“Everything that isn’t illegal is allowed” sounds great until you see what that actually means on the internet.
Anyone who has ever worked in content moderation / trust & safety knows what kind of unrelenting deluge of obnoxious / disturbing / spam-filled / miserable race to the bottom of the lizard brain stuff that is constantly being pushed back on any moderately popular social media site.
It seems so easy from a distance. Just let people say what they want to say, right? Unfortunately the result of that is a place that very few want to spend time in.
Well, I'm going to be somewhat controversial here and say: What the fuck were people expecting?
Moderation is hard, genuinely hard, every order of magnitude increase in community size is not linear to the moderation requirement: it is factorial.
Why? Because every single communication has the potential for abuse, and the number of interactions on a platform do not scale linearly with the increase of users.
This is why things like the "Eternal September" exist, a deluge of new users is basically impossible to moderate at scale.
I think Twitter, Facebook and co. have done a fairly decent job of the mess they made, but crucially they decided that a walled garden where everyone exists together was their business model.
I think this is fundamentally flawed. "Back in my day" (I know it may be glazed with nostalgia, but) smaller close knit forums were much better at moderating communities, because it was still humanly possible.
There did exist some communities which became tyrannical; but the benefit of small communities is that people just go wherever it's "nice enough", and if you don't like the moderation staff or how they moderate you can move on with your friends.
I think people don't want this to be true, people are so financially invested in the centralised model; but ultimately you force a single set of potentially tyrannical moderators and a single culture. -- and people aren't willing (or able) to pay for the correct level of moderation.
I was thinking about forums and how much better they were the other day. One reason, the main reason, was because they learned lessons from Usenet and stopped any flame wars in their tracks. These social media networks didn’t and have let our entire society devolve into a giant flame war. We need systems to force people out of heated conversations and ensure they are engaging in good faith. This requires human moderation but we could surely build tooling to detect if a conversation is heated and force people to take a breather.
> This requires human moderation but we could surely build tooling to detect if a conversation is heated and force people to take a breather.
But who gets to decide what a heated conversation looks like and who should take a breather? This is the fundamental problem that we can't seem to agree on as a society.
We do agree though that almost all political discussions are flame wars, we just think the other guy is the problem. I think just adding cool down periods when two people are rapidly responding would go a long way, as well as giving users a timeout from one another if this happens often.
My personal opinion is that not being able to see the other person's face when you're communicating with them is a big part of the problem. Perhaps a cooldown period would help and I'm all for trying to find ways to improve discourse online, but at the same time, the psychology of being able to say whatever hateful thing you want because the other person is hundreds if not thousands of miles away drives a lot of this behavior. This is true of forums, gaming chats, etc.
> This is why things like the "Eternal September" exist, a deluge of new users is basically impossible to moderate at scale.
Great reference. I used and liked Usenet a lot in the early 90s. I lurked in the comp.* and sci.* groups among several others. Sure, I knew there were nasty crazy things on alt., and as a 13 year old I looked at some porn in there.
Usenet was not "centrally" moderated, and it was fine. There was spam sure, but with a suitable client with spam filter, things where good.
In my view, moderation has to happen at the edge, and not in the center. People should be able to post whatever legal stuff they want in those type of services, in the same way anyone can go to a public park and shout/speak whatever crazy things they want. Now, if you start to pee* in public (illegal) or post something illegal, then the police should investigate and get you for committing an illegal act, but there's not reason why there should be censorship of everyone for the possibility of someone committing a rime.
> every order of magnitude increase in community size is not linear to the moderation requirement: it is factorial.
> Why? Because every single communication has the potential for abuse, and the number of interactions on a platform do not scale linearly with the increase of users.
I'm not sure this is correct. It sounds like your underlying model is "number of users N, number of potential interactions N x N." But people have finite time and resources. Every user can only post a maximum of T times a day, where T is some constant. So I think the number of actual interactions is linear in N.
You’re thinking of 1:1 communications, I would guess.
In reality twitter is 1:n relationships.
Content that is interacted with may lead to new interactions from unrelated people. So it’s really n:n.
A persons posting time, in any event, easily approaches one’s ability to moderate it. It’s very easy to spew content and requires much more effort to analyse and weight it.
I sincerely believe it’s not in step with the growth of users, instead it is exponential.
It's possible. As a thought exercise, I'm sure Amazon has figured out some sort of formula to prevent the sale of illegal items on it's marketplace, which it has to operate at scale. At the very minimum a structure like that could be put in place. Another approach is the community of moderators that Reddit uses.
So many Silicon Valley companies launch products designed to scale without any regard for social impact - it's time to move beyond that myopic pov. It's not someone elses problem.
The incentives for bad actors are also different too. Imagine the difficulty to craft a bot ring that can reach into and affect thousands of forums to reach millions of users. It didn’t ever happen because it was too complex of a problem. Meanwhile, you throw all these millions of users in one basket, suddenly you only need to develop a botring for one network and the incentives are also sky high because how many people you can reach with that one botring.
Money. Moderators are expensive and if you want it to be any good you need people from the culture you're moderating to understand the context of what is and isn't abusive in a given language. It's also an absolutely terrible job because you're just sifting through the absolute worst of the content on the platform all day.
Generally either the users or the admins are unhappy with the moderators of any given subreddit. They're on possible way around it but not a particularly good one and it's less likely to scale because there's not the same "ownership of a community" feeling you can engender in Twitter where there's not really an equivalent to subreddits to give volunteer mods control over.
Moderators on forums had to handle maybe 100 messages per day. They knew the context of each discussion thread and could make pretty well-considered and nuanced moderating decisions.
Twitter receives something like 500 million tweets per day. So if you had a million paid moderators, maybe they would be able to keep up with the sheer volume. And then you'd still get people arguing either side, too much or too little moderation/censorship. Corporate bias would be attributed, rightly or wrongly, just as it is now.
As the GP stated repeatedly, it doesn't scale. Number of interactions scales with the factorial of users. The moderation team itself also doesn't scale. Good moderation is very difficult and requires trust. More moderators spreads the trust thin and greatly increases the chance that you end up with one or more bad moderators, who in turn damage that trust.
