StableAlkyne
In Samuel 1, King Saul visits a medium in Endor in order to speak to the prophet Samuel. Because human dead go to Shoel (in familiar terms, it is similar to the Greek underworld) in the Old Testament. Later on in the New Testament, the idea of super-hell is introduced in the form of Tartarus, which is specifically where the rebelling angels were sent. But that is not the same place human dead go.
The idea that everything "not Heaven" must be Hell is an invention, caused mostly by dogma the rose from bad translations of the Greek and Hebrew in the middle ages (for example, King James refers to all of it as Hell even though the original text most inarguably refers to different places different beings go).
Taking a response at face value is just asking for trouble; they're useful primarily for finding references. The summaries are a good way of prioritizing what to read, and if nothing else the documents it cites tend to be relevant.
The people who get burned by these things are the ones who just blindly trust the generated text.
Also some of the skill is learning to feel out when you can trust it and when you can't, but if you're just starting out, always assume the bullshit machine is generating bullshit.
Though, most of the time I use it to find a reference and ignore the AI summary. Or to put a label on a topic I'm pretty sure exists but don't have a name for (e.g., "<insert problem statement here>. I'm pretty sure this is a known problem in <insert domain here>. Is there a name for it and what is the most recent literature?"
Cigars are still cool though, but those have always been the higher status tobacco due to the cost per stick. Also most people don't go through public spaces smoking a cigar, so it doesn't have the annoyance factor of vapes. Hookah still has a fun vibe too, probably because of the social factor.
If someone wants to vibe code an alternative to a SaaS tool, my gut feeling is "no, this is bad and it will blow up" as vibe-coded projects tend to do.
But then the alternative is often paying $X million per year for licenses to a half-assed program that hasn't had a UI update since 2005.
I just don't have a good argument on why they shouldn't when the outcome will be just as dogshit, only cheaper and flashier.
But what prevents a malicious actor from iterating with their LLM to minimize the score on their text?
I don't think Google is doing this out of some egotistical obsession over "control," it's more that developers and FOSS enthusiasts are a vanishingly small fraction of their customer base.
There are plenty of problems with walled gardens, but IMO this isn't one of them.
The interests of the shareholders diverge quite strongly from the interests of electrical utility customers. It is generally a parasitic relationship where they consume resources that could have been spent improving utility service for the customers, either directly via dividends or indirectly by avaricious behavior on the part of the company in pursuit of their stock price.
Compare that to some of the older stuff like Take 5, Giant Steps, or any of the big band stuff from the 50s. Sure, there was a lot of incredible skill on display, but the music itself didn't seem to take a back seat.
One of my favorite albums of the last couple of years was Against the Fall of Night by Sungazer. Whalefall, the first or second track is probably my favorite song on it - there's still a lot of technical skill, but the composition is also on display. So you get this conversation between the song and soloists, instead of a backing track with a few skilled players. Moon Hooch and Fox Capture Plan are also in this same vein IMO.
So... Any recommendations of newer groups like that? Would love to find more like it.
Not really my point. Malthus was wrong about a future food crisis, but that doesn't invalidate the tools he used to make that prediction. Re-read the first line of my original post, it's pretty clearly about the theory and not his predictions.
The underlying idea is that population growth outpaced food production, and we know it is bad for food consumption exceed production.
> And you missed one point with your discussion of linear and exponential
You're also missing the forest for the trees: agriculture and population growth can be any two functions for the idea to work, so long as the agriculture function starts higher than the population function (technically there is a third function implicitly here connecting population growth to food consumption) and grows slower. Population growth was approximately exponential, and food production was approximately linear. This is a sufficient condition to forecast future changes. Pick any overfit function you'd like for food or population as it was known to Europe in the 1790s, and you will arrive to the same conclusion if you do not "cheat" and allow for magical technology advances in 50 years.
> How much more wrong does he have to be before you are willing to call him "wrong"?
Math would have to change to the point that subtracting consumption from production no longer leads to a negative number for me to consider him "wrong."
Now, I wouldn't say his model was relevant to famine after technology changed, because it no longer was useful for that particular thing. But the core notion of "if demand grows faster than supply, you will eventually hit a crisis when they intercept" is a pretty clear-cut conclusion to draw.
It is a bit odd for a thread to get this heated over a pretty simple economic theory that stopped being relevant a century ago. Is there some political side to Malthus I am missing?
Additionally, it is reasonable in the 1790s to assume exponential population growth. It would not be until the 1830s that Verhulst would even model it as a logistic curve. Moreover, human population growth really was on a roughly exponential trajectory until very recently.
Obviously we did not experience such catastrophe. But it's also a bit wrong to call it "invalid" when it was a completely logical conclusion to make. If we're going to allow him to invent a magical technology that addresses it in 1798, then all reason goes out the window.
The world he was modeling in the 18th century is a very different world than we ended up getting. IMO it's neither right nor wrong, it's just a reasonable observation that made sense for the time.
It's somewhat analogous to Y2K, in that there were obvious problems that were going to happen. We didn't experience those negative outcomes because immense engineering effort solved the underlying problem. That doesn't mean Y2K was "invalid" or "wrong," just that the problem itself got addressed before it could blow up. Same with Malthus, chemistry and mechanization allowed us to solve the underlying issue of food production scaling slower than population growth.
