LessWrongWiki: What the hell are they talking about?

[The Good Reverend Roger]

It feels like some of this applies to PD as well.

Hey, thanks.  Thanks a fucking lot.

Sorry we don't live up to your expectations.

No, that's not what I meant. I meant that in every community there are a small number of innovators and a much larger number of adopters. This isn't necessarily a bad thing (see question about requirements of a community). Hell, I'm an adopter, not an innovator, so why would I insult myself by saying that was a bad thing? It doesn't mean we have crapulism.

Okay, my bad.

Here we have a dozen posters and 4 times that many people who sit and stare at the screen, congratulating themselves for being too smart to watch TV.

Different folks for different folks, I guess.

I really don't want to turn this into a thread about 'what's wrong with PD'. But if I had to sum up why I "sit and stare at the screen", it's because you "dozen posters" have a whole lot of Holy^TM, and I'm just sitting here wondering how the hell I could ever collaborate at your level, not wanting to fuck up the process.

Thing is, this ain't a competition.  It's Discordia.  It's for everyone.

And as far as the Holy™ goes, I recently went back and read my 2003/2004 stuff, and cringed.  You learn to Holy™ by doing it.

Or you can just shoot the breeze.

But for God's sake, let's not let this shit turn into television or a blog or something.

[Mesozoic Mister Nigel]

Here's the thing; my Discordia isn't for everybody, but I try to live my Discordia by the same philosophy Mama from Here Comes Honey Boo Boo lives her life; big, loud, kind of nasty, and don't give a fuck.

This is my Discordia: http://www.youtube.com/watch?v=4hXjFLd1RYg

[McGrupp]

As a relative newcomer to PD.com I will say it can be a little intimidating to post. They always warn you that with discordians you won't be the weirdest person in the room but no one ever points out that you won't be the smartest person in the room. A lot of folks here are smart. Like, really smart, I still have to look up words all the time here and I consider myself pretty well read and learned. Also a lot of folks on here have been talking to each other for a long while. Their maps have been traced out, concepts have been agreed upon (or at least defined in terms of conversation) Catching up on all that takes some time and is a lot of work (good awesome work, but effort nonetheless) The old threads in these forums are a library unto themselves in many ways.

That said, after getting over the, whoa you guys are smart and I've never even heard of a reality tunnel before, I find this place to be a great place to discuss ideas. As long as you're honest, can back up an argument, and recognize when you're A)possibly wrong or B)too ignorant on the subject to continue the conversation, then it's a fairly open place. Sure, sometimes eldritch abominations will ambush you, hold you down and shit in your soul, but you'll have that here. Discordia wasn't meant to be safe. It's a place where everything gets poked with a stick.


Back on topic though, I really really dig the lesswrong movement. Especially the statistics stuff as it's easy to be swayed by statistics that sound good or plausible. An example from a book I just read was a court case where a man was on trial for murdering his spouse who he had previously abused. The defense provided a statistic that showed that of all cases of abused women in a marriage, only 2 percent were murdered by their spouse. 2 percent sounds kind of low but this is not actually relevant. The relevant stat isn't the percentage of battered women who went on to be murdered by their spouses but the percentage of battered women Who were also murdered who were killed by their spouses. This percentage is 90 percent and is the exact opposite of what the defense were trying to show.

The classic girl named Florida problem still makes my head hurt though:

The Girl Named Florida problem is usually preceded by an easier problem that goes as follows:

Suppose you know that a family with two children has at least one girl. What is the probability that this family has two girls?

Now for the Girl Named Florida problem:

Suppose you know that a family with two children has at least one girl named Florida. What is the probability that this family has two girls?

[LMNO]

On the face of it, roughly 50% for both cases. It sounds the same as the "coin flip came up heads ten times" problem.

Am I missing something? Wouldn't be the first time.

[Q. G. Pennyworth]

It's 25% when you have no information, then it's 50% because you've already confirmed one.

(25% because you could have M|F, F|M, M|M, or F|F)

[LMNO]

Wait, what?

Each occurrence has no bearing on the previous. Every pregnancy is a coin flip, with no bearing on any other.

Hmm now I see what you might be saying. I gotta think about this.

[LMNO]

Oh, ok. It's not, "what's the chance of the second child being a girl," it's "what's the chance of the total combination"?

In that case, it is the end result. So you HAVE TO UPDATE YOUR PRIORS. Also, figure out what the question is actually asking.

[Kai]

Oh, ok. It's not, "what's the chance of the second child being a girl," it's "what's the chance of the total combination"?

In that case, it is the end result. So you HAVE TO UPDATE YOUR PRIORS. Also, figure out what the question is actually asking.

As in, the order of birth matters? It's not just the combination, but the order?

[LMNO]

No it's about the sum.

If the question is, "what's the chance this kid is a girl?" Then it's 50%.

If the question is, "if the first kid is a girl, what's the chance of two girls?" Then it's 25%.

I know, it doesn't feel right. But it's looking at the narrative, not the circumstance.

[Kai]

No it's about the sum.

If the question is, "what's the chance this kid is a girl?" Then it's 50%.

If the question is, "if the first kid is a girl, what's the chance of two girls?" Then it's 25%.

I know, it doesn't feel right. But it's looking at the narrative, not the circumstance.

That makes sense. I guess the problem wasn't worded very well. I was wondering how I had so much trouble with that and none at all with the battered wife murder example.

[LMNO]

The more I live, the more the case is, "in life, the question is rarely worded very well."

An argument can be made that this is a common downfall of purely rationalist thinkers.

[Q. G. Pennyworth]

No it's about the sum.

If the question is, "what's the chance this kid is a girl?" Then it's 50%.

If the question is, "if the first kid is a girl, what's the chance of two girls?" Then it's 25%.

I know, it doesn't feel right. But it's looking at the narrative, not the circumstance.

If the first kid is a girl, you have a 50% chance of two girls, because you already have one there and it's just a coin flip for the second. If you don't know either gender, you're doing two coin flips and it's 25%

[LMNO]

Damn. I got it wrong again. This shit is hard.

[LMNO]

Ok, so the questions are:
If it's predicting one kid at a time, it's 50%.
If it's predicting two kids at a time, it's 25%.
If you know the gender of one, it's the same as predicting one kid.

I think I'm getting the hang of this.

[Q. G. Pennyworth]

                                           Male -> Male + Male
                                          /
               Male -> Coin Flip
              /                          \
Coin Flip                             Female -> Male + Female
            \                                Male -> Female + Male
             \                              /
              Female -> Coin Flip
                                           \
                                             Female -> Female + Female

I made a chart!