Posts

Comments

Comment by Sieben on The track record of survey-based macroeconomic forecasting · 2014-04-22T16:44:36.555Z · LW · GW

"For mature and well-understood economics such as that of the United States, consensus forecasts are not notably biased or inefficient. In cases where they miss the mark, this can usually be attributed to issues of insufficient information or shocks to the economy."

Maybe it's the allure of alarmism, but aren't we mostly concerned with predicting catastrophe? This is kind of like saying you can predict the weather except for typhoons and floods.

Comment by Sieben on How long will Alcor be around? · 2014-04-18T20:00:18.200Z · LW · GW

It's fair to assume that a Cryonics company would be set up to endure in the long term. Otherwise they dramatically reduce the number of people willing to sign up. This is different from a startup tech company who does not have to promise its investors and consumers that it will be around for the next 50 years. It's kind of like the opposite of a netflix account. This should give us a lot of hope because even Netflix seems to be pretty robust.

Additionally, just because a company goes out of business doesn't mean that all its capital is thrown away. You liquidate factories and equipment etc. It's been pointed out that frozen people are a "liability", but this is dependent on the contract you pay with the company.

If your estate is set up to pay $X to whatever organization is housing you, then it stands to reason that cryonics companies could move frozen bodies around.

The question is not "will my cryonics company go bankrupt", but whether the entire cryonics industry will cease to exist. That seems pretty unlikely as we become MORE technologically advanced...

Comment by Sieben on The Problem of "Win-More" · 2014-03-28T22:27:29.011Z · LW · GW

The real problem with "win-more" cards is that they're conditional. Being conditionally good is a common criticism of many magic cards. The advice to new players is simple: think about how likely these conditions are to be met. If they can't make this estimate, copy a more experienced player's opinion.

It's also possible that a "win-more" card is also just a decent card even when you're not winning. For example, if you play MTG now, you'll know that 2 desecration demons on the play against G/R monsters is both very good, and that the second demon is "win more".

So I don't really see a problem with playing "win more" cards. The problem is playing too many conditional cards.

Comment by Sieben on What legal ways do people make a profit that produce the largest net loss in utility? · 2014-03-25T17:33:16.307Z · LW · GW

How do you measure arms-race type goods? For example, the world would be better off without any nuclear weapons. However the world is made better off if several reasonable countries can obtain nuclear weapons to serve as deterrents to others using them.

A similar analysis exists for guns.

Comment by Sieben on LINK: In favor of niceness, community, and civilisation · 2014-03-04T18:35:31.923Z · LW · GW

See Yvain's post on Schelling Fences on Slippery Slopes.

This is not a blanket reason to defend all ideologies against censorship. The analysis of many religions also implicitly assumes that there is no cost to tolerating competing religions, whereas there is a definite cost to hearing out many of the worst political ideologies.

It's almost as if the slippery slope works both ways. If you can't filter anything, your energy is drained by a thousand paper cuts.

You do realize that most people have the same opinion about the Singularity?

I wasn't aware that the general public was angry about Singularity nerds. I was talking more about like teenage neo-nazis. Extremely high probability to contribute nothing, piss a bunch of people off, and waste all our time.

Comment by Sieben on LINK: In favor of niceness, community, and civilisation · 2014-02-28T00:31:41.518Z · LW · GW

Therefore it is particularly important that we are able to evaluate all ideas as accurately as we can, and particularly important not to spread lies, etc.

Okay, so, we don't know what the right answer is. But we know what the right answer ISN'T, right? We know that Westboro Baptist Church isn't going to lead the human race into a new golden age. Why not try to limit their influence?

And even if there were some seemingly bad ideas that could, through some twist, actually be good ideas, there are still nonzero costs to considering them. Like if there is a 0.00001% chance it is "the answer", but a 99.99999% chance to waste everyone's time and making some people angry, we should probably discard it. Why waste time when we can pursue that handful of ideas that have a much higher chance of improving the world?

