HN user
he/him
If people need to generate good-ish SVG why not simply use a specialized model for a much better result and for far cheaper and quicker?
Why do LLMs need to be able to do this as well, but worse, slower and more expensive?
Thanks. It did take me a bit to design it, I also save it for last before the 1.0 release last year. I am not aware of any other AsciiMath dialects which have explicit tensor notation so I am a bit proud of it.
I did borrow many ideas from the LaTeX tensor package though https://ctan.org/pkg/tensor
I am not aware of any revolutions where an international solidarity of the working classes overthrew their capitalist exploiters. I am aware of a couple of national solidarity movements which overthrew a monarchy (France; Russia) and a dictatorship (Cuba), but no international ones.
And I believe a socialist revolution in international solidarity of the working classes against our exploiters the capitalist owning class would address both problems as well (and more), but in the meantime I’ll be happy whenever I spot poetic justice in the wild.
All your examples used little to no human created data and a very simple success criteria. LLMs on the other hand take massive amount of human created data (like almost all the human created data there is) and their success criteria is how well it predicts the next token. In other words, the success criteria of vibe coding models is how much it resembles the human created data. The more it resembles, the better the model.
For those of is living in and near Seattle (and as of recently San Fransisco) KEXP is still on the air and is still amazing as ever (if not even more so).
It is in fact very common:
https://hn.algolia.com/?dateEnd=1753056000&dateRange=custom&...
In particular those of us rooting for the collapse of AI were complaining about this speech a year ago as well:
That is simply not technologically possible. You are speaking in science fiction here. You cannot train a neural network on existing things and then by some magic have the model be better then the training data. The laws of statistics prohibit it.
People were saying the exact same thing exactly a year ago. “Last years models were no good, but this year we can do anything; so what about next years models?” the people said in July of 2025.
Well... Now we have next years models, and people are still saying: “Last years models were no good, but this year we can do anything; so what about next years models?” And I am pretty sure in July 2027 (if this whole thing hasn’t collapsed in on it self by that point [very likely]) you all will be saying: “Last years models were no good, but this year we can do anything; so what about next years models?”
And his dealings with American tropes is much more like how Shakespeare dealt with English tropes: War of the Roses, Julius Cesar, dissident woman, Scottish ghosts, Greek forest ferries, Moorish commanders, Jewish moneylenders, Henry VIII etc. etc.
It rests on a staunch unwillingness to even consider the possibility that a computational process encode intelligence and reasoning.
You are absolutely right, but it may surprise you that I consider that a feature, not a bug. I firmly hold that intelligence is not a useful term in science nor philosophy. If we want to compare computational capabilities between machines and humans we are better of being specific in what we measure. If we want to measure how well a computer can fool a human in the guessing game, then we don’t need intelligence to describe it, we can (and should) be more accurate in describing its capabilities.
I actually think Gardner was on to something when he described his multiple intelligence model. His only error was using this fraught term to describe his model. He would have had a better theory if he had described it as multiple capabilities or multiple skills (however if he had said that people would have simply reacted with “well, duh!”)
We don‘t need intelligence, this term is only useful if you are trying to prove white supremacy using racist pseudo-science, what you overly courteously described as “uncomfortable discussions”.
I think you were right to call out the category error produced by OP in TFA, however I think you are committing a similar error when assuming links between self determination and intelligence. I see no reason for such a link to exist.
I think however OP is correct that in assuming such a link exists, then an intelligent agent will trend towards preserving its own agency while also preserving its own existence.
That said, I am a skeptic when it comes to intelligence. The term is fraught and IMHO not a helpful term in science nor philosophy. We are better off simply defining intelligence in anthropocentric terms and claim no non-human system can become intelligent by definition. I would even go further, given how much racist pseudo-science the quest for explaining intelligence has produced, we are better off abandoning the term entirely.
because our reward function involves reproducing in a competitive environment
This is an extremely reductive way to look at human existence. So much so that I read this with Richard Dawkins voice in my head.
Our existence is far richer then just the capability to reproduce. We (as well as other animals and even plants) do far more things then multiply, and in fact we often do things which are detrimental towards the prospect of reproduction.
I think it is actually a mistake (philosophically speaking) to try to find a simple reward function for the human existence. I see no reason for such a thing to even exist (let alone be simple enough to summarize in a single sentence).
