"UYA" notices and face-saving in moderation systems
May 1, 2018 4:12 AM   Subscribe

In the 90s, MIT's Athena system devised a system known as "stopit" to handle online harrassment. One feature of the system was that the notice of reported abusive behaviour sent to the accused user began with the phrase "Someone using your account…". Users were advised to re-password their accounts, as they may have been hacked. Password resets by UYA recipients were taken in good faith by staff, but the most astonishing thing was that the offending behaviour nearly always stopped there.

Gregory Jackson, the Director of Academic Computing at MIT in 1994, explained it thus (poorly-formatted web page):
This is important: even though recipients concede no guilt, and receive no punishment, they stop. [this system has] drastically reduced the number of confrontational debates between us and perpetrators, while at the same time reducing the recurrence of misbehavior. When we accuse perpetrators directly, they often assert that their misbehavior was within their rights (which may well be true). They then repeat the misbehavior to make their point and challenge our authority. When we let them save face by pretending (if only to themselves) that they did not do what they did, they tend to become more responsible citizens with their pride intact
posted by rum-soaked space hobo (58 comments total) 68 users marked this as a favorite
 
It's a bummer the formatting in both of those - at least in Firefox - is not so great. The second article especially appears as one long block of un-paragraphed text, and the writer is far more free speech orientated than me. "Negotiating with terrorists" is a phrase that leaps to mind, "pandering to bastards" is another.

And yet... at work last year I was involved in an initiative to crack down on people abusing their work phone data limits. And let me tell you some of these people were taking the piss, like >500GB on mobile data in a month. We sent the comms, and mostly got no response, however there were a number of mealy-mouthed excuses and denials (when we tell them we can track domains from work phones that tended to shut them up lol). And I can't help wondering if this kind of approach might be worthwhile in a work context.

From a strictly utilitarian view, I suppose the outcome at MIT was good. Of course, from a utilitarian perspective believing women and other minorities, and expelling dickheads would probably also be pretty grand, I think. The only reason this system has to be set up is because of the way society privileges white male voices, I suspect - and by setting this up it helps embed that privilege. The fatigue the writer/s demonstrate at the prospect of long, drawn-out battles between men and their victims kinda pisses me off a bit - it doesn't have to be that way. The reason why it's that way is so men can get away with shit, in the main, and women will give up fighting it. Change the system, I think, rather than working around it.

A win for a tired administrator is more of a band aid I think.
posted by smoke at 4:35 AM on May 1, 2018 [24 favorites]


(view source on the second link to get a much-better-formatted view of it(this link will only work in firefox and chrome); it's named .html, but it's actually a formatted textfile.)

I like the idea of face-saving here; related to, but not the same as, some asshole signing your name to stupid letters. Also, was this part of RIAA/MPAA strategy back when they were fighting piracy by sending C&D letters that referenced only IP addresses?
posted by Fraxas at 4:42 AM on May 1, 2018 [11 favorites]


Also - Problems with stop it. An eternal debate.
posted by unliteral at 4:43 AM on May 1, 2018 [7 favorites]


I accept that there is a gulf of difference between university-run online systems for students in the 90s and the bot-flooded social network systems of 2018. That said, the face-saving mechanism seems like it may still have specific applications that warrant investigating.

A lot of the discussion around poor race relations in the US often highlights how white people describe white offenders in various ways as "good kids" who "deserve a second chance", while vilifying far less damaging offenders of other races. The "second chance" always seems to me like too broad a category. How do we separate "a life lesson learned" from "getting away with it"?

The later phases of "StopIt" involved invitations in good faith to discuss the behaviours in person. I feel this had a useful effect of intimidating a lone student now and again, but I can't imagine a professional online troll like Richard Spencer feeling at all intimidated no matter what carpet he's called out on. What are the limits of shame and face-saving as motivations for staying within community norms?
posted by rum-soaked space hobo at 4:51 AM on May 1, 2018 [8 favorites]


If the only trolls are "pros", it just may stand out a bit more obviously.
posted by sammyo at 5:00 AM on May 1, 2018 [5 favorites]


> Also, was this part of RIAA/MPAA strategy back when they were fighting piracy by sending C&D letters that referenced only IP addresses?

This was more a function of the fact that, unless the ISP was subpoenaed for the name, the IP was all the copyright holder had to go off. Once they filed suit, they would subpoena the names of the legally-responsible party from the ISP.

Of course, this later transformed into the DMCA Notices, where for the first X number of instances the copyright holder would simply notify the ISP, and the ISP would send out the notice to the individual user instead. Those notices followed more closely with the gist of this post, as the ISP ones assiduously avoided placing direct blame, and rather gave recommended actions like changing wifi passwords and checking on kids' activities.
posted by mystyk at 5:17 AM on May 1, 2018 [4 favorites]


Maybe we need to combine this with that other great mid-90s innovation: Clippy.

