Posts
Comments
My complaint is that you don't look at whether countries that are differently effected by the factors you list actually do differently on research output. If you don't do that I think it's hard know whether on of the factors you list is indeed causal for the result.
If you believe that overall research slow down is happening to factors that apply in the US but don't apply in Germany, then the logical conclusion would be that Germany should have a higher research output.
To the extend that you are interested in knowing whether your thesis is true, it would make sense to check. Rationality is not just about making plausible claims but also investigating whether they might be true.
I don't think that it's just social justice across identity groups being at the right place at the right time. As a meme it has the advantage that it allows people who are already powerful enough to effect social structures to argue why they should have more power. That's a lot harder for social justice across social classes.
Deep Work by Cal Newport lays out a framework about how to develop intellectual stamina. On of the aspects is to actually focus for longer periods of time on doing deep intellectual work in an enviroment without distractions and then also allow for some freetime where you don't have constraints about how to spend it.
Cal Newport is impressive in that he managed to do the work required to become an associate professor in computer science while at the same time having a blog and writing six nonfiction books outside of the computer science domain.
As far as c) goes, if your intention is to live right next the university you would have to move. Practically, where I live most people don't move next to university. That in turn means that you also don't gain the social benefits of having your fellow students live next to you.
It seems that your comment tries to take it apart by looking at whether you like the way the system is designed and not by looking at effects of it. That means instead of trying to see whether what you are seeing is true, you expand on your ideas of how things should be.
Most of the features you list of the university experience are not universals of university experience. In Berlin where I live plenty of people don't change their living situation when they enter university and there's also no student debt attached to it.
If you think those factors matter a lot, what outcomes do you expect to be better in a country like Germany?
It feels to be written from a libertarian point of view.
You might argue that the FDA causes a lot more then 45,000 yearly deaths by not allowing valuable medicines to be brought to market. It's debatable whether or not a more government interference or less government interference would be helpful. The same goes for government interventions in the economy. On the other hand there's no other party that might defend the country from attack then the government.
Before the tea party when this was written libertarian slogans were less of an applause light.
So I would interpret the post as saying, immediately afterwards the politicians responding by saying libertarian things but then they overreacted instead of engaging policies that would pass libertarian standards.
(that said I don't agree with the point)
They were couchsurfing with me a few days around the LessWrong Community Weekend when they were still known as Chris. This means that I have a decent insight into them but I'm not one of the people who lived with them in Crow's nest.
At the time there morals were that people should interacted in a way that maximizes utility if they were a timeless decision based agent. I think they were vegan for some timeless decision theory based justification. I don't think the person I might back them would raise an objection.
It's my understanding that at the time of their death they were of the opinions that nothing matters. From that point of view there would also be no objection to his work being used by other people.
A key concept of effective altruism is that you don't ask "what would be effective if everybody does it" but focus on tractability, neglectedness and importance when choosing your own actions.
Apart from that it's a possible feature that everybody eats meat but it's artifical grown meat for which no animal had to suffer. It's one of the approaches that the EA's I know in the field consider tractable.
I see a choice as a type of intention. That choice has effects. The process of feeling certain emotions when interacting an entity with whom one relates is a downstream effect from that choice.
If I do believe reporting I start out with the intention: "I will tell the truth". That's a choice I'm making.
One effect of that choice is that I will relate a certain way to saying something part of me considers untrue. The downstream effect of the choice is that I will feel tension when saying an untruth.
I do agree that the relating that happens downstream from a conscious choice is of a similar nature then the relating that happens without setting a conscious intention but the relating is still another object then the choice.
Other things that now would be a good time to hammer into your head if (like me) you didn't already: coughing into your elbow, sanitising your doorknobs and light switches, and not touching your face.
Instead of sanitising light switches, stop having light switches. Movement sensors/Google Assistent can switch lights without any need for touching switches.
Doorknobs are awful technology to the point that Australia recently outlawed them for new bulidings. Handles are still problematic but have a larger surface area so different people touch at different places and less pressure is also helpful.
When it comes to the advice of not touching my face, that might be the main takeaway I take from this post.
Let me steelman the argument.
The first defense against outside germs is our microbiome. I do take steps like not using shampoo to avoid damaging my microbiome.
If the microbiome on your hand is already full of bacteria who live in happy coexistince with you it's harder for new pathogens to attack you then if you regularly kill the protective layer of the microbiome.
