly enumerating every possible infraction. I just think that this is a good time to consider whether Tildes is at the size where it might benefit from reexamining moderation guidelines, and in doing so considering how those guidelines might be improved.
Fyi, the rule that they were banned for supposedly breaking was right below that:
Not that it takes away from your main point (though I still don't agree with that), but this specific case didn't come down to wishy-washy rules, it was an error in determining what actually happened—if DearDeer was alyaza's account.
Other people have already done a good job of explaining why it's inherently subjective (and has to be), but one other point I want to make is that trying to be explicit with rules often ends up giving more power to people looking for excuses about why their behavior doesn't technically break any rules. It's a lot more obvious that the spirit of the law is the most important when things haven't been defined in minute detail.
This is a blog post that I like a lot on the topic: https://eev.ee/blog/2016/07/22/on-a-technicality/
I wasn't super comfortable with this situation either, but I don't think the solution should be to try and formalize reasons for bans. Allowing moderation to stay human and fluid is something I think we should really value and tightly hold on to.
The issue is more the details of what human decisions are made...which I only got secondhand so I imagine the people involved will have more to say about them.
There is no objective way to measure something as subjective as "too hostile". To expect such a thing is to expect the impossible.
It might be possible to create rules about how many infractions are allowed before a ban, and how many temporary bans, and for how long, should be required before a permanent ban, and other such numeric measures. However, it is absolutely not possible to create objective rules about what constitutes an infraction in the first place. That is always going to be subjective and arbitrary and context-dependent - which is the best argument for having human moderators to make this judgement.
I question the value of making bans public and open to commentary. There is no benefit in having the general public second-guessing every single ban. That is just distracting noise. Too often, it just turns into a slanging match between users and moderators. And the range of opinions about whether someone should have been banned is as wide and numerous as the number of people on the forum.
Imagine when Tildes grows large enough to have 100,000 or 500,000 users, and there's an account being banned every hour (bots, spammers, trolls, and so on). Do we really need to be having dozens of these conversations every week?
Aren't you contradicting yourself here somewhat?
Please see my reply to @emdash further up. I understand the moderation process and the fact that it is inherently subjective. That being said, we already agree that some public code of conduct is warranted, as Tildes already has one. I am simply suggesting that, at the least, we make that code of conduct a bit more explicit (note that this does not mean it has to totally explicit, in that enumerates every possible violation, just more explicit than "Don't be an asshole", which is currently the only guideline directed towards actual user attitude and tone).
This might be the case - at the least, it is a useful discussion to have. Tildes is still at the stage where it is able to experiment with different solutions. Perhaps there can be a separate "mod queue" that is different than the normal ~ feed, that all users can see but only those with a certain level of trust can comment on. Maybe this isn't necessary at all.
I don't think the initial ban should have been publicized because this is not a policy and was enacted arbitrarily, and in this case ended up containing things that were not true. I do think that if this was standard and widely understood, it might help making the moderation process transparent, promote compliance with codes of conduct, and in general increase satisfaction with moderation.
In your time moderating medium sized subreddits, you must have encountered the "rules lawyers": those people who point out loopholes in your rules which means their specific behaviour isn't quite covered. "Your rules ban racist language, but not religious insults, so I didn't break the rules by calling all Muzzies 'towelheads'." I've learned from hard experience that, the more explicit you make your rules, the more detailed they have to become, to cover every possible type of undesired behaviour - because if you don't explicitly list one particular behaviour in your rules, you've left a loophole for the rules lawyers to attempt to challenge your warning/removal/suspension/ban, and then you waste a shitload of time trying to explain something that shouldn't have to be explained. So, you end up writing a full legislative list of all conceivable offences just to cover one rule: "don't be an arsehole". It's a never-ending quest.
@Deimos also linked to a useful blog about this in this other comment.
I agree with @gpl that there might be a middle ground here. I don't need nor want the rules to be comprehensive. In fact, they should state explicitly that they are not. But, I'm also skeptical of the idea of having extremely vague rules with no publication of disciplinary measures, especially once we see more moderators on the site. It feels like we might accidentally swing too far in the opposite direction. While the measures you list had downsides like lawyering, I think their original intention -- to keep mods accountable -- is still worthy of consideration.
Not the original poster, and want to acknowledge this is a thought I havent thought about too deeply and have no moderation experience so it may be obviously bad. But what about (once it is implemented) a subgroup of ~tildes or tildes.offical that has a post when a user is banned, and the group is filtered out of the feed by default? The posts are there and available for those who want to see them, and if documented properly anyone can find the subgroup, but it doesn't clog up everyone's feed. I'm not really sure I am on board with needing bans to be public, but if it is decided they should be, this on the surface seems like a good middle ground to me
Good on you for realizing the mistake and fixing it, although I would have expected it would not have been the type of mistake to happen in the first place.