> Number of interactions scales with the factorial of users.
That is just not true. That's a count of possible relationships. There is a limit to how many interactions a human will perform in a day, and it's not related to how many other users there are on the platform.
The motivations are different, but I think wikipeida is an interesting example where editing / moderation does (mostly) work. Again, I can't see how that could ever work on twitter, but wikipedia is the only large-user-base example of "social media" that isn't horrible.
I think that is because every contributor is responsible for the whole, and thus they are also all moderators, and all responsible for any content digressions.
If someone posts a hateful tirade on twitter, it's no one else's responsibility but twitter's really. It's their account, and Twitters platform.
Of course, if you gave Twitter uses the ability to self moderate, it would be an absolute mess.
The combinatorial explosion mentioned in the above post - it would be bad enough if the requirement for moderators expanded linearly with users, but it actually expands exponentially, more like with interactions.
Plus, Musk tweeted that he's eliminate bots or die trying. If he really means this, it would be a truly great contribution to twitter -- free speech is one thing, but amplified disinformation is another. But, this apparently requires levels of effort beyond all the social media today, especially when it is not just automated bots, but also paid troll farms grinding out disinformation and deliberately undermining communications.
Facebook does this no? I thought the issue was you effectively have to be a psychopath to do the job because of how horrific the unmoderated internet get.
As another commenter pointed out, it actually doesn't matter too much what the users want, from a business perspective.
Twitter's true customers are its advertisers. It's who pays. And they are even less likely then the normal person to want an "everything that isn't illegal is allowed" approach. Advertisers don't want their ads seen next to hate speech and certain other types of speech.
It's been clear that Reddit and YouTube have been censoring at the behest of advertisers for years, and I don't think Twitter is or will be any different.
> As another commenter pointed out, it actually doesn't matter too much what the users want, from a business perspective. Twitter's true customers are its advertisers.
Users matter the most because if non-toxic users flee the platform, Twitter has no value to advertisers.
Sketchy pornographic websites do have advertisers, just extremely low value ones. Twitter wants to maintain a pleasant environment for the medium to high value advertisers, because advertisers are its revenue stream, and primarily it is not an ideological entity, it is a profit seeking one. Ideology is a variable in the equation, but it is primarily and significantly bound by profit seeking.
So first, users are one of the major levers to keep valuable advertisers happy -- not the only one, but also not the only significant one, and they are purely instrumental. Second, Twitter does not want a negative public perception, along the lines of cable news, or much worse, 4chan. The former loses you a subset of a subset of the market and is a constant PR/lobbying headache, the latter loses most medium to high value advertisers altogether and is an existential regulatory threat. In almost no way is the worry that the toxic environment would be a subjectively bad experience for users.
Before Parler was banned from the App store (and before it was re-instated after more moderation was added), its stated policy was roughly to moderate strictly to what's within the bounds of the law, albeit with some significant inconsistencies especially around pornography and nudity. I used it during this time, and it was extremely common to see a lot of seemingly intentionally hateful but legal use of the n-word, etc. P&G, Nestle, etc., don't want to associate their brands with an environment like that. Obviously Apple didn't either.
It's important to always keep these realities in mind for all corporate action. Take Jack Dorsey's bluesky initiative[1], which many viewed in purely ideological or technical terms. My personal opinion is that it was either a way to counter negative PR ("Twitter is a moderated node that's profitable, but technically you can choose another node that's toxic and unprofitable but with less moderation, and good luck with that...", where Twitter gains all the "free speech" PR benefit with none of the significant costs), or a low cost self-soothing action to assuage pangs of guilt for instituting some policies he didn't personally agree with, with the important part that it is designed with a hard constraint to do nothing to negatively affect profits. So it's just an escape hatch for PR or personal angst.
I think we're mostly disagreeing on the meaning of sketchy, but it has many meanings. I'm using it to mean "legally dubious, in the sense of being full of scams, blatantly false promises, malware, etc." (I'm also ignoring issues around copied/pirated content). I'm sure Twitter contains some of the above, but they try to minimize/eliminate it.
But I think this is all a distraction from what I was trying to convey -- my point doesn't depend on getting to the bottom of what is meant by "sketchy". Just define X to be some kind of website that has extremely low value advertisers (scammers, etc.):
> [X] websites do have advertisers, just extremely low value ones. Twitter wants to maintain a pleasant environment for the medium to high value advertisers, [...]
I opened /g/ in incognito mode without an ad blocker, and I was actually surprised by the ads. There are no lowest common denominator "horny babes near you" ads or viagra ads. But there are ads for a crypto lottery, ads for a cryptocurrency, ads for a niche forum, etc. These aren't that different from the shitcoin ads on Twitter or reddit.
Nontoxic users have lower engagement than toxic users. Advertisers want to see average time on site go up and a good way to do that is with rage bait or all the other nonsense they’ve been pumping into these networks for years.
That's simply not true. Think of a slightly different model.
> We don't want our advertisements to be shown next to posts that do X, Y or Z
That's effectively, the same as
> We want to see our advertisements by A, B and C
Just give advertisers different options.
Further, Twitter advertisements aren't much different than billboards. Do you worry about the politics / optics of the drivers or even protesters beneath the the billboards in a city? No, of course not, you just want people to buy your products. Do people care, no.
It's all fake outrage, trust me people will still buy Coke or Pepsi even if someone sees something distasteful while doom scrolling.
In terms of general "toxicity" there's already blocking users, but allow users to "avoid topics" if they need their safe spaces. Then as a business, let users "pay to avoid topics" as it'll require some hand curation -- boom. More money for twitter, more users on the platform, everyone wins.
That only really applies if you believe the advertisers see Twitter as a neatly divided entity, but they don't; they'll see it as a site that does A, B, C, X, Y and Z, all of the above.
And if an advertiser pays a service that, for example, shows terrorist propaganda, then they could be seen as funding and supporting terrorism.
Better to not take the risk. This was another reason why the credit card companies and / or payment providers pulled out (hehe) of Pornhub; they did not verify the age or consent of the people involved, thus were the payment providers complicit in perpetuating this. It would have been morally unjust to keep funneling money into PH if they didn't do anything against them. Not that the payment processors can claim much moral high ground, generally speaking, but you know what I mean.