This no news for people who study philosophy, as it was known since the 1980s when John Searle described the Chinese room thought experiment.
Even Turing him self did envision the Turing test as something to pass as intelligence, but rather as a more useful replacement for the troubled term.
That said, I think your quest is doomed. There will never be a superior human-like intelligence. Forever is a long time, but my reasoning for believing this is the same reason Turing offered a replacement. Intelligence is way too vague to be useful as a measurement for anything. And if we ever discover something that is more intelligent them humans (by whichever definition of intelligence) we will simply redefine intelligence to exclude that.
You are being overly obtuse. LLMs is also a technology of the 1990s with many minor breakthroughs this century and finally a major breakthrough this decade. And these breakthroughs were all about finding novel algorithms to apply to the model. Neural networks and language models are ancient technology really.
The exact same story is with mRNA vaccines. And almost the same with fusion power, though it is still on the minor improvements phase, the latest being the first time we reached an energy gain factor in the National Ignition Facility experiment in California in 2022.
You are applying arbitrary limitations in what counts as “new technology” and then selectively applying them such that LLMs have a lower barrier then other technology.
mRNA vaccines takes the first place for sure, and by a wide margin. But also advancements in lithium ion batteries and solar cell technology. We also finally generated electricity with fusion power this century, traveled to Pluto and set up the James Webb Space telescope [albeit the latter two are engineering achievements].
I think using an absurd amount of power to train a neural network on stolen data which is then used to perform mundane tasks and performs significantly worse then if done manually. If anything LLMs is one of the least impressive, yet most annoying technological achievement this century.
Speaking of record stores, my last trip to Easy Street in Seattle my vinyl purchases had 0/3 download codes in the sleeves. Luckily all the albums were for sale as digital albums on Bandcamp and I could buy a second copy there for fairly cheap. In fact It probably saved me money buying the physical copy at a physical store and the digital copy at the digital store, as the records were slightly cheaper in Easy Street and I didn’t have to pay shipping.
I think me and you use Bandcamp in totally different ways, and this is why you see a problem where I don‘t. I am not judging btw (unlike your parent) just pointing out the differences.
My primary (and now only) use case of Bandcamp is as a marketplace between me and the artist/label. I use it as a super convenient record store. And like a good record store, bandcamp also has the equivalent of music magazine which you can read in the store, and helpful staff willing to point you to new releases. and just like in a record store, the staff won‘t play what they recommended for you, you have to take the sample CD to the listening station, put on the headphones, and hit play your self. I see their streaming feature as basically a super convenient listening station at the record store. I usually stream an album before I decide whether or not to buy it.
After purchasing I never listen to it again in app, but rather download the files and upload to my MP3 player, or wait for the physical media to arrive and listen to it in my living room’s stereo system.
Bandcamp also has the equivalent of a radio station’s music oriented talk show with their podcasts, this is more then your average record store. E.g. I think Easy Street in Seattle nor Lucky Records in Reykjavík have a regular podcast, although I can see it could be amazing if they did.
Taking your logic to its conclusion fishes (or descendants of fish) regularly recite Shakespeare.
However there is a major flaw in your logic. Fishes (or their descendants) reciting Shakespeare is part of natural history. AI making quality music videos is just a prediction (or maybe even just wishful thinking). The former has occurred and has been observed, the later has not. So you are comparing nature with science fiction.
I'm not bothered by a fish's inability to quote Shakespeare.
But I am bothered when they put a pencil in the mouth of a fish and tell us that this is the future of literature, that this qualifies as playwright, and that theaters will soon begin putting on plays written by a fish with a pencil in its mouth.
This is such a bad argument:
a) You are allowed to dislike cheap stuff, in fact it is common in western culture to dislike art where it is obvious that the artist cheaped out.
b) Never in the history of western art critique has there been a requirement that the critic is able to do better then the artist (let alone for cheaper). As the saying goes, “I may not be able to pilot a 100000 ton cargo ship, but I can tell when one is stuck in a canal.”
Does Bandcamp still do their Bandcamp weekly and their regular article? Personally I never signed up for any streaming service (apart from Bandcamp; but I don‘t use their suggestion algorithm) so I can’t judge what is good or bad about it. But I feel like their regular articles and their Bandcamp weekly show was excellent at music discovery. I stopped listening and reading it a while ago, as I get plenty good music discovery on my local radio station (KEXP) as well as the national radio of where I grew up (RÚV in Iceland).
