Starting in December, randomly selected visitors to Regulations.gov may be asked to complete a short customer satisfaction survey. Your feedback will help us improve the quality of this site for all users. We thank you in advance for your participation.
Customers? We're customers now? Does that mean that i'll be able to demand satisfaction or perhaps my money back?
You misunderstand. I don't mean Reiser's case. I meant the case of the media. The media is a device that gives preferential coverage to young white women. And i might note, it's not just racist, it's got gender bias too. You don't see coverage of missing young men typically either.
(and just as a point of note Nina Reiser doesn't fit the perfect heart-breaking dramatic news story either, which again, is part of the reason why it wasn't covered as heavily as other missing persons stories)
Actually, one of the things that i found interesting was having the terran military develop technology that let them actually compete with the alien species surrounding them. This could have been really fascinating, but they developed so quickly, and with so little effort, that it really wasn't that interesting. I think they missed a huge opportunity there. The ground they covered though is fascinating, SG-1 went from running around on foot, and having to flee or leave behind anything they couldn't carry, to being able to teleport buildings, use energy weapons and shields, and travel in hyperspace. I wish they'd been more introspective about it (although i guess O'Neil's character didn't really lend itself that).
Oh well, SG-1 was always soft scifi. Maybe i'm just a hard scifi geek complaining too much:)
The purpose of democracy in the US is to try and do the best for everyone. You don't capriciously let the few abuse the resources of everyone, just because someone feels like it.
Also, people don't understand how expensive it is to execute people. It costs a fortune. Go google execution costs. Capital punishment is not the cheap easy way to remove genuine criminals from society. It's completely permanent but it's not the inexpensive route.
Never the less, even if low new coverage is preferential, which incidentally, i don't believe is the case (as murder investigations go in two phases, discovery then prosecution, the initial part of the investigation definitely benefits from having more eyeballs looking for the victim), a look at the system that selectively picks which stories to investigate does give an interesting view into the biases that the system possesses. And in this case indicates that racism is alive and well.
Also Lacie Peterson was murdered a week before christmas and was 8 months pregnant. See the distinction? Lacie Peterson's family organized a whole bunch of people to go out and look for her prior to a holiday (when people are feeling particularly helpful).
I would have to say that these were the primary things that garnered so much attention for the peterson trial, not the defendant's particularly despicable character:P
Well... since explaining humor is like dissecting a frog (nobody's interested, it's not terribly educational and the frog dies at the end), I'm only gonna toss you one reply.
One of the reasons why i thought it'd be fun to toss my initial reply, is because the exchange above is a golden opportunity for meta-humor. Primarily because the typical reply to the initial comment (i.e. obnoxious navel-gazing about slashdot) is typically "you must be new here".
Also, while i agree that in-jokes are typically not very funny (especially the Soviet Russia thing which i find obnoxious), there is something inherently multi-level about the "you must be new here" jokes that makes it both useful, and funny (since, the implication that you've been around long enough to assess how slashdot is no longer what it once was should also imply that you've been around long enough to know that obnoxious navel-gazing isn't appreciated and typically garners a "you must be new here" reply).
The fact that i got to make a multi-level joke even more complex was worth it:)
Well, if it makes a difference, Sony's current CEO is a British born, US citizen. So if you wanted to be technical about it, their leadership currently is not in fact Japanese.
Er... People seem to have some misconceptions about Talk.Origins. The point of Talk.Origins is to catalogue the arguments that Creationists make, and the appropriate counter argument to any given creationist argument. This is primarily due to the fact that creationist arguments are extremely unoriginal and repetitive (seriously, creationists have been hauling out the same arguments since before Darwin's time, regardless of the change in scientific thought or theory). So i'd say that Talk.Origins does a very good job of presenting both sides. It just happens that the creationist side is out-dated and unsupportable. But that's a matter of fact, not of inaccurate representation.
Also, we can't accurately ascertain whether the earth is the center of the universe or not. It's probably not, but the way space time expands gives no reference point for the point of origin. From any point in the universe, it looks like everything is expanding away from you.
But you are correct, the earth is indeed not flat. It's sort of a squashed sphere.
