BinaryHobo
hauling up the data on the Xerox line
No bio...
User ID: 1535
I don’t know what squishy means sounds like inconvenient fact to me.
My meaning was that the comparison breaks down when any pressure is applied. My apologies for the shorthand, I was in a rush. I probably should have just held off. That one's my bad.
Most of the how to fix policing recommend things like more people from the community or atleast same race but also recommend things like more education. But it’s very tough to hire both qualities in the communities with the most policing issues.
How about devolving autonomy down to these local communities? It inherently deals with the first half. Worst case is that these areas end up being horribly dangerous and known as the place you don't go after dark, and the communities around them place extra police presence just across the boarder. That is to say, roughly the same situation as it is today. The best case is that the new cops that were raised in the community can actually make some sort of difference. Seems like mostly upside to me.
I don't buy the education side of this being a problem though. The average IQ in the areas we're talking about is what? 70-75? Because of the Flynn effect, that's roughly what the US as a whole was working with around 1900 or so, and the US, and each of its cities, managed to recruit plenty of competent police officers. Might have to back down from really abstract crimes and focus on the basics like property crime and violent crime, but my guess is that's what these communities need.
I don’t think any of that disagrees with my point.
My point was that you can't objectively say that policing failed without consulting the local community, unlike with healthcare. Which makes the comparison you were trying to make really squishy and not very useful.
Only if the entire outcome is due to policing, which seems absurd.
I don't think that's necessarily true. I think to do policing properly, you need to have the trust of the local populace*, and failure to earn that represents a very real failure.
That's before we get into the optimal level of safety, which is probably not maximum safety. I did mention "as safe as possible" in a separate post, but I probably should have said "as safe as is practical". That level of safety, and which kinds of safety (e.g. the difference between unsafe worksites and muggings in the street) to prioritize aren't decisions that I, or the police, can make for a community. That's a decision that the community has to make for itself.
*Note: Does not apply if the police in question are controlled by a far away power center. That's better modeled as an occupation force, rather than a self-governing community.
Do we accept that the point is to make it as safe and happy as possible, understanding that it may not be very safe and happy in an absolute sense?
This is a fair distinction, as there is no absolute safety or happiness. Utopia isn't coming, regardless of what the police do.
Yes, the point would be to make the community as safe and as happy as possible. And if the community as a whole doesn't feel that the police are doing that, then they've failed regardless of how many regulations they followed.
It’s just reminds me of killer king hospital where they tried to get “representatives” of the community working there but then a lot of bad care happened. Perhaps, policing isn’t as complicated as medicine.
I don't think how complicated it is matters as much as you're thinking here. I think the difference is subjectivity. The right way to police a community, fundamentally depends on the community. The point of policing is to have safe, happy communities. And if you don't, it doesn't matter how much objective criteria you've ticked off a list, you've failed. It's a very soft-skill heavy, squishy job. Or, as I've heard it said: "The most important thing for a police officer to know, is which laws not to enforce."
Medicine, on the other-hand, has a very obvious objective pass/fail standard. Did the patient live or die? This lends itself more readily toward standardization and academic knowledge. Biology works pretty similar within most bodies, and if it's not working that way, it's often the thing you're supposed to diagnose.
The supreme court nominees.
inventing machines is much easier than altering human biology.
Then imagine a small machine that repairs telomeres or something.
If you want to be taken seriously, you first need to provide some evidence not that such experiments might be run, but rather that they might succeed.
I really don't. If you go look at the thread, TheDag advanced that part of the argument. I'm not responsible for other people's arguments.
I just got annoyed by The_Nybbler's bad shutdown that amounted to "It hasn't happened, yet", which is a completely useless statement during a conversation about future tech, both true by definition but not relevant. I might add, that you also engaged in during your first response to me.
I have advanced plausible interpretations, but I've reached my limit, as I'm not the one that advanced this argument, and so I have no real idea what they were thinking.
So what technology has increased the maximum lifetime of humans so far?
It hasn't. But technology obviously can fundamentally change human limitations. Before the airplane, what technology had increased the maximum flight distance of humans?
than why should we think that future technology will do so?
Because there are several examples of biological immortality in nature (just as there are several examples of flight in nature), and we've gotten pretty good at stealing ideas from evolution. We know it's possible for a biological organism to be immortal.
Now, will the first couple thousand people who try such a genetic alteration probably get some weird form of super-cancer? Sure. Probably. But if we're talking of the elite of the elite pursuing their goals, that's not that many bodies. Elites have thrown way more bodies away in pursuit of mere land.
all concrete science so far indicates that maximum human lifespan is really quite a persistent and narrow range.
Without running crazy genetic experiments, sure. Do you have an argument for why we will never run those experiments in the future?
The essence of that argument is usually something like "look how much the mean (or median even) lifespan has gone up over the years -- surely people will soon live to be 200".
You can't take the average argument for a position, you have to engage with the argument in play, because it only takes one good argument for something to be right. I can generate several hundred bad arguments for any position you want right now.
Even further, that wasn't the case this time. TheDag explicitly called out a technology change.
Quoting them:
I think the most likely path is that the current elite (or the elite of the next generation) will create life extension technology and effectively rule forever...
And when talking about what science is coming, they don't talk about past medical advances, but reference AI, presumably some sort of intelligence explosion.
Sure, but they're talking about a technology change.
Before the internal combustion engine, the maximum human travel speed hadn't changed in centuries either.
Not to say that it's likely, but I think you're ignoring the essence of the argument.
If he does go to jail for a long time, I think this would have an effect on reducing imitation of his lifestyle, but it’s hard to say how much.
Why? Young men with no prospects imitating criminals (and eventually joining them) has been a thing for... like ever.