> if you believe the advertisers see Twitter as a neatly divided entity, but they don't
This is a PR problem and fixable.
Further, this is an issue with advertisers. As Elon is proposing, have people pay to get verified ($5 / year or something). Further, as I mentioned, you can "pay to avoid topics" (have users label / flag stuff they don't want in their topic, like they do now generally for breaking the rules) and advertisers can select topics to promote on AND topics to not promote next to.
The reality, is Twitter would then get a large segment of revenue outside of advertising AND advertisers would feel more confident -- boosting revenue from both sides.
> This was another reason why the credit card companies and / or payment providers pulled out (hehe) of Pornhub
This is different, the advertisers didn't pull out of Pornhub, payment processors did. Twitter could probably lose half it's advertisers short term, but if they grew the user base they'd make more money long-term. Advertisers buy eyes not virtues. Yes, a bad post next to an ad is somewhat damaging, when your choices are literally: Twitter, Google, Facebook; are you really going to cut one of your 3 choices? Not long-term.
> just look at how quickly advertisers abandoned YouTube following the wall street journal reporting.
There's not a chart anywhere I can find that shows advertisers reducing their spending on youtube. I "articles" (potentially paid for advertisments) claiming that was happening; but as far as I'm aware, there's no evidence that it made a material difference.
There’s lots of evidence around from people reporting their YouTube revenue falling 80%+. At the time it was colloquially known as the “adpocalypse” so you can find lots of evidence associated with that search term.
It's hard to explain how much you're understating the difficulty in doing this. It would likely require AGI to do this well. Until then, advertisers buy ads on entire platforms (like a given TV network), and they have to rely on the platform to ensure that the content is advertiser-friendly.
> Twitter's true customers are its advertisers. It's who pays. And they are even less likely then the normal person to want an "everything that isn't illegal is allowed" approach. Advertisers don't want their ads seen next to hate speech and certain other types of speech.
Hell, advertisers don't want their ads to be seen next to Elon's current Twitter feed ( https://twitter.com/elonmusk/status/1517707521343082496 ), let alone what Twitter will look like in a year after everyone starts posting using the owner's timeline as example.
True, but wasn't the porn ban to appease advertisers to try and actually make Tumblr profitable? Of course it probably failed, but it seems like it may have been a no-win situation.
It's true. However, if toxic users drive out all the non-toxic users, will those toxic users pay real money for the platform? Will they even remain at all if they can't get the reactions they (seem to) thrive on?
Advertisers care that the market doesn't primarily see that as the point of twitter - i.e. the average user doesn't think that Twitter is primarily about that.
At this point, he sounds like a politician making conflicting promises that he can't possibly keep... And also like a politician, he probably has his own interests in mind ahead of your average Twitter user's.
"... holding that corporations have a First Amendment right to free speech because they are "associations of citizens" and hold the collected rights of the individual citizens who constitute them." [https://en.wikipedia.org/wiki/Corporate_personhood#:~:text=T....]
> The Supreme Court codified this in CITIZENS UNITED
Actually, the Supreme Court established it no later than First National Bank v. Belotti (1978), despite the frequent claim that this was an innovation in the 2009 Citizens United case.
There’s one obvious flaw in “Everything that isn’t illegal is allowed”: determining what is illegal. Free speech laws are some of the trickiest legal issues we grapple with in the US, and many statues hinge on the intent behind the speech. How is Twitter supposed to implement this (hypothetical) new policy? Do they always give posters the benefit of the doubt? Seems ripe for abuse. Assume the worst? Probably more censorious than it is today. Punt to the courts? Great, moderation now takes years and costs thousands of dollars. What is the standard of proof to take down a tweet? Preponderance of the evidence? What evidence is admissible? Does Twitter just internally recreate the US trial court system to manage this? Do they do that for every country? The point is, the law on these issues is complex and frequently requires significant interpretation. Maximalism is no silver bullet.
This isn't that complicated. Twitter can just take down content in response to court orders, unless their review team decides to fight it on whatever grounds.
This is not only simpler for Twitter to implement, but provides a better level of due process and accountability. How many times has someone been 'abused' on social media and claimed XYZ company did nothing about it?
This is peak modern Silicon Valley: privatize the revenue for the product, socialize[0] the costs (by clogging up courts, in this case). It's not remotely scalable.
0: I don't agree with this vernacular, it's just what the kids say.
Sure, but that relies on someone’s ability to secure the court order. That takes days at minimum, possibly much longer. It costs money, possibly thousands of dollars or more. And what about cases that cross borders? Is someone from from South Africa supposed to seek injunctive relief from a U.S. court? Maybe this is better than the status quo ante, but it’s not obvious to me that that’s the case and it doesn’t seem like anyone is asking the hard questions.
They won't or if they do not for long. Gab, Parlor, and all the other right-wing "we're getting censored on Twitter come here where we won't censor anything (legal)" Twitter clones have all figured out rapidly why content moderation exists and that it's basically a necessity on the web as they get mercilessly trolled and spammed.
All he's really saying is he'd prefer to accept more shitty behavior from people who he aligns with and less from people who annoy him.
The only reason those spaces implemented content moderation was because they couldn't get hosting anywhere. Not because of 'merciless trolling and spam'.
Honestly, I think “Everything that isn’t illegal is allowed” is fine as long as Twitter only shows me content from people I follow. I think the whole "content moderation" problem is something Twitter has brought on itself. I am perfectly happy moderating my own content if they would only let me.
This is all I've ever asked for. Just give me the tools to moderate for myself / choose my own filters, and I could not care less who deserves to have what sized megaphone FFS.
Can you elaborate on your point? Are you talking about Russian bot accounts, or that people in other countries may have views which are not suitable for Americans, or what exactly?
If your only rule for how to moderate your platform is "what is legal", how do you scale that out internationally where there are significantly varying standard for what legal speech is?
