SGU Episode 392: Difference between revisions

From SGUTranscripts
Jump to navigation Jump to search
No edit summary
No edit summary
Line 291: Line 291:
=== Predicting Murders <small>(7:36)</small>===
=== Predicting Murders <small>(7:36)</small>===
* [http://www.dailymail.co.uk/news/article-2260609/Maryland-Pennsylvania-using-Minority-Report-style-computers-predict-crimes-Washington-D-C-next.html Minority Report has arrived: Maryland and Pennsylvania using computers to predict future crimes]
* [http://www.dailymail.co.uk/news/article-2260609/Maryland-Pennsylvania-using-Minority-Report-style-computers-predict-crimes-Washington-D-C-next.html Minority Report has arrived: Maryland and Pennsylvania using computers to predict future crimes]
{{transcribing
 
|transcriber = banjopine
S:  But Jay, you're gonna tell us about predicting murders.  But not just predicting murders, predicting murders like in the movie ''Minority Report.''  Even though it has absolutely nothing to do with that, but go ahead.
}}
 
J:  So, yes. Richard Berk, who is a criminologist at the University of Pennsylvania has developed software that's making predictions about future criminal behavior.  Up until now, the way that this was typically done was that parole officers would have the record of that criminal or whoever was committing the crime.  They'd figure out, okay, how dangerous is this person, how many times have they committed crimes, and all that.  And that person on their own with seemingly not much of a standard would say, okay, I think that they need about this much supervision and they give them, it's like a halfway house type of deal where they're like, okay, you need to live here or you need to come in and check with me every week or every month or whatever.  That variable, though, doesn't really help much.  And it's not effective.  Or so Richard Berk thinks.  So what they're doing now is he has taken 60,000 crimes and he did very smart research on them and he was looking into the patterns that exist.  What Professor Richard Burke found was that a huge factor in somebody's predictability of future crimes isn't a crime that they committed when they were 30 but it was the crime or crimes that they committed when they were in their formative years, when they were young adult, teenagers.  That's a really significant predictor of what they're gonna do for the rest of their life.  So, if you're a 30-year-old, the last ten years, fifteen years, according to Berk, are irrelevant.  It's much more important to take a deep dive into the person's upbringing and all the things that they did when they were young.  Which I found to be very interesting.  It speaks a lot about how, you know, I don't want to get into the whole nature versus nurture discussion, but you think if you provide a good household for your children and you steer them in the right direction you could be saving them from criminal life of the future, right?  Two states in the United States have adopted this software already.  Baltimore and Philadelphia are now using it, and D.C.'s soon to pick it up and start using it as well.  He said that his software is replacing the quote unquote ad hoc decision-making and should identify eight out of one hundred murders. 
 
S:  That's not a lot.
 
J:  No.  But, that's eight more people than what would be predicted by a parole officer.  The thing, I'll go back to what you said, though, Steve, one huge thing about this article is how freaking lame the Daily Mail is.  Not that I have any expectations from these guys, but, I mean, it's so crystal clear that they're using a famous movie that everybody know, throwing in the title of their article just to get people to read it, and it worked on me.
 
E:  Worked on me, too.
 
J:  Yeah, but this is the media using Hollywood to support their readership and show interest.  I found this to be a very interesting article and probably would have read something more clever anyway.  It's just laziness on the author in my opinion. 
 
S:  Yeah, and there's five pictures from the movie ''Minority Report'' in this article. 
 
B:  Oh, my god.
 
S:  Seriously, it's in the title, it's all over the article, all these pictures, you know.  It has nothing to do . . .  in the movie, which would be a good one for us to review by the way,
 
E:  Absolutely.
 
S:  Very good portrayal of near future technology.  But anyway, there were three quote unquote precogs who literally predicted a specific person committing a specific murder at a specific time and place, has absolutely nothing to do with what they're talking about here, which is statistically predicting who is likely to commit a murder again, based on some computer algorithm, taking into consideration evidence-based factors.  Two totally different things. 
 