I don't think libertarians are off the hook that easily. The fact of the matter is that the NeoCons pushed their agenda and their ludicrously ill-thought policies in the name of idealism, and the idealism they borrowed was at least partially taken from libertarian ideals. Don't get me wrong, the out-going republican leadership are opportunists, and have no allegiance to anyone when it doesn't serve their purposes, but if you're being used, your two options are to object to the abuse, or to take your lumps when your benefactors hit the end of the line. And the libertarians who voted for Bush (especially in 2004) can't credibly claim they didn't know what was up.
As I said before, the largest threat group at this time would probably be the Islamic terrorists. You're putting an all inclusive loop around terrorists, but I think for the purposes of the argument in this day and age most people would agree that the biggest threat is Islamic based terrorism.
This set of statements are just so loaded with vague and faulty assumptions.
Largest threat group? You mean largest terrorist threat? Because the likelihood of terrorism is still so absurdly low, that Americans really should be more worried about drunk drivers, carcinogens, and accidental electrocution.
Even if you want to scope it more dramatically to "largest terrorist threat group", that's a statement that depends greatly on your proximity to the middle east. Home grown American terrorist organizations are much more likely to have the access and support network to execute terrorist actions in the US, again, see environmental terrorism, or school shooting. Even if you're an Israeli, and directly neighboring a territory in which terrorist groups reside, the likelihood that you're going to be blown up at a bus stop is not as significant as the likelihood you're going to be hit by a car at the same bus stop.
Al Quaeda is a well funded and smart group. The groups you've mentioned really pose no threat to the US in comparison.
Really? Is Al Qaeda ever going to be in a position to topple the US government? Could they actually destroy US financial markets (that is, any better than bankrupting the US by dragging us into Iraq, and possibly Iran will)? Are they seriously going to damage the operating capacity of the American military? The point of terror is that you terrorise, you're not a serious threat, you just make people really unhappy all the time. And guess what? Everybody's pretty freaking miserable.
And just to dispell another illusion, Al Qaeda, in the context of Islamic radicalism is really rather passe. Al Qaeda, is a single organization, which has in fact been severely damaged by the fall of the Taliban. The major problem from Islamic radicalism, is that it is now distributed, derived from seperate sources, funded from a variety of locations and individuals. The problem the western world is facing is that they have turned economic and social injustice into a conflict of civilizations. Islamic terrorism isn't going to play out until the world changes, either the Arab world is going to change, or the rest of the world will change. Something's going to have to give. Al Qaeda isn't the problem, Arab frustration, and the cycle of violence and economic and social dispossession are going to generate more terrorists than Al Qaeda could ever be responsible for (and those are things that are inside and outside the Islamic world). Playing whack-a-terrorist - i mean mole - ain't gonna solve anything.
Okay, so that's an issue over whether two variables are statistically independent or not. Some factors are just irrelevent to the discussion you're interested in having. A building's exterior paint color is not relevent to the factors that determine its structural integrity.
My point again has been that race is just not relevant. And while i agree that the only thing better than data is more data, wasting your time on data that has such miniscule influence over the probabilities you're interested is again, either a distraction, or genuninely counter-productive. Even if the description of a terrorist requires the consideration of a large number of factors, what i'd float the statement that race, or other purely superficial external characteristics are going to be irrelevent to whatever calculus you can build to sort terrorists from non-terrorists. Again it is important to consider the statistical indicators that you are tracking if you want to get good descriptive or predictive results.
In short the claim i am making is that race, or whatever you want to call it, is in essence, independent, or otherwise so insignificant it can be assumed to be independent of the factors that do contribute to the likelihood a person is going to be a threat to others.
In the scenarios that law enforcement might need to deal with, religious based Islamic terrorism is probably a higher risk than say the Sri Lankan Tamil Tigers.
This still MUST take into account the likelihood that they will encounter religious based Islamic terrorism, which is in fact so ridiculously low, that it's almost not worth considering. Its much more likely that law enforcement will have to deal with homicide, rape, theft, car accidents, etc etc etc, and somewhere far down the line is terrorism.
Police officers should be concerned with other things.
And you misunderstand. I don't believe in race. It's a non-factor, i want people to stop discussing it period. Regardless of that fact, using huge categories, such as ethnicity or geographic origin is stupid, and the reason why it's stupid, like i've pointed out below, is because it catches far too many people to be useful.
