They're not objectively amazing. Friction is not inherently a bad thing when we have models telling humans that their ideas are flawless (unless asked to point out flaws). Great that it made you smile, but there's quite a few arguments that paint your optimism as dangerously naive.
- A queryable semantic network of all human thought, navigable in pure language, capable of inhabiting any persona constructible from in-distribution concepts, generating high quality output across a breadth of domains.
- An ability to curve back into the past and analyze historical events from any perspective, and summon the sources that would be used to back that point of view up.
- A simulator for others, providing a rubber duck inhabit another person's point of view, allowing one to patiently poke at where you might be in the wrong.
- Deep research to aggregate thousands of websites into a highly structured output, with runtime filtering, providing a personalized search engine for any topic, at any time, with 30 seconds of speech.
- Amplification of intent, making it possible to send your thoughts and goals "forward" along many different vectors, seeing which bear fruit.
- Exploration of 4-5 variant designs for any concept, allowing rapid exploration of any design space, with style transfer for high-trust examples.
- Enablement of product craft in design, animation, and micro-interactions that were eliminated as tech boomed in the 2010's as "unprofitable".
It's a possibility space of pure potential, the scale of which is limited only by one's own wonder, industriousness, and curiosity.
People can use it badly - and engagement-aligned models like 4o are cognitive heroin - but the invention of LLMs is an absolute wonder.
Sure - and the people responsible for a new freaking era of computing are the ones who asked "given how incredible it is that this works at all at 0.5b params, let's scale it up*.
It's not hyperbole - that it's an accurate description at a small scale was the core insight that enabled the large scale.
Depends on worldview. If you believe in God, amazing has many dimensions for evaluations. What teaches us more about the the world He created, things that create beauty by expressing righteous thoughts for others to experience, or that which strengthens family.
LLMs certainly teach us far more about the nature of thought and language. Like all tools, it can also be used for evil or good, and serves as an amplification for human intent. Greater good, greater evil. The righteousness of each society will determine which prevails in their communities and polities.
If you're a secular materialist, agreed, nothing is objectively amazing.
You can cherry-pick any two data points like that to make any point you want, all equally unfounded, so the only thing a comment like this actually does is restate your existing preference.
Edit: or perhaps one could say restate-and-reinforce. I think that's (edit: well, might be) largely the function of adding snark and indignation to internet comments.
I think you need to admit HN is failing due to how easy it is to manipulate what content and comments are shown. Allows a single determined group to literally control the narrative. Eliminate upvotes/downvotes. Keep flagging but make it public record. And revoke privileges for those who abuse it.
That's not an accurate description of what's happening, and we're not going to redesign the site in response to political winds, gale-force though they are right now. Actually that would be the worst moment to do that.
It 100% absolutely is, you're just too deep to recognize it.
It's what's happened across Reddit, Twitter, and Facebook over the past decade. You've all built platforms intended for social discourse but instead built extremely brittle systems subject to gamified manipulation by showing people which opinions reward and punish them. And those are then manipulated at-large by outside groups responsible for elevating a particular way of thinking about a given topic.
Conversations do not need a algorithmic popularity mechanism.
That's just not an accurate description of HN based on what I know, but of course it's indeed possible that I'm too embedded to see clearly.
Edit: some of these points can be refuted by data though (albeit not public data, which means people have to take our word for it, which many are understandably reluctant to do). For example, flaggers of these stories are by and large longstanding community members who have participated for years on HN in lots of threads about lots of topics, etc. That doesn't prove they're not flagging at the behest of outside groups, but it does make it unlikely.
That would be a disaster of epic proportions. Every thread, no matter how innocuous the topic, would devolve into perpetual feuds about which kinds of people were upvoting and downvoting which other kinds. These sorts of shitstorms were pretty common on HN, in the long-long-ago, and even then all we were revealing were comment scores, not voting attribution.
Then we remove upvotes and downvote entirely. Web2.0 online discourse (visibility is controlled by recommendation algos keyed on upvotes/downvotes) is fundamentally broken.
That would create a different forum. HN is an experiment in how long this kind of forum can ward off gravitational collapse. There are things fundamentally wrong with these kinds of forums, but that's true of every kind of forum.