"It looks like you accidentally posted a racist meme. Are you sure you didn't mean to post this cute picture of a kitten instead?"
posted by tobascodagama at 5:17 AM on May 1, 2018 [42 favorites]


From that "problems with stop it" link:
> male student. His screen was displaying a graphic image of a sexual
> act. Judy asked the student to remove the image, since it was
> interfering with her ability to work comfortably. He refused - loudly
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Was the image on her screen? If so, it was truly interfering
with her ability to work. No, what was happening was in fact "she didn't
like what he was viewing".
That's some debate there, yep. Seth Finkelstein circa 1994--and probably after, considering his line of work--does not object to this because it doesn't work, he objects to it because he objects to the idea that men should ever have to modify their behavior at all in order to create spaces that women can exist. If he's not going to like literally any plan to counteract hostile work and educational environments because "free speech"? I'm okay with moving forward with ideas that do not make Seth Finkelstein happy.
posted by Sequence at 5:18 AM on May 1, 2018 [63 favorites]


"I never did the heinous acts of which I'm accused, and I promise to never do them again."
posted by clawsoon at 5:23 AM on May 1, 2018 [8 favorites]


I only just read the Finkelstein reply now, and yeah that is some classic 1990s Internet Libertarianism.

And you joke, tobascodagama, but I have to confess I was unaware of the shifting connotations of the pepe frog-face memes until way too late, and everyone was just too dry and ironic to simply explain to me what the deuce was going on.

I actually filed a bug in a work system, which I linked to a patch that replaced all the cute unicode frog faces with rainbow-horned unicorn heads. s/🐸/🦄/g, effectively. I'd just gone all kermit all over the thing, originally.

So yeah, a "You may not know this, but that symbol is now associated with..." popup would have really helped me out there.
posted by rum-soaked space hobo at 5:33 AM on May 1, 2018 [13 favorites]


Sequence- reading through that link also; I think the ignorance that pornography exists in a misogynist environment that tends to use women's bodies for male pleasure and that men are overwhelmingly more likely to be aggressive about sex in the presence of women leaves women very uncomfortable when they realize they're in an environment where men are posting they're favorite porn, (or their dick pics). The inability to understand that context and why that matters left the entire argument a total mess to me. Perhaps they just chose a particularly poor example to start off their point or perhaps their point isn't good. I'm not sure which.
posted by xarnop at 5:34 AM on May 1, 2018 [6 favorites]


> Seth Finkelstein circa 1994--and probably after, considering his line of work...

Ugh, that email is painful! I've been in the same field of Software Engineering for almost as long as him, and it is scary how much that screed of his sounds like it could have been written by most of my CS friends, or even by me, circa '95-'00 (or, minus the computer knowledge, by my father still today). Suffice to say, people can change, but you're right -- given his writings as recently as the last few years -- that he almost certainly hasn't.
posted by mystyk at 6:01 AM on May 1, 2018 [8 favorites]


One of the very first studies of electronic communications in the early nineties set up an electronic bulletin board at a university. It had to be shut down within a few weeks because conversation immediately deteriorated to the point that university professors who knew these communications were being monitored were physically threatening each other and calling each other names, including racial and religious slurs. It wasn't just concern for physical safety, if the experiment continued the experimenters were worried the university would be forced to fire the participants.

Online communication is different. Strategies like these are a big deal to improving user experience, in my opinion.
posted by xammerboy at 6:54 AM on May 1, 2018 [8 favorites]


(1) Good old 1970s behavior modification work tells us that punishment doesn't even have to be noticeably aversive to be effective. "Punishment" is simply the deceleration in a given behavior in response to a given consequence. If you cause deceleration, you've achieved punishment. The punisher doesn't have to look or feel hugely aversive. It only has to work.

(2) More recent advancements in behavior theory (like ACT and RFT) tell us that if you don't step on the mousetrap of self-defensive responding, you don't need to deal with self-defensiveness before achieving the effects you're looking to achieve.

(I don't generally recommend the book RAISING LIONS by Joe Newman because it has a bunch of pseudoscientific cruft and downright dangerous advice in it, but he does include some very nice bits about using essentially micro-timeouts with kids that achieve both these points.)
posted by PsychoTherapist at 6:59 AM on May 1, 2018 [6 favorites]


"Punishment" is simply the deceleration in a given behavior in response to a given consequence. If you cause deceleration, you've achieved punishment. The punisher doesn't have to look or feel hugely aversive. It only has to work.

This is very much a field-specific term of art use of the word, and is rather different to what most people understand by punishment in conversational English.
posted by Dysk at 7:24 AM on May 1, 2018 [6 favorites]


"When we let them save face by pretending (if only to themselves) that they did not do what they did, they tend to become more responsible citizens with their pride intact"

somewhere along the line between shitty behavior and good behavior, forgiveness has to exist??
the face saving pretend-it-was-another-and-sweep-it-under-the-rug can work in lasting fashion?
or true lasting behavior change only happens with admission and then forgiveness?
and we have a bunch of evidence that a lot of punishments make people dig in and get worse??

i'm wondering about the continuum from things like offensive computer screens to war crimes...
recently watched this bit on 2 former child soldiers from opposite sides who have become friends post conflict...
http://www.bbc.com/news/av/world-asia-43867841/crossing-divides-two-ex-child-soldiers-forge-peace-bonds

That is 2 guilty parties so maybe no entirely relevant but it is people REALLY changing their minds and coming to see other points of view. This seems to me the crux of most abusive behavior, lack of the other perspective?
posted by danjo at 7:38 AM on May 1, 2018 [5 favorites]


> "Negotiating with terrorists" is a phrase that leaps to mind, "pandering to bastards" is another.