Sorry, i should be more clear. If a tool has a fatal flaw, all you need to demonstrate the flaw, is that it behaves contrary to its purpose, or that it does no better than randomly guessing. To demonstrate that it behaves contrary to its stated purpose, you only need a single case. To indicate that it doesn't do any better than randomly guessing, then you do need a battery of test cases. In the case i make above, i take the second path. However, that still doesn't invalidate the possibility of the former.
Alright i'll be more specific. (and skip to the bottom if you already understand bayes rule)
Here's bayes rule: P(A|B) = P(B|A)P(A)/P(B)
Bayes rule gives you the likelihood that A is true, if B is true. It does this by asking what's the probability that B is true, if A is true, assessing what the likelihood of A is, and finally putting that in context of the probability of B.
So lets make this more concrete. We want to know what's the probability of a person being a terrorist if they are a Muslim, or in other words what is P(Terrorist|Muslim).
So Bayes rule says this question is equivalent to asking, what the likelihood that a person is muslim if they are a terrorist P(Muslim|Terrorist), and what the probability of a person being a terrorist is, or P(Terrorist), and then whether this matters given the number of muslims out in the world P(muslim)
First, a comment. A lot of people assume that P(Terrorist|Muslim) = P(Muslim|Terrorist) is true. It is not. The fact that some terrorists are muslims, does not make all muslims terrorists, or even that it's more likely that most muslims are terrorists.
So, now, given any human, we first evaluate whether our random human is a muslim (i.e. P(Muslim)). The likelihood is high (i'm not saying it's near 1 or whatever, but it's a big number statistically, naively we'll assume that it's 1/3 for the sake of argument).
So, now pick a number for the likelihood that our random person is a terrorist (i.e. P(Terrorist)). This number is going to be low. The fact of the matter is that there are like >10 billion people on earth, and only a small number of them are psychopathic lunatics who want to send messages to others by killing people. Even if you want to say that there are currently a million terrorists out in the world, the probability that any individual is a terrorist would be 0.0001 (pretty small).
So that puts us at P(Terrorist|Muslim) = P(Muslim|Terrorist) * 0.0001 / 0.33
It's not all that important how many terrorists are Muslims (i.e. P(Muslim|Terrorist)), because even if ALL terrorists are Muslims (i.e. P(Muslim|Terrorist) = 1), the probability of a Muslim being a terrorist is 0.00030303030303 (i.e. not very fucking likely). And since P(Muslim|Terrorist) is definitely significantly under 1, it's even less relevant to know whether a person is a muslim, if you're trying to identify whether they're a terrorist or not.
If you want to find terrorists, you do in fact need to find better indicators which correlate more strongly between the population with that indicator, and being a terrorist. And what i've been saying is that those indicators are not things that a border guard, or a TSA agent is going to be able to find out quickly, with the small amount of time they have with each individual they inspect. This is why immigration requests take months to process. So your implication is correct, i think that law enforcement is not capable of making quick accurate judgements, and that furthermore, quick judgements that they make are going to be inaccurate to the point of ineffectiveness, or counter-productiveness. And just to jump back to your bandana example, lets substitute cars instead. If crimes keep being committed using cars, you're sort of SOL, because everyone uses cars. It may be a common thread, but it's totally irrelevant because there are way too many cars out there. Race is just the same way. It will never be an effective indicator, because there are way too many people with that characteristic, you may as well ignore it. (Also i still do not acknowledge race as anything but an anachronistic fiction made up to divide people)
So, to sum up the nasty mess above, most people don't understand statistics. Most people don't understand the amount of information you need to be able to make good statistical judgements. Most people don't have access to the sort of informat
Right, but regardless of the rules, tazering someone 5 times is excessive in almost any circumstance (unless, for instance, the dude is seriously pumped on stimulants and the size of an elephant), and especially in this particular case, where the dude did not pose a threat, and was definitely not a threat after the first, or second, or third time he was tasered. The rules in this situation are totally moot, this is an issue of proportional response.
Cute, but all i need to show you to prove that a tool doesn't work (even a statistical tool) is a single case. If i wanted to assert that something else did work then i would need to show you a wide variety of cases (preferably all cases). But since i'm not asserting anything like that, i can just tell you to bugger off.