It's easy to point to 'oppressive' countries like China (where twitter is banned), or Turkey[1], or India[2] as examples of where speech can be severely limited, but there are many limitations to speech in other countries like Germany, Canada, France, Australia, UK.
When you make a globally available website, how do you "follow the law" when the laws of various countries are not compatible with each other? Do you geolock all tweets - Canada twitter is vastly different to US twitter? That doesn't sound scalable good!
Indeed; China doesn't like references to a certain children's cartoon / book character, and nearly two billion people do not take kindly to any kind of disrespect aimed at their prophet.
You can also look north to Canada where the American notion of “free speech” does not apply. There are reasonable limits on hate speech, obscenity, defamation and the like.
I agree that people take for granted just how much spam there could be if the flood gates were opened.
Is there a reason you can't turn all of the bans (of legal stuff) into filters? Then users could turn on and off nazis, Q, sex, porn, etc? People could turn them all off and get the same experience as today.
That's essentially a hard AI problem of making something not human that can categorize any post in a rapidly changing cultural environment. For probably the most mutagenic category look at some of the Q nonsense that will rapidly swing from topic to topic and coopt existing conversations like "Save the Children" for their own insane ends.
You can filter accounts instead of posts and recreate the current scenario. Those accounts will complain about it, but it at least brings them back on board without hurting the experience of people who don't want to see them.
That said, they are already auto-flagging posts to some extent.
Maintaining that list is extremely tedious work and pushing that down to the individual user just drives people away. Users don't want to constantly filter and block an ever growing list of spam, bot, and troll accounts. Parlor/Gab/et al essentially tried this "let the users block" method and it failed practically immediately and they've all instituted some sort of content moderation or just disappeared.
The solution to this problem is AdBlock. Everything that AdBlock needs to be the success that it is today is exactly what we need for social platforms as well. Both for moderation(filters), and for the 'algorithm'(sorting).
Want to keep the default list? Good, just do nothing and you will see what Twitter wants you to see. Want to see unfiltered? Just turn the blocking off. Want to add a filter just for yourself? Easy. Upstream that filter such that it eventually becomes the default? Go for it. Have different filters to cater to different cultures? We've got the different lists for you to subscribe to.
They should be made so that you can algorithmically pick whatever subset you want. If you just want Parlor content, go for it. If you want everything except Parlor content, just flip that flag.
Something literally illegal? That gets deleted and removed from everything.
I'm saying let Twitter do the work of blocking those users. Instead of hitting the "ban" button as they do now, they hit the "Nazi" button. Now they're Nazis and everyone who says "I don't want to see Nazis" won't see them. It's not any more work on the user's part.
And I haven't seen a reason that this is any more tedious work on Twitter's part than the work they do blocking it. Maybe one argument is that Nazis are more likely to post illegal stuff, so banning them early preempts the work of looking for reasons to ban them later.
People already abuse flagging systems to try to push people off Twitter and Youtube so Twitter will need something to deal with people false flagging posts or the "no XXXX" filter will be useless.
This is a reality today in the Mastodon world where some self appointed hall monitors created Fediblock to create a list of bad instances which created a shadow network of instances on the list that subscribe to each other and discover new instances to federate with using the Fediblock list
> Is there a reason you can't turn all of the bans (of legal stuff) into filters?
Among other reasons, because Twitter is a brand with an image to maintain. They don't want to be known as a safe hangout for Nazis and other unsavory content.
The old tenant of the internet was “if you don’t like something, block it”. All of these people saying that this doesn’t work in practice on their social media websites have not provided an explanation as to why this is the case.
No, the old tenet of the internet was "if you don't like something, don't subscribe to it". Usenet, Web 1.0 forums, news feeds, email lists didn't have block features; if you wanted to see something, you subscribed to it. If you didn't, unsubscribe. Even if someone on a mailing list was shit, you didn't (and couldn't) block them, you'd either filter out their messages or bail.
That's the actual problem with algorithmic feeds; they want to find ways to put things you don't subscribe to into the your view and the views of people you follow. You can't opt in to the content you see. Even if you studiously avoid algorithmic feeds, the people you follow won't, and they'll share that content onto your feed anyway. Even if you and the people you follow avoid them, nothing's stopping the algorithm from putting you into others' view and effectively inviting them into your feed.
Thus blocking/muting going from being primarily a self-moderation tool against abuse, to a necessity to stop the endless stream of algorithmic content and commentary coming from people you don't subscribe to, or who don't subscribe to you.
If you're in a marginalised community then the objectional content comes to you and can't be avoided. We simply aren't provided with the tools to block this.
There is a fundamental asymmetry in harassment.
My account is important to me; I don't want to abandon it or give it up. If my account is under attack I cannot continue to use the site as I would like.
The accounts used for harassment are either disposable and it doesn't matter to the harasser whether they get blocked or banned. And non-disposable harassing accounts that get blocked can either just move onto the next target or continue to direct the harassment through screenshots etc which encourage the disposable accounts to do the dirty work.
The cost to the victim can be meaningful, but the cost to the harasser is non-existant.
And this isn't just applicable to concerted harassment campaigns.
There is also a lot of "drive-by" harassment from accounts who will just reply to any black/trans/queer/woman who posts online.
It's a question of scale. In the old internet, I subscribed to a few news groups that got tens to hundreds of posts a day. You could easily plonk the few people you didn't like.
It's also bullshit. In the old days people used to file complaints with your ISP or uni or get you removed from distribution on the server. But usually newcomers were brought in line by the community. Now, we have eternal September and it's simply not feasible to educate the hundreds of millions of people who don't know how to behave.
Well, unfortunately we aren't on the old internet though, are we?
Before we talk about this approach we would need the tools to take timeline and personal content moderation in our own hands. This, however, isn't in the interest of Twitter, algorithmic timelines and ad/outrage shoveling in general and thus it probably won't happen.
I wouldn't even be surprised if excessive blocking would lead to your account getting flagged.
You are basically criticizing people for not building their own functional shack while all they have at their disposal is a bunch of timber of varying quality and merely a few rocks as "tools".
Massive social media sites are affecting public discourse, elections, international politics. I don't have a Twitter account, but I don't have any choice but to participate in a society that has polarizing hot takes boiled down to 280 characters.
