I would turn this around- what core part of Less Wrong is actually novel? The sequences seem to be popularizations of various people’s work. The only thing unique to the site seems to be the eccentricity of its choice in topics/examples (most cog sci people probably don’t think many worlds quantum mechanics is pedagogically useful for teaching rationality).
There also appears to be an unspoken contempt for creating novel work. Lots of conjecture that such-and-such behavior may be signaling, and such-and-such belief is a result of such-and-such bias, with little discussion of how to formalize and test the idea.
There also appears to be an unspoken contempt for creating novel work. Lots of conjecture that such-and-such behavior may be signaling, and such-and-such belief is a result of such-and-such bias, with little discussion of how to formalize and test the idea.
Can you think of any ways to formalize and test this idea?
Of course- here is a simple test. Go through discussion threads classifying them as apparently novel/not novel, and attempts to test ideas/conjecture. Check ratios. Compare the karma of the different categories. Obviously, its not a perfect test, but its good enough to form the belief I shared in my post.
And of course, I didn’t suggest that every idea needs to be formalized or tested- merely that there is not enough present here for Less Wrong to develop from a source of popularization of other’s ideas into a forum for the creation of novel ideas about rationality.
This sort of far-mode thinking is usually [1] evidence of an attempt to signal not-”Straw Vulcan Rationality” while simultaneously earning warm fuzzies in those possible worlds in which [DELETED] (ed. Explaining the reason for this edit would either reveal excessive information about the deleted content or require mentioning of true ideas which are considered abhorrent by mainstream society.) and is ultimately the result of having a brain which evolved to have hypocritical akrasia regarding skepticism and to guess the teacher’s password [2].
[1] p(parent post is mere signalling | p-zombie Mary in a Chinese room would claim that “semantic stop-signs are red” is a map-territory-map-mapitory confusion) = .7863, but I may have performed an Aumann update with a counterfactual-me who generalized from fictional fictional-evidence.
I’m not sure if I should upvote this as an amusing parody of f Less Wrong posts or downvote because it seems to actually undermine the implicit claim that LW posts are full of jargon with little content- it doesn’t take much effort to see that the post really is nonsense. Overall, a mildly amusing but not terribly well-done parody.
Applause lights. You should really read the sequences.
It took me a moment to understand that you were creating a parody. I’m not sure if that moment was indicative of EHeller, in fact, being on to something.
Anyway, on the original comment—yes, there was a little bit of tu quoque involved. How could I not? It was just too deliciously ironic. Even when accusing someone else of failing to formalize and test their ideas, it’s easy to fail at formalizing and testing ideas. It’s not meant (entirely) as a tu quoque—just as a warning that it really is easy to fall for that, that even consciously thinking testability isn’t enough to actually get people to make explicit predictions. So, I decided to spend a few seconds actually trying to dissect the claim, and ask what sort of testable predictions we can derive from the “There also appears to be an unspoken contempt for creating novel work.”*
The obvious signs would be a significant number downvotes on posts that deal with original work, or disparaging statements toward anyone presenting work for the first time on LW. Undue or unreasonable skepticism toward novel claims, perhaps, above and beyond what is warranted by their novelty. I have no idea how to formalize this—and, in fact, the more I look at the statement, the more am convinced it really is vague and untestable. I dunno—anyone else want to take a crack at it? EHeller, do you have something more precise that you were trying to get at?
It was an interesting exercise—even if it turns out to be less meaningful or reducible than I thought, it’s good exercising in noticing so.
*Of course, there are good reasons why one might not want to spend time and effort trying to formalize and test an idea. The statement “Lots of conjecture that such-and-such behavior may be signaling, and such-and-such belief is a result of such-and-such bias, with little discussion of how to formalize and test the idea” isn’t as interesting not only because it’s imprecise, but also because it does in fact take effort and energy to formalize and test an idea—it’s not always worth it to test every idea; the entire point of having general concepts about biases is that you can quickly identify problems without having to spend time and energy trying to do the math by hand.
The obvious signs would be a significant number downvotes on posts that deal with original work, or disparaging statements toward anyone presenting work for the first time on LW. Undue or unreasonable skepticism toward novel claims, perhaps, above and beyond what is warranted by their novelty.
I rather doubt that. It’s probably just you confusing the map for the territory.
Hm. EHeller says “LW has too much conjecture without enough discussion of how to formalize and test it!” JScott points out that EHeller’s assertion lacks any discussion of how to formalize and test it.
I would call that tu quoque more than applause lights.
