@ThenElection's banner p

ThenElection


				

				

				
0 followers   follows 3 users  
joined 2022 September 05 16:19:15 UTC
Verified Email

				

User ID: 622

ThenElection


				
				
				

				
0 followers   follows 3 users   joined 2022 September 05 16:19:15 UTC

					

No bio...


					

User ID: 622

Verified Email

(I'm ever curious as to when an AI can write fiction in my style as well as I can, because then I'm going to fucking retire, but that day is not today. Best guess is 1-2 years.)

Shorter timeline than that. Playing with Gemini 1.5 a few weeks ago, I could upload an entire book (substantial ones, e.g. Perdido Street Station, Gravity's Rainbow), give it a basic plot skeleton of a new book, and prompt it to write a couple paragraphs in the style of the author, and it succeeds. There are still some prose issues, but you'd absolutely be able to tell exactly which author it's simulating (sometimes to the point of parody).

Overarching plot structure it's weaker at, though.

Last year, over 800 people died in San Francisco to overdose. Compare that to 56 homicides and 27 traffic deaths. Or, heck, the ~700 COVID deaths from 2020 to the end of 2021.

Addicts have a shockingly low lifespan. And fentanyl is the key component of their mortality: approximately nobody dies from crack or meth, the usual drugs of choice. Which isn't to say they're not damaging or that I don't want to see them off the streets, but fentanyl stands out as particularly evil.

As an alternative theory, fentanyl may be both an extremely pleasurable and extremely addictive substance that (desperate and not especially well-informed or conscientious) people try without grasping the full consequences of what they're doing.

That said, the people who are most hands-off on fentanyl proliferation do not appear to give one crap about the people suffering from addiction to it. It's decentralized MAID. Naively a misanthrope might consider it an effective way to get rid of undesirables, but even that makes no sense: its an addiction that reproduces itself for each new doomed-to-die cohort.

A mechanism isn't really necessary, though: I'm happy to accept mysterious forces at work. But I do wonder how you would differentiate it from a "materialist" world.

I think you might be able to if you managed to successfully model a particular individual human being on some substrate where quantum effects played an even smaller role than they do in the human brain. If you could, that seems like quantum-mediated decision-making couldn't be playing a role, unless you somehow accidentally incorporated the causal influence of the soul of the person you're emulating into the model itself.

ETA: thinking about it a bit more, it would be very easy to accidentally incorporate the soul's causal influence, so you'd need to be very careful not to, even if you did a non-quantum-by-construction neuron-by-neuron emulation.

Materialism and determinism are orthogonal concepts, although popular proponents of one very often hold the other. See: quantum mechanics. Physicists are probably the single most materialist profession there is, and modern physics decidedly rejects strict determinism. (That said, there's a lingering dislike of that rejection, leading to all kinds of abstruse "interpretations" of scientific theories that try to bring something that looks like determinism, if you squint just right, back to science).

It's not clear how nondeterminism helps accounts of free will, though. Suppose you have two universes, one where quantum effects don't play a significant role in cognition and decision making (the one I would argue we live in) and another where random quantum fluctuations make me decide which coffee shop I'm going to this morning. I don't see myself as having any more free will in the latter world, and probably less.

Even unknown or unknowable things don't undermine materialism. Imagine someone designed an experiment that provided strong evidence that ESP existed, and we had no idea how to explain it. The thing is, there are countless things in the world that are mysterious, and ESP would rapidly get a ton of attention. Experimentalists would test it under different contexts (it only appears under heightened emotional states? Then figure out exactly which states. Does distance play a role in the strength of the ESP-effect? Etc), and theorists would come up with testable explanations (maybe physical models that are near isomorphic to each other somehow share causes and exchange psy-particles). Gradually we'd build a model that approximates what's really happening better than random guessing, under the constraints of economic cost and value of building that model. This is nothing new and is constantly happening.

Doesn't this indicate that materialism itself is vacuous, because it can explain everything? Yes, maybe.

It depends on if it's the presence of women that's civilizing, or competition for women. If it's the latter, less intense competition would lead to more brutish behavior.

An internal locus of control gives you better outcomes, regardless of how valid a particular complaint is. Even if it is insanity, it's a useful insanity.

I have no idea if the particular woman in the example above actually faced unfairness or not (she probably has; at some point we all have). But I do know she'd be in a better position, financially and psychologically, if she spent less time introspecting about how mean and terrible and unjust the world is to her and more time embracing her agency.

I agree. But by the same token, too many men are falling into the same trap: "I'm mediocre because the world is biased against me, giving unfair preferences to everyone else."

Sociologically, one or neither (or both!) may be true. But if you embrace victimhood as part of your identity, you're dooming yourself.

Academia is many things, but I don't see people going for (and getting) professorships as slackers. They almost invariably are smart, hard working people who could be making well into the six figures or more in industry. (Note: this is for what I'll just call real fields.)

The big issue is that it's so astoundingly competitive to get any kind of professorship, let alone a desirable one, that intellectual conservatism reigns supreme. Going off on some tangent that has high potential but is unlikely to bear any fruit is just too risky.

I agree with everything you said, and no one should have to worry about making sure there isn't a homeless encampment a block away from a Four Seasons.

That said... 2cim isn't some yokel from Kansas City visiting San Francisco for the first time with her corn-fed husband and kids unexpectedly finding herself surrounded by syringes and shit. She is absolutely aware of the issues with San Francisco, and she's quite capable of finding and staying in parts of the city that are liveable.

SF was (laughably) considered a "Tier 1" city (and had a weirdly cheap Four Seasons), so I stayed in the FS by Union Square, famous for shithole status and close proximity to the Tenderloin.