Alternative view:
"When you surround the enemy
Always allow them an escape route.
They must see that there is
An alternative to death."
And perhaps more succinctly:
"The supreme art of war is to subdue the enemy without fighting."
So the question is, do you want to win? Or do you merely want to fight? Because this looks like winning without fighting.

You're asking a lot of people to tolerate a lot of shitty behavior if you insist on passing up the small but decisive victories against shitty behavior in service to some grand ideological struggle, which frankly you may have far fewer allies in than you do against a particular person who is being shitty.
posted by Kadin2048 at 8:04 AM on May 1, 2018 [59 favorites]


This is very much a field-specific term of art use of the word, and is rather different to what most people understand by punishment in conversational English.

I feel like it's being used with its proper field-specific context here and presented alongside its definition, so I'm not sure what the point of this comment is?
posted by tobascodagama at 8:09 AM on May 1, 2018 [4 favorites]


somewhere along the line between shitty behavior and good behavior, forgiveness has to exist??

No! Blood for the Blood God!
posted by ocschwar at 8:11 AM on May 1, 2018 [10 favorites]


I'm not sure what the point of this comment is?

That there is no punishment happening here, in a conversational sense of the word. There is no accountability.
posted by Dysk at 8:16 AM on May 1, 2018


"When you surround the enemy
Always allow them an escape route.
They must see that there is
An alternative to death."


But an unvanquished enemy can regroup and strike back. Do you want Nazis? Because that's how you get Nazis, waiting for a quorum and then coming out of the woodwork with their tiki-torches and shitty cartoon frog pictures.

The Nazism must be vanquished, annihilated beyond the possibility of it ever returning. Whether the host must die with it is a matter of triage. At best, the alternative to death for the enemy would be comprehensive reeducation, including complete deconstruction of what motivated them to commit their offences, peeling past the easy excuses like “I just wasn't thinking” or affluenza or such; they must be forced to confront the shittiness in their psyche that made them commit the offence. Whether this is possible without a Stalinist/Maoist terror apparatus providing the pressure to make it stick is unknown (witness prison inmates brushing up on “that Jesus shit” for interviews with their parole officers).
posted by acb at 8:38 AM on May 1, 2018 [1 favorite]


So about a year back, I watched this talk by Raph Koster. What struck me from it was how so many of the problems with social software were understood in the nineties and ignored by modern platforms. (The other thing that struck me was how willfully ignorant Zuck seems to be about all of this, but that's another issue.)

In any case, I think what a lot of people are missing here is that this isn't intended for the hardcore trolls. It's for well-intentioned people who don't quite get the social norms of the space. Informing them in a non-blamey way that they've violated a social norm will usually get them to behave better. That's usually enough. The ones who continue to willfully act like jackasses after that are the ones that deserve the hammer.
posted by suetanvil at 8:51 AM on May 1, 2018 [13 favorites]


It's for well-intentioned people who don't quite get the social norms of the space.

No, it's for people who didn't grasp that their misbehavior wasn't going to be quietly tolerated and won't engage in it when they understand that there might be consequences. I understand the tactic here, and think it may be useful in certain contexts, but, when analyzing the situation in the abstract, why, why, why do we have to keep pretending like harassment is some kind of accident the kind-hearted routinely stumble into because of the hopeless impenetrability of social norms?
posted by praemunire at 8:58 AM on May 1, 2018 [27 favorites]


The text of a UYA notice is a sort of polite fiction, which can mean two different things, depending on context:

1: “You made a mistake. Fair enough, we won't make a big deal of it, just make sure you don't do it again.”

Or, if the offender in question is a wilful troublemaker, it can be interpreted as:

2: “You may think you can get away with such things, but we're keeping an eye on you. Modify your behaviour or there will be consequences.”

If someone is deliberately pushing the boundaries to see what they can get away with, interpreting it as 1. doesn't yield a more permissive reading. The regulator may give the benefit of the doubt, but they are watching, and repeated offences will be picked up and escalated. (Of course, this depends on there being steps beyond UYA to deal with persistent jerks, and doing so with the requisite amount of force, precisely directed. This policy isn't for dealing with weev or his kind.)
posted by acb at 9:07 AM on May 1, 2018 [9 favorites]


> when analyzing the situation in the abstract, why, why, why do we have to keep pretending like harassment is some kind of accident the kind-hearted routinely stumble into because of the hopeless impenetrability of social norms?

...because it sometimes is?

I keep coming back to the study that proved that the 'backseat fights' of childhood are based on a sensory-cognitive feedback error. When someone feels attacked, the attack feels 'stronger' than the actual contact. This leads to a problem when the 'measured' response is actually stronger than the actual attack.

...so the feedback is stronger, and felt even stronger, leading to a stronger reply, which is felt even stronger...
posted by CheapB at 9:21 AM on May 1, 2018 [9 favorites]


The "stop it" system seems well-suited to stop bad behavior by inducing private shame. Cessation of bad behavior is the goal.