J:  Yup.
 
S:  You're right. 
 
J:  Lame!
 
S:  Very typical, you know.  I agree, the lazy science reporters always have to find some movie angle that somehow relates to whatever technology or breakthrough or thing they're talking about.  Right, so right, Bob, so every time there's any metamaterial breakthrough, it's
 
B:  Uh!
 
S:  a Harry Potter invisibility cloak. 
 
B:  That's their go-to analogy.  The comparison.  They can't not say it. 
 
E:  Do they feel that the audience isn't sophisticated enough to understand what they might be talking about otherwise unless they have some sort of
 
S:  No!  It's just cheap headlines.
 
B:  No, they want a hook.  It's all about being a hook.  They want to hook you in.  Something to make you read it.  No matter what it takes.
 
J:  Sure.  Evan, think about it, they spend $100 million or whatever to make Harry Potter and everybody knows about it and most people enjoy it and like it and they're just throwing it out there, like
 
R:  I think it makes people more likely to read it and less likely to understand it.
 
S:  Right.
 
J:  Yeah.
 
S:  Right.
 
E:  Yes.
 
S:  So what's the point?
 
E:  The opposite effect, yes.
 
S:  Interestingly, the one movie analogy to technology that I think is actually legitimate that I've never seen a news reporter actually make is transparent aluminum. 
 
J:  Right.
 
E:  That's right.
 
S:  There have been several companies or whatever, researchers that have come out with some form of something that's based on some aluminum alloy that's transparent and hard—it ''is'' freakin' transparent aluminum from the ''Star Trek'' movies and they never make that analogy.  I don't understand that.
 
R:  I guess 'cause nobody saw those movies.
 
S:  I guess not.
 
J:  Yes, we did.  ''(laughter)''
 
B:  ''The Voyage Home,'' that's like, probably like, hugely popular. 
 
R:  Was that the one with the whales?
 
S:  Yup.
 
E:  Yes.
 
R:  Yeah, I like that, that was my favorite.
 
J:  There you go, all right.
 
E:  There was a lot to like in that movie.
 
B:  G-e-e-ek ''(laughter)''
 
R:  Oh my god, I'm such a geek.
 
S:  The other thing that struck me about
 
E:  Wasn't directed by William Shatner.  That's a great thing to like about that movie.
 
S:  The other thing that struck me about this article, and it wasn't mentioned in the article—this is the kind of thing, kind of analogy they should have made—is this is saber metrics, right?  This is ''Moneyball.''  You guys see the movie ''Moneyball''?
 
E:  Yes.  Great movie
 
S:  That would have been the better movie to tie this to.  So this is the same thing.  Instead of a Gestalt gut feeling using parole officers and judges, whatever, to say, yeah, is this guy likely to be a criminal again or not, they use an algorithm based on evidence of predictive factors.  That's the perfect analogy to ''Moneyball'' where instead of using scouts and coaches and owners to see who's the really talented player out there, they use actual statistics to predict their effect on the chance of winning games. 
 
R:  Yeah, but that wasn't a science fiction movie.  That was something that's already happening, so maybe it's not as big of a draw.  'Cause it's not like oh, here's something that we only dreamed of ten years ago in the movies, but now it's real.  You know.  So I think that that's why
 
S:  Yeah, I know, but they're going for the cheap headlines, even though it is
 
R:  Yes.
 
S:  It is confusing and it actually distracts from the actual content of the science.  I'm talking about making an actual meaningful analogy to something else that helps people understand and put it into context. 
 
R:  Right.
 
S:  Which is what they're ''not'' doing. 
 
J:  Yeah, but Steve, they're going by the Rotten Tomatoes rating.  Okay?
 