Simple asymmetry. Scammers can generate junk faster than I as an individual can block it. My choices are to either use a moderated platform or abandon social media altogether.
because community is defined by people not content. A person who posts overtly racist content probably also posts about visiting Disney with their kid, or how much they love Spider-Man. You can’t filter out certain aspects of a persons personality within a community. Either you want to share a space with people, or you don’t, you can’t share a space with people without knowing they’re there.
Why not? It seems like some kind of categorization system (even a really naïve keyword-based one) is already used for things like advertisements. If such a system was surfaced, and you can filter out specific categories, you are then able to see what the hypothetical racist $relation posts about the family, without having to see their hot political takes.
Remember that we are talking about words on the screen 99% of the time. I would be willing to bet most of us are reading on a post by post basis and don't know the full spectrum of every given persons beliefs (in fact, I'm not sure this is healthy or desirable).
Personally I'd consider it an improvement, but that might be because of my naive assumption that people will finally stop mistaking twitter for a sensible way of communicating.
I can't tell you how depressing I found it when Musk described Twitter as the "town square" and his whole rhetorical take on it needing to be a place for free and open discussion as per the "market place of ideas" concept (I don't recall him referencing the quoted term directly- that's my editorial).
Like... we really want TWITTER, of all things, to be the place for important social discourse? I never had a Twitter account, but isn't it still limited to some 200 characters and/or an image per post? And from what I've seen the "threading" of discussions also seems to make replies difficult to follow. Apparently, our society wants important social issues to be discussed in 200 character snippets. I'm going to go cry into my coffee.
Relax, broad social discourse has never been particularly nuanced or long-form, and has always been partially driven by catch phrases, headlines, and incisive, memorable quotes.
Twitter makes it impossible to dig in replies past a very low number of them. Hell, if the tweet is not the last in a chain, you often can't see any replies at all. It is explicitly optimized for very different things than public discussion.
That's not really how it works. The most important tweets are links to long-form discussions. It's a discovery platform. Sometimes the poster will include an executive summary in a thread. That's helpful, too, for deciding if you want to follow the link.
A lot of academics are on twitter, and it's a reasonably good way to find interesting new work. It has problems, but the character limit isn't one of them.
The market place of ideas that fit within a tweet, which is a pretty limited set of ideas. Most of the ideas Elon has don't fit into a tweet, almost nothing of value does.
Its not so much that twitter is a sensible way of communicating its that twitter gives you such wide access to such a massive array of people and topics. For example its how I get my news, I just follow a ton of journalists from different companies and countries. As 1:1 communication you are right its not built for that but it does what I use it for really well.
Musk has a reputation for craziness but if you look at his engineering decisions at SpaceX his record there is highly pragmatic. So the question is: which Musk is buying Twitter?
I personally don’t believe he means that he wants to turn Twitter into 4chan. Rather what he’s saying is that nobody should be censored on Twitter for the content of their (otherwise civil) speech. There is a wide gulf of possible moderation policy choices between current Twitter and 4chan that he could park it at.
Twitter is already a cesspool. I doubt it could get any worse. The only real change I see is more users crying that theres more "transphobia" on the platform.
It's almost like this kind of thing does not exist. Just go to 8chan or whatever it is called now. You will want to pour Clorox in your eyes. Feel free - enjoy "free speech".
The Internet was actually LESS free back in the day, in the sense that we hung out on tightly moderated Perl CGI discussion boards, where idiots and trolls were not tolerated. Usenet was great too, without moderation, as it required one to be pretty technical to get there.
Well, it's still there. Google's Deja probably did not help, "democratizing" Usenet. That said, here we go - another point in favor of strict moderation.
Make it so people can say whatever they want, but it's easy to decide what level of discourse you're personally willing to listen to. You have the right to say what you want and other people have the right not to listen.
It's not an easy system to develop and maintain, but Twitter pays well enough to afford people who can do it. The goal just needs to be set.
i think a big problem is monetization. You have to identify the conversations with the most participants in order to get ads there for sales. You have to design feed algorithms and promote certain users to drive views and adclicks in order to sell more ads.
If Twitter goes private then maybe the income expectations change and therefore the platform algorithms can change. I don't use twitter but it seems like putting more moderation power in user's hands ( chronological feeds, easier management of what you see and what you don't, etc ) becomes possible when Wall St. expectations are no longer a part of the design process.
If you don't have a stock price to answer to then you can do things that create a healthier community but may not be the most profitable.
EDIT: You know, Musk be on to something about leaving a company private. Tesla is public and the shorters almost killed it, he really had to fight them and still does. SpaceX was left as private and is thriving. maybe he's taking lessons learned from both Tesla and SpaceX and trying to apply them to Twitter? just a guess
Yeah it's not even a mysterious lesson. A handful of "we won't censor or moderate except illegal activity" twitter clones have popped up and rapidly learned why moderation exists.
We could just filter that stuff the same way we filter email. Naive Bayes works at email scale why not comment scale? It should all be done on a per user basis so the filters only filter out language you personally find offensive.
It never made sense why try to force me to use the same filter as the Karen that complains about kids skateboarding.
At the very very terrible TED interview last week:
Interviewer: You've described yourself, Elon, as a free speech absolutist, but does that mean that there's literally nothing that people can't say and it's okay?
Musk: Well, I think obviously Twitter or any forum is bound by the laws of the country that it operates in. So obviously there are some limitations on free speech in the US, and, of course, Twitter would have to abide by those rules. [...] No, I think, like I said, in my view Twitter should match the laws of the country
Just to elaborate on how uneducated these people are on the topic of "free speech" and running something like twitter:
Interviewer: Right. So you can't incite people to violence like a direct incitement to violence. You can't do the equivalent of crying fire in a movie theater, for example.
Interviewer: But here's the challenge, is that it's such a nuanced difference between different things. So there's incitement to violence. That's a no, it's illegal. There's hate speech, which some forms of hate speech are fine. I hate spinach.
"I hate spinach" is not hate speech, and importantly, hate speech isn't even illegal.