It does not, however, provide an avenue for justice, by which I mean public accusation, conviction and restitution. That seems to be equally as important to a lot of people as the behavior change.
posted by grumpybear69 at 10:03 AM on May 1, 2018 [5 favorites]


A lot of the lower intensity transgressive sexual behavior by ordinary men is performed in a state akin to sleepwalking, so when you give them a nudge, there is a chance they'll wake up.
posted by jamjam at 10:09 AM on May 1, 2018 [9 favorites]


Do we need accountability, though? Or do we need the behavior to stop?

This is an honest question. For me, I think I'm fine telling accountability to suck it as long as the underlying behavior stops, because the change in behavior will eventually set a new cultural norm that will eventually change the underlying mindsets, at least on a population scale.

I'm all right if people don't grapple with their own ownership of their poor behavior as long as they stop fucking doing it. I know that marginalized folk have to put up with an awful lot of emotional labor demands to soothe people who are terrified to confront their own accountability and desperate to avoid doing so. Is forcing them to do that and incurring even more obnoxious behavior worth it, if it's possible to change their behavior without ever having to invoke that argument?

I'm kind of okay with that trade-off. You mean people will be less obnoxious and I don't have to put in the effort to educate? Damn! Where's the downside?
posted by sciatrix at 10:37 AM on May 1, 2018 [29 favorites]


Which is to say, I guess, this:

Personal accountability requires an inherent energetic cost to achieve, and moreover does not always provide an improvement in outcome either in the short or the long term.

If we can create a system in which harassment or poor behavior is gently body-blocked and potential perpetrators are allowed to consider whether they meant to do that without having to confront social censure, in the certain knowledge that pushing forward will invoke that social censure, and those people then choose to not behave poorly...

has any crime even been committed?
posted by sciatrix at 10:40 AM on May 1, 2018 [5 favorites]


"When you surround the enemy
Always allow them an escape route.
They must see that there is
An alternative to death."

The Nazism must be vanquished, annihilated beyond the possibility of it ever returning. Whether the host must die with it is a matter of triage. At best, the alternative to death for the enemy would be comprehensive reeducation


Sun Tzu's point, of course, is that an enemy that sees no hope of surrender or retreat will fight to the death. Which means you are in much greater danger of losing, or that the cost of winning will be very high.

So yeah, if your position is so strong that you are certain you can wipe out the Nazis without unacceptable casualties to your side, that might be the best course of action. But is your position that strong? I wish it were.
posted by straight at 10:49 AM on May 1, 2018 [7 favorites]


Kadin2048: You're asking a lot of people to tolerate a lot of shitty behavior if you insist on passing up the small but decisive victories against shitty behavior in service to some grand ideological struggle, which frankly you may have far fewer allies in than you do against a particular person who is being shitty.

The "grand ideological struggle" you refer to is the struggle for women and other marginalized people (so, well over half the population) to be afforded full human status and treated with the basic decency that is currently reserved only for Christian cishet white males. It seems kinda important?

Why do you think there would be few people willing to ally themselves to this cause?
posted by Anticipation Of A New Lover's Arrival, The at 10:50 AM on May 1, 2018 [4 favorites]


Face-saving is directly adjacent to one of my hobby horses, guilt vs. shame, and how guilting someone always has a better chance of working than shaming them.

First of all committed trolls need to get tossed after one warning (or immediately, depending on context). There's really no excuse for allowing it on one's platform. However, there are lots of posters who get heated and say something that they wouldn't normally say, but are otherwise functional human beings. I know I've flamed people, or drunk posted, or lost my mind in frustration before (I speak of my long-past mis-spent youth only, of course; I now am a perfect angel).

For the average rulebreaker, a moderator responding with a message that says, in essence, "you are a better person than that, what you said was unacceptable, that's not like you at all" will yield a radically better result than communicating to them (again, in essence) "there's definitely something wrong with your entire everything. You suck, your brain sucks, all your ancestors must have sucked, but we don't have strong rules here so toe the line for a short while and feel free to keep fucking up after that. You absolute smegma of a person."

The difference between, as I'm defining the terms here, is that guilting a person is basically telling them: "you did the wrong thing, and if you stop doing that wrong thing you are welcome to continue participating." It leaves the path to repentance open, and is relatively indifferent to whether the perpetrator is actually sorry (so long as they cut it out). Of course, rules of civility exist because being uncivil hurts other people, but guilting someone doesn't insist they even acknowledge that they did something wrong. Guilting only presents the three choices of: you can follow the rules, you can leave, or you can break the rules again and get suspended/banned. It says that it doesn't matter what's going on inside of you, if you want to participate, then follow the rules and that's the terms of engagement. Even when the rule-breaker feels that they were in the right to hurt someone (because they were angry and wanted to hurt them), the rule breaker can keep thinking that as long as they want to---so long as they post better in the future.

Shaming a person, on the other hand, says something more along the line of: "there is something fundamentally wrong with you; you're not only wrong in this particular instance, but it is very likely that you are just bad person at heart." Which is something that trolls want to hear, but causes otherwise-good-but-currently-upset posters to double down (at least while they're still angry, or whatever the problem is.