=== Lead and Crime <small>(14:49)</small>===
=== Lead and Crime <small>(14:49)</small>===
* [http://www.motherjones.com/environment/2013/01/lead-crime-link-gasoline America's Real Criminal Element: Lead]
* [http://www.motherjones.com/environment/2013/01/lead-crime-link-gasoline America's Real Criminal Element: Lead]

Revision as of 13:32, 30 January 2013

  Emblem-pen-orange.png This episode needs: transcription, 'Today I Learned' list, categories, segment redirects.
Please help out by contributing!
How to Contribute

SGU Episode 392
19th Jan 2013
Quasar.jpg
(brief caption for the episode icon)

SGU 391                      SGU 393

Skeptical Rogues
S: Steven Novella

B: Bob Novella

R: Rebecca Watson

J: Jay Novella

E: Evan Bernstein

Guest

M: Massimo Polidoro

Quote of the Week

No facts are to me sacred; none are profane; I simply experiment, an endless seeker with no past at my back.

Ralph Waldo Emerson

Links
Download Podcast
SGU Podcast archive
Forum Discussion


Introduction

You're listening to the Skeptics' Guide to the Universe, your escape to reality.

S: Hello and welcome to the Skeptics' Guide to the Universe. Today is Tuesday, January 13, 2013, and this is your host Steven Novella. Joining me this week are Bob Novella,

B: Hey, everybody.

S: Rebecca Watson,

R: Hello, everyone.

S: Jay Novella,

J: Hey, guys.

S: And Evan Bernstein.

E: How's everyone tonight?

S: Good, how are you, Evan?

R: Super.

J: Good.

E: Fine. Thank you, thank you, fine.

S: Evan, I noticed you stopped your foreign language hellos.

R: That's true.

E: Well, are you saying that I should have kept that up as a regular segment? It was more of like an experiment.

S: No. I just noticed that you stopped.

R: We were just wondering if you ran out of languages. (laughter)

J: Evan, what are you experimenting with now?

E: (laughing) I'm not necessarily experimenting with anything right now, but, Steve, I'm glad you mentioned it. You sort of rekindled my memory on that. Perhaps I will bring that back, or try something a little bit different.

S: I wasn't trying to encourage you, I was just making an observation.

E: Very astute. Nothing gets by you, Doctor.

This Day in Skepticism (1:05)

R: Speaking of segments, this day in history, on January 19, 1915, one George Claude patented the electrodes used for neon lights, which was what allowed him to perfect the neon sign that quickly became ubiquitous.

B: Very ubiquitous. (laughter)

R: Yeah, and for a while he had monopoly on it because he was so good at the neon sign making. Do you guys know what the first neon signs made in the United States were?

J: Yeah, for beer.

E: Eat at Joe's.

R: Unfortunately

B: Open

R: It's way more boring than any of those. Well, maybe not more boring than "Open." Good guess, Bob. But, it was for a car dealership called Packard. So the sign just read "Packard." Boring. But they cost $1,250 each. Quite pricey.

B: Wow.

E: Wow.

R: They were extremely.

B: They must have been a sight, though.

R: Apparently, yeah.

B: I can imagine.

R: People would just stop and stare at them for hours because people were dumb back then. (laughter) Easily amused.

J: Speaking of cool things that cost thousands of dollars, just on a side note, Kingston came out with a one terabyte thumb drive, costs $2,000.

B: Yes. I saw that.

E: Two thousand dollars!

B: Two grand! Oh, my god.

E: Where are you gonna put that thing? Duhaaa. Doh.

J: It's a terabyte, guys. It's a terabyte. I would wear it around your neck.

E: I would lose that in about three days, I think.

R: Easily, yeah.

E: I should not have that device.

B: I would Velcro it to my body. I would not lose that.

R: I can't even have a one terabyte external hard drive because I'll lose it. God forbid. Yeah. Thumb drive gone.

S: Let's get back to neon lights, because I'm more fascinated by those. So do you guys know how neon lights work?

J: Sure I do.

E: Gases.

S: All right, go ahead, Jay.

J: Well, you electricity.

S: Um hm.

E: Very good.