Beneficial effects of plant-associated microbes on indoor microbiomes and human health argues that if you put plants which aren't sterile into hospitals that has benefitial health effects because it results in a better ratio of harmless/harmful bacteria even if it raises the total amount of bacteria.
You aren't just touching other peoples hand directly. You are usually touching doornobs, light switches and other objects that other people are also touching.
The great shock of the replication crisis to me was that there was a group of scientists who sincerely believed that replication was not important.
The replication crisis is about scientists finally waking up and thinking replication is important. Psychology never had a culture that valued replication.
It seems like your model doesn't factor in legality. We have a lot more laws that add additional burocracy to technology development today then we had 50 years ago.
Nothing in the facts of the scenario layed out in the OP indicates that anybody engages in underestimating. The scenario is completely compatible with everyone estimating the effects of their actions correctly but at the same time are unwilling to pay the price for them.
There are two reasons for jargon.
(1) Developing rationality@LW as it's own paradigma by reusing other concepts from LessWrong.
No field of science can stand on it's own without creating it's own terms and seeing how those terms interact with another.
(2) Defensibly against being able to be quoted in a bad way.
Charles Murray succeeded in writing "The Bell Curve" in a way, where almost nobody who criticizes the book quotes it because he took care with all the sentence to write nothing that can easily taken out of context. Given the amount of criticism the book got that's a quite impressive feat.
Unfortunately, in many controversial topics it's helpful to write as defensibly or even Straussian.
Depending on the goal of a particular post (1) or (2) sometimes matter and at other times it's worthwhile to write for a wider audience.
I do agree with those thesee. I think if you define relating this way you can't at the same time define it as being a choice as the choice is a different ontological object. I took two years to try to figure out what the word relationship means and to me there's now a certain clarity.
If I have a discussion a thousand times I usually do have my sources saved in a way that I can easily share.
I also doubt that you had the same conversation a thousand times because "You make claims about strong effects of parental enviroment and those generally disappear when studies are done in a better way" is likely a different position then when you have the same discussion outside of LessWrong with people who have quite different objections.
Your example about cell phones is a prisoner dilemma. The choice to continue using the cell phone has more utility for each individual participant if they are the only person who would stop using it. At the same time it there would be higher utility for everyone, if everyone would choose cooperate in the prisoner dilemma and stop using their cell phone.
Having a government legislate that everyone picks cooperate in a prisoners dilemma is a way to solve the prisoner dilemma.
Even if a person wants to do something about a problem, it's often much more impactful to donate to an effective charity then to change personal behavior.
The recent founders pledge article on climate change that illustrates that principle for climate change. Animal Charity Evaluators might not be the most trustworthy source but when it comes to the numbers I see from EA's the same principle seems true for that area as well.
My working model for relating is that it's second-order cybernetics. It's not about the first-order effects of interacting with a person (or other entity) but the second-order of what that does with you.
Sometimes that happens as a downstream effect of a conscious choice but plenty of time there's not much consciousness about the formation of relational stance.
If you are the only person touching your smart phone and not give it to other people that should reduce a lot of the potential of them as a vector. If you in addition don't press it against your face, it's not clear to me why it would make sense to worry about it.
(3a) What Are Maze Behaviors? It would be good to have a compact link-to-it answer. But I’m also coming around, after having started a draft of this, to the perspective that this is likely asking the wrong question.
It seems like the question is problematic because there are different kinds of Mazes. The Mazes describe in Immoral Mazes have specific values. There are mazes that have different incentives on how one climbs the ladder within the organization and those mazes will differ in some of the resulting values.
A person who rises in ranks in a big labor union also had to do a lot to optimize rising the ranks but they show a bunch of signs that are different from those described by Jackal.
In Jackal's studies it's a value to avoid sharing information. It's possible to setup incentives to share information in an organization.
It seems like the recommendation is about preventing dirt on your hands to enter your body. If the hands of a woman are contaminated with a virus washing hands after going to the toilet instead of before poses the risk that the virus enters the body.
On the other hand unless you are currently ill, the pee will unlikely be a vector for viruses.
Bill Gates scored 1590 out of 1600 on his SAT which is an IQ of ~150. From what I read about Warren Buffet's IQ it's in the 150 range as well.
The amount of Jewish Americans is 2% at the same time they make 35% of the top 400 richest people. Higher Jewish IQ seems to me like the best explanation for it. If you don't want to argue for Jewish conspiracy, IQ seems to be the prime causal vector.
