yep...regardless of dictionary definition issues this is a bit like trying to trying to argue "Burger King" is trying to redefine the word "King."
I mean come on they're not really the King of burgers?
Because randomly stimulating a single nerve tract to cause the release of neuro -ransmitters is easy while the more precise stimulation needed for perception is very hard (and almost not at all understood). For those unfamiliar with neuro-anatomy the vagus nerve goes throughout the body and (primarily) picks up signals on heart rate etc. that are associate with an increased level of arousal. This nerve projects (again primarily) to the amygdala the brains "emotional center" so your brain knows your scared, excited, elated or any other state of high arousal (because often your body reacts before your brain does, altough this is a simplistic explanation) So all this device seems to do is cause a higher level of arousal. It doesn't seem to do much more than current drugs already do, except maybe that the level of control is more precise than just popping a pill.
McDonough's book "Cradle to Cradle" said pretty much the same thing, that is, current recycling methods don't work. He argues that the reason is that current products (aluminum cans etc.) weren't designed from the get go to be recyclable. Products need to be designed with an efficient life cycle in mind (thus the title "Cradle to Cradle"). Not only will this reduce "waste" but it will eliminate the concept of "waste" and in the long term it will be cheaper than current methods. The point is the traditional "reduce, reuse, recycle" only marginally slows down the invetitable approach to an ecological castastrophe but designing products from "cradle to cradle" stops the approach. Personally, I think he has a little too much faith in the ability of science and engineering to develop ecological life cycles but it's a nice idea.
It seems nobody on Slashdot has considered the other side of the coin . If the alternatives are (for example) Google offering a pitifully filtered search engine or the chinese government forcing it's own search engine on the people then what should google do? I realize that such censorship is unacceptable but change takes time so it would be better if the Chinese people could have at least some access to the outside world while people (i.e. everybody whining on Slashdot) actively fights for social reform in China. I'm sure I'll hear about how no accomadation is acceptable (most likely with many an invocation of Godwin's law) but Rome wasn't built in a day and it's apparent to me that nobody on Slashdot (myself included) understands the situation enough to say that American coporations should allow absolutely no censorship in China.
What many people have pointed out is that we use bad grammar simply because most of the prescriptive rules of grammar come from efforts to distinguish affluent speaker from "commoners" and have no real proven benefit as far as "effectively communicating." In fact there is more of a reason to use "poor" (in the prescriptive sense) grammar rather than "good" grammar because poor grammar can be written and understood more quickly good grammar. For example, in some circles "ur" is written and understood much more quickly "you're" or "your". Even if there is some cost due to confusing "you're" with "your" when typing "ur", "ur" might still end up having an overall higher net benefit than the other two words. In fact I wouldn't be suprised if "ur" some day becomes the dominant form of both words. My point is that language evolves organically and there is not, nor has there ever been, nor will there ever be a "correct" form of any language (just as the there will never be a "most highly evolved species") simply because there in no priviledged vantage point from which to view language. If we followed prescriptive we'd still be pronouncing make as "mach" and for no good reason (well asside from the spelling thing).
I hate these comparisons too. AND they're not even useful for predicting when we can simulate a "fully functioning brain." All of these predictions are based on equating neurons (which for one aren't the only "computations" going on in the brain) with simple transistor-like units (e.g.Perceptrons ). The truth is that when a neuron fires this leads to many possible different chemical cascades resulting in the production or destruction of neuro-transmitter, neuroreceptors and who knows what else. Talk to a neurscientist doing single cell research and they'll tell you that neuron is to perceptron as Boeing 747 is to paper airplane. Maybe you can learn something useful by using the modern computer as an analogy for the brain but it won't get you that far.
On a different note I think from the article it's unclear whether they mean to say that the brain is not like a modern digital computer with ram and hard disks etc. (which is most definitely correct) or whether they're trying to say something as silly as a brain couldn't be modeled by an ideal Turing machine (I think it's a fact that any given physical could be modeled by a Turing machine, though I could be wrong).
The real problem, of course, is that blacklists are needed in the first place. If ISPs did their jobs a little better (aol, hotmail and the likes), the amount of spam would already decrease significantly. And don't speak to me about chinese ISPs, since most spam comes from the US.