So nice try, and thanks for demonstrating that you don't understand statistical reasoning:)
Second, you are selectively picking terrorist incidents. Crimes of terror (or whatever the hell you want to call them) occur more broadly than reported by the media, and perpetrated by groups more widely than just muslims. Just off the top of my head i can remember incidents perpetrated by the Earth Liberation Front (presumably a bunch of fringe west coast hippies), a wide variety of "animal rights" groups, who are predominately comprised of white people, the Sri Lankan Tamil Tigers (against the government of Sri Lanka, who themselves are arguably performing terrorist acts), anti-abortion activists (like the lunatic who drove his car into a medical clinic which did not perform abortions, or even refer people for abortions), and others. Terrorism is not a middle eastern problem. Assuming that it is is just fear mongering, and discrimination. Most middle easterners and muslims are not terrorists, just the same way that most of the Irish are not terrorists, the same way that most Basques are not terrorists. It's not fair to treat Muslims with suspicion, when nobody treats any of these other groups with suspicion, they are just as innocent or just as guilty by association as everyone else.
And if you really want to bring up Israel, remember there are fundamentalist Israelis who go shoot Palestinian children as well as Palestinians who go blow up Israelis.
It's a ridiculous thing to consider. What about all the rest of the people who come from those regions?
At what point does a trend justify profiling people? 100 incidents? 50? 10? 3?
What about consideration for proportionality? For every psychotic killer, how many normal non-psychos are out there? What about all the people in the middle east who aren't suicide bombers? If there is only 1 suicide bomber for every 4 million people, then it seems like you're getting an awful lot of false positives if you are identifying people simply by ethnicity, or race, or nationality. And if those factors are only tangentially relevant to the main factors, why bother including them at all? If you went to psycho bomber school, then it doesn't matter if you're Pakistani, Dutch, Chinese, or Bolivian, you went to psycho bomber school! So why don't we try and identify psychopaths, rather than try and figure out whether brown people are more likely to be psychotic or not? (which itself is sort of an odd question, because nobody ever asks "is group y more likely to be psychotic than group x, and that's what really causes fear mongering, no context.)
Likewise, how many people are there in the US who wear bandanas and don't commit crimes? Should you just throw anybody with a bandana in jail? Or be quicker to taser them than other people?
The problem with profiling, or any sort of statistical reasoning is that you must must must make sure that you have the correct data set, and that you are looking at the correct indicators, or you will get incorrect results. And if there's anything the US has proven in the past 5 years, its that they don't understand the middle east, they don't understand it culturally, they don't understand it psychologically, and they don't understand it politically. I have absolutely zero faith that the US government, and most importantly the people who are being asked to implement what are statistical judgements (police officers, TSA bag screeners, custom officials) are capable of understanding and recognizing the nuances required to make good decisions.
So, to sum up, racial profiling is just garbage, and other forms of profiling need to be done by professionals (like social workers), and so should not be implemented widely, because they are not scalable filters.
Fine, then i hope you understand when i taser every white kid with a crew cut who driving a pickup truck, cause you never know which of 'em is gonna be the next Timothy McVey.
Racial Profiling is stupid, ineffective, and an unjustifiable abuse of statistical reasoning.
It depends radically on the nature of the academic institution you're in. I attended the Ohio State University, which is a publically owned and funded institution. As a result, any resident of Ohio is entitled to use the physical library (for some reason they do not extend the use of their electronic holdings to the public, why i don't really know), as well as a number of other University functions (such as course syllabaries, and what texts are being used for a course).
Private universities are entitled to do whatever the hell they feel like (within the realm of the law of course), which is why students at private universities don't technically have the right to protest on university grounds, unless the university approves (no right to public free speech if you're not in public).
Now, as to what policies UCLA (which is part of the public University of California system) has on it's libraries, i don't know, but i can certainly say that this particular incident is not indicative of all institutions of higher learning in the US.
Slashdot Mirror
You misunderstand. I don't mean Reiser's case. I meant the case of the media. The media is a device that gives preferential coverage to young white women. And i might note, it's not just racist, it's got gender bias too. You don't see coverage of missing young men typically either.
