Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Proposal 0004: User Blocklists, Concentrations of Power, Threats, and Solutions #36

Open
enn-nafnlaus opened this issue Jul 5, 2023 · 3 comments

Comments

@enn-nafnlaus
Copy link

enn-nafnlaus commented Jul 5, 2023

Let me preface this, before diving in, with the following.

  • If no action is taken, sooner or later, someone is going to die because of it.

Let's start at the core, well-known issue:

Internet Mobs and Targeted Harassment

This is a problem as old as the internet. One that Twitter turned out to be particularly vulnerable to. And one that, I'm afraid, Bluesky is heading in a direction of even greater risk. If you're not familiar with the problem, I strongly recommend this article:

https://www.nytimes.com/2015/02/15/magazine/how-one-stupid-tweet-ruined-justine-saccos-life.html

image

The short: perceived slights go viral. The mob is encouraged to, and finds joy in, "righteously" punishing the perpetrator. The perpetrator suffers in gross disproportion to the degree of the slight. In some cases - when the target is particularly emotionally vulnerable - targeted harassment can even end in suicide.

The Benefits of Blocking and the Appeal of Lists

Interpersonal conflicts are one of the fundamental sticky issues that social media networks encounter. Perhaps the most effective tool to combat this is the ability to mute or block perceived troublemakers. Being harassed? Block your perceived harasser. Often, when there's a fight, the best solution is that both sides block each other and move on.

However, many people feel particularly vulnerable - as a threatened group or a past target of group harassment. A new social media site mainly made of refugees from another site is particularly prone to worries that their potential harassers will arrive and harass them at the new location. This is understandable.

image

Moderation seeks to create tools to eliminate harassers and troublemakers. But a common perception is that moderation will not help them, and that people should preemptively block their threats. And various individuals have stepped forth, introducing lists with names like "Nazis", "TERFs", "Bigots", "Trolls", and so forth. One can mute all people on the list with a single click, although more commonly, people are encouraged to go down the lists and block them all.

What person, feeling threatened, wouldn't like to block all Nazis in one fell swoop?

Vigilantism and Pariahs

image

With the advent of "popular blocklists", certain individuals gained immense power over Bluesky. With a few clicks and no due process, a person is now publicly declared a Nazi, a TERF, a Bigot, a Troll, or any other thing. Almost nobody who looks at the list will question it. A sizable percentage of Bluesky will mute or block the individual.

How will the person even know they're now a pariah? It's immediately obvious. Bluesky breaks for them . For the major lists, something like one in four threads that show up on their feed will suddenly have part of the conversation missing, making it unintelligible, and constantly reminding them that they are now despised by literally thousands of people who they've never met, who now believe that they're a monster. Because the list said so.

image

image

image

It's at these times that one should be reminded that people on Bluesky are not simply text and pictures - they're real people. And so when, say, a queer non-binary disabled Jewish woman gets put on someone's Nazi list and is mass blocked by the network, just uncritically accepting "The List Says She's A Nazi".... that this isn't something that just gets easily brushed off.

image

I'll repeat: sooner or later, this is going to happen to someone who is particularly vulnerable, and it could end a life.

image

These lists also serve as convenient target lists. For anyone who wants to seek out revenge against Nazis, TERFs, Bigots, or whatnot, you now have a nice convenient public list of everyone who has been declared such in one place, to doxx and take things into the physical world.

People live in fear of having said the wrong thing to the wrong person and ending up on one of the lists, and thus mass rejected by their own communities and with no way to get off it.

image

image

image

image

Segments of the black community have in particularly felt attacked by the Contraption list.

image

image

image

At one point, the list's maintainer was literally encouraging people to block everyone that one of the targets was following.

image

Checking out some of the victims, I decided to dig into a random one.

image

A biracial Black German/Jewish man. I took the time to read literally everything he has ever written on Bluesky, and I can say: the man is a giant ball of cheer. Literally the "worst" thing he ever wrote on Bluesky was:

image

The rest of the time it was just stuff like:

image

image

But how he's mass-blocked by thousands of people, a pariah, declared a monster because.... a list maintainer decided that a different person who they had been arguing with was bad, and so went down the list of people that that person was following, and listed them as monsters.

