I experiment with many things, as do those around me. Some of this is randomization and explicit records, more is just trying different things, muttering ‘VOI‘ and repeating what felt good.
I have a feeling that much of humanity are born satisficers, and for them, experimenting doesn't hold much reward. Experimentation (the desire for new and varied experiences) is the driver of capitalism, so capitalism will in turn try to pressure us into experimentation. What is an ad, if not an encouragement to experiment, to take a chance on something unfamiliar which you might like? No expense is spared at connecting a readiness to experiment with high status. By historical standards, we have a hyper-experimentalist society. I would guess that we are the fist human society to know pangs of regret/guilt when we do something "ordinary", because we suspect that something more optimal could have been done instead.
"Another point is that information transmission is costly, attention is scarce, and dishonesty is incentivized, so even if something actually works and was known in ancient times, there's no guarantee that you've heard of it or consider it credible."
I think a lot it is this. There (at least historically) hasn't been a good structure for reliably and credibly transmitting that kind of information all the way to the average modern person in a way that individuals could recoup the cost of both discovering the information and transmitting it.
I read recently that the idea of physical comfort is a relatively recent thing - at least in English, that usage of the word in writing started sometime in the 1700s, in descriptions of cottage houses. (From Luxury in the 18th Century by Palgrave MacMillan, if I remember correctly.) I also think I remember hearing that the idea of adapting work environments to people instead of expecting people to adapt to the equipment (ergonomics) is even more recent. If the things people value in everyday life really have been that different for most of history, that would be part of the reason why there is still low-hanging fruit.
In some domains, differences between individuals might also be part of it. Especially effects of food on health - there seems to be a lot of variety in how diet affects people. (Though I may be basing this on atypical observations, since bizarre food intolerances run in my family.)
One might discover good things by experimentation if you are not a normal person, and can't cheaply sample from a group of similar peers. Or if the environment is changing really, really rapidly. Or if powerful others are attempting to manipulate you. Or if you are a very smart person. Otherwise, tradition is a powerful and important friend.
Some people might find it useful to have a wider comment area and greater depth of replies. But this is just a superficial thing and many websites only allow a few levels of comments.
I do think you are correct, at least in many domains. The solution to the puzzle is to deny the assumption that generates it ("if such and such is so useful, it would have been discovered and spread already"). Man is indeed an imitating animal, so innovations spread. But I see no convincing evidence that in general the mechanism (if it can be described as such) sorts good from bad, useful from stupid innovations. To the contrary, innovations more often than not can be thought to spread via an information cascade. People will adopt or believe anything if it gets a critical mass of followers. This is especially true if the subject matter doesn't belong to an established authority of experts. The disposition to trust authority is probably its twin; both are cognitive biases that probably have evolutionary origins. Your desire to experiment is an improvement over our atavistic tendencies (and yes, it even manifests itself in intelligent men) to pooh pooh the experimental. Experimentation can benefit yourself and society in a land of homo sapiens, if not the fantasy land of homo economicus.
But there are exceptional areas where Hanson's thesis is correct, namely those where a mechanism clearly exists to weed out crappy innovations. To take two arbitrary examples: venture capital (spending money on a biz plan with no cash flow is a terrible investment) and programming conventions in open source software (they reflect the optimum). In both cases innovations are vetted through a process where bad ones cannot be silently absorbed. Energy wasted (programming time, investor capital) is not easily substituted from other domains. In such domains, conservatism wins.
> Not sure how to reply to gwern below ... is there a limit to depth of replies?
Yes. It's a bizarre misfeature of OB, and constant technical losing like that is a major reason most former OB commenters migrated to LW - why put up with a crappy platform when OB is now just the Robin Hanson show?
> But why is there are no attempt to even acknowledge that some knowledge might be a good thing for young people to have?
Because what would they tell you? Again, Near/Far: anything that might be empirically useful like Game or The Rules would come off as awful and cynical and misogynist and/or misandrist.
Not sure how to reply to gwern below ... is there a limit to depth of replies?
In any case ... yes, "love education" might not be realistic in practice. But why is there are no attempt to even acknowledge that some knowledge might be a good thing for young people to have?
Well, to use your sex education example - what makes you think any 'love education' would be useful and not turn out to be Far and equivalent to abstinence-only sex ed? Love is about as Far a topic as you can get....
Thank you, that first link was the post I remembered. My take is a little different: not so much about an old person passing on lessons (although that would be great!), but an overall lack of societal interest in having young people know.
Schools teach sex education, right? Why don't they teach "love education"? I think there would be opposition to that, perhaps for some of the reasons Robin mentioned in that post.
could we get all of these weirdos in one place, though? then at least we could benefit each other
"why hasn’t it been worth it for others in the past to collect it and put it where I can see it?"
I have been troubled by this, too. It seems to me self-evident that there should be a wiki for "soft" data, anecdotes, stories... why isn't there?
