brhsagain

Liberals reading into this and seeing racism is itself an example of this phenomenon. They hear something like "diversity of ideas" and instantly a big red moral stop sign shows up in their head. It shows how when everyone has fundamentally different values, and everyone's different sets of moral landmines are overlaid on the same map, it becomes basically impossible to walk through the space of ideas without sooner or later stepping on one.

What's stopping someone from starting a distributed effort to train an open source ChatGPT model, like Folding@home but for AI? Is it a technical limitation with how models are trained, difficulty of coordination, something else?

I mean, I agree with this in the abstract, but I’m very skeptical of the claim that this process is aided by meetings.

Honestly, my experience after nearly a decade in the industry has led me to believe that the “democratic” approach to management that underpins meeting culture… basically doesn’t work. The idea that the whole team should convene and everyone should give their input and that this process converges on the truth, or the optimal path forward, or whatever… is just not rooted in reality. What we actually need are individual smart leaders who take responsibility for the vision and just tell everyone else what to do. The actual thing that makes a group of people productive is for everyone to know their place/role, and the truth is that the majority of people do not need to be at or contribute to “knowledge-making” meetings.

I also don’t know if I buy that programming is knowledge work in the usual sense. It requires knowledge but the output is not just a report or something, it is the actual product. The only reason it seems more like knowledge work than it is, is that we’ve set up big tech companies in a way where your work is like 40% being an information router, 40% being a politician, and maybe 20% programming. In this context I’m sure it would seem like “meetings are the work.” But maybe we should change that, and not just shove people into more meetings.

The dean who wrote this message doesn’t care about the students. She doesn’t know them. She’s fulfilling a role where she acts as a figurehead that students can project feelings of compassion onto. Like any other job, this one also involves some degree of drudgery, such as writing emails like this. It’s good that ChatGPT can make her life easier by automating that task. Her actual fault was accidentally admitting to it and ruining her image as a figurehead, which is her actual job. But if you’re looking for a referendum on the dangers of ChatGPT on society or something, this story isn’t it.

Is that not the case?

What I am in favor of is recognizing, as a society, the value of free speech, not just as a constitutional technicality but as a principle. I believe we ought to value it as a good. And I think that as a society, we should stop and be like, hey, wait a minute, online communication is now dominated by a handful of sites, doesn't it violate the spirit of free speech if, in practice, all those sites enforce roughly the same overton window, and your options are to (a) get in line or (b) not use the 5-7 sites that everyone else uses?

In fact, I think the conversation about the constitutional limits of 1A, or governmental rights, or whatever, is a distraction. I'm not suggesting the government step in. I'm just saying we clearly have a situation where the principle of free speech is being violated — where people are not able to share certain ideas on the platforms that everyone else uses. I'm aware we don't have a law protecting that. I'm making a prescriptive argument that society ought to view this as a bad, dangerous thing.

So in terms of a practical agenda, I dunno, I would like to see a stronger shared appreciation of free speech, and bottom-up pressure for the big communication platforms to stop enforcing their own capricious filters. From this perspective — not the 1A perspective — I think the argument that "private enterprises can do whatever they want" is pretty weak and irrelevant. The actual question is what externalities it has on society when the platforms used by everyone decide what you can say on them. I think the answer to that is not obviously "nothing," and probably "something."

No, it's just gloating, right? We get it, he hates Elon, and when you hate someone you'll throw whatever dart you've got on hand at them.

One of the ways this feeling comes up for me is when I realize that almost no one is actually engaging in actual thought in order to solve specific problems. First you have maybe 90% of people who are just collecting a paycheck, or trying to maintain social harmony, or playing empire-building games, or in general playing a game that is a different game from building good things and figuring out the truth. Of the remaining, most people think using metaphors, heuristics, social consensus and other forms of indirect thought. Very few people have the ability to squint and stare directly at a problem and engage with the subject itself in a high-decoupling "autistic" way.

