If you weigh just one penny, and use that as the average, then you have some total inaccuracy X. If you instead weigh 10 pennies and divde the weight by 10, the inaccuracy is much less: roughly X/10.
Er, not if your error is asymmetric and nonlinear, which is usually the case for dodgy lab equipment. Suppose your scale was introducing a positive error of +5 for each coin and this error increased by +1 for each five coins on the scale. With one coin, the error is +5. With ten coins, the error is +5 for each of the first five coins and +6 for each of the next five coins. When averaging five coins, the error is +5. When averaging ten coins, the error is +5.5. 100 coins is left as an exercise for the reader.
(Warning: verbage follows!)
Well, we can certainly agree that the presentation of equations in physics lectures without a proper discussion is a poor method of teaching. In my experience, a professor who does this is either too brilliant to understand that students don't live and breathe mathematics in the same way was he does (unlikely) or is not familiar enough with the work to present a rigorous investigation (much more likely).
It is very annoying when professional scientists dismiss queries with equations. You may or may not be familiar with the plight of the late Eric Laithwaite and his presentation to the Royal Society of the non-Newtonian properties of gyroscopes. Laithwaite, a famous electrical engineer who pioneered the linear motor, essentially suggested that the behaviour of gyroscopic systems was, under certain cirumstances, non=Newtonian. Laitewaite was laughed out of the scientific establishment for this claim and he never received an explanation of his phenomena from any of the many physicists who attacked him. It took Laithewaite some years to derive and understand the (rather complicated) system of equations which demonstrated that the observed behaviour was completely Newtonian (of course). His fellows at the Royal Society had such faith in the equations of Newtonian physics that they did not need such an explanation.
However, I still disagree with your implicit statement that the equations are different from, or less efficient than, a more verbose (and less mathematical) explanation of the "reasons things actually happen". Clearly, the math does not drive the physical process - I suspect that even Wolfram would agree that his automata are a human construction (although he may claim that they are the best possible construction). However, I am confident that, given a physical process, math can be used to describe that process to an arbitrary degree of accuracy and that this is the simplest way of doing so.
However, I am hopelessly biased because my local university closed its physics department, leading me to do a math degree instead.
Given that you're not really going to memorize most of the equations -- they're not really the "take home" knowledge that you retain at the end of the semester, but the concepts are, it's better to pay attention to those.
Rubbish. If the equations you have don't express the concepts, you're using the wrong equations. A good set of equations will provide all the concepts in a handy, compact form, provided your mathematics is good enough to understand them.
Look, this whole 5 senses thing goes back to Aristotle. He was just trying to find some order in a chaotic world. So the dude was wrong. Give him a break, he's dead, ok?
Yeah. He was wrong. That's OK. Trouble is, he was wrong about just about every single thing he tried, and then got cited as an unassailable authority by just about everyone in Europe for over a thousand years.
I very much doubt that there were no skeptics of Aristotle in this time. I also very much doubt that these same skeptics were not at the pinnacles of their respective disciplines of their time.
For starters mathematics isn't a science - it tells us nothing about the physical world, mathematics is a tool, nothing more, nothing less. I have the highest esteem for mathematics, but it's not science.
Stephen Wolfram and many other mathematicians would disagree. Mathematics has moved on since you took Calculus 101.
Recent work on complex systems indicates that mathematical fundamentals have a more direct influence on the shape of the physical world than the current generation of physicists accepts. This is expected because mathematics is formulated using our experiences of the physical world. Wolfram's book was righly criticised in Nature and is extremely long winded. However, it is the most significant book which attempts to demonstrate this direct influence which is not written by Mandlebrot.
Many physicists think that mathematics is nothing more than a tool, albeit an essential one. The closest analogy to this assertation that I can think of would be a technician describing a physics textbook as 'just a tool' because he is using it to force down a switch on a broken oscilloscope. In this situation, the textbook reveals how the oscilloscope is constructed but it can also be used to get things working.
Ok. So using your own approach, the fact that your fellow human, Jeff Dahlmer, ate young boys parts means that you have no sense of morality.
See the problem with your line of reasoning?
You'll need to work a little harder on your argument if you want to be taken seriously. Making up fragile, poorly reasoned analogies won't cut it.
You are right - that makes no sense. I will therefore refine my argument with your help.
I propose that it is possible to divide a group of animals into two categories: moral and immoral. I will define a moral animal as one which makes a moral choice. According to this definition, the morality of any group of animals can be tested by testing the morality of a large number of individual animals. On completion of this testing, if immoral behaviour is the exception rather than the rule, the species of animal is deemed immoral. In this way, the test is not affected by the actions of unrepresentative members of the species.
