6 ms·
I got a 7 day ban for a tweet that included “political suicide pact” as an idiom. I then got permanently banned for including the line, “sending their children
by Waterluvian 4y ago
I got a 7 day ban for a tweet that included “political suicide pact” as an idiom.
I then got permanently banned for including the line, “sending their children to die in Ukraine” in a tweet.
The “appeal” button causes me to get a denied email within one minute. This tells me no humans are in the loop on any of this.
I don’t think these tweets were controversial or require any diversity of viewpoints. They just require appreciation that you cannot automatically moderate anything accurately unless you are prepared to be very very VERY relaxed about the rules.
I’m not sure Elon is even interested in fixing this kind of problem. He seems focused on the politics and “cancel culture” type issues (whether they’re real or imagined).
- concinds 4y agoTwitter ignored multiple reports of personal threats directed at me specifically (involving concentration camps and general death threats, etc.) "This tweet doesn't violate Twitter rules." People I didn't know, who I had zero negative interactions with before. That was way before Musk. You just can't have human moderation at that scale. They have to brainstorm how to prevent these tweets in the first place, rather than how to catch and delete them.
- Waterluvian 4y agoThat’s awful. And I’ve also witnessed the same kind of abuse that goes ignored. Has any large social media site ever solved this? 4chan does by having no rules: if you show up, expect the absolute worst of humanity. Reddit does by having countless volunteer moderators who basically do 98% of the moderation. Facebook suffers from the same problems as Twitter.
- concinds 4y agoThe way I'd solve it is here: https://news.ycombinator.com/item?id=33377423 https://news.ycombinator.com/item?id=33377423 Basically, more siloed converations would ironically create a better "public town square", because it would reduce the need for moderation and allow freer speech. > That’s awful. And I’ve also witnessed the same kind of abuse that goes ignored. Honestly, it wasn't. Maybe due to my temperament (and I certainly received far less hate than high-profile people do) but I could never empathize with the "online bullying" concerns. It's the nature of the internet, and you just develop the mental strength to tune it out, or use it to your advantage as fuel. Even after those experiences I'd still be much happier with Musk's content policies, or Dorsey'd blockchain-free-for-all with users being able to choose the moderation algorithm they want. I have friends who were suspended for BS reasons and that's a far bigger concern to me.
- Waterluvian 4y agoHmmm. Would you consider that to be similar to the Reddit model of cloistering themes so that unlike-minded people aren’t artificially pushed together?
- concinds 4y agoThe problem is that reddit clusters based on topic, rather than social group. If I'm a bad state actor who wants to influence America on a mass-scale, I buy a bunch of accounts, and buy upvotes, and can shift the hivemind. Or I just infiltrate the mod team. Obviously you'd need to be relatively sophisticated to not get caught. But Twitter is based on social circles, not topics; those concerns don't apply. Instead of having an entire topic/subreddit be biased one way or another, you'd have small friends groups and small communities, that have social dynamics that are as close to real life as possible. Hence all the scale-related problems don't occur. The radicalization rabbit hole? Doesn't happen with siloes. Harassment? There'll be much less with siloes, where someone needs to "add" you (like on Facebook) to interact with you, rather than a drive-by quote-tweet. Ideological censorship? Again, with siloes, there's less need for moderation, so people who aren't harming any others get to communicate with each other. It's more organic. Obviously implementing that isn't as easy, but there's good ideas for clustering in this thread.
- tqi 4y agoTBH I don't think clustering on social groups is the panacea that you hope. Misinformation spreads insanely fast on WhatsApp, even after limiting Group thread size and # of forwards. We'll never get back to the way things were before the internet became ubiquitous. Back then, we had a limited number of sources, and normal people had very limited reach. But now, it's too easy for individuals to pass information along, and with more sources of it the juiciest / most outrageous information will always go the furthest.
