Well, I'm sure they could shed a lot of light on the problem if they would just count people who didn't answer the poll, as verified by an auditor. That is, require *any* attempted poll solicitation that doesn't get an answer to be included in the poll results on par with the other answers, and list it as a "didn't respond". If no one picks up the phone, it goes in the results. If a person refuses to take the poll, it goes in the results. Only dead numbers should be ignored, and even that's questionable.
I suspect that if they followed that procedure, polls would look something like:
"How effective do you think Obama has been at advancing his plans for health care reform?"
Results: 1% Very effective 1% Somewhat effective 1% Very little or not effective 97% No reponse Margin of error +/- 3%
Does the government or Tesla know that money is fungible?
Is the government enforcing this requirement against the Big Three companies that took this loan? I mean, they pretty obviously pissed away this "capital investment loan" on operating costs just to keep their heads above the water.
Can Tesla shift all its assets to Tesla2, leave the debt with Tesla, and then have Tesla declare bankruptcy and stiff the government on its loan, like GM and Chrysler got to do with their assets and debt?
I thought they got a big loan from the government at obscenely low interest rate for a venture capital startup? (It was the program intended as a giveaway to the failing Big Three but dressed up as a program to increase fuel efficiency, and Tesla apparently called their bluff.)
From my understanding the Greek ideas were excellent but they lacked the organisation and ambition to create great scientific or technological works....
No, the idea of all celestial bodies including the "fixed stars" revolving in perfect circles around earth, "because a circle is the most perfect thing", and then not bothering to collect the empirical data to check its validity... no, that wasn't a good idea.
The idea that everything's made out of earth, fire, wind, and water, and don't bother testing the implications systematically? Not good either. Etc, etc.
And no, you can't excuse them on the grounds that they didn't have the resources to "get shit done" -- even when governments had such resources, it took centuries of Enlightenment scientists ("natural philosophers" as they called themselves) to correct the errors of the Greeks *before* they made progress in any of the areas on which the Greeks spoke. The Enlightenent would have made progress a lot faster if they never knew about the Greeks' science and had to start from scratch.
Of course, their *math* was great, and Archimedes was great, but that's about it.
Frankly, it's an embarassment that philosophers still take their ideas seriously. It's like, somehow they got everything wrong that could be tested... but they *must* have been correct about philosophy, right? Uh, no.
I'm *not* one to defend the Greeks, by any means whatsoever, and I agree that their mistakes steered Western science in a very wrong direction that required ages to correct.
Still, I'm not sure that's a fair criticism you made of the Socratic method. That sounds like a teacher implementing it poorly, if they don't intervene in the bullshitting to bring it back to reality. The Socratic method is a great way, nevertheless, to guide students through the thought process of how discoveries were made, and thus to better ground their knowledge beyond a collection of facts.
If students understand the grounding behind the knowledge, they can better remember it and adapt to it. Think about e.g. being taught the formula for a cone ex cathedra vs. having to consider different ways to derive it, what's the right and wrong way, and finally being gently guided to the answer.
Great points, but I've got to defend the Bayesian order;-)
There can be an objective Bayesian answer to this, if you infer the prior you're supposed to have from the problem statement. Here, since nothing else is said about it, you should go in with the prior that a person with two children has a 1/4 chance that they're both boys. Then, you update on the information given in the problem ("one of my two children is a boy born on tuesday"). I showed a Bayesian solution matching the correct answer in this post.
Also, frequentists statistics, contrary to received wisdom, do use priors, it's just that they obscure them so that it's harder for students using frequentists tools to understand what assumptions are being made and how they affect interpretation of the results.
Alternate method, using my favored version of Bayes's Theorem:
O(H|E) = O(H) * P(E|H) / P(E|~H), where O = odds, i.e. P = 1/(1+O)
Here: H = I have 2 boys. E = I have (at least) one boy who was born on a tuesday
Prior should put 1/4 chance on a given person with two children having 2 boys. O(H) = 1/(4-1) = 1/3
P(E|H) = one minus the probability both of the boys were born on non-Tuesdays = 1 - (6/7)^2 = 13/49
P(E|~H) probability of E given that you're in some part of the outcome space that does not have two boys. 2/3 of this space has one boy and one girl, and each boy/girl combination has 1/7 chance of the boy being born on tuesday.