Not to pile on what tptacek has already nicely said, but I agree and still think there is plenty of room for new forums. The vast majority of possible permutations have not been tried yet, and I wish more people would.
I think many users would be willing to hear a long-form version of your take on the subject (maybe in a fresh meta thread for visibility?), but your comment as-is feels like a shallow dismissal of legitimate concerns.
You’re not really offering any insight into why or how you reached the conclusion that you posted. As a user on the outside, it sure looks like the site structure has begun to buckle in the wind.
Sorry - I spent hours yesterday writing about this at length, so in my mind I've already done that, but you're right, of course. Let me dig up the links for you...
That’s entirely understandable - I’ll dive into your profile to read those thoughts.
I second the suggestion to have some sort of metathread, stickied comment on hot political threads, or other consolidated way to present your collected thoughts so that a larger number of users see it and can benefit. Like you said, the winds are blowing hard, and you might go a long way to quelling the moderatorial waters by addressing the whole site as a collective.
Past efforts to do that kind of thing haven't worked, so I'm a bit down on the idea. What ends up happening is (1) the meta communication stirs up a flurry of objections, some relevant and many not; as well as people passionately restating what they feel about current affairs and/or how much they dislike what some other comment said; and at the same time (2) most people still don't see it, so it ends up not having the desired effect and just taking a ton of time.
Somehow, no matter how often we repeat something, the set of users who hear it always has measure zero (https://news.ycombinator.com/item?id=25787443). That's why I mostly stick to answering specific comments with detailed explanations, and hope that at least some other users will see it.
There's something a little interesting in the measure zero observation; we've run into it in a couple places too. For instance: we now do voice calls ahead of our last hiring exercise (it's a little bit counterintuitive and needs some prep) because no matter how carefully we wrote about it, even with highly motivated candidates (many of whom we statistically end up hiring!) nothing we write seems to sink in. We have an API quirk that works the same way, too.
These aren't "political winds". This is all a bit too much "we're all looking for the guy who did this", when it comes to Y Combinator and "tech bros" in general.
> Balaji then revealed his shocking ideas for a tech-governed city where citizens loyal to tech companies would form a new political tribe clad in gray t-shirts. “And if you see another Gray on the street … you do the nod,” he said, during a four-hour talk on the Moment of Zen podcast. “You’re a fellow Gray.”
> The Grays’ shirts would feature “Bitcoin or Elon or other kinds of logos … Y Combinator is a good one for the city of San Francisco in particular.” Grays would also receive special ID cards providing access to exclusive, Gray-controlled sectors of the city. In addition, the Grays would make an alliance with the police department, funding weekly “policeman’s banquets” to win them over.
That HN is for stuff not discussed in the news is basically disproven daily. The other thing I keep hearing that "if we allow this, the front page would be only about politics", I don't believe it anymore. It's like the stone that keeps tigers away. People may genuinely believe in that stone and that they're guarding the village, but I think it's bull and having to entertain this superstition takes up more resources than the occasional tiger attack would.
I know disabling flags is probably not feasible, since real spam does get posted. but we have "showdead" for comments, why can't we simply have the same option for threads? Then those who want to discuss these things can do it, at the price of also having to wade through actual spam. Anybody else would be unaffected. If the goal is not to suppress awareness and discussion, that can be very easily proven with such a feature, and the best time to implement it would have been a long time ago IMO.
To be honest, it's such a long article I couldn't find a good passage that summarizes it, so I went for the most stark bit. Because it's really quite mad.
But it wasn't my intent to claim you run HN like this either way; but I'm making the educated guess that people who think the above isn't mad, but quite exciting, would be both likely to use this site, and prone to abuse flagging to suppress discussion, and/or awareness of the article that would be discussed.
Ah I see, and yes that's a mistake I make rather often (assuming that a comment is about moderation when it's actually about the community). But tbh I don't have this impression of the HN community. I don't think we have that many users who feel that way, and certainly we have many more users who would strongly disagree.
I don't know what "many" would be here, but it's obviously too many, since the flagging of these topics is really constant. And that flags are used by people because they are "annoyed by" or "don't care" about these things is stated by people all the time, who never followed up when I challenge them. So that doesn't give me the impression of good faith. If the reasons are benign and not to suppress awareness and discussion -- in short, not people thinking themselves the lords of others -- then nobody would have a problem with people opting out with something like "ignore thread flags" option.
