Very nice thread by the former CEO of Reddit about online behavior, moderation, etc: x.com/yishan/status/1586955288061452289 Per the thread, it’s possible this problem could’ve been solved by burying the site in affiliate links…

I think this is important enough to highlight. Tweets are very different from other forms of communication on the internet. You can see it even here on HN in the comments section. Twitter railroads the discussion into a particular type by the form of discourse. Each tweet, whether it's meant to or not, is more akin to a self contained atomic statement then a paragraph relating to a whole. This steers tweets into short statements of opinion masquerading as humble, genuine statements of fact. Often times each tweet is a simple idea that's given more weight because it's presented in tweet form. An extreme example is the joke thread of listing out each letter of the alphabet [0] [1]. When tweets are responding to another tweet, it comes off as one of the two extreme choices of being a shallow affirmation or a combative "hot take". Compare this with the comments section here. Responses are, for the most part, respectful. Comments tend to address multiple points at once, often interweaving them together. When text is quoted, it's not meant as a hot take but a refresher on the specific point that they're addressing. The HN comments section has its problems but, to me, it's night and day from Twitter. I basically completely avoid responding to most everything on Twitter for this reason. Anything other than a superficial "good job" or "wow" is taken as a challenge and usually gets a nasty response. I also have to actively ignore many tweets, even from people I like and respect, because the format over emphasizes trivial observations or opinions. [0] twitter.com/dancerghoul/status/13273612… [1] twitter.com/ChaikaGaming/status/1270330…

I think this gets almost all the way there but not quite — there is one more vital point: How we act depends on our environment and incentives. It is possible to build environments and incentives that make us better versions of ourselves. Just like GPT-3, we can all be primed (and we all are primed all the time, by every system we use). The way we got from small tribes to huge civilizations is by figuring out how to create those systems and environments. Yes, the algorithm is not the problem alone, but a good algorithm can help fix the problems — since it creates the "loss function" (the incentives) for the humans using the platform (I go into that in more detail here twitter.com/metaviv/status/152987979986… and here belfercenter.org/publication/bridging-b…... for those who are curious). So it's not about "reaching for the stars" or complaining about how humanity is too flawed. It's about carefully building the systems that take us to those stars!

If you were to say "hey, we should all fight for foobardom, a new term I just made up" but didn't give a good definition for then, yes, you'd have a problem: talking about foobardom would be pointless since it's an ill-defined term and the discussion would serve no purpose. However, this discussion about philosophy and freedom is the reverse: people are already fighting and in many cases dying for what they think is this thing called 'freedom', and yet it proves elusive to find a definition. The real-world, societal impact is already there. So it's worth studying. And if you study it and find that it's elusive and nearly impossible to pin down... well now that's interesting, and should make you think critically about anyone who uses the term as a low-nuance bludgeon in an argument. (In the other article around Twitter here today - this thread - twitter.com/yishan/status/1586955288061… , I think this point is made pretty clear early on by the spam example: perfectly legal, yet "bots" are a self-professed problem for pro-whatever-speech-is-legal,don't-get-involved-further Musk...)

Here we go, hope this helps: twitter.com/MaxKennerly/status/15883781…

> but how would you go about categorizing ideas into those that are worthy of equivalency and those that aren’t? obviously this is one of the most complicated questions that would require actual libraries full of books and experienced actual experts to properly come to any solid conclusions, but my suspicions are that some of the pieces that would be important would circle around: - we start by recognizing that most of us are dumb in most areas. if we're lucky, we have expertise in one or two areas. if we're really lucky, peers in our field will publicly recognize our expertise. outside of our areas of expertise, compared to the experts in those fields, we're dumb. and thats ok. comparatively im an idiot in fluid dynamics, soil sciences, and millions and millions of other areas. i might have some hobby level interests, but compared to recognized experts in those fields, im an idiot. collectively we seem to have forgotten our limitations. i mean, the old phrase is there for a reason: the smartest person is the one who recognizes what they don't know. - i dont know shit about anesthesiology, if an algorithm weights my comment discussing anesthesiology with the same weight as an actual widely peer-respected anesthesiologist, something is _severely_ broken. this type of thing is happening across the board. - often, more speech is not necessarily better. if the more speech is all nonsense babble, the conversation is just DDOSed and the situation is absolutely worse. as yishan said in that incredible HN post the other day [1], its about managing signal-noise. if its just simply "more" speech, we end up with noise. - if we've already collectively decided that 2+2=4, yet someone keeps screaming they're not convinced, maybe its ok to ignore them. i can't tell you how many times ive seen people's arguments get thoroughly dismantled, then we'll see them screaming the exact same arguments an hour later pretending the dismantling didn't happen. - this is gonna be a hard pill to swallow for many but, when we're building something with social as the primary, we _need_ to have _more_ people deeply involved in the building process who have expertise in humans. people who understand the human condition. some of us don't understand people very well and its absurd how many of us (myself included) turn our noses up when our projects involve human complexities. could you imagine a party planner building a bridge without heavily including engineers? this is a hard problem, but the very first place id start is, no, not all of our opinions have equal weight in all topics. i know how difficult that is for some people to take. but its just true. [1] twitter.com/yishan/status/1586955288061…