(and just as a point of note Nina Reiser doesn't fit the perfect heart-breaking dramatic news story either, which again, is part of the reason why it wasn't covered as heavily as other missing persons stories)
Actually, one of the things that i found interesting was having the terran military develop technology that let them actually compete with the alien species surrounding them. This could have been really fascinating, but they developed so quickly, and with so little effort, that it really wasn't that interesting. I think they missed a huge opportunity there. The ground they covered though is fascinating, SG-1 went from running around on foot, and having to flee or leave behind anything they couldn't carry, to being able to teleport buildings, use energy weapons and shields, and travel in hyperspace. I wish they'd been more introspective about it (although i guess O'Neil's character didn't really lend itself that).
:)
Oh well, SG-1 was always soft scifi. Maybe i'm just a hard scifi geek complaining too much
Sir, i applaud you for the most humorously capricious invocations of Godwin's Rule i have ever encountered. Bravo.
Er, that's not a joke.
The purpose of democracy in the US is to try and do the best for everyone. You don't capriciously let the few abuse the resources of everyone, just because someone feels like it.
Also, people don't understand how expensive it is to execute people. It costs a fortune. Go google execution costs. Capital punishment is not the cheap easy way to remove genuine criminals from society. It's completely permanent but it's not the inexpensive route.
Way to go, Sir False Dichotomy.
Obviously if you believe in due process, then clearly you support O.J. Simpson.
Never the less, even if low new coverage is preferential, which incidentally, i don't believe is the case (as murder investigations go in two phases, discovery then prosecution, the initial part of the investigation definitely benefits from having more eyeballs looking for the victim), a look at the system that selectively picks which stories to investigate does give an interesting view into the biases that the system possesses. And in this case indicates that racism is alive and well.
Uhh...
:P
Also Lacie Peterson was murdered a week before christmas and was 8 months pregnant. See the distinction? Lacie Peterson's family organized a whole bunch of people to go out and look for her prior to a holiday (when people are feeling particularly helpful).
I would have to say that these were the primary things that garnered so much attention for the peterson trial, not the defendant's particularly despicable character
Well... since explaining humor is like dissecting a frog (nobody's interested, it's not terribly educational and the frog dies at the end), I'm only gonna toss you one reply.
:)
One of the reasons why i thought it'd be fun to toss my initial reply, is because the exchange above is a golden opportunity for meta-humor. Primarily because the typical reply to the initial comment (i.e. obnoxious navel-gazing about slashdot) is typically "you must be new here".
Also, while i agree that in-jokes are typically not very funny (especially the Soviet Russia thing which i find obnoxious), there is something inherently multi-level about the "you must be new here" jokes that makes it both useful, and funny (since, the implication that you've been around long enough to assess how slashdot is no longer what it once was should also imply that you've been around long enough to know that obnoxious navel-gazing isn't appreciated and typically garners a "you must be new here" reply).
The fact that i got to make a multi-level joke even more complex was worth it
*ribbit*
I think you mean, "You must be new here."
:)
Although your number is higher than his.
So, perhaps i should say:
You must be new here, because i think you mean, "You must be new here."
Well, if it makes a difference, Sony's current CEO is a British born, US citizen. So if you wanted to be technical about it, their leadership currently is not in fact Japanese.
Er... People seem to have some misconceptions about Talk.Origins. The point of Talk.Origins is to catalogue the arguments that Creationists make, and the appropriate counter argument to any given creationist argument. This is primarily due to the fact that creationist arguments are extremely unoriginal and repetitive (seriously, creationists have been hauling out the same arguments since before Darwin's time, regardless of the change in scientific thought or theory). So i'd say that Talk.Origins does a very good job of presenting both sides. It just happens that the creationist side is out-dated and unsupportable. But that's a matter of fact, not of inaccurate representation.
Also, we can't accurately ascertain whether the earth is the center of the universe or not. It's probably not, but the way space time expands gives no reference point for the point of origin. From any point in the universe, it looks like everything is expanding away from you.
But you are correct, the earth is indeed not flat. It's sort of a squashed sphere.