For cats, my test of morality is inspired by our discussion. It is known that in certain circumstances, some mother cats have eaten some of their kittens. Regardless of the circumstances, I will define this as an immoral act. My test of morality is to put a mother cat in a situation where it is known that cats have eaten their kittens in the past and observe whether the mother cat behaves immorally.
As with any test, the many variables such as the exact nature of the situation and breed of cat can be chosen to obtain a desired result. However, I predict that when a significant proportion of the many possible values of these variables have been tested, it will be found that moral behaviour is the exception and that cats are immoral.
To perform tests of this nature would be immoral. I suspect that it will never be done. However, I think you will agree that while this remains a thought experiment, this is a strong argument that cats are immoral.
Eh? What part of that are you unfamilar with? Any cat owner who has been around birthing cycles knows about mothers not eating the kittens. Do I really have to explain this to you?
It's rather simple, really. You attempted to assert that cats have a sense of morality by observing that "Mothers rarely eat their young.". The fact that mothers eat their young at all is clearly immoral behaviour and implies that cats do not have any sense of morality.
However, the basis of your misconceptions is a confusion of morality (which is a property of decisions of choice) and instinct (which is a property of uncontrollable decisions).
"Cats have morals. Mothers rarely eat their young. Cats rarely eat their owners, unless the owner dies. Even then, some cats cannot overcome that predjudice, though they will eat other animals."
Top tip: follow your assertations by statements which support them.
Actually, the answer is "Nobody leaves." Since they cannot determine their own eye colour by communicating with anyone else and the Guru's statement provides no information about the colour of their own eyes, they cannot be certain of the colour of their eyes. No amount of 'logical deduction', or indeed any other action, will change this because there are no constraints on eye colour.
Given this situation, it's hard to define the boundaries that separate artificial intelligence, pattern recognition, statistical inference and classification and the rest.
Actually, it's quite easy - you just start with proper definitions. Classification separates data into categories which fit given criteria. Statistical inference works out the chance of an event occuring given a data set of previous, related occurences. Pattern recognition labels given patterns in data.
Classification separates out the data. This is different to pattern recognition which matches sequences of data to ones which have been previously defined. This is again different to statistical inference which must involve the use of chance.
Slashdot Mirror
Hasn't this been done already?
You mean to tell me that an animal that can navigate using sonar can control the pitch of sound? This is groundbreaking science!
"...they are getting paid very little in comparison to bloated unionized factories"
:(
OK, I can't be the only person who read this as un-ionized and read the entire article looking for a reference to vapours or high voltages? Can I?
What a dumbass. You intentionally bypassed the schools internet filter by setting up your own proxy server (dumb in and of itself).
Only on Slashdot could bypassing network security by setting up a proxy server be described as "dumb".
If you weigh just one penny, and use that as the average, then you have some total inaccuracy X. If you instead weigh 10 pennies and divde the weight by 10, the inaccuracy is much less: roughly X/10.
Er, not if your error is asymmetric and nonlinear, which is usually the case for dodgy lab equipment. Suppose your scale was introducing a positive error of +5 for each coin and this error increased by +1 for each five coins on the scale. With one coin, the error is +5. With ten coins, the error is +5 for each of the first five coins and +6 for each of the next five coins. When averaging five coins, the error is +5. When averaging ten coins, the error is +5.5. 100 coins is left as an exercise for the reader.
Know your errors.
(Warning: verbage follows!)
Well, we can certainly agree that the presentation of equations in physics lectures without a proper discussion is a poor method of teaching. In my experience, a professor who does this is either too brilliant to understand that students don't live and breathe mathematics in the same way was he does (unlikely) or is not familiar enough with the work to present a rigorous investigation (much more likely).
It is very annoying when professional scientists dismiss queries with equations. You may or may not be familiar with the plight of the late Eric Laithwaite and his presentation to the Royal Society of the non-Newtonian properties of gyroscopes. Laithwaite, a famous electrical engineer who pioneered the linear motor, essentially suggested that the behaviour of gyroscopic systems was, under certain cirumstances, non=Newtonian. Laitewaite was laughed out of the scientific establishment for this claim and he never received an explanation of his phenomena from any of the many physicists who attacked him. It took Laithewaite some years to derive and understand the (rather complicated) system of equations which demonstrated that the observed behaviour was completely Newtonian (of course). His fellows at the Royal Society had such faith in the equations of Newtonian physics that they did not need such an explanation.