If you have this ability to take a problem, break it down into first principles and just think hard about it until you find a solution, most everyone will probably seem stupid to you, because they are not doing that. Sometimes you watch two people argue about something and understand within ten seconds what the central point is and you want to shake them both and make them just address that directly, but instead you listen to them ramble on for another hour. Sometimes you watch people stand around being puzzled by some bug, and then you go in, actually investigate the thing itself, build an understanding of what is happening and what exactly is going wrong and push a fix in half an hour. Sometimes the truth about a topic is just obvious to you if you stare at it with laser vision and ignore all irrelevant considerations and then you wonder why it isn't obvious to everyone else.

It may not even necessarily be an intelligence thing. Raw brainpower may be a prerequisite but it is more of, I guess, an actually-thinking-about-the-thing-itself thing. My raw IQ is probably not that high but I make up for it by spending a lot of time thinking and caring a lot about the truth to the exclusion of everything else. Then I wonder why other people don't do this.

The feeling of frequently being the only person in the room doing this is very isolating, and people in this position seem to end up with either super high or super low self esteem. I was the latter. I felt like I would never be able to communicate with or relate to anyone and that society would always see me as a loser, and it took a lot of time for me to overcome this. Nowadays I have a personal project that I believe in, that I can apply all of my problem-solving skills to, and I have a strong sense of the things I value and have specific tracks/plans to further each of them. I’m at peace with my psychological profile and have accepted my path in life. (If you’re one of the ones with high self esteem, you can fortunately skip this whole section.)

On to the practical side of things… society heavily pushes the message that no matter what, you will inevitably have to deal with people on some level. I think this is true and that there's no use fighting it. Everyone has people they can’t get along with; for you that list may simply be bigger. But also, I think that social interaction can be a tricky thing and there are many mental traps to fall into here. For example, you may believe that you just want to be left alone or not have to deal with anyone, but then you get what you asked for, and your all-too-human brain rebels and you realize social interaction is healthy after all. Or you may adopt a cynical view where you view human interaction as “just saying the right things” or “playing the game” to get along with others, and it works for a while because you are smart, but then you realize how ultimately unsatisfying it is. The point I'm making is that humans have a complicated relationship with other humans (or at least I do), and it can take lots of trial and error to figure out what kind of interface with others will make you happiest.

What worked for me was to figure out and be explicit about the type of relationship I’m able/willing to have with each person. I can usually tell based on what they say, what beliefs/values they express, how they behave, etc how well I can get along with them, and then I keep them at the right distance accordingly. I have acquaintances, activity friends, shared-ideology friends, and close/"real" friends. The “stupid” people in your question usually get pushed to acquaintance or activity friend.

Work is trickier. I am a terrible employee so I don’t really have good advice here. I kind of cheated and just went the self-employment route. I have a part-time contract gig that pays the bills and then I spend the rest of my time working on my own project. My coworkers for the most part are also just trying to pay the bills so I have a pretty good shared understanding and working relationship with them. I carefully structured my life so that I don’t depend on what you call “stupid people” for anything that I care about, and that has worked out pretty well. But I am also in a very fortunate position where I don't have kids and I work in an industry that allows for relative autonomy.

This has been pretty stream-of-consciousness, and maybe 80% of this doesn’t even apply to your situation, so I’ll stop it here. Hope everything works out for you.

The whole argument seems like a discussion-ending cover for people's emotional attachment to alcohol. It completely doesn't engage with the cost-benefit analysis of how alcohol affects the functioning of society at large or navigate any of those tradeoffs.

This really seems like a case where the existing categories in the law created hundreds of years ago are no longer adequate to capture present realities, similar to how they didn't have AR-15s in mind when they wrote 2A. If you ask whether Reddit is a public or private company, sure, it's private. But if you look at what intuitions back then were around "public" and "private", and for that matter around what "speech" was, it seems like the real answer is that (in regards to this debate) Reddit is a third type of thing that doesn't have a name yet.

> Get a blog! The very worst people on Earth have managed to keep blogs running.

This might illustrate my point better. I assume you don't support people getting banned straight off the internet. Is it simply because no one owns the internet, but Reddit is owned by Reddit (the company)? Maybe it's time for that boundary to be reexamined.

