guajalote
2yr ago
Do you believe that your smartphone could become conscious and experience qualia, with no hardware modifications whatsoever, if you could just find the right software to run on it?
I don't see how I could rule out this possibility. If you believe you can, why?
If you had instructions for a Turing machine that perfectly simulated the behavior of a human, and you instantiated that Turing machine by moving around untold trillions of rocks in an infinite desert - would the resulting system of rocks be conscious?
I don't see how I could rule out this possibility. If you believe you can, why?
guajalote
2yr ago
Yeah I feel like in Houston money is a means to the end of doing whatever the fuck you want, whereas in Dallas money is an end unto itself.
guajalote
2yr ago
Houston says "do whatever the fuck you want." No zoning, no gun control, no real enforcement of drug or traffic laws. Sort of an anarcho-libertarian paradise and hellscape rolled into the same package.
guajalote
2yr ago
Yes, from what I understand private schools and "better" public schools (i.e. the ones in more affluent areas) tend to use phonics more.
guajalote
2yr ago
I think this may be true if we start out with all children learning via Method A, then switch to all children learning via Method B. But due to sorting effects, in the real world the kids near the top are already learning via Method B and the kids near the bottom are learning via Method A. So if we switch everyone to Method B the gap may well close to some degree.
I would also argue that switching to better methods that make learning easier may tend to close gaps simply because smart kids are more able to learn via any method, whereas less intelligent kids will struggle more with suboptimal methods. If you take a cohort of kids with different intrinsic skiing abilities and have them start on a black diamond (difficult slope) then you will see a big delta in performance between kids. The weakest skiers will fail and give up quickly, the strongest skiers will figure it out and get better. If you start them all on the bunny slope, you'll see less of a delta between the best and worst skiers since the worst skiers are at least able to make progress.
guajalote
2yr ago
The idea that different teaching methods have zero impact strikes me as just as implausible as the blank slate position. Yes, a lot of it is selection effects, but not 100%.
guajalote
2yr ago
Years ago, I made a bunch of money in the stock market, based on recommendation from a friend. That friend deserves my thanks, and perhaps even recompense (reparations, if you will), even if some other person might have given me even better advice.
Let's say your friend tells you to buy Apple stock and you make a 3% return. But the market as a whole went up 5% in the same period. If you had just given no thought to the matter and bought a total stock market index fund like VTSAX, you would have performed better. In that case, I don't think it's correct to say your friend deserves any thanks or credit for his recommendation. He didn't really help you in any meaningful way, since your default option was better than his suggestion.
guajalote
2yr ago
On the other hand, this essay notes that US cotton provided something like 75% of British textiles. That’s potentially a lot of money flowing into the US.
No one denies that slavery brought in money, but the claim is that far more money would have been brought in if there had been a market based labor system. As compared with the alternative, slavery was a net loss.
guajalote
2yr ago
I think it's clear that slavery was economically bad for both the northern and southern states. If slavery had never existed, both the north and the south would have been economically better off in the long run (even if we ignore the economic losses caused by the civil war).
An enslaved person has no incentive to invest in the future; their incentive is to have as low a time preference as possible. There is no point accumulating assets or wealth, since you cannot legally own them. There is little point in accumulating skills, education, or other forms of human capital because you do not own your own labor. This is a system that massively disincentivizes investment and long-term growth. The system may have economically benefitted slave owners, but it was a loss for the US economy as a whole.
guajalote
2yr ago
I think it's cool and impressive when someone can play the violin well. In general, I think being able to play the violin is "better" than not being able to play the violin. That said, if there was a pill that made people want to practice violin 8hrs a day like a professional violinist, I wouldn't support giving that pill to everyone (or even most people). The fact that something is "good" doesn't necessarily mean everyone should do it.
I think being gay is morally neutral. I think most liberals feel the same way. A small subset of activists probably think being gay is better than being straight. But even this latter view doesn't necessarily imply a belief that everyone should be gay.
guajalote
2yr ago
Physically, it works the same way a transformer works -- induction. When you plug your phone (or computer, etc) into the wall, there is a transformer in the circuit stepping down the voltage. There is no basis to be worried about risks from inductive chargers any more than you worry about risks from transformers. Also this isn't some "new untested science" this is Maxwell's Equations, stuff we've understood since the 1860s.
guajalote
2yr ago
I don't have a constantly running internal monologue. I "turn it on" when I'm thinking about highly verbal concepts, like the particular way I want to phrase something, but in general my thinking is more conceptual and less verbal.
This got me thinking about what my default internal monologue "sounds" like, and I guess it sounds like myself when I was a little boy, like 10 year old me. I feel like the sound of my internal monologue hasn't changed since I was a kid. I can deliberately make it sound different, like I'm currently deliberately narrating this sentence in the voice of Jimmy Stewart as I type it. But by default it's little kid me.