P(E|~H) = (2/3)( (1/2)(1/7) + (1/2)(1/7) ) = 2/21
O(H|E) = (1/3) * (13/49) / (2/21) = 13/14
Convert back to probability: P(H|E) = 13/(14+13) = 13/27
Really? I thought they were supposed to avoid hypotheticals like the plague. At least, that's the line that lawyers and lawyer wannabes spout at me: "No, you can't expect a judge to consider hypotheticals, it's all just too complicated, they just have to rule on what's their and let the plebes guess what they'll rule next based on as few datapoints as possible."
I think that's referring to Amazon defending its exemption from sales taxes (as cross-state sales typically are, at least in practice), the argument being that it's not bearing its share of e.g. road upkeep costs for the products its delivering.
Of course, I disagree with that argument, since taxes are normally completely decoupled from provision of the government service they fund, but I'm just trying to clarify what (I think) it's referring to.
Except for the bit about "soon, in the same place". Wholesalers transfer the goods from varied places to a more convenient place for the consumer and over a regular, long-term, lengthy process.
Yes, in a trivial sense. But it's still important to distinguish the *kinds* of perceived values, because they have a much different impact on the world and on the (mis)allocation of resources. Here's a simple attempt at the distinction:
Type I perceived value: Bob pays X gallons of milk for a cow because he believes that, over time, it will yield a certain amount of milk and meat.
Type II perceived value: Bob pays X gallons of milk because he believes other people will soon, in the same place, pay him more than X gallons for the cow (and cows have been trading for higher and higher prices over the past day).
Type II causes severe, costly swings in the marketplace, while type I does not. It may be hard to externally distguish which type of perceived value is going on, but there is a real difference.
sparking interest. I would bet that more than a few students would see the computer model and think, "wow! I wonder how other shapes would interact with airflow?".
Right and he'll get interested in learning how to make shapes in the program and learning the magic number it spits out. Science is about learning how the "magic" works, not playing with its fruits.
Unless they're learning something about *how* the model generates the results -- which takes a lot of explaining even if you minimize the formula use -- all they'll get out of it is:
"Magic computer gives magic results that we compare to some experiment."
The most important thing to impart on the students is not any fact itself, but that nature is not magic, that we use models to understand and predict nature, and that you can learn how the models actually work if you try.
Since they won't have the time to learn how the CFD software works, even at the pseudocode level, I don't see this as being a very helpful demonstration.
Now, in a perfect system, students would have all the background to jump right in to CFD by age 16, but we don't have one, and I don't think that this school is an exception.
Are we talking about the same Michigan? I was on a flight that landed in Detroit at about 2pm CST, and on the way to my destination I saw NUMEROUS buildings just devastated by the earthquake, a lack of essential services, disruption of civil order, severe deterioration of roads and infrastructure, looting in broad daylight... you name it.
They must have been near the earthquake's epicenter.
Er, cancel that. If you look at the PDF disclosure form, it sure ain't from owning mutual funds. He holds a lot of stock in energy and drilling companies, though it's hard to compute the relative value of them compared to his entire portfolio because it only lists value codes (e.g. between 15k and 50k), but there are a lot of them.
You're not alone in suggesting that, but that seems like a strange mentality that's taking hold.
I work as an aerospace engineer. Let's say stringent new rules about aviation go in force because of some disaster or environmental concern, reducing demand for my skillset.
I can haz compensation for lower salary? Remember, it's not *my* fault for whatever caused those rules to be implemented.
Why do people seem to think they have a right to do exactly what they're doing right now, for the same wage, for as long as they want, demand be damned? (Hey, neat anagram.)
Slashdot Mirror
That's a "no", then?
All of them, or just those who survived being wrong to tell about it?