But if the fate of the human race hangs in the balance, then you can afford the luxury of that assumption.

I'm going to assume you meant that you can't afford the luxury of that assumption, and actually yes I can. In fact, I have no choice. I have a finite amount of computational power and if I go through all possible permutations of ideas then the probability of me coming out with The Right Answer becomes vanishingly small. Instead, I can apply some very defensible heuristics to write off huge sections of thought wholesale. I should focus my efforts on ideas that are not obviously wrong.

Comment by Sieben on LINK: In favor of niceness, community, and civilisation · 2014-02-26T15:04:11.644Z · LW · GW

Simple examples of playing dirty:

  • Someone links a URL but it is broken in an obvious way. If you truly interested in arguing for the sake of argument, you could fix the URL and go to their link. But you could also take the opportunity to complain that they are just wasting your time and aren't really serious.

  • Sometimes, there is a finite amount of time or space for your opponents to reply to you in. You can pick arguments whose articulation is economic, but whose rebuttal is not. This puts a huge volumetric burden on them such that they will be unlikely to be able to reply to all your points. Later you can point out that they "ignored many of your best arguments". This is an old debater's trick.

  • You're going to have a live debate online for a public audience. 45 minutes beforehand, you receive an e-mail from your opponent indicating that they are having difficulty connecting to Skype and suggest the debate be moved to Omegle. You can play nice and get the debate to happen, or you can pretend that you didn't see the e-mail in time and then gloat that your opponent didn't show up because of "technical difficulties" har har har.

  • Abuse the last word. If you're in the final stretch of a debate, bring up new issues that your opponent cannot address because they are out of time. This technique is actually heavily penalized in high school debate competitions, but people get away with it regularly because adults are more biased than teenagers.

Comment by Sieben on LINK: In favor of niceness, community, and civilisation · 2014-02-26T14:48:52.691Z · LW · GW

Addressing the most stupid of opposition's arguments is not an enlightened way of discussion, but it's still way better than manufacturing and spreading widely false statistics.

You seem to be confused. Both of the things you mentioned are examples of "playing dirty".

If the other side played equally dirty, we would see articles like: "Did you know that 95% of violent crimes are committed by Social Justice Warriors?" or "Woman is most likely to get raped at the feminist meeting (therefore, ladies, you should avoid those meetings, and preferably try to ban them at your campus)".

But this is a very stupid way to play dirty because it is transparent and can backfire. Making a public example of the other side's inarticulate idiots is extremely unlikely to backfire.

Just leaving a hint: Imagine how a successful support for a false statistics could be used to design an ironic revenge at the very person who supported it.]

Just a hint: If you are using consequentialist arguments against playing dirty, then you are open to playing dirty if you can be shown it works. I submit to you that you have a failure of imagination.

I hope this sufficiently illustrates that the belief that the other side already is fighting as dirty as they can, and you cannot give them ideas by fighting dirty yourself, is completely false.

Strategic mimicry is not one of my arguments. You seem to be arguing with someone else. Regardless, see the "consequentialist" point above.

Comment by Sieben on LINK: In favor of niceness, community, and civilisation · 2014-02-26T00:13:53.629Z · LW · GW

Just think about how much more persuasive fighting dirty sounds if the whole fate of the human race hangs in the balance. As is, there is an underlying assumption that we have infinite time to grind down our opposition with passive logical superiority.

Comment by Sieben on LINK: In favor of niceness, community, and civilisation · 2014-02-26T00:12:15.807Z · LW · GW

I don't understand all the consequentialist arguments against playing dirty. If your only objections are practical, then you're open to subtle dirty maneuvers that have very high payoffs.

A really simple example of this would be to ignore articulate opponents and spend most of your energy publicly destroying the opposition's overzealous lowest-common-denominators. This is actually how most of politics works...

... and also how this conversation seems to be working, since the Scott Alexander side seems more intent on arguing through hyperbole than addressing the actual spirit of what is being suggested. A simple example could be to deliberately misinterpret what the other side is saying, and force them to clarify themselves ad nauseam until they run out of energy, "conceding" the issue by default.

