Iconochasm
2. Bootstrap the rest of the fucking omnipotence.
No bio...
User ID: 314
not at women being afraid of being approached somewhere without a clear escape route.
This inverse of this was the many people saying at the time that they understood why he took his shot when in the elevator - it meant he didn’t have to get shot down and potentially embarassed in public. It's not likely that he chose the elevator because of the implication.
Rephrase option 3 as "Sorry, I'm narrowmantic." Now it sounds logical and scientific and pushing further would make you a bigot.
They'll also just give them to you to take home if you ask. Or, at least they did when I would ask for my scouts.
What about just getting them fired and banned from using banks?
This is basically my attitude, with more aggressive schadenfreude. EA is a community built by autistic nerds with more money than sense; this is just whining that some other group of sociopaths was running cons on the quokkas first. You don't want to deal with the weird nerds? Go build your own mountain of grant money.
If the groundhog sees its shadow, we get six more weeks of winter. If the groundhog doesn't, then spring comes early.
So what does it mean when the groundhog just dies?
(This is why we need the BLR back. That headline is just chefskiss.)
And note my original post expressed quite a bit of skepticism about the general claim; it might indeed be incorrect!
The point is that this doesn't matter. Sure, assume it's true. When you use that truth to set a target ("We want more AP students"), you lose ceteris paribis; all else is no longer equal, there's a new incentive structure in place.
Remember, these are social "rule of thumb" "laws" we're talking about here, not natural laws of physics. Maybe this is some weird situation where there was the pedagogical equivalent of the $100 bill lying on the ground, and everyone manages to dodge all of the obvious and unobvious ways the attempts to reach the target could backfire or go wrong.
My contention is just that it's still the kind of situation that Goodhart was warning about.
In contrast, the claim re taking an AP class is that, even when controlling for student quality, it provides students with things -- skills, knowledge, the exposure to college-level expectations -- that themselves make college success more likely.
The specific claim was "controlling for test results", which I would argue is a poor proxy for "controlling for student quality". "People who take AP classes" is still a heavily selected subset of "high school students". Even the ones who bombed the test still had teachers and counselors thinking they were a good fit for the class in the first place. Going back to the diploma example, schools didn't just start off by giving out free high school diplomas. What they did was lower standards incrementally until we get the situation with grade inflation. Which feeds into why I think "slapping an AP label on an existing class" is exactly what we will see moving forward. That AP label is worth an extra 1.0 on a GPA, which will help students when applying to college, which makes the school look better. And even if every kid in that honors AP class bombs the test, well, bombed tests don't count against you, and the College Board themselves said that taking the class at all helps with college success! It's not like adding the AP class in AAS is going to add time to the school day; every kid taking that class is taking it instead of some other class. If that class isn't something like another AP history class, then you're going to be pulling from the kids who were specifically taking honors classes instead of AP classes.
If that is true, then increasing enrollment by members of group X will indeed result in greater success in college for those students.
And in theory, making every kid finish HS will increase literacy and general education. In the real world, "increasing literacy and general education" is actually very hard (especially when educators are hobbled with ideological bullshit), and gaming the system and targeted statistics is much more achievable.
It is absolutely incorrect to say, as you do, that the College Board is using "Being an AP student" as a measure of college success.
Earlier, you cite the College Board saying:
"Research consistently shows that AP students are better prepared for college than students who don’t take AP, regardless of their exam score. They’re more likely to enroll and stay in college, do well in their classes, and graduate in four years."
This is the measure. The correlation between being an AP student and doing well in college. So they made a target of "number of AP students", under the assumptions that "being an AP student" causes "doing well in college". But by making it a target, they change the incentive structure around "becoming an AP student" which means that the old correlation doesn't necessarily hold anymore, and given general trends in the incentives of large organizational structures, that change will probably be in an undesirable direction (probably "more AP students don't do well in college").
An alternative phrasing of Goodhart's Law might be "There are no cost-free optimizations in matured systems. The very act of attempting an optimization imposes costs elsewhere in the system." If you want to increase the number of AP students, there will be side-effects somewhere else in the system (because the system consists of people who react to the rules change) that will hamper or ruin the purpose of the increase. Check the wiki page for the alternative formulations and corollaries, I think the concept quite widely applicable.
On the more general topic of "large organization logic", this is probably driven by someone(s) in management who needs a measurable goal to point to the next time they apply for a promotion, and this is a number whose increase can be justified with facially plausible logic. Those people probably don't much care if "being an AP student" becomes less predictive of "doing well in college", because that almost certainly won't come up during the VP interview.
Disclaimer: I am generally not a city fan, and probably coming at this from a place of motivated reasoning. Nevertheless.
The gist of the comment was that whenever you hear somebody talking about how they want to live in a city because of museums, or a symphony orchestra, or lots of rock concerts, etc., what they're really saying is "I want to live next to other smart, cultured, cool people like me". And what they hate about the suburbs isn't so much the lack of those cultural touchstones, as much as it is having to live next to people who are perfectly happy with just a house that has a yard and a garage and a grocery store and a few chain restaurants within an easy drive.