- concinds 4y agoMay not be a panacea, but it's the healthiest option. > Misinformation spreads insanely fast on WhatsApp, even after limiting Group thread size and # of forwards. That's true, but it's relatively easier to solve: really viral messages are a minuscule minority, and can set off internal alerts and be reviewed by humans; to prevent hysteria and murderous rhetoric. And the problem also existed with word-of-mouth, centuries ago, before the masses ever startied reading "the press"; rumors could spread across entire nations. Fake stuff will spread, and people will become more skeptical and adapt, as they did before. Skin in the Game makes a related point.¹ The whole problem with censoring "disinformation" is that it removes society's self-healing mechanisms. Public discourse has always been messy. Louis Farrakhan is labelled hateful, but he ran the Million Man March and is credited getting many African-American men to register to vote. Malcolm X would have been considered an extremist and banned. But it would have been wrong; he may have been flawed on the micro level (that's subjective), but he was beneficial and heroic on the macro level. Censorship has huge second order effects, including self-censorship, introducing status-quo bias. It also promotes conspiratorial thinking, since people who think information is being hidden from them are more likely to seek fringe sources. How do you censor real disinformation, without ending up with things like the Hunter Biden story being censored too? It also allows governments to lie with impunity, because accusations that the government is lying are treated as disinformation, while actual government lies (like the intel agency staffers who said the Hunter laptop was a Russian psyop) are spread everywhere.² People need to be able to spread "conspiracies", since while they may be factually incorrect, they may be directionally correct, while the "official" narrative may be total BS. Note I couldn't care less about Hunter or his laptop, but it's a clear recent example. ¹ https://medium.com/incerto/the-facts-are-true-the-news-is-fake-5bf98104cea2 https://medium.com/incerto/the-facts-are-true-the-news-is-fa... ² https://greenwald.substack.com/p/biden-the-media-and-cia-labeled-the https://greenwald.substack.com/p/biden-the-media-and-cia-lab...
- 0x00000000 4y ago4chan does have rules. Low quality posts, spam, racism and nsfw are against the rules on most boards.
- Danin 4y ago4chan has rules, albeit incredibly lax and with biased enforcement, but they are enforced (see /bans) eventually. Many of the larger boards seem perpetually and woefully understaffed which gives the impression that it's a lawless website to a casual observer, but depending on the time of day it's likely just the case that there's an insufficient amount of jannies to deal with the volume of posts, with posts being deleted upwards of an hour after being posted. One of the more notable examples of rules being enforced on 4chan was its stance and subsequent banning of anything relating to Gamergate, causing a large part of the community to leave for 8chan. Or more recently the removal of the /qa/ board for cross board raiding.
- MonkeyMalarky 4y agoMake people pay for a subscription and use some of the money to fund moderation?
- Waterluvian 4y agoOooh I forgot! The Something Awful forums approach. I think this is the best one.
- matheusmoreira 4y ago> 4chan does by having no rules It does have rules. Lots of them. To the point the place changed significantly over the years. You probably mean 8chan which had exactly one rule: don't do anything considered illegal in the USA.
- freejazz 4y ago"You just can't have human moderation at that scale" oh you definitely can. It's just not remotely cost effective and would ruin the bottom line of any tech company.
- kazinator 4y agoBut the scale may just radically shrink under the new boss.
- freejazz 4y agoBy what? all the users leaving?
- Kiro 4y agoNaive comment ignoring other obvious problems with human moderation: https://www.thebureauinvestigates.com/stories/2022-10-20/behind-tiktoks-boom-a-legion-of-traumatised-10-a-day-content-moderators https://www.thebureauinvestigates.com/stories/2022-10-20/beh...
- freejazz 4y agoI'm not advocating for human moderation so much as pointing out that the profits obtained by these social media organizations that eschew what would be reasonable moderation in any other medium come from their externalization of this cost.
- lelanthran 4y ago> Twitter ignored multiple reports of personal threats directed at me specifically (involving concentration camps and general death threats, etc.) "This tweet doesn't violate Twitter rules." I want a site where I can see all these screenshots of tweets with Twitters "This tweet doesn't violate Twitter rules" in the shot. Pair it with another site that has screenshots of tweets that were banned, that obviously don't violate the rules as shown in the first site.
- anonymousab 4y ago> You just can't have human moderation at that scale Effective human moderation - or at least, followup to moderation - should be the necessary cost of scale. If you can't operate at a certain number of users under that constraint then you shouldn't.
- matsemann 4y agoI got a 48 hour ban recently for sharing a quote from an article we were discussing. The quote was nothing special, just a statement from a spokesperson for a company. The "problem" was that I included the name+title of the person I quoted, and that somehow got flagged as me "doxxing" that person and their place of work. Whose job it was was to be a public spokesperson for this company.. It wasn't any political or contentious topic, so no idea what set it off. Basically along the lines of "as X working at Y said in the article, they want to do Z soon".
- RetpolineDrama 4y agoMy SIL got permanently suspended for posting pitbull attack statistics. Some pitmommy group (what they called themselves) found her tweets and mass-reported her. I doubt a human was ever in the loop, if you get enough reports off with your head.