AND JUST LIKE THAT WE'RE BACK TO THE IRONY

Comment by Sieben on Don't rely on the system to guarantee you life satisfaction · 2014-02-19T17:47:37.285Z · LW · GW

so it should be emphasised that this is not a Fully General Counterargument against learning anything.

I think it's okay to tell kids that if you're incompetent you'll still do fine in life, because it's true. The function of telling them that the material doesn't matter could be to reduce their anxiety over the obvious tension between valuing coursework and real world pragmatism.

Another takeaway from the argument could be that adults are generally pretty incompetent. Most people don't use math or calculus, but in my experience this hurts them quite a bit. If they have a bunch of data they won't be able to model it as well. If they have to design an experiment they won't even think to go look at the constitutive equations. All of this cross-applies to evaluating other peoples' work... Like professional engineers will frequently look at academic literature but only be able to read the abstract/conclusions section. They aren't punished for it per se, but they aren't exactly rewarded for it either. So you could go the other way and say: "look, if you want to be in the top 0.1% of achievers, learn this stuff because it actually is applicable if you go out of your way to apply it, and it will be awesome".

I also just don't have any sympathy for the people who are hurt by informing them about their biases :)

And if you tell them to listen to smart and successful people, you might be preparing them to fall for the next MLM scam.

Above I told them to google or wiki stuff. Just start doing it. There are so many things the average person believes that are just false upon reading the wiki article. For example, eating breakfast is the most important meal of the day because you've heard it in Kellog's commercials since you were 5. Except maybe that's not true and you should look deeper into the issue since you'll be eating every day for the rest of your life. Similarly it boogles my mind that parents can't summarize expert opinion on corporal punishment for children. Your children are insanely important, and you can't even spend 5 minutes reading wikipedia to double check what you already "know"?

Regardless of how they get their information though, it would be good for them to get in the habit of discussing their ideas with strangers (on the internet) to serve as kind of a check on them getting too out of whack. You don't necessarily have to convince other people, but they shouldn't be able to destroy your belief by just linking a wiki article :P

Maybe just... give them your e-mail and tell them to feel free to ask anything

I think we should emancipate 15+ year olds.

Comment by Sieben on Don't rely on the system to guarantee you life satisfaction · 2014-02-19T00:46:20.497Z · LW · GW

The points made seem very vague and trivially true. For example: "Consider homeschooling and online school. Depending on your situation, these may be superior alternatives to regular high school for you." Yes, of course, for the "right situation" homeschooling is good. Kind of like how you shouldn't drink "too much" water or breath "too much" oxygen.

Try giving specific examples of the way the system is bad for them. For example, point out that most adults can't actually do algebra or calculus, and that these skills are almost never used even by professionals who CAN do them, such as engineers, on a regular basis.

However it is pretty easy to get students to admit that at least some of the classes they take are a waste of time. You could try getting them more concerned about having hundreds of hours of their lives wasted following instructions from an unambitious middle aged authority figure. Do we really want children learning from or respecting people who chose one of the easiest careers possible and conceded to be lower-middle class for the rest of their lives? Pfft. Although this point will probably receive a lot of pushback because "education" is a sacred cow and teachers should be worshipped because a few of them claim to have a "passion for teaching".

Instead of focusing on why high school is bad though, maybe you could get more mileage out of telling teenagers what they SHOULD be doing with their time. Most kids don't work out or eat properly because their parents don't have those values. Maybe since teenagers should practice being less biased, because certainly they don't get any sort of pressure to be rational. Suggesting that they get in the habit of googling the expert consensus on issues before they make up their mind COUGH politics COUGH is a step in the right direction.

Or maybe direct them to radical cultural outlets like RSDNation so they can get some perspective on the monogamous suburban middle class life they're planning on living.