How Thousands Of People Started Being Told I'm A Sex-Worker Rapist

Let me back up for a minute. Way back.

I was in my early thirties. I was just divorced from my partner of over a decade - we had been one of the first lesbian couples to get married in Iowa, but ultimately, we couldn't make it last. I was newly moved to Iceland as well. Alone. Emotionally vulnerable. I went out to a club that night, and was walking back to my car, after midnight. A guy saw me and started chatting me up. He seemed nice, took my hand, and we walked together for a while, almost like a date. But after a while, something started to feel off. I started trying to find my car, but I didn't know the city yet, and couldn't find it. He kept encouraging me to keep walking with him. I was sure I had gone past it, but he kept driving me forward every time I tried to turn around. Finally we got to the statue.

image

I knew this was wrong. I just stopped walking, and he couldn't urge me forward. I could easily have run. Or shouted. But I didn't. Because you don't do that, it's so ingrained into you, that even when you sense something is wrong, you stay silent, you don't make a scene.

Even when he literally picked me up and started carrying me to where he sexually assaulted me, I didn't fight back. I kept saying no, constantly in shock that the word wasn't sufficient, like I was dealing with a rational actor in a normal situation. For several years, I couldn't find the place where it happened, always lost at the statue, and started gaslighting myself as to whether it happened at all, starting to doubt my own sanity (until eventually, years later, I found the place - he had carried me caddy-corner through a parking lot to a non-connecting street).

The lesson people needed to learn was... if something feels off, FUCKING RUN.

Then, a couple weeks ago on Bluesky, a post popped up randomly on my feed, with someone saying that exact message: FUCKING RUN.

image

Only they limited it to only sex work. Little did I know that their posts were supposed to be "sex-worker only" - it's just a post that came up on my feed, there's no disclaimer. So I commented:

image

... trailing off in ellipses because I didn't feel comfortable talking about my experience.

The conversation continued:

image

... and ultimately:

image

And ultimately, with the added barb that my assault was "a much siller, personal reason":

image

Let me back up again, briefly: several of my friends have done sex work. One in particular. Back in 2007, my partner and I met a woman who had just fled from Chicago. She had been groomed into sex work when she was a teenager, and had just left. She was homeless, so we offered her our spare bedroom. Her behavior was odd, to say the least. Every time she needed to go to the bathroom, she would timidly ask for permission to do so - it over a week to get her to understand she didn't need to. And some of her behavior around us felt unusual. Eventually she confided in us that when she moved in, she expected to have to pay in sex, and was confused that we never propositioned her. The very concept of charity was alien to her life.

She slowly learned to relax, but never fully. Eventually met a really nice guy, who she started dating, and eventually moved in with. They were together something like a year (and we chatted remotely), but eventually, she stopped. We later heard from her boyfriend, who sounded devastated - she had run off. Eventually, a few weeks later, her body was found, of an apparent suicide.

I sometimes wonder if I could have done more to save her.

But now here in the present, I - with my Bluesky feed mainly known for posting plant science threads...

image

...was added to a list used by thousands of people with the following description:

image

Ignoring the irony of a transwoman misgendering someone: I am now declared to be "a man" who is "a general creep to, if not worse, to sex workers." The implication being, a probable sex-worker rapist.

Risks

Before you ask, I'm fine. This whole thing made me relive some pretty awful stuff, and I'm now constantly reminded by the constant daily stream of "BLOCKED POST"s in my feed that thousands of people are being told that I'm a probable sex-worker rapist. And yes, I've cried, and had some pretty awful days because of it. But I'm not a vulnerable teenager anymore, who doubts her self worth. I'm also offered some limited degree of security against physical vigilantism by living on a remote rock in the middle of the North Atlantic.