That's a bit unfair to Malthus though; sure it was the enlightenment, but all he could rationally go on was to assume that the last 2000 or so years of civilization would be similar to the next, in terms of food and population growth.
What would be abstract would be to make the assumption that unprecedented technological advancement would be made in the next hundred years, bordering on ( to what Malthus would be) magic. You can assume any problem will be solved if you also assume a technology will be invented to fix it. But that's not a reasonable approach to make if you have no reason to believe it is possible.
It was never proven wrong, as much as the world changed in a way that invalidated it.
The entire idea was that population growth was an exponential function, whereas agriculture output was a linear function. There would have been famine if nothing had changed, because there is only a certain amount of physical labor a person is capable of doing, which has to be divided between the farming, logistics, and cooking of the food.
What happened was that a century later we got mechanized farming and synthetic fertilizer, decoupling agricultural output from the population. Now a single person can do the work of thousands, and every year that number grows due to technology advances. It is too expensive to set up fully automated farms on the industrial scale, but it is technologically possible.
But none of that invalidates his idea. If you believe that an exponential function and linear function sharing a point at (0,0) must cross each other (subject to other reasonable conditions like not having crossed yet, the linear having a positive finite slope, etc), then it is not logical to reject Malthus. It just no longer applies because we decoupled our food production capacity from the population size.
Granted, I'm also someone who heavily encourages the use of Black as the Python formatter in projects I'm on, specifically because it has no configurable options - can't have a holy war about formats if nothing can be changed.
Then you ignore everything the robot says, and just click the Wiki link (or journal article or whatever). Skimming the AI text and just finding the citations is the most important thing to do to save time with this method. You're only using it because it is good at ranking search results, not because you want a bad AI summary.
Beyond some corporate manager seeing a dashboard that will cause them to say "oh neat" in a meeting once, there is zero reason to put a refrigerator's cooling system online. Or frankly any network for that matter. Is there really such a massive productivity increase from being able to remotely change the temperature when a human has to physically visit the machine to restock every day anyway?
I don't think anyone™ goes out of their way to explicitly break things for old Python versions.
It's usually more like "oh cool I can make this code faster/more readable if I use this feature. I don't have to care about the old version anymore so I will do that."
If you really to super duper need a backport, you're in luck: python is an interpreted language whose source is in plaintext, on your local machine.
Why would I care about vibe coded a demo when I could just ask Claude to make the same thing for me?
If you have two demos side by side, one AI and the other human, the human version is immediately more interesting because it showcases a high level of skill that you may aspire to achieve one day. The AI version is as forgettable as the person who prompted it, because it does not demonstrate that person's skill.
Same reason why we generally find weightlifters to be impressive even though machines have been able to out-lift anyone for thousands of years.
That said, I don't think most people dislike using it for tedious work or information searching. Similar to how most are still impressed by a lifter using a Smith machine to stabilize an absurd weight.
This is a bit of an overcorrection to the post-conquest myth about Quetzcoatl (which doesn't show up in pre-contact sources).
This stems from the word "Teotl" which subjugated tribes started calling the Spaniards when they began to openly defy the Aztecs. The translation is complicated, as the word means both "the magical force underlying the world" (our closest word is mana) and "magical creatures" (which in Western mythological terms includes deities, demons, devils, saints, angels, kobolds, elves, etc.) with the implication that they have some power over the world.
This word was translated to Deos by the Spaniards, as their mythology was very strongly influenced by Christianity (modern mythology is similar this but with the extensive expansion that Tolkien popularized), and this was the natural word they would use. Plus it makes them look good at home to be able to say "these people were so primitive they thought us gods" - so mistranslation serves them well.
But there certainly was a larger-than-life view of them due to the immense technology difference and sheer audacity of the expedition. That's why they got the nickname teotl: similar to how an extremely effective military unit nowadays might earn the nickname "demons" or "devils" even if it doesn't literally mean that.
What is a usanian?
* The clunky turns of phrase and jarring introduction of analogies reads strongly of Claude to me, at least
For example, if your weights were trained as 32-bit floats and you need 1TB of RAM, you could reduce that to around 256GB by quantizing to 8-bit floats. You also make the model faster in the process because there is less data to process to calculate the next token.
The game is to balance between the savings of quantization and making the model dumb enough the people notice
I think this post is missing a lot of context and background. I haven't the foggiest what an "established line theory" is
At most universities as well, if you decide the PhD isn't for you, there is always the option of leaving with a Master's. Because you get free tuition and a stipend, you also come out ahead of if you had gone straight for the MS (which you generally would pay for).
lite.cnn.com is their text only subdomain. Probably meant for accessibility, but since it's only text it loads quickly and is very pleasant to use.
One example strongly tied to culture is cliched writing.
Things become cliche as they are get overused; no amount of cleverness in the cliche itself can overcome this. Claude's writing has rapidly become cliche because it uses the same pool of stock phrases and words. People pick up on this and have the same reaction as they do to any other cliche piece.
Let's take two works generally seen as good: if Lord of the Rings or Foundation released today, they would probably be panned as cliche because they created those tropes in the first place, which future authors copied or built upon. So yes, most would not think this is poorly written 20 years ago because this style didn't exist yet and it wasn't cliche.
Claude vibe-writing isn't "bad" because it is grammatically or structurally bad. It's bad because it's overused and signals to the reader - whether intentional or not - certain things about how much the prompter values the written piece.
...as opposed to a dishonest score?