Once you figure out what you want to do, high school and college just become tools that are very easy to optimize for your goals. But most high school kids are just going through the motions, passively expect to get , passively expect to marry , have kids when they're 27-30, etc. This is how a lot of people wind up unhappy. If I could tell a teenager anything, it would be for them to become 10,000x more strategic and independent.

... of course convincing someone to ACTUALLY become these things is very difficult.

Comment by Sieben on Googling is the first step. Consider adding scholarly searches to your arsenal. · 2013-05-08T17:05:18.095Z · LW · GW

Yes, except that I am the only expert on what music I like.

Oh, so you agree there are can be good reasons to discount the "expert" establishment, no matter how much "peer review" or citations they have.

Are we talking about degrees here. I am pretty sure Ive been talking about top level articles. Or can anyone with an IQ above 110 publish one of those?

Yes. But getting a degree is normally a prereq for publishing, and everyone who gets a degree publishes something. And yes, you can publish in the "top" journal articles in grad school.

No winning out here. The research will be closer to the truth than a random answer because the accuracy of the theories gets compared to reality buy doing experiments for example. Or because not every single person is completely biased and blind to the results that they get.

Not every single person has to be biased. Just enough of them.

Hey, that's why they are correlations. I am not stopping you from believing that being predisposed to diabetes and cancer or whatever makes you more likely to eat red meat for example.

But the researchers conclude that red meat increases your risk of heart disease simply because it is associated with heart disease. That is dishonest. If they can get away with blatantly unsubstantiated statements like that in epidemiological papers, what can't they get away with buried in their SAS databases and algorithms?

Comment by Sieben on Googling is the first step. Consider adding scholarly searches to your arsenal. · 2013-05-08T17:01:36.993Z · LW · GW

Doctors not researchers in the top peer-reviewed papers.....

Researchers who got there because other researchers said they were good. It's circular logic.

Haven't been interested at all in the subject and have never looked into it. And anyway if you are right and they are completely fake and wrong, this would not be general evidence that papers are always as good as coin flips.

It's prima facie evidence. That's all I hoped for. I haven't actually done a SRS of journals by topic and figured out which ones are really BS. But of the subjects I do know about, almost all of the literature in "top peer reviewed" papers is garbage. This includes my own technical field of engineering/simulation.

I am leaving this conversation. If you really believe that the most-cited, accepted, recent articles etc. are as accurate as a coin flip because people have biases and because the statistics are not perfect and if nothing that I've said so far has convinced you otherwise then there is no point in continuing.

Straw man. I did not say the statistics were not "perfect". And I did not say they were "as accurate as a coin flip". In the red meat example, they are worse.

Also, not to be rude, but I do not see why you would join LessWrong if you think like that. A lot of the material covered here and a lot of the community's views are based on accepted research.

A lot of LW is analytical.

The rest is based on less accepted research. Either way, the belief that research (especially well peer-reviewed research) brings you closer to the truth than coin flips on average is really ingrained in the community.

Research is a good starting point to discover the dynamics of a certain issue. It doesn't mean my final opinion depends on it.

Comment by Sieben on Googling is the first step. Consider adding scholarly searches to your arsenal. · 2013-05-08T14:50:37.051Z · LW · GW

You do understand that scientist don't just look for correlations but form a bit more complex models than that. Do you seriously think that things like that are not taken into account!?

Yes. You should read the papers. They're garbage.

Remember that study on doctors and how they screwed up the breast cancer Bayesian updating question? Only 15% of them got it right, which is actually surprisingly high.

Okay now how much statistical training do you think people in public health, a department that is a total joke at most universities, have? Because I know how much statistical training the geostatistitians have at UT and they're brain damaged. They can sure work a software package though...

Hell, I am willing to bet that a bunch of the studies test those correlations by comparing for example smokers who eat more red meat versus smokers who eat less/none read meat.

"A bunch of" ~= the majority. I'm sure there could be a few, but it wouldn't be characteristic. I'm not saying ALL the studies are going to be bad, just that bulk surveys are likely to be garbage.