Claiming that you can get a dean of Harvard or the chairman of the CCP with an IQ of 100 seems to me a pretty implausible claims and you do nothing to argue why your readers should believe it.
Gwern's summary of the IQ research suggests that IQ generally does correlate with the good things and there's no mention that that stops at 100.
Writing a long article on how intelligence isn't important without engaging with the research literature that supports intelligence being important feels out of place for LessWrong.
It's not clear to me why you aren't linking to quality sources if you think the quality sources for your claims exist.
It seems like "real expert" is here used in two different senses. In the one sense there's an expert is someone who spend their 10,000 hours of deliberate practice and developed strong opinions about what's the right way to do things that they can articulate. That person will likely have convictions about what public misconceptions happen to be.
In the other sense being an expert is about an ability to produce certain quality outputs.
You can tell whether a person is an expert in the first sense by seeing whether they try to correct your misconception and have convictions about what's the right way to act or whether the person just tells you what's popular to say and what you want to hear.
Yes, I don't see how it changes the fact that this setup causes stronger incentives to Goodhart the milestones given that they decide about whether the companies dies or can continue to function.
Milestones are set. If they are met, funding is unlocked at pre-negotiated levels, locked in for both sides in advance.
It seems to me like the pressure to Goodhart are higher when you agree to be only funded by a single entity with specific pre-negotiated milestones then if you are in a state where you are going to seek capital from a bunch of people with at least slightly different evaluation criteria.
It seems to me like it would be good to have an easy way to only make posts (and maybe also comments) visible for logged-in users with >10 karma for those political discussion that we want to happen.
Aren't there also people for whom 3 points is a strong upvote that you can't distinguish from those where 3 point is a weak upvote?
Anyway you do science can turn out to be usually good but systematically wrong in some way you didn't test. Most placebo-blind studies are build on questionable assumptions about how blinding works.
jessicata does according to their profile work in "decision theory, social epistemology, strategy, naturalized agency, mathematical foundations, decentralized networking systems and applications, theory of mind, and functional programming languages".
In a field like theory of mind there's not knowledge that verified to standards that would satisfy a physicist. The knowledge you can have is less certain. In comparison to the other knowledge sources that are available increasing your ability of self introspection is a good help at building knowledge about the field.
It sounds to me like you are arguing that humans can't improve machine learning models by studying them. That seems to be far from our reality. Most successful machine learning models we have today get tuned by humans for the task they are doing.
Why is not all the data for tackling this issue publicly available ?
Because people value privacy and don't want their medical data being public.
The single model that was tried on this dataset claims to perform about as well as a radiologist. But imagine if we didn’t have only 1.2 million images, imagine if we had over 100 million and image if we didn’t have a single model, imagine if we had hundreds of them,
If the trainings dataset happens to be about how the average radialogist rates an image you don't get better then a radiologist by looking at more images with the same labeling.
if we didn’t have a single team but rather hundreds of teams working on this problem.
If you look at Go as a casestudy, the progress when hundreds of teams were working at the problem was relatively slow. The progress took of when the single well-funded team at DeepMind spend it's resources on the problem.
A few well-funded teams will likely do a better job at analysing cancer images and produce a commerical product then academic teams.
It is not a coincidence that rather than propose a systematic policy, the companies singled out as targets by this candidate are the relatively new companies, not politically well connected or seen in a good political light, that have best created value for us over the last decade. To the extent that life in these parts has gotten better, it is hard to think of more central reasons for this than Amazon and Google.
Google and Amazon both as politically well connected. They are both in the top20.
Good-governance advocates in the EU succeeded at reducing the ability of individual cities or countries giving subsidies to get investors. The process is more burocratic then desireable, but it's a huge improvement over the status quo from before and also from what happens in the US.
It's also part of the Washington Conensus and countries that take IWF money often has to agree to pay less subsidies.
Domestically in the US, states rights are an argument that likely prevents and politician from the left from pushing through similar limits as we have in the EU.
A debate about a recent search for a location where a new headquarter can be build however suggests to me that it will be difficult to get majorities for a reduction in corporate subsidies as it's not something that people consider to be important.
I think a person who has trained awareness about thier own cortisol levels is likely to have some useful knowledge about cortisol.