I don't think libertarians are off the hook that easily. The fact of the matter is that the NeoCons pushed their agenda and their ludicrously ill-thought policies in the name of idealism, and the idealism they borrowed was at least partially taken from libertarian ideals. Don't get me wrong, the out-going republican leadership are opportunists, and have no allegiance to anyone when it doesn't serve their purposes, but if you're being used, your two options are to object to the abuse, or to take your lumps when your benefactors hit the end of the line. And the libertarians who voted for Bush (especially in 2004) can't credibly claim they didn't know what was up.
As I said before, the largest threat group at this time would probably be the Islamic terrorists. You're putting an all inclusive loop around terrorists, but I think for the purposes of the argument in this day and age most people would agree that the biggest threat is Islamic based terrorism.
This set of statements are just so loaded with vague and faulty assumptions.
Largest threat group? You mean largest terrorist threat? Because the likelihood of terrorism is still so absurdly low, that Americans really should be more worried about drunk drivers, carcinogens, and accidental electrocution.
Even if you want to scope it more dramatically to "largest terrorist threat group", that's a statement that depends greatly on your proximity to the middle east. Home grown American terrorist organizations are much more likely to have the access and support network to execute terrorist actions in the US, again, see environmental terrorism, or school shooting. Even if you're an Israeli, and directly neighboring a territory in which terrorist groups reside, the likelihood that you're going to be blown up at a bus stop is not as significant as the likelihood you're going to be hit by a car at the same bus stop.
Al Quaeda is a well funded and smart group. The groups you've mentioned really pose no threat to the US in comparison.
Really? Is Al Qaeda ever going to be in a position to topple the US government? Could they actually destroy US financial markets (that is, any better than bankrupting the US by dragging us into Iraq, and possibly Iran will)? Are they seriously going to damage the operating capacity of the American military? The point of terror is that you terrorise, you're not a serious threat, you just make people really unhappy all the time. And guess what? Everybody's pretty freaking miserable.
And just to dispell another illusion, Al Qaeda, in the context of Islamic radicalism is really rather passe. Al Qaeda, is a single organization, which has in fact been severely damaged by the fall of the Taliban. The major problem from Islamic radicalism, is that it is now distributed, derived from seperate sources, funded from a variety of locations and individuals. The problem the western world is facing is that they have turned economic and social injustice into a conflict of civilizations. Islamic terrorism isn't going to play out until the world changes, either the Arab world is going to change, or the rest of the world will change. Something's going to have to give. Al Qaeda isn't the problem, Arab frustration, and the cycle of violence and economic and social dispossession are going to generate more terrorists than Al Qaeda could ever be responsible for (and those are things that are inside and outside the Islamic world). Playing whack-a-terrorist - i mean mole - ain't gonna solve anything.
Okay, so that's an issue over whether two variables are statistically independent or not. Some factors are just irrelevent to the discussion you're interested in having. A building's exterior paint color is not relevent to the factors that determine its structural integrity.
My point again has been that race is just not relevant. And while i agree that the only thing better than data is more data, wasting your time on data that has such miniscule influence over the probabilities you're interested is again, either a distraction, or genuninely counter-productive. Even if the description of a terrorist requires the consideration of a large number of factors, what i'd float the statement that race, or other purely superficial external characteristics are going to be irrelevent to whatever calculus you can build to sort terrorists from non-terrorists. Again it is important to consider the statistical indicators that you are tracking if you want to get good descriptive or predictive results.
In short the claim i am making is that race, or whatever you want to call it, is in essence, independent, or otherwise so insignificant it can be assumed to be independent of the factors that do contribute to the likelihood a person is going to be a threat to others.
Great. 'Cause being hot is the quality we should be concerned with when identifying the best geeks.
In the scenarios that law enforcement might need to deal with, religious based Islamic terrorism is probably a higher risk than say the Sri Lankan Tamil Tigers.
This still MUST take into account the likelihood that they will encounter religious based Islamic terrorism, which is in fact so ridiculously low, that it's almost not worth considering. Its much more likely that law enforcement will have to deal with homicide, rape, theft, car accidents, etc etc etc, and somewhere far down the line is terrorism.
Police officers should be concerned with other things.
