WRL62

[Higus]

I'll give you >= 0.5%. In?

let me think about it. I really have not been putting a lot of attention into the polls to start placing bets on them.

Elections are over a year away.
Ugh, it feels bad to just think about that.

 

[JDragon]

He didn't say that you were stupid. He said that you were less intelligent than Trump.

 

[Andy Capp]

I think you're right on the confirmation bias thing for the average person unaware of it. However, I think that people are capable for minding that bias and correcting for it. For example---this might be a bit of overkill but just for sake of argument---a careful observer of the NYT headlines I mentioned could assign bias scores weekly for the entire four-month period in question in order to address the potential for an increase in confirmation bias through time.

One might be capable, but one is unlikely to consistently apply that ability, which, I suspect, was Jack's point, [edit] or rather that one should not assume it was consistently applied without evidence, such as you have failed to provide ITT.

 

[Tycho- Taylor's Version]

Interesting that you read Yudkowsky. He and Robin Hanson were the original two writers at Overcoming Bias, the blog that @Jack V Savage and I were discussing the other day.

I don't follow him too closely. I followed a guy whose podcast I used to listen to back in the day over to LessWrong and spent a brief period there. A little hyper-rational for my tastes, but often fun and with some good reading suggestions.

 

[Madmick]

I know. Some people think he's a genius.

To me, they seem as deluded as those who think he isn't capable of any calculation or insight at all.

 

[Jack V Savage]

I think you're right on the confirmation bias thing for the average person unaware of it. However, I think that people are capable for minding that bias and correcting for it. For example---this might be a bit of overkill but just for sake of argument---a careful observer of the NYT headlines I mentioned could assign bias scores weekly for the entire four-month period in question in order to address the potential for an increase in confirmation bias through time.

First, you're going to run into problems scoring the bias for any individual NYT headline. But second, such an approach would probably lead to better results than relying on memory. Have you, in fact, taken that approach?

 

[Higus]

Edit:
and another,
https://www.studentnewsdaily.com/example-of-media-bias/why-headlines-matter-2/
Fortunately, there is good news,
Younger Americans are better than older Americans at telling factual news statements from opinions

I think younger people, having been raised online, are generally more aware that someone is constantly trying to sell them something in digital space.

 

[waiguoren]

One might be capable, but one is unlikely to consistently apply that ability, which, I suspect, was Jack's point.

I think that's probably true, but depending on how unlikely, we still expect a core group of people who do consistently apply that ability.

Either way, we're talking about whether or not the NYT headlines have been biased against Sanders, an empirical question for which the data set is readily available---it doesn't really matter if most people's impression is correct.

@Jack V Savage To show that "the mission" of the channel really isn't important here, consider The Hill's politics show on Youtube. No one would argue that The Hill's mission is to promote Sanders, but there is no doubt in the mind of anyone familiar with that show that it has a pro-Sanders bias in titles of its programs (and in the content of its shows, but that's not the topic here). People don't need to know an outlet's "mission" to detect bias correctly.

 

[JDragon]

I know. Some people think he's a genius.

To me, they seem as deluded as those who think he isn't capable of any calculation or insight at all.

Not sure if I can respond without danger of getting carded.

 

[Madmick]

Not sure if I can respond without danger of getting carded.

Life must be hard living in such an unsafe space.

 

[waiguoren]

First, you're going to run into problems scoring the bias for any individual NYT headline.

Yes, we already touched on the measurement issue. Your best argument is probably down that route, actually. You could posit that the individual's biases would affect the scoring. I would agree, but I think there are ways around that (e.g., have a larger group of politically aware people score the headlines).

But second, such an approach would probably lead to better results than relying on memory. Have you, in fact, taken that approach?

No.

 

[Jack V Savage]

I don't follow him too closely. I followed a guy whose podcast I used to listen to back in the day over to LessWrong and spent a brief period there. A little hyper-rational for my tastes, but often fun and with some good reading suggestions.

I've read a book of his, and I've read through LessWrong (even cited some things from it here, and I've talked to my daughter about a lot of it), but I think he's also a little nutty and wrong about his primary interest (AI in general, but also the AI threat).

 

[Andy Capp]

I think that's probably true, but depending on how unlikely, we still expect a core group of people who do consistently apply that ability.

Either way, we're talking about whether or not the NYT headlines have been biased against Sanders, an empirical question for which the data set is readily available---it doesn't really matter if most people's impression is correct.