Well, I'm sure they could shed a lot of light on the problem if they would just count people who didn't answer the poll, as verified by an auditor. That is, require *any* attempted poll solicitation that doesn't get an answer to be included in the poll results on par with the other answers, and list it as a "didn't respond". If no one picks up the phone, it goes in the results. If a person refuses to take the poll, it goes in the results. Only dead numbers should be ignored, and even that's questionable.
I suspect that if they followed that procedure, polls would look something like:
"How effective do you think Obama has been at advancing his plans for health care reform?"
Results:
1% Very effective
1% Somewhat effective
1% Very little or not effective
97% No reponse
Margin of error +/- 3%
A few questions, though:
Does the government or Tesla know that money is fungible?
Is the government enforcing this requirement against the Big Three companies that took this loan? I mean, they pretty obviously pissed away this "capital investment loan" on operating costs just to keep their heads above the water.
Can Tesla shift all its assets to Tesla2, leave the debt with Tesla, and then have Tesla declare bankruptcy and stiff the government on its loan, like GM and Chrysler got to do with their assets and debt?
Seriously, how do they even get polls to work now, given that people try to avoid them like the plague?
I thought they got a big loan from the government at obscenely low interest rate for a venture capital startup? (It was the program intended as a giveaway to the failing Big Three but dressed up as a program to increase fuel efficiency, and Tesla apparently called their bluff.)
From my understanding the Greek ideas were excellent but they lacked the organisation and ambition to create great scientific or technological works. ...
No, the idea of all celestial bodies including the "fixed stars" revolving in perfect circles around earth, "because a circle is the most perfect thing", and then not bothering to collect the empirical data to check its validity ... no, that wasn't a good idea.
The idea that everything's made out of earth, fire, wind, and water, and don't bother testing the implications systematically? Not good either. Etc, etc.
And no, you can't excuse them on the grounds that they didn't have the resources to "get shit done" -- even when governments had such resources, it took centuries of Enlightenment scientists ("natural philosophers" as they called themselves) to correct the errors of the Greeks *before* they made progress in any of the areas on which the Greeks spoke. The Enlightenent would have made progress a lot faster if they never knew about the Greeks' science and had to start from scratch.
Of course, their *math* was great, and Archimedes was great, but that's about it.
Frankly, it's an embarassment that philosophers still take their ideas seriously. It's like, somehow they got everything wrong that could be tested ... but they *must* have been correct about philosophy, right? Uh, no.
I'm *not* one to defend the Greeks, by any means whatsoever, and I agree that their mistakes steered Western science in a very wrong direction that required ages to correct.
Still, I'm not sure that's a fair criticism you made of the Socratic method. That sounds like a teacher implementing it poorly, if they don't intervene in the bullshitting to bring it back to reality. The Socratic method is a great way, nevertheless, to guide students through the thought process of how discoveries were made, and thus to better ground their knowledge beyond a collection of facts.
If students understand the grounding behind the knowledge, they can better remember it and adapt to it. Think about e.g. being taught the formula for a cone ex cathedra vs. having to consider different ways to derive it, what's the right and wrong way, and finally being gently guided to the answer.
But was he impious because a court ruled it, or did a court rule it because he was impious? ;-)
Great points, but I've got to defend the Bayesian order ;-)
There can be an objective Bayesian answer to this, if you infer the prior you're supposed to have from the problem statement. Here, since nothing else is said about it, you should go in with the prior that a person with two children has a 1/4 chance that they're both boys. Then, you update on the information given in the problem ("one of my two children is a boy born on tuesday"). I showed a Bayesian solution matching the correct answer in this post.
Also, frequentists statistics, contrary to received wisdom, do use priors, it's just that they obscure them so that it's harder for students using frequentists tools to understand what assumptions are being made and how they affect interpretation of the results.
Alternate method, using my favored version of Bayes's Theorem:
O(H|E) = O(H) * P(E|H) / P(E|~H), where O = odds, i.e. P = 1/(1+O)
Here:
H = I have 2 boys.
E = I have (at least) one boy who was born on a tuesday
Prior should put 1/4 chance on a given person with two children having 2 boys.