And you misunderstand. I don't believe in race. It's a non-factor, i want people to stop discussing it period. Regardless of that fact, using huge categories, such as ethnicity or geographic origin is stupid, and the reason why it's stupid, like i've pointed out below, is because it catches far too many people to be useful.
Sorry, i should be more clear. If a tool has a fatal flaw, all you need to demonstrate the flaw, is that it behaves contrary to its purpose, or that it does no better than randomly guessing. To demonstrate that it behaves contrary to its stated purpose, you only need a single case. To indicate that it doesn't do any better than randomly guessing, then you do need a battery of test cases. In the case i make above, i take the second path. However, that still doesn't invalidate the possibility of the former.
Alright i'll be more specific. (and skip to the bottom if you already understand bayes rule)
Here's bayes rule: P(A|B) = P(B|A)P(A)/P(B)
Bayes rule gives you the likelihood that A is true, if B is true. It does this by asking what's the probability that B is true, if A is true, assessing what the likelihood of A is, and finally putting that in context of the probability of B.
So lets make this more concrete. We want to know what's the probability of a person being a terrorist if they are a Muslim, or in other words what is P(Terrorist|Muslim).
So Bayes rule says this question is equivalent to asking, what the likelihood that a person is muslim if they are a terrorist P(Muslim|Terrorist), and what the probability of a person being a terrorist is, or P(Terrorist), and then whether this matters given the number of muslims out in the world P(muslim)
P(Terrorist|Muslim) = P(Muslim|Terrorist) * P(Terrorist) / P(Muslim)
First, a comment. A lot of people assume that P(Terrorist|Muslim) = P(Muslim|Terrorist) is true. It is not. The fact that some terrorists are muslims, does not make all muslims terrorists, or even that it's more likely that most muslims are terrorists.
So, now, given any human, we first evaluate whether our random human is a muslim (i.e. P(Muslim)). The likelihood is high (i'm not saying it's near 1 or whatever, but it's a big number statistically, naively we'll assume that it's 1/3 for the sake of argument).
So, now pick a number for the likelihood that our random person is a terrorist (i.e. P(Terrorist)). This number is going to be low. The fact of the matter is that there are like >10 billion people on earth, and only a small number of them are psychopathic lunatics who want to send messages to others by killing people. Even if you want to say that there are currently a million terrorists out in the world, the probability that any individual is a terrorist would be 0.0001 (pretty small).
So that puts us at P(Terrorist|Muslim) = P(Muslim|Terrorist) * 0.0001 / 0.33
It's not all that important how many terrorists are Muslims (i.e. P(Muslim|Terrorist)), because even if ALL terrorists are Muslims (i.e. P(Muslim|Terrorist) = 1), the probability of a Muslim being a terrorist is 0.00030303030303 (i.e. not very fucking likely). And since P(Muslim|Terrorist) is definitely significantly under 1, it's even less relevant to know whether a person is a muslim, if you're trying to identify whether they're a terrorist or not.
If you want to find terrorists, you do in fact need to find better indicators which correlate more strongly between the population with that indicator, and being a terrorist. And what i've been saying is that those indicators are not things that a border guard, or a TSA agent is going to be able to find out quickly, with the small amount of time they have with each individual they inspect. This is why immigration requests take months to process. So your implication is correct, i think that law enforcement is not capable of making quick accurate judgements, and that furthermore, quick judgements that they make are going to be inaccurate to the point of ineffectiveness, or counter-productiveness. And just to jump back to your bandana example, lets substitute cars instead. If crimes keep being committed using cars, you're sort of SOL, because everyone uses cars. It may be a common thread, but it's totally irrelevant because there are way too many cars out there. Race is just the same way. It will never be an effective indicator, because there are way too many people with that characteristic, you may as well ignore it. (Also i still do not acknowledge race as anything but an anachronistic fiction made up to divide people)
So, to sum up the nasty mess above, most people don't understand statistics. Most people don't understand the amount of information you need to be able to make good statistical judgements. Most people don't have access to the sort of informat
Right, but regardless of the rules, tazering someone 5 times is excessive in almost any circumstance (unless, for instance, the dude is seriously pumped on stimulants and the size of an elephant), and especially in this particular case, where the dude did not pose a threat, and was definitely not a threat after the first, or second, or third time he was tasered. The rules in this situation are totally moot, this is an issue of proportional response.