I think the fact that I don't always internally verbalize allows me to think better/faster, but who knows. If I'm working on math then I think in terms of a visual or conceptual image of the math problem. If I'm thinking about music then I just hear the music in my head. Music is probably the form of thinking that's easiest for me, I can internally play back a song with multiple instruments and vocal parts. I have written original songs entirely in my head that I'm not good enough to actually play on an instrument but I can easily play them back internally as though I was hitting "play" on a recording. I have some musical training but am by no means a professional.
guajalote
2yr ago
I've posted here before about how my wife is considered a "person of color" by employers and universities because she's from Mexico, even though her skin is pale white. Outside of the really clear-cut cases, American notions of race are pretty incoherent. It's not like sex where 99%+ of people can be reliably classified as male or female.
guajalote
2yr ago
I suspect its going to be hard to find this stuff for products that are still on the market since these processes are closely guarded trade secrets. But if you find any good sources let me know, it sounds interesting.
guajalote
2yr ago
I shower every day but only use soap/shampoo every 2-3 days since (in my anecdotal experience) excess soap use causes excess oil production and damages skin and hair. So I'm sympathetic to what you're saying. But if someone isn't at least rinsing themself off with water every day or two, they are going to stink.
guajalote
2yr ago
It is possible that millennials are becoming conservative at exactly the same rate as previous generations even though they are not voting for conservative parties at the same rate as previous generations. Parties are not static ideologies or sets of policy goals. Parties compete for voters, modifying and triangulating their platforms and rhetoric over time.
In the US, Republicans and Democrats have fought each other to a pretty consistent 50/50 stalemate in nationwide votes for several decades, modifying their views on all sorts of issues along the way to maintain that equilibrium. As the overall population has gotten older over time, the equilibrium "stalemate age" (the age when a voter is equally likely to vote for either party) also gets older over time. To put it another way, as the electorate ages it becomes increasingly "cost effective" for the Republican party to cater to older voters, which makes it easier for the Democratic party to attract and retain people in their 30s who might otherwise be potential Republican voters.
guajalote
2yr ago
If you literally have no clue how to do it, then I can see how more time would not be helpful. But I wouldn't consider that being a "fast test taker" I would consider that simply being unprepared for the test.
In my experience as someone who majored in physics and minored in math, there is almost always some way to use additional time productively on math tests, even if you're stumped by a problem. Re-write the problem in a different form and see if it looks more familiar. Change the coordinate system and see if it makes things easier. Try out various mathematical tools/techniques and see if they work.
guajalote
2yr ago
Can you be more specific about the type of test you're talking about? I initially envisioned a g-loaded multiple-choice test like the SAT, but what you're describing sounds more like a long-form written answer test based on subject matter knowledge. As another commenter pointed out, maybe you're just a fast writer/typist?
As someone who has always been an extremely fast test taker (and a consistently high-scoring test taker), I still struggle to think of situations where more time would not be beneficial for me. Unless the test is pure regurgitation of memorized lists, or unless the test is so easy that you're scoring close to 100% on the initial pass, having more time to review and refine your answers should tend to result in a higher score.
guajalote
2yr ago
There's no good reason you should do worse after going back and reviewing answers. More time to think should always be better, ceteris paribus. The only explanation I can think of would be if you're biased toward assuming your initial answer was wrong (e.g. because you doubt your own intelligence or doubt your test taking abilities) and so you consistently talk yourself into changing answers you shouldn't change.
guajalote
2yr ago
In some contexts, yes, it's used as a sort of a formal or impersonal version of the 2nd-person singular. Especially in customer service contexts you'll sometimes hear a singular "y'all" as a sign of respectful distance between the speaker and the person being addressed.
(Note: my comment is referring to standard usage in Texas and isn't directly addressing the point above about black American vernacular).
guajalote
2yr ago
I think I'm speaking pretty plainly. I'm asking OP to consider how ChatGPT performs in relation to an average human. This is a pretty common question people consider when talking about AI performance. After all, the Turing Test is one of the oldest and best-known tests of computer intelligence.
I am asking OP to consider these questions as a way of pushing back against statements like the following:
Seeing it underperform so much in my field is giving me a sort of Gellmann Amnesia effect for people touting how it can write code on its own.
"Underperform" is an interesting choice of words here, because it seems that the bar for performance is being set at "subject matter expert." Obviously ChatGPT is not at that level. To paraphrase Arnold Kling on the most recent EconTalk episode, "it's about at the level of an undergrad BS artist who didn't study for the test." But consider how much training and skill it takes a human to reach the level of "undergrad BS artist" and how few humans are able to attain even that level of performance. I think OP should be more impressed with how far we've come. We don't need to go a whole lot further to close the gap between "undergrad BS artist" and "skilled electrical engineer." The former often becomes the latter with just a few years of additional education.
guajalote
2yr ago
What fraction of human beings alive today do you think could generate something that plausibly looks like SPICE code? What fraction of those could "with some handholding (or, rather, explicit statements of how to fix the circuits) ... get closer to something functional?" What fraction could give "some justification for how it connected the nodes of each individual device?"
guajalote
2yr ago
my biology teacher was a nun and we certainly never skipped the chapter on evolution.
The Catholic church officially endorses evolution. It's a subset of evangelical protestants who don't believe in evolution, because they believe in biblical literalism and sola scriptura (the Bible alone is the highest authority). So to these kinds of evangelical protestants, "seven days" literally means seven days, end of story.
guajalote
2yr ago
Because race is an extremely noisy and inconsistent proxy for other things we might care about like culture and genetics. As I asked above, I'm genuinely unclear what OP means by "race" and what qualifies as interracial. This is because the concept is so un-rigorous.
It's like if I decided to separate all dogs into four "races" of dogs as follows:
Race 1: yellow labs
Race 2: black labs
Race 3: dogs with short tails
Race 4: all other dogs
Technically this is a valid way to classify dogs into 4 categories. And these categories are undeniably correlated with things like genetics. But the correlation is tenuous and arbitrary to the point that this classification schema has minimal utility.
I think it's very unlikely that individual rocks are conscious, just as I think it's very unlikely that individual neurons are conscious, but a collection of neurons, rocks, or transistors arranged in a particular way and executing a particular algorithm may well be conscious.
More options
Context Copy link