O(H) = 1/(4-1) = 1/3
P(E|H) = one minus the probability both of the boys were born on non-Tuesdays
= 1 - (6/7)^2 = 13/49
P(E|~H) probability of E given that you're in some part of the outcome space that does not have two boys. 2/3 of this space has one boy and one girl, and each boy/girl combination has 1/7 chance of the boy being born on tuesday.
P(E|~H) = (2/3)( (1/2)(1/7) + (1/2)(1/7) ) = 2/21
O(H|E) = (1/3) * (13/49) / (2/21) = 13/14
Convert back to probability: P(H|E) = 13/(14+13) = 13/27
What's more, he was probably just a lying Cretan.
So, then, upper courts can say, "If this were the case, that would be my ruling", but lower courts can't? Why the difference?
Really? I thought they were supposed to avoid hypotheticals like the plague. At least, that's the line that lawyers and lawyer wannabes spout at me: "No, you can't expect a judge to consider hypotheticals, it's all just too complicated, they just have to rule on what's their and let the plebes guess what they'll rule next based on as few datapoints as possible."
Any examples of them doing what you say?
James Cameron called. He wants the plot to The Terminator back.
I think that's referring to Amazon defending its exemption from sales taxes (as cross-state sales typically are, at least in practice), the argument being that it's not bearing its share of e.g. road upkeep costs for the products its delivering.
Of course, I disagree with that argument, since taxes are normally completely decoupled from provision of the government service they fund, but I'm just trying to clarify what (I think) it's referring to.
Really? They sell an entire delivery within minutes of getting it and they only did so because of a brief upward trend in the price of that good?
Except for the bit about "soon, in the same place". Wholesalers transfer the goods from varied places to a more convenient place for the consumer and over a regular, long-term, lengthy process.
Yes, in a trivial sense. But it's still important to distinguish the *kinds* of perceived values, because they have a much different impact on the world and on the (mis)allocation of resources. Here's a simple attempt at the distinction:
Type I perceived value: Bob pays X gallons of milk for a cow because he believes that, over time, it will yield a certain amount of milk and meat.
Type II perceived value: Bob pays X gallons of milk because he believes other people will soon, in the same place, pay him more than X gallons for the cow (and cows have been trading for higher and higher prices over the past day).
Type II causes severe, costly swings in the marketplace, while type I does not. It may be hard to externally distguish which type of perceived value is going on, but there is a real difference.
Yes, we could definitely use more Jobs right now...
Unless they're learning something about *how* the model generates the results -- which takes a lot of explaining even if you minimize the formula use -- all they'll get out of it is:
"Magic computer gives magic results that we compare to some experiment."
The most important thing to impart on the students is not any fact itself, but that nature is not magic, that we use models to understand and predict nature, and that you can learn how the models actually work if you try.
Since they won't have the time to learn how the CFD software works, even at the pseudocode level, I don't see this as being a very helpful demonstration.
Now, in a perfect system, students would have all the background to jump right in to CFD by age 16, but we don't have one, and I don't think that this school is an exception.
Are we talking about the same Michigan? I was on a flight that landed in Detroit at about 2pm CST, and on the way to my destination I saw NUMEROUS buildings just devastated by the earthquake, a lack of essential services, disruption of civil order, severe deterioration of roads and infrastructure, looting in broad daylight ... you name it.
They must have been near the earthquake's epicenter.
Er, cancel that. If you look at the PDF disclosure form, it sure ain't from owning mutual funds. He holds a lot of stock in energy and drilling companies, though it's hard to compute the relative value of them compared to his entire portfolio because it only lists value codes (e.g. between 15k and 50k), but there are a lot of them.
You're not alone in suggesting that, but that seems like a strange mentality that's taking hold.
I work as an aerospace engineer. Let's say stringent new rules about aviation go in force because of some disaster or environmental concern, reducing demand for my skillset.
I can haz compensation for lower salary? Remember, it's not *my* fault for whatever caused those rules to be implemented.
Why do people seem to think they have a right to do exactly what they're doing right now, for the same wage, for as long as they want, demand be damned? (Hey, neat anagram.)