But that can't be said about everyone that this happens to. And if this keeps up, as Bluesky continues to grow, sooner or later, this is going to end in tragedy. And that's a liability for Bluesky.

It should also be mentioned that there's potential legal liability here. And not because people already broadly mistaking some of these sorts of vigilante lists as being run by Bluesky itself:

image

Rather - and I'm not a lawyer here - but it sure seems that if Bluesky knows that someone is being libeled on their platform, is informed, but refuses to take action, that they might be party to a libel suit. Someone falsely being publicly labeled a Nazi or whatnot absolutely threatens significant damages.

Solutions

This is a challenging problem, because we need to acknowledge the reason why people turn to these sorts of blocklists in the first place, while simultaneously recognizing the risk that having this sort of unchecked power to throw people into Kafkian situations, in the hands of random users, poses to the network. There are a handful of solutions - some straightforward, others involving weighing pros and cons.

1. Unhide blocked posts

While this sort of papers over the underlying issue, it would restore consistency with Twitter, where a blocked post can be viewed with an incognito window. After all, at a protocol level, nothing is truly blocked anyway, and eventually, Bluesky will go fully open rather than behind a invite-only login wall. The primary intent of blocking remains to stop a person from replying, which is the key functionality that needs to remain intact. The person would still be mass-blocked, but would not be as constantly reminded of the fact, or face a broken user experience where a large chunk of conversations on their feed are incoherent because one voice is blocked out.

2. Better Define Targeted Harassment

Made it clear that targeted harassment can apply to lists, because in practice, it's a particularly devastating form of it. Make it reportable. Have moderators review reports of targeted-harassment-via-list. If someone gets into an argument with a person who has great power via managing such a list, and the person takes revenge by listing them as a Nazi - and they're not one - the repercussions for declaring (to thousands of people) that said person is a Nazi should be no less severe than any other mass-targeting on the platform.

3. Discourage Vigilanteesm as a Matter Of Policy - Promote Moderation Instead

People need to feel safe on the platform. If they feel that moderation will protect them, they won't feel the need to turn to vigilantism. The Bluesky team needs to press this fact publicly - that people who are legitimately proving to be habitual harassers or avow hateful ideologies won't be tolerated on the platform, and that moderation will defend the platform's users from them. At the same time, users should be publicly discouraged from taking part in vigilantism, but to instead to be diligent about reporting threats to the moderation team.

4. Add "block reasons"

Honestly, this applies to any type of block - if someone sees "Why am I blocked by this person?!?" and has no clue, they'll get angry, but if they click on the block and there's a screenshot of them being an arse to the person in question, most will go "oh yeah" and move on with their life. And indeed, blockers commonly forget why they blocked a person in the first place, and it would be useful for them as well.

5. Add mute and block timeouts

Not every mute or block should be forever. Sometimes people just need a cooldown in their interactions. Users should be able to choose for how long.

6 Require that lists intended to be shared with the public meet basic standards of judicial review

Aka, that anyone added have the specific context of why they were added be publicly stated; that timeframes for inclusion match the severity of the infraction (e.g. single one-time incidents != forever inclusion); and that an appeals process be available.

Ideally, though, #3 is preferred to #6. Vigilantism is in general not a thing to promote, even in a "regulated manner".

None of the above should be construed to apply to individual blocks - only vigilante lists that encourage the targeting of third parties. All individuals should retain the right to block whomever they want, for whatever reason they want, to curate their experience on the network.

@pfrazee
Copy link
Contributor

pfrazee commented Jul 24, 2023

Thank you for sharing and offering very specific concerns and suggestions. I've read this multiple times and have thought about it a lot.

I can tell you that recommendation 5 is definitely happening, and that I'm giving a lot of thought to 3 and 6. I share your concerns about vigilantism and unaccountability. I also worry about the dynamics around list-management and how that can become extremely toxic for the people who run the lists. I'm spending a lot of my evenings (when I'm not coding) working on these issues and finding the overall framework that we can apply. I'll share more when concrete, but I wanted to be sure to share that I'm looking closely at the issues you're raising.