They might have hundreds of experiences where they did X and then they noticed their cortisol rising. If you talk with them about stress they might have their own ontology that distinguishes activities in stressful and not-stressful based on whether or not they raise their own cortisol level. I do think that such an ontology provides fruitful knowledge.
A decade ago plenty of psychologists ran around and claimed that willpower is about how much glucose one has in their blood. Professor Rob Baumeister wrote his book Willpower for that thesis. If Baumeister would have worn a device that gave him 24/7 information about his glucose levels I think he would have gained knowledge that would have told him that the thesis is wrong.
(I guess that when you wrote "piano" you meant "violin".)
Yes, I changed the instrument and forgot to change all instances. I corrected it.
It seems like you only pick the time-frame where there was international attention on the issue. I would expect that this attention raised the discovery rate.
Where did you get the idea that the number of infected people doubles every two days? https://forum.effectivealtruism.org/posts/g2F5BBfhTNESR5PJJ/concerning-the-recent-wuhan-coronavirus-outbreak suggests that each person might infect 2.6 people on average and the incubation time is relatively long.
The fundamental difference seems to be that most biology teachers have not signed up for teaching religious fundamentalism and their university training didn't teach them that ideal but critical theory is deep in the university curriculum for teachers.
Because a lot of power isn't projected through laws. If you take Hitler's Enabling act, parliamentarians that didn't vote for it, did get punished but there was no need to explicitely write that into the law.
Comparing epistemic conditions over time is very hard because different communities have different epistemic conditions.
I found the Robert Moses biography The Power Broker by Robert Caro very informative when it comes to understanding how an epistemic environment was shaped in the past. Robert Moses effectively collaborated to get any opposition to himself censored while he build parks and bridges in New York.
He had his bloodhunts that digged out dirt on his critics and ended up without anybody speaking up against him for decades. Of course Robert Moses only cared about a more narrow area and didn't prevent people from speaking outside of that field but it still impressive how much power he managed to weld in the supposedly democratic New York in the middle of the last century.
Caro writes that being against Moses and being against parks was like being against motherhood. It wasn't a tenable position even when it might have made more sense to build other infrastructure then parks with the same money.
Being against parks is a bit like being against equality in our times. Parks are nice and equality is valuable, but it's always a question of the price that's payed.
The notion of Straussian knowledge that can't be expressed directly is older then this recent debate but it's hard to accept for previous times because the official narrative of a time conveniently leaves it out. It's very hard to reason about it.
One great feature about our times it's that it's much easier to get the knowledge of what happens. The internet gives us a way to reason about what's happening with us that wasn't available in the same way for someone dealing with Moses in the 40's or 50's.
An expert violine player who never read any academic papers about violin playing but has thousands of hours of practice doesn't only have correct idea about violine playing but can reasonably be said to have some knowledge. The violine player has metis. That metis can't easily transfered.
The practices that jessica described also produce a kind of metis.
Metis is by it's nature hard to transmit and it's easy to search for some techne when making a claim that one believes because of metis.
From the twin studies we know that parental enviroment has relatively little effect on educational outcomes compared to genetics.
A quick Googling brings me to statistics like single fathers are more likely to be black. Given that blackness does correlate with poor educational outcomes, it wouldn't be surprising to see a correlation of single motherhood with poor educational outcomes simply because of the sampling.
To be convinced of any effect of parental enviroment I would want to see an effort made to clear up the data to be convinced and most Google hits do't do that.
Editing posts afterwards to remove statements is not a great way to have rational debate.
You do it to the extend that you have a causal model in your head that links the two. If you take the issue of toxic pesticides, new pesticides got used and a lot of our bees died. Whether or not there's a correlation is subject to public debate. That's how real-world examples look like.
The map is not the territory. In the process of measurement the deaths due to a pesticide you need a complex model about causality. That model means you have an abstraction.
If you get your drug unblinded by giving it strong side effects it will perform better against placebo. It's a way to Goodhart the gold standard in our way to establish the causality of whether a drug helps a patient.
Any model of the causality of deaths due to your pesticide will be subject to Goodharting.
How can object level results be used as metrics, instead of proxies? (And how can 'metrics being gamed' be measured, particularly automatically?)
The chapter called "The Goodhard Trap" seems to be about this being principly impossible. Anything you use to measure is by it's nature a proxy and subject to Goodharting. The map is not the territory.
Are there specific statistics about the metrics you are referring to?