There's some sense in this, if you're talking like Boston or SF, but cities have normies too, a fucking ton of them. Plus an enormous number of underclass people who are even less nerdy than normies. If you just want a large enough total number of like-minded people, and you're willing to search out the diamonds in the rough in a massive, alienating metroplex, I guess? I have friends who commute 40-60 minutes out from the local major city for D&D night. Traversing NYC might take just as long, and you'll spend all of it packed in a subway with normies instead of isolated in a nice, normie-proof car. If you can't find a dozen friends in a 500k county, your odds don't seem much better in a 5M city unless you're looking for something super niche; the problem is more likely with you.
and if you're the kind of person who hates suburban normies.
I think this sort of thing is usually projection, and indicates the sort of "I think attending cultural events means I have a personality" hipster whose whining about cities is tiresome.
the suggestion appears to be to change the weighting until it evens out the races, regardless of the impact on the efficiency of detecting lost revenue.
The most obvious stuff seems like it wouldn't even require much of an algorithm. If (Times SSN Claimed As Dependent >1), Then Audit (All Taypayers Claiming SSN As Dependent). If (Claimed Income) != (Reported Income), then Audit.
If this sort of thing is really the source of the discrepancy, then it's not even some AI algorithm thing. It's just basic computerized logic checking. The sane solution is to try to teach the black community to not commit easily detectable tax fraud, and instead engage in incredibly based tax avoidance.
Goodhart's Law is
When a measure becomes a target, it ceases to be a good measure
That is exactly what's happening here. "Being an AP student" is a measure, and they want to increase the thing that it measures, so they set a target of "more AP students". This will probably mean that "being an AP student" will be less useful as a predictor of future success. Because people involved, who are responsible for meeting the target, will be heavily incentivized to cut corners and take the easiest path to meeting the target.
but rather that the College Board thinks it does
Or is willing to pretend it does, etc, etc. It's still a solid example of Goodhart, unless there is rock solid reason to think that AP classes actually increase college success. Because the most likely scenario (because it's the easiest, most reliable method) is that more AP students will come from the pool of more marginal students who will do less well in college. Just like "high school graduate" no longer reliably implies basic literacy or numeracy; the easiest, most reliable way to bump the numbers for "college graduate" is just to stop having standards.
It depends on what is meant by socially conservative. In my experience, I see normalized antipathy towards homosexuality, open hostility towards trans, and very redpill sort of views towards gender.
OTOH, the churches seem more blatant about being scams, in the Heinleinian "snake oil shamans" sense.
That's exactly what Goodhart's Law is for. Being an AP student seems to be a good measure, in that it appears to predict college success, so we make a target of "number of AP students" under the assumption that there's a causal relationship in this correlation (AP classes cause better college outcomes), while failing to account for other explanations like selection effects (students more likely to do well in college are more likely to take AP classes).
Also, the liberal media does dominate in someplace people forget about - the nightly news. Twenty million people still watch the nightly news daily. Yes, most of those people are over 55, but even w/ that 3.4 million in the demo (25-54) still watch it daily averaged, which dwarfs even Tucker's 25-54 ratings.
As one of the 25-54 wierdos, I encourage people to try it sometime. The morning news is even worse.
I did see that. It seems like criticism of the course for waging the culture war. Unless I misunderstood you, and that's what you originally meant?
Did they? Iirc, the stated motive was an apparent lack of academic rigor.
Whether it's "weak sauce" was not the point. The point was whether progressives unanimously love Pfizer
Oh, I popped in after that, just to talk about the regulatory capture aspect.
But the "culture war" aspect is why Florida rejected this particular course.
Did they explicitly say that?
I saw a lot of this sort of thinking on those EA forums that were linked over the Bostrom controversy. It's a fundamental problem with collectivist, authoritarian thinking. They can't just accept a descriptivist fact, it must be twisted into a prescriptivist dictate.
While this specific article is about Pfizer, regulatory capture absolutely is nothing new to progressive discourse.
This is really weak sauce. It's a book review about a novel featuring cartoon capitalism that mostly misses the point of regulatory capture, and whose proposed solution is doubling-down on the exact stuff that enables regulatory capture.
This is blaming every bad thing that ever happens to black people on white people.
Yes. This is unironic white supremacy, the core belief that black people are not agentic, are not capable of being responsible for themselves in the way that white people are. If black people do something wrong, be they criminals or criminal cops, the true blame goes to the nearest relevant white men, even if those nearest white men are actually hilariously far away. In much the same way you blame the parents for the anti-social antics of unsupervised middle-schoolers. This belief, as an unexamined, unadmitted premise is fairly common among the more ideologically-dedicated woke people, especially white ones.
I remember once puttering about the house in the morning, not really paying attention to the news on the TV, and recognizing his voice. I had an "Aw shit. Here we go again." moment. And sure enough, that was just after the extremely disingenuous shooting of Jacob Blake, and just before the Kenosha riots that catapaulted Rittenhouse to infamy.
DePape's own testimony
So, this is the kind of careful wording that puts my hackles up. He wanted her to tell the truth about what? She was the "leader of the pack", we get that as a purported direct quote, but "of lies told by the Democratic Party" is not a direct quote. Later direct quotes have him wanting her broken kneecaps to be a symbol to "other members of Congress". Did the police just not bother to ask what sort of "truth" he was looking for? It seems kind of pertinent to the attempts at partisan spins.
One potential issue here: what is the rate of violence women are intuitively expecting vs the rate at which nerdy tech dudes actually lash out? I would expect a very large discrepancy between "ancestral environment" or "feminist paranoid take" versus "low T nerd convention".
More options
Context Copy link