J: And it hyper-stimulates the gas

S: (laughing) Hyper-stimulates!

J: Yes. It stimulates the gas in there, which is neon gas and the particles hit, they hit each other, right, from their moving really fast they hit each other and they produce light when they hit each other.

S: No. Good try, good try.

R: That sounded really good, though.

E: Yeah, it was pretty plausible.

S: The electrical current, which has to be at a very high voltage, it ionizes the gas. It strips an electron out of the outer shell of the neon gas.

J: That's what I said. Yup.

S: Yup. So then you have ionized gas, and with the, electrons essentially get kicked up into a higher energy state. And then, but that's unstable. So when they come back down to a lower energy state they give off a photon of energy, that's how they get rid of that energy and because the electron shells are quantal, right, it's a quantum, you have to go from one very specific state to another very specific energy state. It gives off the exact same amount of energy in each photon and that equals the frequency of light, and so neon gives off, always gives off that exact same frequency of orange-y red light. Other gases of course will give off different energies and they give off different colors. Right.

J: That's right, exactly right, Steve.

E: Argon.

S: Thank you, Jay.

B: That's kind of how a laser works, in a sense. Where you have the population version where the electrons jump to higher levels and then they spit out the electrons when they go back to their ground state. But they're all in the same frequency, though, very very specific frequency, and it's like a collimated beam, so that's what makes a laser light. Similar.

S: Right.

J: So, Steve, are you saying that the electrons rock down to Electric Avenue and then they take it higher?

S: That's exactly what happens, Jay. Now,

B: Oh. My. God.

E: You should write a song about that, Jay.

S: You can tell what gas is in

R: I gotta give that one credit, actually. That was good.

B: Wait, yeah.

R: That was really good.

B: Didn't you just like totally riff off of that?

J: I'm a fan, you know.

R: Credit where it's due.

S: So you can tell what kind of gas is in a neon tube, they're not, of course, all neon, based upon what color it is. So helium, for example, gives off an orange-y white light, neon of course is the orange-red, argon is a violet pale lavender blue, krypton (gotta love krypton) grayish, then a really pretty blue light is given off by mercury vapor. But mercury vapor also serves another purpose. Do you guys understand the difference between a neon light and a fluorescent light?

J: Aboslutely. (laughter)

S: So, this is interesting, too.

E: There's fluoride in the fluorescent light.

S: No. So a fluorescent light is essentially a neon light that has mercury vapor

J: Mercury vapor.

(laughter)

S: As the gas, and it's charged enough so that the mercury vapor gives off ultra-violet light.

R and J: Ultra-violet, yeah.

E: Ultra. Ultra-violet.

S: And then you can coat the inside of the glass

J: Insides (laughter)

S: with a fluorescent substance. The ultra-violet light from the mercury vapor hits that and

J: It hits it, yeah.

S: that fluoresces in whatever color it is. So that you can

B: That. Yeah, that, yeah.

J: That's good stuff right there, Steve.

E: Wow. So a one-terabyte thumb drive. (Everyone pretty much loses it)

J: Yup. How 'bout that, Evan. You could literally put your entire music catalog in your thumb drive and take it with you. Not that you'd need it. But you had it.

B: Wait, wait. How big is your library?

R: You could literally put an entire textbook about neon lights on a terabyte.

E: Fascinating stuff.

S: Multiple.

J: That is cool, though, Steve, and I wanna test you one year from now to see if you remember all that stuff you just read to us.

S: Okay. Do it. (laughter)

B: I can hear Steve talking to Cerie: "Cerie, remind me in three hundred and sixty-three days to look up fluorescence again and neon lights."

E: All I'm hearing is Electric Avenue in my head.

J: All right, so who's the guy that sang the song? Who remembers his name?

E: The Jamaican guy, ummm

R: Yeah.

B: Wink Martindale.

E: Eddie something. Eddie….

J: Oh, you're so close!

R: Eddie Capitan.