Maybe I should have chosen "Theologians' opinions on God" rather than "Middle aged/classed suburban nutritionists' opinions on red meat". I thought everyone here would see through frakking EPIDEMIOLOGICAL STUDIES, but I guess not.

Comment by Sieben on Googling is the first step. Consider adding scholarly searches to your arsenal. · 2013-05-08T14:42:45.005Z · LW · GW

Wow. This is a pretty far-fetched claim..

This is a pretty solid argument.

My theory is that respected papers are done in a method more resembling the scientific method than coin flip on average and thus they get more accurate results than a coin flip. There, happy?

Thanks for clarifying. I disagree. See the systematic bias/complexity arguments.

I did answer your question - the answer was yes.

Do you really choose your music based on the average opinion of "experts"? Give me a break. Look, if you could randomly draft 20 people who had demonstrated independent rationality and objectivity, assign them to a problem, and take the majority opinion, I would be fine with that. But that's not what we have at all. Anyone with an IQ above 110 can get any degree they want.

He is biased. So is the guy that went into grad school with anti-vegetarian views. If those guys are not changing their opinion based on the evidence then the chance is smaller (not nil though) that their papers will be highly cited.

Why would the best research win out? Why not the most fashionable research that confirms everyones' worldviews? Why not the research that has the punchier abstract title? Why not the research that was fudged to show more impressive results?

You call studies that find correlations between things fear mongering? Oh my.

They could probably find a correlation between eating red meat and watching action movies, but that's not exactly publishable.

Oh my. Okay, first of all you can die of pretty much anything and pretty much anything has some dangers. Or at least that's what does fear mongering scientists claim. The studies show you some numbers to guide you in how much danger X (in this case red meat) poses to specific individuals.

I mean sure, if you consumed more red meat than was physiologically possible to scarf down without choking, you'd die. But that's not unique to red meat. They're claiming that there is a unique property of red meat which causes all these health problems, so not it doesn't fall under the same category as "pretty much anything can kill you".

And no, they technically don't even show danger. All they do is show correlations. Would you also conclude that wearing XXL t-shirts makes you fat?

Do you have any specific reason to think that those studies are fabricated and that in fact red meat has none of the effects that they claim?

Confounding variables mentioned above. Lack of replication/opposite findings in controlled studies. Testimonies from thousands of people on the paleo diet who have reversed their blood chemistry. Fat doctors/nutritionists, etc.

Furthermore, if I tell you that drinking a large amount of water can kill you and do a study to prove it then am I a fear mongering scientist?

If you try to publish dozens of studies on it in the year 2012, yes you are.

"Hey guys I just did ANOTHER study showing that drinking 82 gallons of water in one sitting will kill you (p<0.05)"

That would be fear mongering, although people probably wouldn't take it seriously.

Comment by Sieben on Googling is the first step. Consider adding scholarly searches to your arsenal. · 2013-05-08T12:39:15.106Z · LW · GW

People who eat red meat tend to:

Smoke

Eat cheeseburgers and drink coke

Exercise less

etc etc...

Do you understand why it's not... entirely honest... to blame red meat? It shows up as a statistical correlate. It can be used to identify people at risk for these conditions, but then researchers make a leap and infer a causal relationship.

It's an ideological punchline they can use to get published. And that's all.

Comment by Sieben on Googling is the first step. Consider adding scholarly searches to your arsenal. · 2013-05-08T12:33:07.764Z · LW · GW

Or that information in studies is more accurate than made-up information?

This is exactly my point. Studies on many many subjects may not contain information more useful than coin flip, let alone an educated guess.

All I am advocating is to look for 'respected' studies and look at them. If you don't think that looking at studies 'approved' by the field gives you more accurate information than not doing it I can't really do much.

This is question begging. You have to have a theory about why a "respected" study is likely to be correct. I've already provided theories explaining why they're likely to be incorrect a large portion of the time.