I mean, it's very, very rare that someone ever asks themselves, "am I the baddie?" (Youtube "Are we the Baddies," Mitchell and Webb's Nazi sketch). Let alone when someone is trying to shame you. However, a good portion of people, even when they're angry, will respond in a pro-social way if they're told (in essence): "That wasn't like you. Please don't do that again" (if delivered in a non-condescending way, of course). Either somebody is capable of guilt or they're not, but for minor transgressions, whether they're capable of change or not, nothing is lost by erring on the side of their behavior being a misstep rather than a character trait.

Humans are social creatures (ostensibly), and in face-to-face social interactions, shame actually can result in behavior change (if and only if they don't want to be ostracized, of course). Now, you might shame someone with the result that they find other assholes like themselves to pal around with, and strut off to go be assholes together elsewhere, but if they're gone... the immediate problem is solved.

It's much easier to justify bad behavior to oneself rather than admit one did wrong. I know I do it. We all do it. It's vastly easier to rationalize away the knowledge that we erred, because admitting that one hurt other people feels awful. I think this kind of rationalization is even more than easy, it's reflexive.

So since there's a way to nip some bad behavior in the bud, it's going to save a whole lot of victims a lot of grief. Concerning oneself with the tender feelings of the transgressor doesn't feel satisfying. They've transgressed, and tit-for-tat retaliation is a default/natural reaction when you're hurt by someone. But would scolding them lead to a change of heart anyways? No. either they're trolls and would love it, or they're normal people who'll double down at least temporarily. Giving the chance to all rule breakers to save face is a good short & medium term solution. Changing assholes into better people is out of scope of web moderation in the first place. It's not impossible, but that kind of change takes a willingness on the part of the jerk, and lots and lots of emotional labor from the asshole's friends/family/mentors. It's just functionally impractical for forums. Not in scope.

When I think more about it, I feel that face saving is also a good long term solution. It leaves the door open to a deeper change of heart in the future. After all, if you're called an asshole enough times, you might conclude that maybe you are an asshole, that the people who're calling you an asshole are assholes themselves, that everybody is an asshole, so why not keep hitting back? I know I've fallen into that pit (temporarily) before, and but for the grace good friends go I.
posted by wires at 11:14 AM on May 1, 2018 [82 favorites]


Oh, flagged as fantastic, wires.
posted by sciatrix at 11:17 AM on May 1, 2018 [3 favorites]


I was an undergrad. At MIT. In 1994. And spent entirely too much time on Athena. And have friends who were and are MIT IT staffers.

I don't know of any behavior at the time that qualified an Athena user as a nazi who should be encircled and offered no quarter, but I wasn't on every Zephyr conversation.

I'd be amused at the way this discussion is going, but right now mostly tired and somewhat disgusted with the Mefite spirit working its way here.
posted by ocschwar at 12:32 PM on May 1, 2018 [1 favorite]


...because it sometimes is?

I'm going to invite you to consider why it is that it is so easy for you to default to the option that denies the pervasive existence of sexism and racism in the texture of our daily lives in favor of the idea of a huge population of good-hearted (mostly) men who somehow just can't figure out, darn it, why having to look at the image of a blowjob on your co-worker's screensaver is harmful. If only someone had told them! These are the same men, mind you, who somehow miraculously manage not to transgress norms when it comes to the people who have authority over them, or those whose respect they want to gain. Geez, it's a mystery why they keep stumbling when it comes to those groups they've been taught to consider themselves superior to. Women, who can even tell what they want, am I right?

People can most definitely do hurtful things without realizing it or intending to. That is rarely the case with harassment, except in the sense that men are trained not to admit that harm done to certain others is harm. What the UYA technique does is signal--or at least, give a signal--that this is an environment where that kind of misconduct is treated as misconduct rather than as an opportunity for building group solidarity amongst the harassing class. For a fair number of people, the conduct is so thoughtlessly entitled that a mere discreet reminder to that effect may change their behavior in that environment. This is a fairly low-effort means of conduct correction and I do think it has its place. But it is, to some extent, trading justice for compliance. And, possibly more importantly, it's not clear whether it serves to change behavior once the person is outside that particular forum.
posted by praemunire at 12:39 PM on May 1, 2018 [18 favorites]


I can buy that this may have been a sensible approach to take back in 1995, but somewhere around 2014 I permanently expended my supply of benefit-of-the-doubt when it comes to dealing with people being jerks on the internet. (Maybe gamergate is what finally did it.) If you're doing this stuff nowadays, you know people don't like it. Neither guilt nor shame is effective, because these people don't feel guilty and they have no shame. They do it on purpose because they think they are in the right, and they are praised (sometimes even paid) for doing it. Online harassment is an industry now. It's an organized effort. It elected Donald Trump to the presidency, for fuck's sake.

"You know this is wrong, now let's just pretend it wasn't you as long as you don't do it again" doesn't work anymore. These people don't know that it's wrong, they think they are right to get someone's private phone number and then publish it so that hundreds of people can call the victim up and explain to her in graphic detail how they are going to rape her to death in front of her children. That's where we're at, that's the world now, and no amount of offering people a face-saving out is going to help because they are proud of themselves for what they do.