Surely "wow, this Four Seasons is priced like a motel, what a steal!" should have raised some red flags? Silly games, silly prizes.

Next time do the 1 Hotel, which is better situated and where my company suggests visitors stay. (Which isn't to excuse our municipal decay.)

In the case of San Francisco, hotels were originally set up in what were then nice areas, which then collapsed. Unfortunately, most SF hotels are now in the most unpleasant part of the city (or, more accurately, two or three blocks from the most unpleasant part of the city). You could throw a dart at a map of San Francisco and hit a better area than the Four Seasons.

In San Francisco, we got a homeless pedophile advertising free fentanyl to kids from pre-k to 8th grade outside their school:

https://abc7news.com/san-francisco-free-fentanyl-sign-child-molester-adam-moore-found-guilty-stella-maris-academy-sf/14215982/

(In fairness, he did end up arrested and convicted, but only after public outcry.)

The issue with full time pundits is that they become more beholden to their audience, both in terms of having to constantly produce and in terms of playing to their audience's biases. The best things about Tracing are that he only writes about things he finds interesting and that he isn't motivated to suck up to any particular viewpoint.

I do like the independent investigative journalist for hire angle. Give him $10k for 80 hours of work (does that sound right for pricing?), let him nerd out on some obscure corner of the world for a few weeks, and at the end you get something interesting.

It's absolutely fair to say that, if you're doing something the government places a high priority on detecting and punishing, Google is not the place to put digital evidence of that something. And that's a certainty.

The issue comes in when someone believes that there exist digital safes that no one but they can open. You're not going to build one in your spare time, and you're certainly not going to find one in other well-known third-party services (which are equally compromised by the government and less secure than Google) or in unknown fly-by-night services (half of which are government honeypots, and the other half are people waiting to do a rug pull to steal all your bitcoin and which are probably breached by the government anyway).

Gin, mdb, rpcsp... Security there is taken very seriously. There're always potential holes in the system, but I trust Google much more to keep my data safe against realistic adversaries than anyone's homelab duct taped together with VLANs and reverse proxies. (And at least 90% of alternative non-Google third party hosts are honeypots, either out of incompetence or malice.)

From the original Vice article:

The document says that Google terminated 36 employees in 2020 for security-related issues. Eighty-six percent of all security-related allegations against employees included mishandling of confidential information, such as the transfer of internal-only information to outside parties. Ten percent of all allegations in 2020 concerned misuse of systems, which can include accessing user or employee data in violation of Google's own policies, helping others to access that data, or modifying or deleting user or employee data, according to the document.

So, it's a bit hard to parse without the actual numbers, but it appears that of 36 security incidents, 31 (86%) were Google employees leaking confidential corporate information (ironically, including the document leaked to the Vice reporter). 4 of them were misuse of systems (which includes but is not limited to accessing user or employee data). This is actually pretty amazing, considering how many Google employees there are and the scale of data that Google collects. You might say "well, that's how many were caught," but it's very likely the majority of cases are caught (all major systems at Google have every user data access logged and audited, though I suppose some minor systems that no one uses might not have that set up).

I consider myself a terrible slob, and I'm probably a bit below a 2. Even the worst of my friends don't get close to 3.

My suspicion is that within a couple weeks it will be clear who has the technological edge, and whoever does will win the war. And no one in the world has concrete knowledge of even how one side's weapons will fare, let alone both's.

It will certainly be interesting to see what technologies both countries have up their sleeves when pushed to the limit.

China has a material advantage in the local theater, but the best it can hope for is getting its neighbors to commit to neutrality (and at least Japan will not, and it still has meaningful shipyards). The US also can shut down Malacca.

Everyone's economy will be f'ed, but if China can't win in the span of ~9 months, it has lost. That said, I don't reject the possibility of it winning in that duration: there are just too many uncertainties to call an outcome.

For that matter, did the US genocide Japan?

Something like a million Japanese civilians died in the latter years of WW2 in retaliation for Pearl Harbor. And this was long after Midway: the US was clearly the dominant power by that point and not under existential threat, if it ever was.

since the level of pre-meditation and personal desire makes it qualitatively different to say, treating someone who tried to commit suicide because of mental illness.

Isn't something like nullification a pretty solid indicator of mental illness? FWIW I agree that we should give more sympathy/pity to people who attempt suicide, but I have a hard time identifying the difference.

Maybe the usually higher level of pre-meditation and planning plays a role, but I'd still sympathize more with someone who planned a suicide attempt over months than someone who planned and received a nullification over the course of a day or two.

Trump is an unprincipled egotist who is unable to work with the Establishment: he'll do whatever he wants, and he has no incentive to work with the Powers That Be because they despise him and would never cooperate with him (and the feeling is absolutely mutual). No other candidate comes close to offering that.

It's not particularly likely to lead to anything good, I think, but if you're broadly anti-establishment, he's the closest thing to a sure bet to do things differently than how the Establishment wants things to be done.

I don't see it as providing any new information: this was almost so certain as to be predetermined. The MIC is a lucrative and stable system to be integrated into, and the dynamics of capitalism were going to inevitably drive OAI into its arms. Little different from Google pulling out of China for being a totalitarian regime to, five years later, begging Daddy Xi to please let them make money in China.

Safetyists can draw some minimal level of comfort from the fact that OAI priorities will marginally shift from improving capabilities towards AGI to building tools that the MIC desires. More profits, less fundamental/deep research. (And I think that's genuinely good for safetyists: a smarter drone swarm is not going to destroy humanity.)

It makes me wonder: if every Cabinet official disappeared overnight, how long would it take for it to become common knowledge among top level officials in the departments they manage? How long until it hit the news media?