J: That song was written by—Eddie Grant.

S: Eddie Grant.

E: Eddie Grant.

J: I almost forgot him.

R: Never heard of 'im.

E: Grant.

B: Is that Amy Grant's brother?

S: Jay, did you know that when you ask a question and nobody has the right answer, that you're supposed to wait a very specific amount of time before you give the answer yourself? You know how long that is?

J: How long is that?

B: Three seconds.

S: Three seconds. That's right. Wait three seconds.

E: That's the courtesy window?

R: I think it depends on how funny we're being about the wrong answers. (laughter)

E: Right.

S: That's true. That's a variable that's often not included.

News Items

Predicting Murders (7:36)

S: But Jay, you're gonna tell us about predicting murders. But not just predicting murders, predicting murders like in the movie Minority Report. Even though it has absolutely nothing to do with that, but go ahead.

J: So, yes. Richard Berk, who is a criminologist at the University of Pennsylvania has developed software that's making predictions about future criminal behavior. Up until now, the way that this was typically done was that parole officers would have the record of that criminal or whoever was committing the crime. They'd figure out, okay, how dangerous is this person, how many times have they committed crimes, and all that. And that person on their own with seemingly not much of a standard would say, okay, I think that they need about this much supervision and they give them, it's like a halfway house type of deal where they're like, okay, you need to live here or you need to come in and check with me every week or every month or whatever. That variable, though, doesn't really help much. And it's not effective. Or so Richard Berk thinks. So what they're doing now is he has taken 60,000 crimes and he did very smart research on them and he was looking into the patterns that exist. What Professor Richard Burke found was that a huge factor in somebody's predictability of future crimes isn't a crime that they committed when they were 30 but it was the crime or crimes that they committed when they were in their formative years, when they were young adult, teenagers. That's a really significant predictor of what they're gonna do for the rest of their life. So, if you're a 30-year-old, the last ten years, fifteen years, according to Berk, are irrelevant. It's much more important to take a deep dive into the person's upbringing and all the things that they did when they were young. Which I found to be very interesting. It speaks a lot about how, you know, I don't want to get into the whole nature versus nurture discussion, but you think if you provide a good household for your children and you steer them in the right direction you could be saving them from criminal life of the future, right? Two states in the United States have adopted this software already. Baltimore and Philadelphia are now using it, and D.C.'s soon to pick it up and start using it as well. He said that his software is replacing the quote unquote ad hoc decision-making and should identify eight out of one hundred murders.

S: That's not a lot.

J: No. But, that's eight more people than what would be predicted by a parole officer. The thing, I'll go back to what you said, though, Steve, one huge thing about this article is how freaking lame the Daily Mail is. Not that I have any expectations from these guys, but, I mean, it's so crystal clear that they're using a famous movie that everybody know, throwing in the title of their article just to get people to read it, and it worked on me.

E: Worked on me, too.

J: Yeah, but this is the media using Hollywood to support their readership and show interest. I found this to be a very interesting article and probably would have read something more clever anyway. It's just laziness on the author in my opinion.

S: Yeah, and there's five pictures from the movie Minority Report in this article.

B: Oh, my god.

S: Seriously, it's in the title, it's all over the article, all these pictures, you know. It has nothing to do . . . in the movie, which would be a good one for us to review by the way,

E: Absolutely.

S: Very good portrayal of near future technology. But anyway, there were three quote unquote precogs who literally predicted a specific person committing a specific murder at a specific time and place, has absolutely nothing to do with what they're talking about here, which is statistically predicting who is likely to commit a murder again, based on some computer algorithm, taking into consideration evidence-based factors. Two totally different things.

J: Yup.

S: You're right.

J: Lame!

S: Very typical, you know. I agree, the lazy science reporters always have to find some movie angle that somehow relates to whatever technology or breakthrough or thing they're talking about. Right, so right, Bob, so every time there's any metamaterial breakthrough, it's

B: Uh!

