I hope it’s not purely ai generated, but who knows, maybe it is and it’s still interesting and informative. It could still be with such huge volume and high signal basis. Wish I’d thought of this actually.
One where all uuids are always different whenever the person who wrote the implementation tries it, regardless of computer. And whenever it’s someone else, always the same.
Well said. I think we all shouldnt be too quick to assume that the problem starts with the person doing the bullying, nice and simple as that would be.
> I think we all shouldnt be too quick to assume that the problem starts with the person doing the bullying
I don't think anyone is making that assumption, but being ok with corporal punishment likely comes down to three things:
1. We should care more about victims of violence than perpetrators, and all measures should be taken to protect victims and prevent victimization, even if it hurts perpetrators. Meaningful consequences for violent behaviour is critical.
2. The belief the physical deterrents work, if applied consistently and not abused to the point where it doesn't provide clear guidance as to acceptable behaviour.
3. That the primary job of schools and educators is to provide a safe and effective learning environment. Being therapists that get to the root of problematic behaviour is neither in their training nor in their job description.
How about when the perpetrators are also victims? If child A is bullying child B because they themselves are suffering abuse at home (as is often the case), don't both kids deserve help and support? Just beating up child A is no more productive a solution than throwing people in jail.
I grew up in a place where teachers could and did beat children. School kids there are far better behaved than kids in US public schools. I’m not making an argument for it, just an observation of effectiveness. It works.
> If child A is bullying child B because they themselves are suffering abuse at home
Experiencing hardship doesn't excuse violence against others, just like it wouldn't excuse breaking the law. You can say "here is the punishment for your bad behaviour, now let's ALSO have child services remove you from that environment AND have the justice system punish your parents' bad behaviour". Everybody has their job and if they do their job, then what's the problem?
> Just beating up child A is no more productive a solution than throwing people in jail.
Firstly, there's no "just do X" for multifaceted problems. Secondly, people these days dramatically underestimate the value of prison. Over 60% of violent crime is committed by under 5% of the population. Don't underestimate the value of simply removing repeat violent offenders from society.
> Experiencing hardship doesn't excuse violence against others
I totally agree, but I don't agree that forgoing violence as a punishment is the same as excusing the bad behaviour. The best outcome for everyone is surely rehabilitation, no? There are other punishment options if you still insist on inflicting some hardship.
> Over 60% of violent crime is committed by under 5% of the population. Don't underestimate the value of simply removing repeat violent offenders from society.
That neatly avoids the question of why they reoffend, which is precisely my point. If prison is effective as a deterrent then why do they keep coming back? "Simply removing them" for a period of time simply perpetuates the problem, thus helping to ensure more violent crime in the future, not less.
> I don't agree that forgoing violence as a punishment is the same as excusing the bad behaviour. The best outcome for everyone is surely rehabilitation, no?
The dirty secret is that we have absolutely no idea how to rehabilitate anyone. Even our best therapies for people who desperately want to change their behaviour are only 40% effective at only slightly modifying behaviour, and most violent criminals unfortunately have no such desire.
This is partly psychology's fault for doing such poor science for decades (35% replication rate!), but partly also the false premise that there are no innate biological factors at play.
> That neatly avoids the question of why they reoffend, which is precisely my point. If prison is effective as a deterrent then why do they keep coming back?
This is already known but ignored in "polite" society: poor impulse control. Most repeat offenders only stop offending once they age out of impulsive behaviour, not because they had some kind of revelation or personal growth; stories like this conflate correlation and causation.
Age is the best behaviour modifier we know, because hormonal profiles change, which ends up changing strength and frequency of the impulses we have to overcome. Imprisoning repeat violent offenders until they age out of it poor impulse control is totally a policy that should be on the table.
At the point a parent is beating up their own kid I wonder what options are available. If they're removed from the family then placing them in foster care almost always leads to worse outcomes than leaving them with the abusive family. The state doesn't know how to raise children.
Then surely the focus should be on solving that problem? Just clamping down on the proximate cause doesn't really help - as others have pointed out, it seems likely to incite revenge attacks rather than stopping the bullying.
That's the thing, it's unclear if it's a problem that can be solved. It has to do with fundamental benefits of staying with biological family, and avoiding the extremely negative consequences of lack of attachment.
My bully had two much older brothers and I guess that's how he learned to communicate, so I communicated back. We became friends afterwards.
Looking back it's not the physical bullying that was the most damaging, but social. I went to a different middle school and without a support network it was difficult to say the least.
If I can just stand up for the nitpicker - arguably in the uncanny valley it’s more natural to point out it’s not reading (by their definition) than outside it (ssd’s).
makes sense in a philosophical debate or when you're talking to your confused grandparents, but does anyone on hn not know how LLMs work, at least on the level of "tokens, matrices, data, sgd"?
otherwise, that reminder must imply that people do know how it works, and yet they still ascribe to these models some property like qualia, i.e. something other than "being able to turn english into code and compute into shareholder value";
but then if you disagree, why even mention it in the first place? do atheists randomly proclaim "btw god isn't real!" in unrelated conversations with strangers of unknown religious beliefs?
Well said. And similarly, it always seems to be the simple, bottom up, “let’s just build something simple and minimal that works” projects that get iterated on that do can do well, and start to strain when the technical debt and complexity accumulate.
I found it thoughtful and thought provoking. I agree that it seems too unlikely and irrational and unnecessary - and quite rude to speculate, given Occam’s razor - that Altman actually had someone killed.
I joined the waiting list.
I hope it’s not purely ai generated, but who knows, maybe it is and it’s still interesting and informative. It could still be with such huge volume and high signal basis. Wish I’d thought of this actually.
reply