I would turn this around- what core part of Less Wrong is actually novel? The sequences seem to be popularizations of various people’s work. The only thing unique to the site seems to be the eccentricity of its choice in topics/examples (most cog sci people probably don’t think many worlds quantum mechanics is pedagogically useful for teaching rationality).
There also appears to be an unspoken contempt for creating novel work. Lots of conjecture that such-and-such behavior may be signaling, and such-and-such belief is a result of such-and-such bias, with little discussion of how to formalize and test the idea.
I sometimes think a quote I’ve heard in reference to Wolfram’s “A New Kind of Science”, might apply equally well to the sequences:
Can you think of any ways to formalize and test this idea?
Of course- here is a simple test. Go through discussion threads classifying them as apparently novel/not novel, and attempts to test ideas/conjecture. Check ratios. Compare the karma of the different categories. Obviously, its not a perfect test, but its good enough to form the belief I shared in my post.
And of course, I didn’t suggest that every idea needs to be formalized or tested- merely that there is not enough present here for Less Wrong to develop from a source of popularization of other’s ideas into a forum for the creation of novel ideas about rationality.
Applause lights. You should really read the sequences.
This sort of far-mode thinking is usually [1] evidence of an attempt to signal not-”Straw Vulcan Rationality” while simultaneously earning warm fuzzies in those possible worlds in which [DELETED] (ed. Explaining the reason for this edit would either reveal excessive information about the deleted content or require mentioning of true ideas which are considered abhorrent by mainstream society.) and is ultimately the result of having a brain which evolved to have hypocritical akrasia regarding skepticism and to guess the teacher’s password [2].
[1] p(parent post is mere signalling | p-zombie Mary in a Chinese room would claim that “semantic stop-signs are red” is a map-territory-map-mapitory confusion) = .7863, but I may have performed an Aumann update with a counterfactual-me who generalized from fictional fictional-evidence.
[2] The password is Y355JE0AT15A0GNPHYG.
I think you mean Y355JE0AT15G00NPHYG.
(This comment is a thing of beauty.)
This coupon entitles the bearer to one free internet.
Also, a related comic and essay.
I was attempting to parse that comment seriously at 8AM in the morning before drinking any coffee. Never again.
I’m not sure if I should upvote this as an amusing parody of f Less Wrong posts or downvote because it seems to actually undermine the implicit claim that LW posts are full of jargon with little content- it doesn’t take much effort to see that the post really is nonsense. Overall, a mildly amusing but not terribly well-done parody.
It took me a moment to understand that you were creating a parody. I’m not sure if that moment was indicative of EHeller, in fact, being on to something.
Anyway, on the original comment—yes, there was a little bit of tu quoque involved. How could I not? It was just too deliciously ironic. Even when accusing someone else of failing to formalize and test their ideas, it’s easy to fail at formalizing and testing ideas. It’s not meant (entirely) as a tu quoque—just as a warning that it really is easy to fall for that, that even consciously thinking testability isn’t enough to actually get people to make explicit predictions. So, I decided to spend a few seconds actually trying to dissect the claim, and ask what sort of testable predictions we can derive from the “There also appears to be an unspoken contempt for creating novel work.”*
The obvious signs would be a significant number downvotes on posts that deal with original work, or disparaging statements toward anyone presenting work for the first time on LW. Undue or unreasonable skepticism toward novel claims, perhaps, above and beyond what is warranted by their novelty. I have no idea how to formalize this—and, in fact, the more I look at the statement, the more am convinced it really is vague and untestable. I dunno—anyone else want to take a crack at it? EHeller, do you have something more precise that you were trying to get at?
It was an interesting exercise—even if it turns out to be less meaningful or reducible than I thought, it’s good exercising in noticing so.
*Of course, there are good reasons why one might not want to spend time and effort trying to formalize and test an idea. The statement “Lots of conjecture that such-and-such behavior may be signaling, and such-and-such belief is a result of such-and-such bias, with little discussion of how to formalize and test the idea” isn’t as interesting not only because it’s imprecise, but also because it does in fact take effort and energy to formalize and test an idea—it’s not always worth it to test every idea; the entire point of having general concepts about biases is that you can quickly identify problems without having to spend time and energy trying to do the math by hand.
I rather doubt that. It’s probably just you confusing the map for the territory.
Okay, okay, I’m done. >>
Hm.
EHeller says “LW has too much conjecture without enough discussion of how to formalize and test it!”
JScott points out that EHeller’s assertion lacks any discussion of how to formalize and test it.
I would call that tu quoque more than applause lights.
That was a joke. ;p
Oh.
Oops.