- yrgulation 4y agoYeah buggy ais plague twitter and facebook. Something similar happened to me on facebook, apparently i violated a french law when using a harmless idiom. Was unbanned within the hour tho because it was obvious i wouldn't really sell my own kidney to buy ethereum if it dropped at 200 usd despite its ai thinking i am involved in human traffiking. They even apologised which was nice. Having said that i think relying solely on the judgement and discretion of rich powerful ceos to regulate political discourse is dangerous. Its how oligarchies are born. We need clear laws on how such platforms and news sources or distribution mediums work.
- smcl 4y agoOn the other hand, I’ve reported posts for using racial slurs (you know the one) and for calling for genocide … and been cheerfully told they didn’t break any rules but that I can block the user if my feelings were hurt. These sites aren’t moderated ideologically, they’re moderated randomly
- gpm 4y agoAnd really, random moderation is inevitable at scale. It's always going to be a judgment call made by different people in different moods. You can spend more money to reduce the std deviation, and spend less while increasing it, but you can't eliminate the randomness.
- Beltalowda 4y agoEvery time I've reported something it's taken well over a week for Twitter to respond; all denied, and all of them were very clear-cut cases IMO (I don't have the exact phrasings at hand, but all involved people fantasising about smashing someone's face in with a brick because they didn't like something they said). I don't expect moderation to be 100% consistent or 100% perfect, but now it just seems like such a random unpredictable crapshoot that it's pretty broken IMO.
- ghostpepper 4y agoI'm not sure what country you're in, or what the law is in the US, but I think there's a difference between saying "I want to to smash their face in with a brick" and "someone should smash their face in with a brick" It seems to me that the call to action or lack thereof is salient. That being said I have no idea if this was the reason why it was allowed. There's no transparency.
- Beltalowda 4y agoI agree there's a (subtle) difference, and legally this difference may certainly matter, but everyone knows what's intended and IMO neither have their place on a platform where "people have a free and safe space to talk" (from Twitter's "About" page).
- rubyist5eva 4y agoI got a permanent ban for quoting a video which contained "when do start killing white people" and said that this was bad. Banned for inciting violence, upheld on appeal and told it would not be looked at again. Twitter's current moderation policy is extremist ideological garbage.
- hot_gril 4y agoYeah, people self-censor nowadays. Common to see TikTok videos where a benign use of the word "kill" is crossed out or spelled differently just to avoid running afoul of automatic filters. You do what you gotta do.
- fleddr 4y agoI reported a tweet where somebody threatened to kill somebody. One day later they came back to say that it's fine, actually.
- aaron695 4y ago
- alxmng 4y agoA few years ago I tried to reply to a discussion with a link to the Anti-Federalist Papers, and my tweet was censored/blocked. If I recall correctly, it was labeled as misinformation or something. The censorship on Twitter goes far beyond hate speech or threats of violence.
- yodsanklai 4y ago> This tells me no humans are in the loop on any of this. At that scale, it's very hard to achieve. 500M tweet a day. Let's suppose 1% of the tweets are flagged by users or by keyword, that would be 5M of tweets a day to review. Let suppose that one person can review 1000 tweets a day, that would be 5000 people working full time. And even with humans, there'll always be some flagging that will be contested. Besides, different countries have different legislation regarding what can be be published on the web. It's a hard problem to solve, even with the best intent. I don't think Musk is going to fix this. I think it'll be more about reinstating people like Trump on the platform.
- pnt12 4y agoWhere I'm from, minimum wage is around 15k per year. So hiring 5k people would cost 75M in salaries, and let's just double it for other operation costs, 150M. People don't work 24/7, so we must quadruple it for triple shifts + weekends. That leaves us at around 600M: yes, that's terribly expensive. 1 But maybe we don't need to review all the reports! What if we rank them? Correct reports grand a user positive 'report karma', and they climb higher. Posts that are 'trending' in reports should also climb higher. New posts take priority over old posts, ie let's police new content, not everything that has been ever written. And of course, we can also use modern techniques, ie AI to score the reports. If Twitter did that and could review the top 10% of reports, at 60M per year, would that be worth it? If successful, I think so. 1: this number is biased because it assumes all reports are independent, but I suspect a significant amount of reports would concern a low number of posts, eg controversial or political posts from accounts with a large follow count.
- pcurve 4y agoDid you have a lot of followers? Im guessing manual and auto moderation approach differ based on follower count?