To me the big glaring central point here is that everyone uses big tech platforms. It’s just a fact of life. If you tell people, sorry, I don’t use Facebook, I use some wonky alternative instead, can you add me there? you’ll get weird looks. I think it’s insane to just ignore that elephant and continue studiously down the path of, well, they’re a private company and therefore are entitled to publish whatever they want.

When we established the precedent that private entities can say what they want, we clearly didn’t have Facebook in mind. Like there is just intuitively an obvious difference between a private journal deciding what to write about and a huge platform everyone uses deciding what to pass through or not. The state of the law doesn’t currently reflect this but that doesn’t mean the difference doesn’t exist.

I’m arguing the “ought” rather than the “is” here. I disagree with the philosophical stance that simply because Reddit is a private website, they are entitled to control the content as they want and that no level of qualitative difference between Reddit and the average website is enough to change this.

In general, we regulate private enterprise when it has negative externalities on society, and I think there is a discussion to be had about what those externalities are here. I don’t think “private companies can do whatever they want” is a sharp enough tool to engage this issue with.

> The view of the law that you're espousing is one where, simply by dint of letting people sign up to a service you run, you take on an obligation to carry their messages. That's an intense amount of authority to give the government over private enterprise, and a stultifying precedent for anyone who wants to stand up a site of their own.

Couldn't you use this formulation to say that anti-discrimination laws make it so that, simply by dint of providing a service, you take on an obligation to provide that service to everyone? That's an intense amount of authority to give the government over private enterprise, and a... you get it. We accept government intervention when we believe it benefits society. The argument being made here is that it would benefit society if these massive pipes of information that in practice everyone uses were similarly regulated in the types of discrimination they can engage in.

There's a calculus to be made about what level of control platforms having over their content would best serve the interests of society. It might be that despite all the things I listed above, the result of that calculus remains the same, but it is not immediately obvious that that is the case, as your "private sites can do what they want" formulation would have one believe.

It would falsify this if you found examples of our culture supporting guys just going through their daily lives and crying as an accessible reaction to random things that make them feel bad, almost as a sort of banality. (I think this is the bar, since culturally we're okay with women doing this.)

Even just the category itself kind of makes the point. The original claim way upthread was that women find guys who cry repulsive. I would assume this is roughly referring to guys crying when they run into a problem in life instead of dealing with it, that sort of thing. The examples in your link are, like, guys who gave their whole lives to a sport, everything's on the line, maximum skin in the game, and their crying is a show of strength, the raw emotion of someone at a peak/climactic moment.

> Boys don't cry. It doesn't matter the circumstances; shedding any tears is the ultimate no-no in terms of what you can and can't do as a man. But sometimes... sometimes... sometimes there comes a time when a man's emotions deny any other form of expression, and they pour forth—prerequisite impassioned speech may or may not be present—by cascading down his cheeks. And crying does NOT make him any less badass in the eyes of the audience, but in fact may even reinforce this status. In other words, Manly Man-Tears Of Manliness.

...supports rayiner's point. Men aren't generally allowed to cry; when they are it's for exceptional circumstances, when the beauty of life so overflows one's cup that it becomes an expression of peak experience, or something like that. But the subtext is that everyday crying is taboo.

Also, I think this is a bit overstated. Programmers (and smart people in general) like to think that their real job is high level system design or something, and that "mere regurgitation" is somehow the work of lesser craftsmen. When in reality what GPT shows is that high-dimensional regurgitation actually gets you a good fraction of the way down the road of understanding (or at least prediction). If there is a "buried lede" here it's that human intelligence is less impressive than we think.

Aren't these kind of mutually exclusive, at least directionally? If the interview is meaningful you'd expect it to predict job performance. If it can't predict job performance then it is kind of useless.

I guess you could play some word games here to occupy a middle ground ("the coding interview is kind of useful, it measures something, just not job performance exactly") but I can't think of a formulation where this doesn't sound pretty silly.