So the argument is that, as the aristocracy moves to extracting wealth from human capital, the propaganda naturally moves towards methods designed to protect that human capital (and the related heavy infrastructure)?
That seems plausible on the face of it. I guess, does anyone know if we see more glorification of war in official propaganda from resource-extraction based economies (e.g. petro-states, countries that make most of their money on diamonds, etc)?
Even if so, that seems to conflict with the recent rise in pro-war messaging we've seen over the Ukraine issue, though. The left has been pretty gung-ho on it, as well as being the political side that generally benefits from higher quality human capital.
Sure, but that's not... exactly my point.
Why is the current aristocracy behaving differently than past ones?
Treaties are US law, assuming they're ratified by 2/3 of both houses of congress.
war isn't positive and beautiful
That's entirely fair, but for most of human history, the rhetoric around it was. So was the vision. War was painted as glorious.
Obviously war and the culture war are different, If we're going to compare them, why are they different in this particular manner?
I've made the joke since I joined Reddit an embarrassingly long time ago that nobody hates Reddit more than Redditors.
I feel like that was more of a post ~2012ish stance, maybe post 2010 in certain places (I was also on reddit for an embarrassingly long time).
Around the time of the Great Digg migration, I remember the site being really proud of itself (I'll let others decide if that was deserved or not).
Thinks like this were emblematic of the time. Reddit just really seemed like it was a better place to be than everywhere else. This is also the era when I heard it described as "4chan with a condom", so take that for what you will.
But that was a long time ago. If you'll excuse me, I need to go be nostalgic for a while.
Why we find ourselves in an orderly world can be explained by the anthropic principle of "if the world was not orderly, we would not be here asking the question".
I've never liked this as a rebuttal to the point made. It definitely answers the question as you've phrased it. Why do we find ourselves in an orderly world? Because if it weren't orderly, we wouldn't find ourselves anywhere. I get the line of reasoning, but it gives no insight into why the world is orderly, which is what question is really being asked. It merely asserts that it is the case, which wasn't really up for debate.
To rephrase the point in a slightly less charged light. When discussing the question "Why does the necessary precursor to A exist?", answering "A exists, therefore the necessary precursor to A exists" doesn't answer the question. It completely ignores the "Why" part of the question.
Now subtract all of the men who
Perhaps some of them have power as thought leaders or community leaders. Perhaps people who are thought leaders or community leaders think they might get bought off with sinecures if they can produce some big riots.
This seems unlikely. To the extend that older women have sway in urban minority communities (especially as community leaders), it tends to be through church involvement. The folks that get DEI jobs generally aren't the type that attend church services regularly, so far as I'm aware.
It seems more likely that there were already tensions in these communities, and a spark set them off and people just rioted. Humans do this occasionally under certain circumstances. The DEI consultants seem more like outside opportunists, capitalizing on the riots and internal progressive (mostly white) status games.
Someone who really is happier at home with his family and his pursuits is not a traitor; better that he go home, than stick around distracted and half-assing the job.
I mean, on one level, sure. Some random person who is happy at home isn't a traitor, but that's ignoring the context of a (possibly secret) society devoted to $GREAT_WORK.
Generally when you join such a group, you would take an oath to put the group above all else. It's the betrayal of this oath that constitutes treason, which requires that you be in the in-group before you can commit it. The in-group membership is important. The same way someone born in Bolivia cannot be a traitor to the country of Iraq, someone who never joined the secret society can't be a traitor. An enemy? Sure. But not a traitor.
Presumably the takeaway is that someone who is happier at home should never take such an oath and join such a society, which seems reasonable. But the context of the quote seems to specify someone who has already made such a commitment.
That doesn't seem like something the US didn't like. It seems like something the US didn't care too much about, and Pakistan was useful during the Cold War.
They specifically talk about "American unipolar hegemony", so I think they're specifically thinking about 1991-now (or whenever you want to argue that American hegemony breaks down).
Specifically, the dynamic is different because pre-1991, countries could get away with things the US didn't like as long as they were willing to suck up to the USSR.
or is there some logic here that I'm not seeing
Best guess I've got is that it will last exactly until a deal is signed with some big AI company for access to the dataset.
Otherwise, what is the metric for wealth distribution?
How useful you are to the people with the money to pay you. Which generally implies being willing to take on extra work, or having done a bunch of work yourself for free to build expertise, so that you're useful later. The children of the wealthy are generally considered extreme outliers, who tend to lose the money within a couple generations, and don't really have much of a long-term effect.
Is roofing in the summer as a redhead especially lucrative?
Suffering doesn't buy the boss anything. Working hard may require suffering, but the suffering itself does not have any value, unless the boss is a (literal) sadist, in which case it may provide quite a bit of value. The value is generally in the work, and if you can find a way to do the work without suffering (example: the redhead wearing a wide-brim hat to protect their neck), basically nobody would say you should charge less.
That only seems to present a problem if we assume all labor is the same value.
If we assume some labor can be more scarce than others, say an expert or those with exceptional natural talent, the situation resolves itself.
But, what is scarcity, but a description of how much effort (labor) is required to obtain something? The record companies probably do put in 1000x the effort to recruit someone like Taylor Swift. Talent scouts, marketers, etc. Of course the LTV requires people be willing to expend labor in exchange for the product. But, the record companies only expend the labor cultivating talent because they expect to receive value in return. The labor required to create a talent acts more like a price of production on a standard economic graph. Products that return less value than required to create simply flop and don't stay on the market very long.
If we've described scarcity in terms of labor, I think the labor theory of value holds together, although I think it does become one of those equations where you can solve for any of the variables.
More options
Context Copy link