> And in fact the line from Justice Holmes in Schenck v. United States is “The most stringent protection of free speech would not protect a man in falsely shouting fire in a theatre and causing a panic.” That “falsely” is what’s doing the work, both in Justice Holmes’s hypothetical, and in how such a false shout would be treated by First Amendment law today. Knowingly false statements of fact are often constitutionally unprotected — consider, for instance, libel, fraud, perjury, and false light invasion of privacy. That would presumably apply to knowing falsehoods that cause a panic.
The spinach example is so bad it's either such gross misunderstanding as to disqualify that interviewer from any future interviews, or it's an example deliberately designed to conflate issues.
The whole difficult part of 'hate speech' is that it often is highly offensive and nasty, but what people find highly offensive and nasty differs (e.g. an atheist stating there is no god is deeply hateful and offensive to many people, or being pro-abortion). Ugh. I can't even.
Well, in US there are certainly attempts to make it so. There is definitely framework in place and there is the mores. The time seems ripe too given how people seem afraid of what people might say if you let them.
There are definitely hate speech laws in other countries.
For the record, I am sympathetic to your stance, but I am not sure it is accurate.
My stance is "these people don't know what they're talking about".
I'm not actually American, but my understanding is that there have been prior attempts, especially in some US states, to make hate speech illegal, but each time it just gets knocked down because it violates that pesky first amendment.
Now, if you're saying "All twitter should just follow the laws of the country", now you've got to decide which country?, because most Twitter users are not in the US, and Twitter operatates - has business entities and employees - in other countries apart from US.
Really, what are the attempts to make hate speech illegal in the U.S.? That would come up against the first amendment really quick. New constitutional amendments are unlikely to succeed in todays polarized political environment.
You pose an interesting question and I might not have sufficient information to give you a full picture, but I might try to show a glimpse of what I think may be happening.
Someone somewhere decided it may be a good idea to expand existing framework ( hate crimes[1] ), which was relatively easily adopted in America due to historical ( slavery ) and political factors ( combating racism ).
With that in mind, the first step is saturating the media with opinions indicating some sort of support for a policy/law change ( in this case hate speech - links with sample articles follow - note how old some of those are ):
Whatever efforts are happening here will be shut down by the courts unless they managed to do something about the first amendment. I am not saying these efforts are good or bad, just that they are (I guess in my opinion) extremely unlikely to succeed.
That would come up against the interpretation of the First Amendment espoused by the Supreme Court in 1969, in the case Brandenburg v. Ohio. Before then, the First Amendment was not interpreted nearly so broadly, and speech regulations at the federal and state levels were common, beginning with the Alien and Sedition Acts soon after the Constitution was ratified.
If/when the Democrats restructure the Supreme Court to have a liberal majority, overturning Brandenburg with a single ruling becomes on the table, and the door is open for hate speech legislation.
First part I agree with. The second part, maybe I guess? But that is, if ever, decades away so we can worry about that when it becomes a realistic prospect.
I don't think any of the current sitting life time appointed supreme court justices have expressed any will at overturning that decision.
In his TED interview with Chris Anderson last week, Elon said:
"If in doubt, let the speech exist. If it's a gray area, I would say let the Tweet exist. In a case where there's perhaps a lot of controversy, you would not want to necessarily promote that Tweet. I'm not saying I have all the answers here, but I do think that we want to be very reluctant to delete things..."
I'm quickly discovering that people see in Elon Musk whatever they want. Somehow he's going to both stop freedom of speech suppression AND shit-posting.
But what if - and just hear me out here - there are not ways around the fundamental problems of social media platforms as they exist today?
I think we've had plenty of years to demonstrate a way to make it work without the toxicity and damage to society, yet we've not done so - even with the most scrutinized platforms in history and the world's most capable software engineers.
We are social creatures. Social media is not at all an incarnation of 'society' in which we can function.
> "the most extreme 10% of the left and the right are equally frustrated."
yeah that sounds great but the devil is in the details and they have always been. I feel like Musk gets tunnel vision in his thought process. Like he goes "A leads to B, B leads to C, and then C leads to D and done." without contemplating the complexities along the way. It's interesting because he's not naive to business and how things work. He's certainly gotten things done in timelines that people thought impossible and even laughed in his face but other times he has wildly missed.
let people filter for themselves. someone is offended by pornography so chooses an algo that accounts for that, another is offended by anti-trans sentiment and another algo accounts for that. everyone should be more broadly free to speak but we are not all forced to listen.
> unrelenting deluge of obnoxious / disturbing / spam-filled / miserable race to the bottom of the lizard brain stuff that is constantly being pushed back on any moderately popular social media site.
I keep hearing this, and it makes no sense.
1) Add a block button.
2) Don't surface random content to users.
3) (Optional) hard restrictions on private messages.
This solves the issue entirely, with no need to moderate anything that isn't illegal.
The simple fact is that this is a really easy problem to solve. Companies just don't want to implement 2).
Trending topics, your friend liked/replied to this, your friend follows this person, we think you'll like this tweet... all of these things show up in your Twitter timeline now, without having followed any of the people they're showing you.
I mentioned this in another reply but maybe taking Twitter private enables Musk to make changes to the timeline that hurt profit (ad sales) but produce a better community or "public square". If you don't have to answer to a stock price then options open up.
You are 100% correct and most of us think that way. There are some people though that think just the ability to say terrible things or even just things they strongly disagree with is proof of the collapse of society and will lead to the end of all things. This is not meant as an attack on these people that feel this way as I think they are just unable to control themselves, and I think have just never been able to adapt to the concept of social media. They ruin their lives obsessing over what others may be posting or saying and how dare the rest of us not be outraged as well. I had a sibling disown my entire family over something like this.
I like this article on the trouble with not censoring:
But once you remove all those things, you’re left with people honestly and civilly arguing for their opinions. And that’s the scariest thing of all.
Some people think society should tolerate pedophilia, are obsessed with this, and can rattle off a laundry list of studies that they say justify their opinion. Some people think police officers are enforcers of oppression and this makes them valid targets for violence. Some people think immigrants are destroying the cultural cohesion necessary for a free and prosperous country. Some people think transwomen are a tool of the patriarchy trying to appropriate female spaces.