@Jack V Savage To show that "the mission" of the channel really isn't important here, consider The Hill's politics show on Youtube. No one would argue that The Hill's mission is to promote Sanders, but there is no doubt in the mind of anyone familiar with that show that it has a pro-Sanders bias in titles of its programs (and in the content of its shows, but that's not the topic here). People don't need to know an outlet's "mission" to detect bias correctly.

Of course it matters when the point is you made a claim and have not provided any evidence.

 

[JDragon]

Life must be hard living in such an unsafe space.

The fact of the matter remains that I am convinced that Trump is not a very intelligent man, and I believe this to be very obvious. Therefore, I both believe that @Trotsky 's retort was like for like (as the relative claim becomes an absolute one) and that his claim - that his counterpart was not able to detect intelligence properly - is not only evidently correct, but also necessarily must be applied to anyone disagreeing or sanctioning this assessment.

 

[JDragon]

Double

 

[Jack V Savage]

I think that's probably true, but depending on how unlikely, we still expect a core group of people who do consistently apply that ability.

Either way, we're talking about whether or not the NYT headlines have been biased against Sanders, an empirical question for which the data set is readily available---it doesn't really matter if most people's impression is correct.

@Jack V Savage To show that "the mission" of the channel really isn't important here, consider The Hill's politics show on Youtube. No one would argue that The Hill's mission is to promote Sanders, but there is no doubt in the mind of anyone familiar with that show that it has a pro-Sanders bias in titles of its programs (and in the content of its shows, but that's not the topic here). People don't need to know an outlet's "mission" to detect bias correctly.

The mission of the NYT is not related to advancing any particular candidate, and it executes its mission at least roughly as well as any similar organization. That matters. I don't really watch anything regularly on YT, and I'm not sure if I've ever seen the show you're referring to. Your claim seems surprising given what I know about the publication, but I'm not in a position to address it, except in the general way that identifying bias in someone else is always complicated because you must first correctly assess your own bias and then record observations in a way that doesn't introduce more bias (as relying on memory would).

Yes, we already touched on the measurement issue. Your best argument is probably down that route, actually. You could posit that the individual's biases would affect the scoring. I would agree, but I think there are ways around that (e.g., have a larger group of politically aware people score the headlines).

No.

That would get around the individual's bias. Your approach, if well-executed, would approximate the processes that produced the headlines in the first place (highly informed people trained to root out political bias--excluding bothsidesist bias--working together to prevent it).

 

[Sketch]

Yeah that's pretty fucking thin. Bad call, blue.

My advice is to report everything that is a derail or an insult not directly related to the topic. If the standard is being used by mods and the dogshit right-wingers to play a reporting game, then blow out the game.

If you listen to this guy ^ it will be a very bad idea. Think of someone from the Simpson’s who already tried this.

 

[Tycho- Taylor's Version]

I've read a book of his, and I've read through LessWrong (even cited some things from it here, and I've talked to my daughter about a lot of it), but I think he's also a little nutty and wrong about his primary interest (AI in general, but also the AI threat).

Which book? I grabbed Inadequate Equilibrium or whatever when he was giving it away, but I didn't make it very far. That unedited prose (see also: N.N.Taleb) is a chore to navigate.

I have some issues with his AI stuff too, but I'm a little green on the subject still. Superintelligence was more accessible but still felt a bit off somehow. I never thought I'd say this, but I was actually hoping Sam Harris would put out a book on the topic (following up his TED Talk) because I need the ideas formulated a little more clearly to work with them.

 

[Jack V Savage]

Which book? I grabbed Inadequate Equilibrium or whatever when he was giving it away, but I didn't make it very far. That unedited prose (see also: N.N.Taleb) is a chore to navigate.

I have some issues with his AI stuff too, but I'm a little green on the subject still. Superintelligence was more accessible but still felt a little off somehow. I never thought I'd say this, but I was actually hoping Sam Harris would put out a book on the topic (following up his TED Talk) because I need the ideas formulated a little more clearly to work with them.

Inadequate Equilibria is the one I've read. I didn't have a problem with the prose, but maybe that's because it was so short. NNT is the most punchable writer working today, I think.

Haven't seen Harris' TED Talk, but AI doesn't really interest me. Keeps coming up in stuff I do read, though.

 

[Tycho- Taylor's Version]

Inadequate Equilibria is the one I've read. I didn't have a problem with the prose, but maybe that's because it was so short. NNT is the most punchable writer working today, I think.

Haven't seen Harris' TED Talk, but AI doesn't really interest me. Keeps coming up in stuff I do read, though.

Darn, I was going to link you to my AI thread lol. I could see the topic overlapping with general economics and/or risk management types of discussions.