Curious, what process did you go through in determining the user had an alt-acct?
Why is "DearDeer" still banned?
You are doing a fantastic job by the way, keep up the good work!
I won't go into much detail about it, since a lot of it's personal/private, but there were multiple things that I found that were suspicious around the account's creation and behavior that made me think it was an alt. When I was able to look into it in more detail (and locate the inviter), all of these just turned out to be coincidences and/or bad luck with what I had (and hadn't) looked at.
Just as one example, when I banned DearDeer (which logs the user out), they tried to log back in, got the "This account has been banned" message, and just immediately left entirely, and never touched the site again. That's very weird if it's not an alt account. People almost always go back to look at the threads they were posting in, look for a way to appeal their ban, try to register again, something. But it makes sense if it's an alt: they know the account is dead and just close the incognito window or whatever they were using for that account, but still use the site from their other account.
I left them banned because I don't think they're going to do well here based on how they basically went from zero to openly hostile immediately. If they send in an appeal or something I'd be willing to reconsider, but like I said, they seem to have abandoned the site instantly, so I don't think it's likely.
Ah got it, thanks for that info.
Makes perfect sense to me. I think we all appreciate the openness and honesty from you, makes me believe in this site/community.
I am glad to see that you realised a mistake and immediately corrected it. This is in a stark contrast with what has happened on Reddit in some subs, where users got banned for rules violations that did not happen, appealed, and the mods' response was
Oh yes, you didn't violate rules. However, your ban stands. LOL.I am also glad that it looks like the primary reason for the original ban was (alleged, is it has turned out) alt account abuse rather than the extremely subjective “comment hostility.”
I also completely agree with this comment by Whom.
Finally, I need to note that while “routine” bans of spammers or trolls might have gone unnoticed, banning a prominent user such as alyaza would be definitely discovered by the userbase, so such bans should be publicised (after it is absolutely clear to you or, eventually, other users with a banhammer, that the ban is deserved).
Thanks for doing a fantastic job!
I may be in the minority, but I believe that once the guidelines are posted, bans and other disciplinary actions don't need to be anything more than 'this user was
%action%on%date%--%guideline broken%posted to their former profile page.The banned user can make their case, etc -- but I don't think its anything the community really needs to be part of. This is under the assumption that users are given warnings or criticism regarding their approach to discussions (heated or just trolling.)
thanks! I agree that having the guideline posted along with the 'deleted by site admin' is great.
Well this is a fuck-up. Your honesty is commendable though. Welcome back, alyaza.
This is an interesting phenomenon, to me anyway, about the way language works. Say you have a data set of people, all of whom speak English. Then you divide that data set into multiple pieces (for the sake of simplicity let's say two pieces), based on a topic of varying polarization. For very non-polarized topics you will find use of language varies wildly, save for common idioms and the like. As you divide the two based on more and more polarizing topics, however, the language used begins to congeal into a blob of sameness due to a variety of psychological effects including (but not limited to) the Baader-Meinhof Phenomenon and a dash or two of Confirmation Bias.
It takes a very broad view of language as a whole, so obviously as you get more specific with that data set it begins to fall apart, meaning it's all probably just a statistical anomaly, but it helps me rationalize the idea of a shared human consciousness.
Also, if anyone is interested in how our language differentiates us and brings us together, I highly recommend The Secret Life of Pronouns by James W. Pennebaker. Super good book.
Nah, this only captures phrasing, syntax and word choice.
I don't on Tildes in specific, but what's wrong with it? Most people do on twitter and Discord.
not really sure why you're saying this, because as @Whom stated upthread, my ban was apparently only done under the assumption that deardeer was me and that therefore deardeer's comments were mine. specifically it would have been done under the "don't be an asshole" rule plus these two subpoints of the alt rule:
at least as far as i'm aware, nothing that i actually said in the thread was objectionable to @Deimos (i assume if there was he'd have told me at some point in this process), and the only comment that was ever removed by Deimos accordingly during all of this that is mine was actually the single one that seemingly implicated me as using an alt in the first place (which i think has been restored, and can be found here if so).
Yeah, just for the record, I never saw anything you said as objectionable in that topic. To my recollection it was only DearDeer being inflammatory, and it was only my mistaken impression that it might be an alt of yours that led to this mess. So again, I sincerely apologize for that.
p.s. @Loire
I'm only in favor of this if @alyaza starts capitalizing! j/k ;-P
Also glad to see she's actually back, I've read many interesting contributions from her.