Yes, I believe in science no matter what scenario I am in. You don't need to blindly trust it or anything, I put different weights on different claims etc. but I would still take into account information from recent, well-cited meta-analyses or whatever I can get my hands on.

I believe in science too. But "science" and "science articles" are different things. But you didn't answer my question, and I really want to drive home that almost no one thinks it's a good idea to trust "majority expert opinion" in all sorts of areas.

So I should worry that researchers are interested in the topic that they are researching. What douchbags, eh?

Don't be dense. You know exactly what I mean. A vegetarian goes to grad school and does research on nutrition. What do you think is going to happen?

Okay. Citation? And remember we are not talking about 'most studies' or anything. The studies that we are talking about are well cited, by known researchers if possible and systematic reviews if possible.

Citations above where you commented. You can also just punch "red meat" into google scholar and it's all about how you can die from it.

Comment by Sieben on Googling is the first step. Consider adding scholarly searches to your arsenal. · 2013-05-08T01:56:04.653Z · LW · GW

Yes, there is A LOT of garbage. This is why I am recommending using heuristics such as numbers of citations - to maximize the accuracy of the information. And, yes, peer review is not perfect but compare journals/fields that rely on peer-review to those that do not...

My argument really boils down to 2 things. Researchers being systematically biased (ex: red meat), and researchers having a very low probability of actually knowing the right answer but publishing something that fits some narrow set of data (ex: "advanced" simulation). To be sure, I've used research to make a lot of informed decisions over my lifetime, but it's always been straightforward, pretty much unanimous, and with lots of testimonials from online groups to give it statistical mass.

This post is not so much about academically controversial issues but even in those cases if you don't have any reasons not to, then siding with the majority will bring you to the truth more often than the alternative.

Would you adopt this heuristic in any other scenario where the "right answer" isn't obvious? Music, books, diet, politics, etc? Even when you restrict your sampling pool to "experts only", the results are still pretty bad. These people are self-selecting to do research. It's not like you're picking a random disinterested intelligent person and asking them to study the problem. No one becomes a nutritionist because they have no opinion on food.

This is the type of thing that you see if you do a normal google search instead of a scholarly search. I have not checked but I bet that the most cited recent review articles on those issues can provide you with some pretty good information.

The overwhelming trend is fear mongering coming out of epidemiological studies.

Comment by Sieben on Googling is the first step. Consider adding scholarly searches to your arsenal. · 2013-05-08T01:44:10.419Z · LW · GW

I punched in "red meat" to google scholar.

http://care.diabetesjournals.org/content/27/9/2108.short 197 citations - concluding that eating red meat "may" increase your risk of type II diabetes.

http://ajcn.nutrition.org/content/82/6/1169.short 173 citations - Shows more "correlations" and "associations" for the "beneficial effect of plant food intake and an adverse effect of meat intake on blood pressure."

Comment by Sieben on Googling is the first step. Consider adding scholarly searches to your arsenal. · 2013-05-07T17:52:46.095Z · LW · GW

As someone who actually does academic research and has spent countless hours reading the fine details of "prestigious" publications, 90% of the material out there is total garbage, and it is difficult to know if a paper is garbage just by reading the abstract. Peer review doesn't help either because review boards are lazy and will never double check any of your actual footwork. They will never read your code, rerun the algorithms you claim you used, etc. A simple glitch can change the sign of your answer, but you typically stop looking for glitches in your 10,000 lines of code once the answer "looks right".

So, if there is any controversy on the issue, remain agnostic. Like a 90/10 split in academia is totally reasonable once you factor in the intellectual biases/laziness/incompetence of researchers and publishers. All an article tells you is that somewhere out there, someone with an IQ >= 110 has an opinion they are publishing to further their career. I don't place very much weight on that.

Don't become one of these "reserch sez..." people who just regurgitate abstracts. You'll wind up worried about the dangers of red meat, getting too much sunlight, doing 45 minutes of cardio every day, etc.