Also, I feel like we don't need to be plumbing the depths of social psychology in search of a more effective way of dealing with harassers than what we do now. You know what the standard way of dealing with online harassers is now? Nothing. It's nothing. Twitter and Facebook et al generally do nothing when even quite extreme examples of harassment are reported, even when they are reported by multiple people. Same goes for law enforcement—they generally do nothing, even in the face of very clear evidence of dangerous harassment. You know what would be an improvement? Something. Anything!
posted by Anticipation Of A New Lover's Arrival, The at 1:01 PM on May 1, 2018 [11 favorites]


Anticipation Of A New Lover's Arrival, The: I agree entirely. 99% + of the problems on the social web today are the result of there being no consequences for literally criminal behavior. My big response re: guilt and shame is for that remaining 1% -- like, for young teenagers, for behaviors that where the victims of whatever the bad behavior is actively/positively consent to giving the transgressor another chance (as opposed to being a "cool girl" and feel uncomfortable and not having to insist that Derek gets booted from the study group, for examp).

You know what? you're right. If someone is sexually assaulting someone via pictures or rubbing shoulders or similar, fire them, expel them, shoot them into the sun. I am more than tired of it just like everyone else.

Still think it's better to guilt vs. shame, for example, one's racist uncle... if you have the energy for that emotional labor. If it's easier to plant seeds than to not show up to family gatherings. Maybe.

Full disclosure: I haven't had a substantive conversation with my family since 2012, so that's where I stand on that one...
posted by wires at 1:57 PM on May 1, 2018 [5 favorites]


The example given in the second piece is someone displaying porn on their desktop in a computer lab, and digging in when confronted about it.

This is hardly an accident and digging in when confronted by a woman is taken normal somehow? Note in the second piece the author implies that both parties are in error with this only-natural conflict, and he kinda implies both parties would get the StopIt email?

Im not cool with that shit, and if that means grandstanding so be it. In 1994 it was gross, in 2018 its beyond gross. I want to create norms by making unacceptable behaviour unacceptable. And deconstructing systems that are set up to protect men and burden women.

Oschwar, whilst you're questioning mefi's nature, maybe question if it's possible that you as a dude (you are a dude right? Apologies if not) didn't maybe see all the harrassment women endured or didn't notice it at the time as a young male undergraduate? Guys saying that they didn't see harrassment therefore it doesnt happen is not a great look.... Maybe someone hacked your account? ;)
posted by smoke at 2:02 PM on May 1, 2018 [8 favorites]


Guys saying that they didn't see harrassment therefore it doesnt happen is not a great look.... Maybe someone hacked your account? ;)

Did I say I did not see harrassment? To reiterate: " I don't know of any behavior at the time that qualified an Athena user as a nazi who should be encircled and offered no quarter, but I wasn't on every Zephyr conversation. "
posted by ocschwar at 2:27 PM on May 1, 2018


Fair enough, I just found the example in the article quite infuriating, and thought it was very minimising, so I'm a bit sensitive about it.

Language about offering no quarter etc often privileges perpetrators at the expense of victims. I feel like as a society, we have a long way to go before we can legitimately say "oh were taking racism and sexism too seriously", and I reckon MIT in the nineties had even further to go, so I'm not inclined to be generous.
posted by smoke at 2:55 PM on May 1, 2018 [4 favorites]


"That there is no punishment happening here, in a conversational sense of the word. There is no accountability."

Punishment is hollow, if this encourages a dipshit to not act like a dipshit in the future, I'd say it seems like a good alternative to something that will make the problem worse or continue to spread hurt.
posted by GoblinHoney at 3:45 PM on May 1, 2018 [6 favorites]


I suppose it’s a bit like the question of what do we want out of our justice system? Punishment or rehabilitation? Recidivism or betterment?
posted by Thorzdad at 4:05 PM on May 1, 2018 [2 favorites]


It does not, however, provide an avenue for justice, by which I mean public accusation, conviction and restitution.

I find it ironic that the people who advocate the most strongly for restorative justice as an alternative to prison are often the people who are the most harsh on (perceived) Internet assholes.

No, this doesn’t provide access to justice. That’s because nothing will. People will be assholes for most of their lives and then will be seen as being redeemed. People with truly horrible views will never understand the harm they’ve done.

Because that’s life. And the best we can hope for is that people will be better than they used to be.
posted by steady-state strawberry at 4:10 PM on May 1, 2018 [7 favorites]


I'm finding it hard to feel so absolutely certain about what it meant to be showing porn on a screen in a computer lab at a private university in 1994.

I understand that many people don't see it this way, but as a queer kid at the time, porn on the internet felt liberating, and a point of connection. At the time, visual culture felt like it was dominated by conservative "family-friendly" values, and two years later, Bill Clinton would sign the Communications Decency Act. So internet porn felt rebellious and anti-establishment. And at the same time, everyone having their own private window into cyberspace was new. Confronting someone in a public library about their choice of book seems absurd and offensive, so I can see how someone off at a private university with new access to the internet might not respond well to the confrontation. And standing up to the objection might have felt like standing against the Jerry Falwells and Anita Bryants of the world.

That said, 2018 me thinks that the guy was probably just an asshole, but an asshole who might still have inadvertently given cover for 90s queer kids to feel a little more ok about having queer-looking web pages on screen.
posted by Lempel-Ziv at 4:23 PM on May 1, 2018 [4 favorites]


I suppose it’s a bit like the question of what do we want out of our justice system? Punishment or rehabilitation? Recidivism or betterment?