@NetWalker108
Copy link

Just read through the entire issue and the only practical solution in my opinion is #5.

1 takes away the purpose of the conventional block.

2 is an issue that can, at best, be handled at the client level, without infringing on the independence and free speech properties of every Bluesky user.

3 runs similar infringement risks as 2 but also encourages the heavy reliance of users on central actors/parties.

4, much like 2, can be best handled at the client level. There are many who won't want to give a reason for blocking, the same way they won't want a reason to make a post.

6 doesn't scale well and also infringes on the freedom of users.

In a nutshell, all 6 solutions seem viable under a dedicated PDS+Client endeavour that pursues these goals and properties. This is the path to achieving them without drastically altering Bluesky as we know it.

@EstrogenEmpress
Copy link

EstrogenEmpress commented Jul 29, 2023

I wanted to take a moment to address some of the statements made here as many of them do pertain to decisions made surrounding and pertaining to myself and my mute list operations.

  1. The situation pertaining to your initial listing was one where, among other things, additional context may have been necessary to make a more informed decision on my part. This decision has, effective today, with this additional context, been reversed. You should no longer appear on any of my mute lists. However, the personal block will remain in place due to reasons I will get into shortly.
  2. The naming of the latter mute list has since been changed to "Users Who..." as I realized in hindsight the potential for misgendering and if there was any confusion to this end, I apologize.
  3. I have stressed on multiple occasions that the intent behind my personal mute lists - as that is what they are, both personally curated and intended as a mute button rather than a hard block - is to limit access to communities who may be made to feel unsafe due to statements made by a user. These can be tricky, and as I have also stressed, I do not have moderator training nor any official enforcement capability. I, and any other user who operates these, will be 100% susceptible to the biases that come with them. I have stressed time and time again that anyone has the capability to choose whether or not to use said lists, or to subscribe carte blanche, and that at the end of the day, I'm just some girl. I am not affiliated with Bluesky nor Skywatch in any official capacity.
  4. The screenshot where I am advocating for a carpet block on a user's followers list was taken out of context, as the original post (I merely commented in it, another user made the original) emphasized that users following said account may OR may not be prone to targeted harassment against transgender individuals, particularly one with whom I am a mutual. I blocked indiscriminately and without question as a hard barrier against any harm coming to my friend via my account. This is the reason why I am retaining the personal block. none other.
  5. I'm going to respectfully disagree with item 5 on that basis - I do not wish to ever have contact with these individuals in my online experience, and I trust that all 509 of the users who have blocked me have their reasons.
  6. Regarding point 6 I would disagree as well, solely on the basis that multiple mute lists I am on (and as the 4th most muted user by number of lists on the protocol I should know what that feels like) are, at the end of the day, completely arbitrary - if you were to believe everything said about me via the lists I am included on, I would be shuffled off to the Hague in a heartbeat, which is to say, simply: people are going to make completely biased decisions regarding these lists. I recall one user had a list entitled "Nazis" which was comprised entirely of Jewish users. The list name has since been changed, but you get the idea. Secondly, not every individual using the lists feature intends on it being an objective categorization - I have been added to mute lists by users whom I have never interacted with, for reasons I will never know, and that is something I will have to deal with.
  7. Regarding point 3 - I do sincerely hope that it hasn't been seen that my role is in any way officially endorsed by Bluesky or any other entity besides myself. I, like Skywatch, operate in an unofficial capacity, without endorsement or any moderation capabilities.

In closing: I would like to issue an official apology on the part of myself, and myself alone. I cannot speak for anyone else. My Discord is in my Bluesky bio if you wish to reach out, though I understand that the grievance may still stand and perhaps this apology seems insincere. Regardless of that, I am sorry for any harms done, and I hope you understand that I hold no personal grievance against you.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants