I'm totally fine with that, but I think there are still quality standards that if a bit higher, would let the skill at least be much smaller and thus less token-consuming. One place where I draw the line is when somebody, with minimal effort using an LLM, could have improved the quality of the LLM generated output. Also, I treat skills like programs and skim all external skills to make sure that they're not doing things they shouldn't be doing.
Unfortunately, gamified gambling on your phone is so bad for society that I think it should be banned or heavily limited even at the expense of the freedom for consenting adults making transactions with each other.
Same argument around banning hard drugs. They just make society worse. Obviously hard drugs are more clear cut and definitely worse for you, just saying, similar situation in principle.
But US has to pass https://en.wikipedia.org/wiki/Unlawful_Internet_Gambling_Enf... in 2006 to outlaw PokerStars and all sort of online casinos, so clearly a state gambling regulation is not sufficient. Now tell me why does that not cover Kalshi / Polymarket?
I have given out a certain of points and bubbles in my life under harm reduction programs. These efforts (which I've worked at but not been central to) have had a measurable and positive impact on syphilis rates in rural Colorado among a number of other metrics.
While I think it's not great folks use, the worst parts of many of these things are created by states sanctioning things. People mess their lives up all kinds of ways and creating a black market pretty much just changes the incentive structures. It just makes it more likely folks will do fent because that's what they can find, but if they can't find that they'll drink hand sanitizer.
Unless you think that these folks should kill themselves off using tainted drugs (which was a policy pursued by the US government during alcohol prohibition), then legalization program makes some sense.
Introducing a government monopoly on things like heroin would have a lot of up-sides with the downsides largely being slightly skewed versions what we have right now.
I don't think that marketing heroin in social media apps the way gambling does is a sane idea, but I also don't think gambling apps should be allowed to market themselves.
Like a lot of things, I believe having a government undercut the market for things would have a positive impact, but depending on your view of the relationship between the DEA and the CIA you might think that is already happening but in the worst of all possible ways,
Harm reduction is a double edged sword and should be assessed on the actual merits of the evidence - I am skeptical that legal heroin would be net mortality reducing.
There's already causal inference evidence out there indicating that needle exchange programs reduce rates of HIV and other bloodborne infections, but increase the risk of fatal overdose. If simply making fresh needles more accessible increases overdose rates, I suspect legalizing heroin would have a much larger effect.
Harm reduction could be mortality neutral and still be an improvement on other metrics.
But I personally am not sure what the causal relationship between access "clean needles" and "fatally overdosing" is, though I would be interesting in hearing that account.
You used to - like a century ago - be able to buy things like cocaine and methamphetamine OTC at pharmacies and they were not as big of a problem as they are today. A lot of today's problems come from things outside the drug itself. Most overdose deaths, for example, come from the drug user being wrong about the amount they're injecting because their dealer lied to them, or didn't know themselves because their dealer's dealer's dealer changed supplier, and didn't notify those downstream because reliable communication is difficult because of the police.
jfc this is just terminal contrarianism. do we really need to debate whether it's harmful for extremely addictive drugs to be widely available? the Chinese literally fought wars over this
Extremely addictive drugs are already widely available. People at the lowest rungs of our society have very little difficulty obtaining them.
If we can create a situation where those people aren't breaking into cars, stripping the wire out of buildings, or dying in the street - that would be a net positive.
Prohibition has five decades of abject failure under its belt. Try something else.
The War on Drugs people aren't actually serious. They think too small or are kids wanting to play with toys. Fighting the supply side gets you narco subs and secret passageways and cartels. Fighting the demand side with housing and jobs and hope for the future and families and therapy and happiness just doesn't look as cool on tv as rolling up with a SWAT team with GUNS mowing down bad guys.
Straw man. Nobody is calling for banning gambling nationally.
For hard drugs, the expert consensus around harm reduction is supervised injection. Casinos are, in one sense, supervised injection for gambling. Kalshi, on the other hand, is the local dealer selling fentanyl to teens [1].
The Swedish national anthem is entirely defined by common consensus, there was a proposal to establish it by law in the 2000's but that was rejected.
Sweden doesn't have a state created e-id system, instead a consortium of banks jointly funded development of a solution in the early 2000 and this is basically the default way you prove your identity online in Sweden.
It's a nice thought. I doubt it's true in reality. If it were true in this specific case for example, the industry wouldn't have the freezer situation to begin with (it implies non-effective regulation is already the context).
1. many countries have higher regulatory burden than the US
2. that does not mean regulatory gaps don't exist
The person above was simply pointing out that litigation is not the only way to solve a problem. Which is plainly true, you can simply require compliance via other legal means. For example, a government could create a regulatory standard for security and require vendors to follow it. Whereas in the US the political appetite is to be more hands-off and allow people make vulnerable products if the market allows.
Litigation is the last resort in the US system, too. It is the threat of fines/damages that incentivizes a company to do right by its customers/other stakeholders.
Parent is not correct. Litigation certainly is a thing, but it is less prevalent than in the US. Perhaps the food chain will take the loss, be covered by insurance, and the insurance company will sue.
I started playing the rs-sdk (https://github.com/MaxBittker/rs-sdk) with Claude after seeing it on HN the other day. I haven't looked at the Minecraft experiment but just letting an agent loose in the game and telling it to play did not work well. At first it really did not have a good feel for how the game actually plays and would routinely waste tons of time doing something that makes no sense — very reminiscent of the "four agents spent 11 minutes appointing a ‘sole fertilizer operator’ for a single piece of bone meal on a single wheat block" anecdote in this article.
But, after several hours of correcting and directing Claude (build modular, re-usable snippets of useful code rather than one-off scripts for driving, document your learnings for future agents, etc., explicit guidance on playing more optimally) it can now play the game fully on its own. I just gave it a goal to play through the end of the "Legends quest" (nearing end game content, requires playing through lots of complex quests ) and it's breezing through it.
I haven't experimented with swarms of agents yet, but I know if I had tried to create a swarm in 2004scape before the agent was actually good at playing the game, it would have failed horribly. So, I wonder how good the agents they were experimenting with are at Minecraft when playing alone.
How do you get Claude to continually work at it? I find it will do a number of steps and then just stop, no matter what I tell it. Then I prompt “continue on” and it will work a few more steps on its own.
I just tell it to work until described result. It works unless it gets stuck at a necessary step I had explicitly forbade earlier, or an unreachable port.
You can also tell Claude to setup scheduled tasks.
This is very helpful when you want Claude to do something specific, like analyze the next phase. Review past performance etc. I have found this to be a little better than Goal. However, cron tasks only continue for 7 days.
I considered that but I also considered that I don’t come close to maxing a normal $20 a month Claude pro plan. So I think “incredible waste” is probably too far.
I also see total garbage slop upvoted on this site every day. At least with this, I am experimenting and learning a little about how to use agents effectively, I’m not sure I can say the same happened for other slop on this site.
How banal. The first action after achieving "AGI" is not taking over the world, curing cancer, or anything fantastic like that, but rather "lowering operating expenditure"..
Leaving your “math is for humans” take alone, if the purpose of math is to “gain a precise and clear understanding of the world”, why is clarifying results a contemptible way to spend time?
Because the best way to do that is in a sufficiently minimal enviroment so people can just think and go their own way. Being bombared with machine-generated results is antagonistic to this process.
This is one of the most exhausting (and frankly, exhausted) recurring arguments that comes up over and over again on HN.
Some bit of open source software does something bad or unwanted which causes people point out that it shouldn't do that bad thing. In this case, not even "the authors of this software should be held liable for the software doing the bad thing", just, "Hey, the right thing to do would be to update the software to not do the bad thing."
Why does this always lead a zillion people to come out of the woodwork to point at licenses and warnings or whatever? Like, yes, there's a warning. Your software having a warning doesn't mean people can't criticize you and your software for doing bad thing. Your software license does not give you immunity from criticism or from people saying you should change your software.
No, you don't have to change the software. Yes, other people are within their rights to fork the software themselves. But they can also point out that they told you that your software was doing the bad thing and you didn't fix it or change it, and that as a result they don't like you or your software or both — whatever. Nothing at all wrong with that.
I actually think you framed it really well. No parties involved have any particular obligations to each other. Nor should there be expectations otherwise without support. It sucks to lose data, everyone knows that, and no one wants it. And identifying things which can be improved is important. I stop at expecting someone else to do something because I want them to. I might hope they would, and try to convince them.
reply