I would settle for just making them stop. Whether that means pretending like maybe their account was hacked so that they stop on their own out of embarassment, banning them from an online forum so that people don't have to listen to them, or imprisoning them so that they cannot continue to harm people, they need to be stopped. If they can be rehabilitated, great. As far as vengeance, I personally would rather not. But they should be made to stop by whatever means necessary, for the safety of others.
posted by Anticipation Of A New Lover's Arrival, The at 4:36 PM on May 1, 2018 [2 favorites]


One of the biggest rules of interface design is to minimize error messaging. You never want an error message to convey the user is making a mistake. Similarly, here it sounds like one never wants a social ruling type of message to convey the user is a bad person.

Obviously, there are exceptions in both cases, but 90% of the time at least I can see it making sense that you don't want to call someone out as inherently bad / wrong for something they did unless you want them to dig in for a long term fight.
posted by xammerboy at 7:07 PM on May 1, 2018 [5 favorites]


> ocschwar:
"No! Blood for the Blood God!"

SKULLS FOR THE SKULL THRONE!
posted by Samizdata at 7:26 PM on May 1, 2018 [1 favorite]


> jamjam:
"A lot of the lower intensity transgressive sexual behavior by ordinary men is performed in a state akin to sleepwalking, so when you give them a nudge, there is a chance they'll wake up."

I concur on the nudge. Try to avoid "the kick to the face" or "nuclear options" though. If they are worth warning, then the nudge will be enough. Otherwise, they tend to skew the other direction and get defensive which defeats the whole point of the exercise.
posted by Samizdata at 7:29 PM on May 1, 2018


I'm finding it hard to feel so absolutely certain about what it meant to be showing porn on a screen in a computer lab at a private university in 1994.

It meant fetching 4 separate posts from alt.binaries.*
Cutting out the content.
Pasting them together,
UUdecoding the result.

Voila: a still JPEG image of N people dong something to attain some kind of sexual gratification, displayed on the screen. WHich you then discard because you have no room on any disk drive of your own to put it on.

Really it was easier to just to to Boston's combat zone and buy a VHS of something at a porn store the way God intended. But if you're at a computer lab at 2 AM and nobody else is there, and you just discovered alt.binaries.*, well, people did this. And some were stupid enough not to make sure they were alone first. So, you're logged in to m56-155-2 and 4 screens down someone else is doing this. You type "finger @m56-155-6" and email stopit, and that person never does it again.
posted by ocschwar at 8:58 PM on May 1, 2018 [3 favorites]


Why do you think there would be few people willing to ally themselves to this cause?

Because there are a lot of people who believe very strongly in a nearly absolute right to free speech and, perhaps because they are not themselves on the receiving end of typical shitty behavior, are not going to willingly curtail free speech in any meaningful way when the issue is framed that way (and they are extremely receptive to that framing). Otherwise known as "why we can't have nice things", at least in this context.

There are a lot of people who may see shitty behavior going on and identify with the victim in the individual case, but as soon as it is framed as an ideological issue over speech rights, they are more than happy to sacrifice someone else's right "not to be offended" (or harassed, as it were) in order to preserve or not restrict those rights.

If you insist on duking it out and trying to win over all those people ideologically, or even a majority of them, you are in for a tough slog and I don't think your victory is assured, at least within the context of the US or US-dominated part of the Internet. Which is not to say that it's not a worthwhile thing to try to do, but I think it is a serious mistake to pass up an opportunity to make someone cut out their crappy behavior because doing so would be somehow ideologically impure or infinitesimally deleterious to some greater struggle.

But an unvanquished enemy can regroup and strike back. Do you want Nazis? Because that's how you get Nazis

I am fairly certain that the tiki-torch Nazis did not arise out of some unvaniquished Boys From Brazil sect who somehow crawled out of the woodwork when the time was right, because granddad failed to hang enough Germans back in 1944. You can't "vanquish" an ideology. (Though not for lack of trying. See: Global War On Terror, 2001-present.)
posted by Kadin2048 at 9:38 PM on May 1, 2018 [4 favorites]


And, possibly more importantly, it's not clear whether it serves to change behavior once the person is outside that particular forum.

I also question the extent to which this would work outside of other forums. MIT has a few things going for it as an institution that help make this effective, most notably that the authority emailing you to stopit has meaningful power over you. They can not only lock you out of ATHENA, they can begin formal disciplinary hearings against you as representatives of university administration. They can throw your entire life upside down, by - in an extreme case - getting you thrown out of university, and out of university accommodation. Even if you've not done something criminal.

This simply doesn't apply to Facebook or Twitter, or Reddit, or wherever. The worst they can do is ban your account. Which forces you to register a new one if you want to keep being an asshole, a minor speed bump at best. The person telling you to stopit is a joke, not someone who's authority necessitates at least grudging respect for the power they have over you.
posted by Dysk at 1:46 AM on May 2, 2018 [1 favorite]


There's effectively an implicit and potent "...or else" at the end of the stopits sent out by MIT, that would be missing from stopits from pretty much anyone else (except perhaps employers to their employees).
posted by Dysk at 1:56 AM on May 2, 2018 [2 favorites]


I find it ironic that the people who advocate the most strongly for restorative justice as an alternative to prison are often the people who are the most harsh on (perceived) Internet assholes.

