fasterik

4 ms·

31.
by fasterik 1mo ago
I'm not dodging discussing working for a paycheck, that's included under what I called working to meet the need for survival. You asked me what basis I have for saying that many people enjoy working on things, and I gave you examp
32.
by fasterik 1mo ago
You're using a narrow definition of work as exchanging labor for a paycheck. I don't agree with that definition. I would define it as putting time and energy into a project to meet personal or organizational needs. One of those ne
33.
by fasterik 1mo ago
The case for the "bullshit jobs" theory seems extremely weak to me. If a job is pointless and not adding value, there is an incentive to eliminate it. Markets aren't perfectly efficient of course, but they're not so inef
34.
by fasterik 1mo ago
I'm totally in favor of the pursuit of knowledge for its own sake. But if a non-human agent had been one of those participants, does that subtract from the value of the result? I don't think so. >It will be a truly sad world i
35.
by fasterik 1mo ago
That's a strange characterization of an article that spends the bulk of the text referencing studies and talking about research practices in psychology. Apart from the No Country for Old Men reference, the images are mostly mental heal
36.
by fasterik 1mo ago
>But what's a "recommender system"? Are my friends and family who understand my tastes at a deeper level than an algorithm not a "recommender system"? Not really. "Recommender system" has a pretty nar
37.
by fasterik 1mo ago
The article is about improving recommender systems, not replacing them. I generally agree that the best way to discover things is through other information channels like friends and family, reviews, curated critics' lists, etc. Prime V
38.
by fasterik 1mo ago
The claim comes from Milch himself. I don't know how much he was exaggerating, or how much of it may be true but didn't make it into the final shooting script. E.B.'s soliloquy isn't in strict blank verse, but there is c
39.
by fasterik 1mo ago
I understand the frustration with current recommender systems, but what are the alternatives? Adding a bunch of arbitrary buttons or sliders isn't likely to actually improve the experience. "People who liked this also liked this&q
40.
by fasterik 1mo ago
I'm also reminded of how David Milch supposedly wrote a lot of Deadwood in iambic pentameter. I only learned about this after seeing the show, but I did pick up the "Shakespearean" feel of some of the dialogue. Constraints of
41.
by fasterik 1mo ago
The encyclical is not anti-AI, so there's not really any contradiction or hypocrisy if AI was used to write it. It's more of a criticism of the "technocratic paradigm" and the potential for abuse of power and lack of acc
42.
by fasterik 1mo ago
I don't understand your point. The phrase "when you got the software it got the source code" makes no sense. And arguments were just as common decades ago on forums, mailing lists, and IRC. What exactly are you trying to say?
43.
by fasterik 1mo ago
The article is talking primarily about Eastern Europe. It's also not inconsistent to say that it's raining during a drought. That's why we distinguish between weather and climate.
44.
by fasterik 1mo ago
>in that scenario you need nuclear capable of supplying ~100% of your power. Where are you getting that from? Again, nuclear doesn't need to be a full backup, and nobody is suggesting we ditch renewables for nuclear. A perfectly v
45.
by fasterik 1mo ago
Natural gas and hydro are firm power. You're correct that nuclear is not a good backup power source, but that's not the suggestion. The question is whether adding high-capacity-factor firm generation reduces total system cost. In
46.
by fasterik 1mo ago
The problem with 100% solar+batteries is that you need to massively overbuild for the system to meet demand. It's far more cost-effective to have some portion of generation come from a firm, weather-independent source. The main options
47.
by fasterik 1mo ago
That's why we use different tools when teaching/learning something vs. doing real work. Another example is computer algebra systems like Mathematica. It would be self-sabotage to use them on your calculus homework, but they'r
48.
by fasterik 1mo ago
The cogito argument is about cognitive processes (doubt, deception) which in my view don't depend on phenomenal consciousness. You may be right that the distinction isn't that important, since there's a parallel argument that
49.
by fasterik 1mo ago
Ironically, by predicting the most chaotic outcome you're falling for the narrative that AI will transform the economy overnight, which is exactly what AI companies' marketing divisions want you to believe. >More than 90 perce
50.
by fasterik 1mo ago
We don't know yet because we don't have a fundamental theory of consciousness. My money is on that it's an objective property of the structure of computation happening in the brain.
51.
by fasterik 1mo ago
Sure, but staking a claim in that debate is taking a position on what is and isn't conscious.
52.
by fasterik 1mo ago
Blaise Agüera y Arcas is a computer scientist at Google. But I agree, it's an opinion piece and it's not the Economist's job to grade it as a philosophical essay. Generally, I think the Economist is a high quality publication
53.
by fasterik 1mo ago
It claims that there is no objective fact about whether something is conscious. From the second paragraph: >Consciousness is, in other words, a model or a belief, not an inherent property. It is a belief about what kinds of entities hav
54.
by fasterik 1mo ago
It's hard to read this article and keep track of all the philosophical confusions at once. Whether or not something is conscious (has subjective experience) is independent of whether humans care about it. Whether or not something is a
55.
by fasterik 2mo ago
An auteur is just the "author", the primary creative force behind a project. Sustained success is orthogonal to that and depends on the person. Francis Ford Coppola didn't make anything of note after the 1970s, but he was sti
56.
by fasterik 2mo ago
What you're describing is the inability to maintain quality over a career, which is a separate issue from whether auteurs exist. Often it's because fame and success remove the constraints that were forcing the work to be tight and
57.
by fasterik 2mo ago
What "interest group" are you referring to, and how do you know that it's not individual users with a legitimate good faith disagreement?
58.
by fasterik 2mo ago
It's not that surprising if we remember that the model is trained to be a generically useful next-token predictor, not necessarily an agent or a chatbot. It needs to know about the environment it's embedded in and what assumptions
59.
by fasterik 2mo ago
You could be right, but you're making a lot of assumptions about how complexity, scientific understanding, and explanations scale. One of the features of a good scientific discovery is that it often simplifies and compresses things
60.
by fasterik 2mo ago
I think this is exactly what we should expect. LLMs are engineered to produce the most likely output given the training data. When you give it a generic design task, you're going to get a distillation of design cliches and tropes from

More ›