Cute, but all i need to show you to prove that a tool doesn't work (even a statistical tool) is a single case. If i wanted to assert that something else did work then i would need to show you a wide variety of cases (preferably all cases). But since i'm not asserting anything like that, i can just tell you to bugger off.
:)
So nice try, and thanks for demonstrating that you don't understand statistical reasoning
First: See my reply to the sibling post below.
Second, you are selectively picking terrorist incidents. Crimes of terror (or whatever the hell you want to call them) occur more broadly than reported by the media, and perpetrated by groups more widely than just muslims. Just off the top of my head i can remember incidents perpetrated by the Earth Liberation Front (presumably a bunch of fringe west coast hippies), a wide variety of "animal rights" groups, who are predominately comprised of white people, the Sri Lankan Tamil Tigers (against the government of Sri Lanka, who themselves are arguably performing terrorist acts), anti-abortion activists (like the lunatic who drove his car into a medical clinic which did not perform abortions, or even refer people for abortions), and others. Terrorism is not a middle eastern problem. Assuming that it is is just fear mongering, and discrimination. Most middle easterners and muslims are not terrorists, just the same way that most of the Irish are not terrorists, the same way that most Basques are not terrorists. It's not fair to treat Muslims with suspicion, when nobody treats any of these other groups with suspicion, they are just as innocent or just as guilty by association as everyone else.
And if you really want to bring up Israel, remember there are fundamentalist Israelis who go shoot Palestinian children as well as Palestinians who go blow up Israelis.
It's a ridiculous thing to consider. What about all the rest of the people who come from those regions?
At what point does a trend justify profiling people? 100 incidents? 50? 10? 3?
What about consideration for proportionality? For every psychotic killer, how many normal non-psychos are out there? What about all the people in the middle east who aren't suicide bombers? If there is only 1 suicide bomber for every 4 million people, then it seems like you're getting an awful lot of false positives if you are identifying people simply by ethnicity, or race, or nationality. And if those factors are only tangentially relevant to the main factors, why bother including them at all? If you went to psycho bomber school, then it doesn't matter if you're Pakistani, Dutch, Chinese, or Bolivian, you went to psycho bomber school! So why don't we try and identify psychopaths, rather than try and figure out whether brown people are more likely to be psychotic or not? (which itself is sort of an odd question, because nobody ever asks "is group y more likely to be psychotic than group x, and that's what really causes fear mongering, no context.)
Likewise, how many people are there in the US who wear bandanas and don't commit crimes? Should you just throw anybody with a bandana in jail? Or be quicker to taser them than other people?
The problem with profiling, or any sort of statistical reasoning is that you must must must make sure that you have the correct data set, and that you are looking at the correct indicators, or you will get incorrect results. And if there's anything the US has proven in the past 5 years, its that they don't understand the middle east, they don't understand it culturally, they don't understand it psychologically, and they don't understand it politically. I have absolutely zero faith that the US government, and most importantly the people who are being asked to implement what are statistical judgements (police officers, TSA bag screeners, custom officials) are capable of understanding and recognizing the nuances required to make good decisions.
So, to sum up, racial profiling is just garbage, and other forms of profiling need to be done by professionals (like social workers), and so should not be implemented widely, because they are not scalable filters.
Fine, then i hope you understand when i taser every white kid with a crew cut who driving a pickup truck, cause you never know which of 'em is gonna be the next Timothy McVey.
Racial Profiling is stupid, ineffective, and an unjustifiable abuse of statistical reasoning.
It depends radically on the nature of the academic institution you're in. I attended the Ohio State University, which is a publically owned and funded institution. As a result, any resident of Ohio is entitled to use the physical library (for some reason they do not extend the use of their electronic holdings to the public, why i don't really know), as well as a number of other University functions (such as course syllabaries, and what texts are being used for a course).
Private universities are entitled to do whatever the hell they feel like (within the realm of the law of course), which is why students at private universities don't technically have the right to protest on university grounds, unless the university approves (no right to public free speech if you're not in public).
Now, as to what policies UCLA (which is part of the public University of California system) has on it's libraries, i don't know, but i can certainly say that this particular incident is not indicative of all institutions of higher learning in the US.