S: a Harry Potter invisibility cloak.

B: That's their go-to analogy. The comparison. They can't not say it.

E: Do they feel that the audience isn't sophisticated enough to understand what they might be talking about otherwise unless they have some sort of

S: No! It's just cheap headlines.

B: No, they want a hook. It's all about being a hook. They want to hook you in. Something to make you read it. No matter what it takes.

J: Sure. Evan, think about it, they spend $100 million or whatever to make Harry Potter and everybody knows about it and most people enjoy it and like it and they're just throwing it out there, like

R: I think it makes people more likely to read it and less likely to understand it.

S: Right.

J: Yeah.

S: Right.

E: Yes.

S: So what's the point?

E: The opposite effect, yes.

S: Interestingly, the one movie analogy to technology that I think is actually legitimate that I've never seen a news reporter actually make is transparent aluminum.

J: Right.

E: That's right.

S: There have been several companies or whatever, researchers that have come out with some form of something that's based on some aluminum alloy that's transparent and hard—it is freakin' transparent aluminum from the Star Trek movies and they never make that analogy. I don't understand that.

R: I guess 'cause nobody saw those movies.

S: I guess not.

J: Yes, we did. (laughter)

B: The Voyage Home, that's like, probably like, hugely popular.

R: Was that the one with the whales?

S: Yup.

E: Yes.

R: Yeah, I like that, that was my favorite.

J: There you go, all right.

E: There was a lot to like in that movie.

B: G-e-e-ek (laughter)

R: Oh my god, I'm such a geek.

S: The other thing that struck me about

E: Wasn't directed by William Shatner. That's a great thing to like about that movie.

S: The other thing that struck me about this article, and it wasn't mentioned in the article—this is the kind of thing, kind of analogy they should have made—is this is saber metrics, right? This is Moneyball. You guys see the movie Moneyball?

E: Yes. Great movie

S: That would have been the better movie to tie this to. So this is the same thing. Instead of a Gestalt gut feeling using parole officers and judges, whatever, to say, yeah, is this guy likely to be a criminal again or not, they use an algorithm based on evidence of predictive factors. That's the perfect analogy to Moneyball where instead of using scouts and coaches and owners to see who's the really talented player out there, they use actual statistics to predict their effect on the chance of winning games.

R: Yeah, but that wasn't a science fiction movie. That was something that's already happening, so maybe it's not as big of a draw. 'Cause it's not like oh, here's something that we only dreamed of ten years ago in the movies, but now it's real. You know. So I think that that's why

S: Yeah, I know, but they're going for the cheap headlines, even though it is

R: Yes.

S: It is confusing and it actually distracts from the actual content of the science. I'm talking about making an actual meaningful analogy to something else that helps people understand and put it into context.

R: Right.

S: Which is what they're not doing.

J: Yeah, but Steve, they're going by the Rotten Tomatoes rating. Okay?

Lead and Crime (14:49)

Biggest Thing in the Universe (25:27)

Million Dollar Challenge (30:11)

Turkey Bans Evolution Books (39:12)

Quicky With Bob: Apophis Update (42:31)

Who's That Noisy? (44:40)

  • Answer to last week: John of God

Interview with Massimo Polidoro (47:20)

  • Massimo Polidoro, director of the Italian Skeptics

Science or Fiction (1:00:13)

Item number one. A recent study finds that subject's memory for Facebook posts were significantly greater than for book entries or faces. Item number two. New research finds that some children diagnosed at a young age with autism may outgrow the diagnosis entirely. And item number three. A new study finds that graphic cigarette warnings have minimal effect, and are no more effective than text-only warnings.

Skeptical Quote of the Week (1:16:53)

No facts are to me sacred; none are profane; I simply experiment, an endless seeker with no past at my back.

Ralph Waldo Emerson

Announcements

NECSS (1:17:18)

Podcast UFO (1:18:00)

Template:Outro1

References


Navi-previous.png Back to top of page Navi-next.png