However, I still disagree with your implicit statement that the equations are different from, or less efficient than, a more verbose (and less mathematical) explanation of the "reasons things actually happen". Clearly, the math does not drive the physical process - I suspect that even Wolfram would agree that his automata are a human construction (although he may claim that they are the best possible construction). However, I am confident that, given a physical process, math can be used to describe that process to an arbitrary degree of accuracy and that this is the simplest way of doing so.
However, I am hopelessly biased because my local university closed its physics department, leading me to do a math degree instead.
Rubbish. If the equations you have don't express the concepts, you're using the wrong equations. A good set of equations will provide all the concepts in a handy, compact form, provided your mathematics is good enough to understand them.
FOREIGNER
I don't have a computer, you insensitive clod!
MOD PARENT er SOMEHOW?
Recent work on complex systems indicates that mathematical fundamentals have a more direct influence on the shape of the physical world than the current generation of physicists accepts. This is expected because mathematics is formulated using our experiences of the physical world. Wolfram's book was righly criticised in Nature and is extremely long winded. However, it is the most significant book which attempts to demonstrate this direct influence which is not written by Mandlebrot.
Many physicists think that mathematics is nothing more than a tool, albeit an essential one. The closest analogy to this assertation that I can think of would be a technician describing a physics textbook as 'just a tool' because he is using it to force down a switch on a broken oscilloscope. In this situation, the textbook reveals how the oscilloscope is constructed but it can also be used to get things working.
You are right - that makes no sense. I will therefore refine my argument with your help.
I propose that it is possible to divide a group of animals into two categories: moral and immoral. I will define a moral animal as one which makes a moral choice. According to this definition, the morality of any group of animals can be tested by testing the morality of a large number of individual animals. On completion of this testing, if immoral behaviour is the exception rather than the rule, the species of animal is deemed immoral. In this way, the test is not affected by the actions of unrepresentative members of the species.
For cats, my test of morality is inspired by our discussion. It is known that in certain circumstances, some mother cats have eaten some of their kittens. Regardless of the circumstances, I will define this as an immoral act. My test of morality is to put a mother cat in a situation where it is known that cats have eaten their kittens in the past and observe whether the mother cat behaves immorally.
As with any test, the many variables such as the exact nature of the situation and breed of cat can be chosen to obtain a desired result. However, I predict that when a significant proportion of the many possible values of these variables have been tested, it will be found that moral behaviour is the exception and that cats are immoral.
To perform tests of this nature would be immoral. I suspect that it will never be done. However, I think you will agree that while this remains a thought experiment, this is a strong argument that cats are immoral.
Eh? What part of that are you unfamilar with? Any cat owner who has been around birthing cycles knows about mothers not eating the kittens. Do I really have to explain this to you?
It's rather simple, really. You attempted to assert that cats have a sense of morality by observing that "Mothers rarely eat their young.". The fact that mothers eat their young at all is clearly immoral behaviour and implies that cats do not have any sense of morality.
However, the basis of your misconceptions is a confusion of morality (which is a property of decisions of choice) and instinct (which is a property of uncontrollable decisions).
"Cats have morals. Mothers rarely eat their young. Cats rarely eat their owners, unless the owner dies. Even then, some cats cannot overcome that predjudice, though they will eat other animals." Top tip: follow your assertations by statements which support them.
Actually, the answer is "Nobody leaves." Since they cannot determine their own eye colour by communicating with anyone else and the Guru's statement provides no information about the colour of their own eyes, they cannot be certain of the colour of their eyes. No amount of 'logical deduction', or indeed any other action, will change this because there are no constraints on eye colour.
Someone set up him the $$$!
As an administrator, I'm not particularly intrested in a distribution that will "teach me the inner workings of Linux".
Surely, as an administrator, your job is to know the inner workings of Linux already?
It's artificial intelligence that's poorly defined, which is inevitable because intelligence is poorly defined.
I agree absolutely.
I wonder if anyone has done any research into whether men pretending to be women are worse at maths than men?
Hell they are probably watching your pr0n collection over your shoulder RIGHT now!
Well the joke's on them - I get off to goatse!
Given this situation, it's hard to define the boundaries that separate artificial intelligence, pattern recognition, statistical inference and classification and the rest.
Actually, it's quite easy - you just start with proper definitions. Classification separates data into categories which fit given criteria. Statistical inference works out the chance of an event occuring given a data set of previous, related occurences. Pattern recognition labels given patterns in data.
Classification separates out the data. This is different to pattern recognition which matches sequences of data to ones which have been previously defined. This is again different to statistical inference which must involve the use of chance.
How could it be simpler?
I got the bloopers straight away when I played the DVD and I had to play my old VHS copy to remove them. :7(
...obscure software engineering techiques review you?
I think that's the kind of laptop I've got but I can't seem to remember.