Now define "outgoing attack". You're trying to keep it defined to "ddos" only but surely you would agree that the school filtering a student actively attempting to hijack school infrastructure to send active shooter threats to the student population should also be filtered yes? Keep going a few steps and you'll realize that's exactly the problem, there's no level of specificity you can define here that works without some level of subjectivity. Try and see. Good luck. Suggested reading: twitter.com/yishan/status/1586955288061…

Yishan did this same sleight of hand trick with twitter before Elon bought it[1]. Then his point was that twitter doesn't have a bias, it's simply a normal moderation problem (noise-signal ratio). Now that twitter is owned by Elon, everyone can see it obviously moved right. What normies are fairly noticing in these corps is moderating not by the overton window of the normal consumer, but by the overton window of the managerial class who run the business. 1. twitter.com/yishan/status/1586955288061…

It's very nice. Apparantly it doesn't follow pages of an HN thread. It summarized this news.ycombinator.com/item?id=34579175 to mostly about staying warm. See labs.kagi.com/ai/sum?url=https://news.y…... This is how it summarized Yishan's thread on content moderation labs.kagi.com/ai/sum?url=https://twitte…... It obviously got confused when he side stepped into his CO2 mantra middle of the thread. From these two links, it feels like this engine mostly focuses on the beginnings of the URL, things that come later on a page are not given the same priority when summarizing.

This CEO did the same thread 6 months ago and was blasted off the internet. You can see his thread here: twitter.com/yishan/status/1514938507407… edit/ Guess it is working now? The most important post in his older thread: twitter.com/yishan/status/1514939100444… He never ever justifies this point. The world absolutely has not changed in the context of censorship. Censorship apologetics notwithstanding. The realization is the world changed is a reveal. He as CEO learnt about where the censorship is coming from.

The core is really: > twitter.com/yishan/status/1586956650455… "Hereʻs the answer everyone knows: there IS no principled reason for banning spam. We ban spam for purely outcome-based reasons: It affects the quality of experience for users we care about, and users having a good time on the platform makes it successful." And also this Chinese Room argument: "once again, I challenge you to think about it this way: could you make your content moderation decisions even if you didnʻt understand the language they were being spoken in?"" In other words, there are certain kinds of post which trigger escalating pathological behavior - more posts - which destroy the usability platform for bystanders by flooding it. He argues that it doesn't matter what these posts mean or whose responsibility is it for the escalation, just the simple physics of "if you don't remove these posts and stop more arriving, your forum will die".

This is likely because threads are a "high engagement" signal for Twitter and therefore prone to being gamed. There are courses teaching people how to game the Twitter algo. One of those took off significantly in the past 18 months. You can tell by the number of amateurs creating threads on topics far beyond their reach. The purpose of these threads is for it to show up on people's feeds under the "Topic" section. For example, I often see see random posts from "topics" Twitter thinks I like (webdev, UI/UX, cats, old newspaper headlines). I had to unsubscribe from 'webdev' and "UI/UX" because the recommended posts were all growth hackers. It wasn't always that way. I'm not the only one, others have commented on it as well, including a well known JS developer: twitter.com/wesbos/status/1587071684539…

In this thread, Yishan promotes an app called Block Party that lets you block anyone, or even all accounts that liked or retweeted something. twitter.com/blockpartyapp_

I mean that's the claim, the counter-claim would require a social network banning topics and not behavior. Note: As a user you can see topics, you can't see behavior. The fact that some users flood other users' DM's is not visible to all users. So how do you know? "I don't trust left-y CEO's", is a fair enough answer, but really that's where the counter-claim seems to end. Now that we have a right-wing CEO, looks like the shoe is on the other foot[1] [1] twitter.com/AOC/status/1588174959817658…...

> This is a very good way to pitch your afforestation startup accelerator in the guise of a talk on platform moderation. /s Also for anyone have Deja-Vu, he had almost the same thread a few months ago (and probably a few more times before/after. twitter.com/yishan/status/1514938507407…

Polarization predates social media. Polarization, partisanship, and radicalization receive a great deal of academic attention. (I can't speak to the public mind, or political hobbyists.) The book Why We Are Polarized [2020] by Ezra Klein surveys what was known at the time. Nationalization of politics, consolidation of media, death of journalism, election reform, campaign spending. TLDR: Politics used to have weird coalitions, where party was not strongly associated with positions. Then people sorted. Then people's identities (positions) fused into a kind of super identity. Now party is a strong predictor of position. Rachel Bitecofer twitter.com/RachelBitecofer researched (negative) partisanship and used to talk about it. I don't know of an otherwise good intro. People have always self-radicalized. Pamphlets and zines. Social media made it a self renewing pandemic. Root cause is probably a combo of algorithmic hate machines and social isolation. (People are increasingly anxious and isolated. These para-social relationships fill the void.)

Lets be clear - twitter favors right-wingers - says their own engineers: twitter.com/MaxKennerly/status/15883781…

FWIW, you can use an extension like Redirector (einaregilsson.com/redirector/) to automatically redirect any Twitter URL to Nitter with the following: Redirect: twitter.com/\* to: nitter.net/$1 Hint: Twitter to Nitter Example: twitter.com/yishan/status/1586955288061… Applies to: Main window (address bar) I'm also using this for 'fixing' referral URLs but that's another story.