Each of these views has adherents who are, no offense, smarter than you are.
I would like to give people another perspective on events like Tumblr banning female-presenting nipples or Patreon dropping right-wing YouTubers or Twitter constantly introducing new algorithms that misfire and ban random groups of people. These companies aren’t inherently censorious. They’re just afraid.
I think an important thing about that CW thread is you get a strong group of regulars. Any fora with a strong group of regulars can survive almost anything thrown at it - except an influx Eternal September style of newcomers that overwhelms the group.
It’s been seen time and time and time and time again and I’m not sure there’s any real way to preserve it.
You have a similar thing with tiny tourist places - everyone wanting to go there forces it to change, it cannot remain a tiny tourist place and almost always ends up being a place that is “on show”.
But in here, even the author seems to point to testimonials that indicate that value of unique discussions gained just by allowing the conversation to happen.
It is sad that it ended partially because author was harassed for hosting it, but I suppose this is the price you pay in 2022.
As it pertains to Musk and article linked to Musk, I am relatively certain he can handle the mob.
Yeah, maybe Musk can take it on. But I'm not certain.
Tim Wu has a book "The Master Switch" also about this. I think his argument was once you become big enough, the rules change and you end up being forced to censor. You become valuable target for people who want censorship and close to a common carrier.
Then have filters that allow people to customize their experience. Don't want to see content about race? Filter it out. Only interested in science? Filter for only that subject.
The user should be the one who is given the tools of censorship.
Isn't that what the block, mute, and muted wordlist options are for? If these were not available options I could see the argument, but they are, so why isn't this sufficient for users to moderate their own content?
Just because you want to allow people to speak, doesn't mean you need to force people to listen. The current moderation scheme, which is to provide essentially 1 moderator with no ability to customize it.
The obvious solution to me is just let people pick their own moderation tools, or at least configure it to how they would like. Moderation is great, but what I want moderated is different than what you might want moderated, and we should support broader implications.
This also allows Twitter to gain some independence from the wrath on both sides who want more or less moderation. They can respond by saying "Use the X moderation tool if you want more moderation or "Y if you want less".
Except it already is exactly that way, anyone can block anyone else they want. It turns out blocking people does nothing when you're being harassed by big groups of people, the damage was already done before you got the chance to hit block. And also it doesn't stop people from trash talking you behind your back and furthering the damage.
And there is the other dimension where you actually want to remove the ability to block from some users, like public businesses and brands probably shouldn't be able to block complaints from customers and shareholders, politicians shouldn't be able to block their constituents, etc.
> obnoxious / disturbing / spam-filled / miserable race to the bottom of the lizard brain stuff
What you are describing is the heavily moderated, arbitrarily censored social media experience the entire planet experiences today.
> Unfortunately the result of that is a place that very few want to spend time in.
Perhaps social media is a place that we should not encourage people to spend inordinate amounts of time in. Perhaps the idea of combining all the world's people into a single centralized location to communicate with one another, should carry with it the very explicit notion that such a place would be chaotic, high-energy, and risky --- just like going to the casino for the night.
Michael Saylor had an idea he discussed on a recent Lex Fridman podcast. People should be able to post $10 collatoral via the Bitcoin Lightning network for safe passage on the web. This can be done by your browser via HTTP. Bad behaviour (such as causing abuse) results in a penalty being applied to the collatoral. You get what's left back when you leave. Because it uses Lighting, it's super fast and cheap, far faster and cheaper than a typical 2.5% credit card.
His point was that there's no conservation of energy on the internet right now. Requiring a small ransom for safe passage would fix that, retaining anonymity.
The problem with this idea (aside from the gratuitous use of bitcoin, and the pricing out of people who can't pay) is that punishing people isn't the hard part of online moderation. For most simple cases, a system of warnings, temporary suspensions, and permabans mostly works. Yes, you've got to deal with sockpuppets, but my understanding is that they're usually fairly easy to recognise, and the extreme cases are usually rare and notorious figures. Nothing in this process is so complex that it requires a stake system to be fixed.
No, the problem is that determining bad behaviour (or at least, doing so in a way that is fair, just, and broadly accepted) is very difficult. There's no objective naughtiness meter that detects when someone's behaving in bad faith. There's no set of words that always, regardless of context indicate undue rudeness. There's no perfect demarcation between just asking questions, and harassment.
So basically any system is going to have to make a series of judgement calls, and those judgement calls are going to be specific to the context of the person (or people) making the judgement. And most importantly: other people will disagree with that judgement call! Not least the poor victim of your unfair banhammer (or the poor victim of the abusive spammer you choose to do nothing about).
Money's not going to fix that problem - if anything, it'll make it worse when people have more on the line.
It's not trying to solve the problem of assessing whether rules were broken, it's attempting to add friction to bad behaviour. If you get banned and lose your deposit, it's going to cost you another $N to try again. Right now there's practically zero cost to signing up to a website again with a new account to continue the bad behaviour. At least this way, only the richest and most determined trolls will continue to have an impact on services.
Saylor discussed this mainly in the solution of DDOS mitigation. Perhaps he didn't intend for it to extend as far as content moderation and I'm taking his idea too far, but I think he probably did.
Though it will also exclude people that can't afford it. I'm thinking of poor people and children. The kind of people who use computers at the local library, because they can't afford their own device (and more expensively) the Internet connection to make it useful.
And yes, I realize that some of the above groups are also absolute shits, who I would keep off the Internet (if I could) until they decided to grow up.
It's not $10 per site, it's $10 one off. You would get the money back almost immediately if your request is not abusive. I'd be interested to know how many twitter users couldn't scrape together a once off $10.
You should ask Michael Saylor to be sure how he proposes this will all work, but he did mention his company (MicroStrategy) spends $1M/year on DDOS mitigation. I expect if Lightning ends up as cheap and fast as he and others are saying it will be, it could be funded by companies as a cost of doing business in the same way they fund the electricity and hardware for serving their websites.
It's actually the sort of thing I could see CloudFlare etc. providing, so each website doesn't need to implement it themselves.