Votes are a far bigger problem than flagging because it takes a pavlovian approach to what ideas are allowed. Flagging is important to remove spam and wholly inappropriate content. And if you revoke the ability of users who flag content based on a personal disagreement, that user simply loses that ability.
I suggested to add the ability of users to have their view of HN unaffected by these flags. Without that, HN simply has no ability to defend itself against abuse of users who take it upon themselves to play censor.
Some of us are sick of politics all over HN all the time.
To answer the three replies that are already here:
No, it's not because we're Elon fanboys. "Aiding Musk's government takeover" means it isn't going to be an article about Musk per se, but about US politics.
No, it's not because we're closet fascists that want to silence all dissent. We don't care what you think of Musk, we don't care what you think about US politics, we just don't want it to be the dominant topic on HN. Yeah, we know, tech is bleeding over into politics, but... how many Musk stories do we need in one day? However many the number is, we're past it.
"Elon" brings out the worst in us? Yeah, that may be true... but it's true of the people who post the articles as well as of the people who comment.
> Some of us are sick of politics all over HN all the time.
That was never the situation.
> how many Musk stories do we need in one day? However many the number is, we're past it.
So the number is zero. Having one story for each major separate event is one story too many. This is still what it is. The longer the rationalizations for it get, the more sad it gets.
> Yeah, we know, tech is bleeding over into politics, but... how many Musk stories do we need in one day?
You should Ask Musk to cool down then. We didn't vote him in. We didn't ask him to break the law and compromise american security. We didn't grant him access to the US treasury. "We" voted this in. Those who didn't want this are 3 months too late.
And I see this excuse on every platform. I see a story I don't want to engage with... I just move on. Maybe you browse new, but I've never seen politics be "the dominant topic on HN".
Since there's a question mark about brutalism, here's a link or two to read up on. tl;dr Brutalism in graphic design today has two distinct in popular usage meanings:
1. Bare bones. Pure utility, any styling is simply for readability. Craigslist style.
2. Big, garish, bold; sometimes called Nu-brutalism or Sportsbrut.
I should note that each of these has very little, if anything, to do with the 20th century architectural trend, which focused on applying usage of basic shapes and raw materials at a large scale. On the web, the "raw materials" part is the only real connection to the origins of the term, with the first one focused on lofi design with basic tooling, and the second highlighting garish things you can do with basic forms.
Neo-brutalism is another trend to note; think of it as a focus on raw shapes with some niceties added on top to make things more interesting.
Seems to me that a truly minimalist website would have to be set in Times Roman (is that not still the default typeface). And that a truly brutalist website should have a gray background.
Both utterly pedantic quibbles that are completely intended to be ignored.
Truly minimalist: Accept the defaults. Whatever they are. Count your bytes.
Aesthetic minimalism: A non-serif font. We may have strong opinions here on which. I shall name none to avoid firey flames ;-)
By pointing out Times New Roman by name you do not imply minimalism. Rather a passive aggressive swipe at good old Times New Roman which I find too practical and beautiful to be lumped into nu-brutalism nor aesthetical minimalism. To show good faith I will let the missing "New" slide. Gentlemen shall be given room to misspeak. Happens to the best of us.
This model was pioneered in Oakland by Steve Aoki's "Pizzaoki" location, also known as Lorenzo's, Thick & Tasty, Happy Slice Pizza, Gabriella’s New York City Pizza, Chubby Pie and each of other pizza joints listed at 536 Lake Park Ave. They used DBAs to surface under different keyword searches on various delivery apps and cycle out poorly rated brand names. Pizza was pretty mid too.
Thanks for the info I always wondered what was going on there.
I've gone to Gabriella's in the mission a number of times. It's one of the only authentic New York slices you can get in the city for people who crave the taste of home
Okay, why be down-voting a perfectly valid question? Just because I (or you) knew the answer don't mean everyone does, and down-voting the question "ghosts" the answer as well, so other folks who might have the same question will end up havin' to ask it again instead of benefiting from the (correct) answer already here.
reply