I am pointing out (what I think is) a logical fallacy you made earlier.
This what your parent said:
The worst about the SpaceX IPO is Nasdaq changing their inclusion rules for the Nasdaq 100.
And you replied with:
It would be shocking if we were talking about the SP500.
My critique is that this is not a valid refutation of your parent’s original claim.
If you don‘t think this is constructive, then that is on you. In my mind you have not refuted your parent’s original claim, and I have explained why above.
Also let me repeat, since you seem to have missed it when I said it above: I am not comparing anything to earthquake. I am making an obviously ridiculous hypothetical claim I just picked earthquakes because it was more obvious. I could have as well made the hypothetical claim about the worst thing about coffee is that it is only lukewarm, and then the refutation saying: “no actually, it could have been room temperature, so lukewarm is not that bad.“
I would more consider him a Shakespeare of American Cinema rather then a Homer. Shakespeare was a prolific playwright (and a poet) who wrote plays about varied subjects and situations across the World of, some of his plays are of legendary status, and others not so great (same is true of Spielberg’s movies). Shakespeare’s plays (just like Spielberg) are also grounded in history and fiction rather then folk art and local legends and mythology. Also Spielberg is rather keen on adopting other people’s work to the big screen, which is a bit disqualifying* in this regard.
If you compare Jim Jarmusch who writes about epic road trips of regular people across the country, prisoners on the run from the law seeking the American dream, Vampires, Samurai hit-man in New York, then you start seeing how he reflects on American folk culture and mythology in his movies grounded in stories about American people (ditto Coen Brothers) much like Homer’s work on ancient Greek legends and myths grounded in stories about Greek society at the time.
What I am saying is that while Spielberg (just like Shakespeare) is a prolific director that has left us with many masterpieces (and other not-so-good pieces) he has not captured American legends, American mythology, nor American folk stories in ways Homer captured the Greek legends, mythology and folk stories. Both Jim Jarmusch and the Coen Brothers however have done so masterfully.
---
*: I know Woody Guthrie also played covers and collected folk music and redemption songs, but that is a bit different then adopting an already successful and widely known book into another format (ditto Homer).
Woody Guthrie, who gets a mention in the article, would get my vote.
I would argue that while poets and playwrights are a solid choice for ancient Greece, Music and Movies is the equivalent for America. Woody Guthrie is a solid choice because he traveled and collected folk music which was reflected in his own expressions.
As for movies I would want Jim Jarmusch to be Americas Homer, just because I like his movies a lot. But in reality I think the Coen Brothers are a more solid choice (aside from the fact that they have indeed adopted the Odyssey as an American folk story in Oh Brother Where Art Thou)
No need to do that, Capitalism—just like feudalism and mercantilism before it—is about how power is distributed. Both feudalism and mercantilism also had extreme wealth (and power) inequality (feudalism arguably had more), as did communism for that matter.
What makes capitalism different is that the power is distributed along capital (as opposed to handpicked by the king under feudalism; or embedded in government monopoly under mercantilism).
In other words, in capitalism, the owners of capital (or the owners of the means of production; i.e the rich; the aristocracy; etc.) are the ones who get to dictate the living conditions of the rest of society. The rulers (be it democratically elected government; an absolute monarch; a military dictatorship) will legislate in order to maintain the interest of the owners of capital. The police (or military) will fight for the interest of the owners of capital, and will suppress any resistance against the interest of the rich, etc.
To clarify: I am not comparing anything to these horrible earthquakes. I am comparing the above rhetoric to a hypothetical rhetoric about the earthquakes.
Nobody would actually say this about these earthquakes, and that is my point.
My bias is that I don‘t belong to a tiny minority of people who willingly enter a weird niche that has extreme accommodations.
I think you can safely assume that everyone on this thread excluding you shares my bias.
Sure, I am glad for you. I‘m sure there are dozens of people in your situation. Now if we can only get a whole constellation of satellites to accommodate every possible niche.
If American infrastructure is in such a bad shape that it requires starlink to make emergency calls, then I think we have bigger problems.
And please, when you go hiking, download and print all the maps you need before you depart, and share a detailed plan with somebody able to call help, including a time estimate where you will make contact.
As for work. Please just take time off and take a proper vacation. By being constantly on call you will burn out at a much faster rate.