I have a feeling that much of humanity are born satisficers, and for them, experimenting doesn't hold much reward. Experimentation (the desire for new and varied experiences) is the driver of capitalism, so capitalism will in turn try to pressure us into experimentation. What is an ad, if not an encouragement to experiment, to take a chance on something unfamiliar which you might like? No expense is spared at connecting a readiness to experiment with high status. By historical standards, we have a hyper-experimentalist society. I would guess that we are the fist human society to know pangs of regret/guilt when we do something "ordinary", because we suspect that something more optimal could have been done instead.
"Another point is that information transmission is costly, attention is scarce, and dishonesty is incentivized, so even if something actually works and was known in ancient times, there's no guarantee that you've heard of it or consider it credible."
I think a lot it is this. There (at least historically) hasn't been a good structure for reliably and credibly transmitting that kind of information all the way to the average modern person in a way that individuals could recoup the cost of both discovering the information and transmitting it.
I read recently that the idea of physical comfort is a relatively recent thing - at least in English, that usage of the word in writing started sometime in the 1700s, in descriptions of cottage houses. (From Luxury in the 18th Century by Palgrave MacMillan, if I remember correctly.) I also think I remember hearing that the idea of adapting work environments to people instead of expecting people to adapt to the equipment (ergonomics) is even more recent. If the things people value in everyday life really have been that different for most of history, that would be part of the reason why there is still low-hanging fruit.
In some domains, differences between individuals might also be part of it. Especially effects of food on health - there seems to be a lot of variety in how diet affects people. (Though I may be basing this on atypical observations, since bizarre food intolerances run in my family.)
One might discover good things by experimentation if you are not a normal person, and can't cheaply sample from a group of similar peers. Or if the environment is changing really, really rapidly. Or if powerful others are attempting to manipulate you. Or if you are a very smart person. Otherwise, tradition is a powerful and important friend.
Let me try again. People are afraid of having to talk about things like this:http://jobcreationplan.blog...
(Read at your own risk)
http://tvtropes.org/pmwiki/...
More specifically, they would be afraid of having to talk about this: http://occupywallst.org/for...
Some people might find it useful to have a wider comment area and greater depth of replies. But this is just a superficial thing and many websites only allow a few levels of comments.
I do think you are correct, at least in many domains. The solution to the puzzle is to deny the assumption that generates it ("if such and such is so useful, it would have been discovered and spread already"). Man is indeed an imitating animal, so innovations spread. But I see no convincing evidence that in general the mechanism (if it can be described as such) sorts good from bad, useful from stupid innovations. To the contrary, innovations more often than not can be thought to spread via an information cascade. People will adopt or believe anything if it gets a critical mass of followers. This is especially true if the subject matter doesn't belong to an established authority of experts. The disposition to trust authority is probably its twin; both are cognitive biases that probably have evolutionary origins. Your desire to experiment is an improvement over our atavistic tendencies (and yes, it even manifests itself in intelligent men) to pooh pooh the experimental. Experimentation can benefit yourself and society in a land of homo sapiens, if not the fantasy land of homo economicus.
But there are exceptional areas where Hanson's thesis is correct, namely those where a mechanism clearly exists to weed out crappy innovations. To take two arbitrary examples: venture capital (spending money on a biz plan with no cash flow is a terrible investment) and programming conventions in open source software (they reflect the optimum). In both cases innovations are vetted through a process where bad ones cannot be silently absorbed. Energy wasted (programming time, investor capital) is not easily substituted from other domains. In such domains, conservatism wins.
Gwern, no better platform is available to me. OB is supported by the same group who made the LW platform.
> Not sure how to reply to gwern below ... is there a limit to depth of replies?
Yes. It's a bizarre misfeature of OB, and constant technical losing like that is a major reason most former OB commenters migrated to LW - why put up with a crappy platform when OB is now just the Robin Hanson show?
> But why is there are no attempt to even acknowledge that some knowledge might be a good thing for young people to have?
Because what would they tell you? Again, Near/Far: anything that might be empirically useful like Game or The Rules would come off as awful and cynical and misogynist and/or misandrist.
Not sure how to reply to gwern below ... is there a limit to depth of replies?
In any case ... yes, "love education" might not be realistic in practice. But why is there are no attempt to even acknowledge that some knowledge might be a good thing for young people to have?
Well, to use your sex education example - what makes you think any 'love education' would be useful and not turn out to be Far and equivalent to abstinence-only sex ed? Love is about as Far a topic as you can get....
Thank you, that first link was the post I remembered. My take is a little different: not so much about an old person passing on lessons (although that would be great!), but an overall lack of societal interest in having young people know.
Schools teach sex education, right? Why don't they teach "love education"? I think there would be opposition to that, perhaps for some of the reasons Robin mentioned in that post.
If you have never made any experiments yourself how are you supposed to evaluate the experiments that other people do?
I have nothing to say except: vegemite and marmalade on toast.