The stakes are so very very different. Putting someone in prison and giving them a permanent criminal record is a very very big deal, and should not happen lightly. Banning someone from an Internet service is not, and the bar to clear before such action is warranted is and should be drastically lower as a result. You don't need a fair trial in front of a jury of your peers to be thrown off a forum, and nor should you. Carrying over conventions based on the stakes being so much higher in the real world serves in effect to help the assholes.
posted by Dysk at 2:02 AM on May 2, 2018 [7 favorites]




The UYA system would be not the entire system of dealing with trolls, but merely the guard hairs at the front end of it. Its purpose would be to allow those who might be persuaded to change course of their own volition to do so without it escalating to punitive discipline. (It's the same logic as not throwing teenage shoplifters in prison to mix with hardened criminals.)

The next stage would presumably include a permanent banishment that can only be reversed by the offender completing self-motivated rehabilitation. The extent of the rehabilitation and proof would be proportional to offences: a serious troll would need to write a Masters-level thesis on the issues around their offences and why they've changed.

Of course, you can't tell if somebody remains, in their heart of hearts, a Nazi. But you can make them keep their heads down. As the old adage goes, thoughts become actions, actions become behaviours, behaviours become character. If insincerely rehabilitated crypto-Nazis are forced to maintain at least an Ingvar Kamprad-level charade of liberal-minded enlightenment for the rest of their lives, that is a victory of a sort. Every day, the facade becomes more of a reality. Perhaps, should there come a quorum (using the analogy from cancer-cell biology) where enough crypto-Nazis realise that they can publicly sieg-heil again, attrition will have wiped out some of their commitment.
posted by acb at 6:28 AM on May 2, 2018 [5 favorites]


Which forces you to register a new one if you want to keep being an asshole, a minor speed bump at best.

I dunno. I guess there are many subspecies of Internet troll, and there are certainly people who create accounts simply for the sole purpose of being assholes and will do so until they are literally unable to do so. But it's actually fairly easy to detect and automatically boot people like this if they're unsophisticated; the usage pattern of signing up a new account and then immediately posting a bunch of stuff that gets reported stands out like a flaming bag of shit on the service's collective doorstep. To get around this the troll would have to carefully curate and build sockpuppet accounts using VPNs or Tor, with a clean browser fingerprint, etc., which takes time and has to be done carefully, and that's a different arms race that eventually leads you down the road to nation-state-level/APT actors and stuff. (No joke, cutting edge ML/AI stuff is being done in this field, absurdly enough. If you are into ML, you can take your pick of working on self driving cars, stock trading... or detecting sockpuppet bots.)

Any popular service that isn't tracking and re-banning average morons when they create new accounts after a ban isn't even trying. So either they are purposely leaving the barrier to reentry low (possible; some services don't really regard harassment as a problem and wave their hands while looking the other way—that's a different issue) or they're incompetent enough that they're probably well on their way to being nothing but an echochamber of nation-state-sponsored propaganda bots chittering away at each other. They're dead services walking; it's like being a webmail service without a spam filter in the late 90s.

But I think there is a different, IMO more common case of drive-by shittiness that's especially endemic on Twitter and Facebook and these are not people with throwaway troll accounts. Some of the shittiest stuff I've seen posted on FB certainly gives every appearance of being posted by an actual person, and probably not a nation-state bot either. My guess is that these people, despite being assholes, are probably just as attached to their Facebook/Twitter/whatever accounts (with their associated gamified networks of friends, followers, etc.) as a normal person. Maybe more so. Threatening these sort of people indirectly with account termination, but not doing it in such a way that triggers their martyr-for-the-cause complex, might be pretty effective.

At the very least I would be interested to see some Internet-scale service give it a try as part of some sort of controlled test and see how the results worked out.
posted by Kadin2048 at 8:16 AM on May 2, 2018 [1 favorite]


i wish it were possible to look at a comparable school that put similar amounts of effort into preventing harassment on their computer systems, but used a different policy. unfortunately MIT is a very weird institution, and I think Athena was basically unique at the time, both in the technology and in its scope and role on campus. so there may not be any natural experiment, other than the fact that harassment complaints and recidivism dropped after the policy was put in place, which seems good but isn't all that easy to interpret.
posted by vogon_poet at 10:38 AM on May 2, 2018 [3 favorites]


Yes, a policy like this lowers the punishment a bad actor may receive.

But I am not sure that’s wholly a bad thing; in ocschwar’s example, if you know the person will receive a reprimand but not (immediately) lose anything, then your worry about retaliation is much lower. The doofus with the porn is a lot less likely to try to figure out who told on him if they aren’t directly accused.

More low-bar, low risk reporting is a good thing, I think; if the only way to report harassment is to go through some complicated scary reporting process then more people will choose not to report.

Also, occasionally it actually would be someone else— I left my computer logged in and unattended for too long once when I was an undergrad, and well it’s not like *I* wanted my desktop set to goatse.
posted by nat at 12:56 PM on May 2, 2018 [5 favorites]


« Older Antcheater   |   Árvakr ok Alsviðr þeir skulu upp... Newer »


This thread has been archived and is closed to new comments