The site. They stake their reputation on being fair. Unfair sites will lose users. This is basically what happens already with moderation, it's just adding a small monetary penalty to bad actors to discourage them.
It's not linked to your identity. You can make a new wallet any time, but it'll cost another $10. In that sense it makes a pretty awful social credit system.
We're still at the stage of trying to apply the crypto solution to more non-problems when the existing solutions would completely suffice. Crypto adds nothing to the original thought outside of shilling.
You're talking about a 'universal' ban system predicated on crypto. Whatever system you think is a good or bad solution, I don't think I want a universal social credit score system on the internet that lets all sites ban me on the credibility on individual sites with different moderation abilities. Whatever it is you think it solves, it creates huge problems on societal level.
This is actually pretty common. Won’t be long before we see another HN post about someone getting banned from a platform with no idea why and no recourse forward.
Actually similar happened to me recently. I woke up to find my PayPal account close and no one can tell me why.
It's neither a completely crazy nor completely new idea. E.g. Metafilter charges (charged? I haven't checked in a while) a $5 one-time sign-up fee that's less for revenue generation and more intended to make it sting a bit if you're banned (which Metafilter is notoriously hesitant to do).
Problem being that Metafilter is unpopular compared to many of the free alternatives. Correlation is not causation, but there is a plausible causation mechanism here.
This is a straw-man argument. Everybody knows there is an issue with a deluge of spam, obnoxious, content etc. The issue is trust/control. Whom do you trust to have control over who sees what content? A public platform should be beyond reproach in terms of its political bias, which is not the case for Twitter. A way to get closer to that would be to devolve the control over content moderation as close to each user as possible.
I believe there's a middle ground that doesn't involve the forced adoption of a single subjective viewpoint of "correctness", while preserving the benefits of moderation to whoever desires it. I refer to the following approach as "moderation lenses", although it's certainly been thought of before under some other name.
"Moderation lenses" would operate as follows:
Any person or entity can be a moderator. Moderation could be done manually or in an automated fashion, it doesn't really matter. Users would be able to opt-in (and later opt-out) of whatever set of lenses they want. The lenses they opt-in to would affect what posts they are able to see.
Online communities (for example, a subreddit), could have a default set of lenses applied to newly-joined users, but as with any other lens, they would be removable by the user. One would imagine that most users would leave a "spam" lens in place.
Governments would be able to produce their own lenses for things such as "misinformation", "hate speech", or whatever the evil-of-the-day happens to be. And if people want a government filtering what they see, they can add those lenses. And if they don't, they don't.
Human opinion is inherently subjective, and agreeing on what is "correct" or "appropriate" across large groups of people is rarely possible. Trying to impose a single viewpoint of "correct" across millions of people is laughable.
Yaknow, I've helped build sites for Nazis and furries and fisting fans and others who were not anticipating a broad welcome; and their boards were remarkably civil places. Even with the inevitable and never ending "yall are sinners and need jesus" crews such places attract.
Just maybe its the enforcement of orthodoxy that makes "content moderation" so toxic?
Those places are civil because they are homogenic. Most minor boards with a narrow subjects are civil. Mix in anti-nazis/furries/etc. and everything will quickly devolve. It requires surprisingly few hostile users to destroy the tone of a board for some time. As anyone that have taken part in raids, or seen them happen, can attest to.
Yes. The discourse was courteous and polite, usually; folks would calmly discuss the most heinous horseshit and could actually in that setting be open to education about facts where available.
Its very hard to get people to learn things by shouting at them
It's weird how "change their mind or kill them before they kill you" is now widely considered a more just and equitable solution than simply not giving such people a platform to politely discuss your murder on to begin with. Much less the biggest possible platform.
Twitter's already a huge shit hole, it's just that the fascists are going to rejoin the Marxists so everybody can excrete into the common hole together. Three is no reason to believe anything will happen other than more "journalists" getting abused, and let's face it - they deserve it. Learn to code being bannable was, for me, the last straw and I stopped using Twitter after being banned for saying it to a journalist.
The implication here is Tesla is doing something about climate change.
The irony is that Tesla worsens climate change by selling billions of dollars of carbon credits which enable polluters.
If it actually wanted to make a difference it would forgo those credits which would force polluters to actually reduce pollution. Instead it's just a wash.
What if I think Tesla is a great company, but also that its stock is overpriced (its P/E ratio is > 200, so that's not a stretch). Am I allowed to short TSLA or does that automatically mean I'm "trying to cash in on the failure of Tesla"?
This kind of thing doesn't give me confidence that Musk will respect differing opinions as the owner of Twitter.
>Am I allowed to short TSLA or does that automatically mean I'm "trying to cash in on the failure of Tesla"?
Anybody is allowed to short TSLA, and yes, if you short them you want the value of the company to drop and you're trying to cash in on that. TSLA should continue growing if it's a successful company. I'm not aware of companies that lose stock value year after year and are considered successful.
>This kind of thing doesn't give me confidence that Musk will respect differing opinions as the owner of Twitter.
I don't think Musk's goal with the Twitter buyout is to respect opinions. He's buying Twitter so disrespectful opinions won't be taken down.
Now if he starts banning things he personally disagrees with, like the famous plane account, I'll turn on him and call him a bad guy. But I'm not going to speculate nefarious reasons he's buying Twitter because I have no reason to believe he'll do anything other than what he said he'll do.
I am aware what his twitter is like - but he can just shoot those off from the hip with no oversight - would he just be able to ban some person without someone pushing back, or explaining to him why it might be a bad idea? I would imagine he would be at least a couple of steps removed from being able to do that.
Every politician recorded betraying their professed morals is supporting evidence that power, intelligence and resources don't automatically transmute into shrewdness. Also, he likely believes that sort of consideration is beneath commanding his attention.
Anyone who has ever worked in content moderation / trust & safety knows what kind of unrelenting deluge of obnoxious / disturbing / spam-filled / miserable race to the bottom of the lizard brain stuff that is constantly being pushed back on any moderately popular social media site.
It seems so easy from a distance. Just let people say what they want to say, right? Unfortunately the result of that is a place that very few want to spend time in.