I think that the most effective way to filter spam (in terms of maximum hits and least false postives) would be to have individual messages marked as spam. For example when a message is received a user can mark it as spam. The message would then be sent to some central server that would find statistical regularities (so that slight variations of the same message are also marked as spam) in the messages and block the message from being received by other users. (Doesn't Gmail already do this?) Heck individuals you could could SET the hit/false alarm ratio (based on how the algorithm has worked in the past) to whatever they think is optimal.
Of course this means that some people will have to receive the message first but after spammers see a dastric decrease in the amount of return they may stop spamming as much. Of course the spammers might try to counter the filter by having clever algorithms that generate unique messages or by hiring super poor people to compose a variety of different spams but I seriously doubt they'll be able to keep up with the awesome power of Google!!!
true enough for this article. I guess I'm just commenting on a more general trend I see of bad neuroscience and psychology on slashdot and the mainstream media (as well as in the field). These studies aren't neccessarily bad but the way they are discussed often is.
as a graduate student in psychology I am upset by the frequent unscientific articles posted on slashdot relating to my field. Here are a couple of points that I think need to be made:
1. All of you probably know this but I'll just emphasize the point: "correlation is not causation."
2. All these "facts" are based on very questionable statistical techniques. I won't go into these techniques here but the idea in these experiments is that you want to make sure the effect of some manipulated variable is above chance (statistical significance). There are two ways to do this: have a large effect (e.g. eating breakfast causes a 10% increase in IQ) or just get lots of people so that even a small effect is not very likely by chance (e.g. eating breakfast causes a.000001% increase in IQ but in 10,0000) people. I can assure most of the results reported in this article are based on studies that are closer to the later method than the former. Furthermore this.00001% increase (is in the previous example) is an AVERAGE. That it is, for you in particular eating breakfast may even decrease your IQ. That is, infering something is true about an individual from the group that individual belongs to is a fallacy. There's even a name for this fallacy (the ecological fallacy).
3. Finding a brain area that is "activated" (fMRI) during a certain cognitive task is about as unexciting as learning that people use their hands to perform a physical task. OF COURSE, there is some area (or mostly likely areas) of the brain that is used in cognitive task just as of course there are areas of the body that perform "physical" tasks. Furthermore, it's unclear if "activation" should mean anything at all. Activation is defined as some small change in blood flow (which is correlated with neural activity) as measured via the oxygen levels picked up by a fMRI machine. Again, these difference results just need to be statistically significant to get published so that who knows what that meeds for individuals.
Little offtopic but shoddy science is the bane of all true geeks damn it.
....you are all wrong. People are notoriously bad http://www.analytictech.com/mb021/notes_on_gilovic h.htm at finding patterns in certain things and the economy happens to be one of those things. The problem is two fold: One) We're likely two see random occurences as evidence for our position. For example, if all our firends lose their jobs at once this is a rare occurence but not unlikely to happen to at least a couple people given that there are some many people in the US. If you flipped a coin a million times you might end up with a run of 100 heads in a row. The second problem is that even if people do use real evidence (ie unemployement rates) they are likely to associate it with the wrong cause. For example, Andrew Jackson ruined the economy by devaluing paper money by making gold the only way to buy land. He was long gone before the the disastrous effects of his decision were apparent. So why don't we blame clinto, bush sr., or even FDR instead of Bush jr.? It's sad to say but human beings are cognitively unable to associate fluctuations in the economy with the current present. (at least without a huge amout of studying or in rare a situation where it is completely obvious to everyone that the current president screwed up).
Just as a side note on blaming presidents for things they had not much to do with (ie the economy), incubent presidents are way more likely to lose in areas were there has been natural disasters. apparently they get blamed for not fixing (and sometimes not preventing) natural disasters. Looks like bush might not win florida this year.....
Slashdot Mirror
yep...regardless of dictionary definition issues this is a bit like trying to trying to argue "Burger King" is trying to redefine the word "King." I mean come on they're not really the King of burgers?
Because randomly stimulating a single nerve tract to cause the release of neuro -ransmitters is easy while the more precise stimulation needed for perception is very hard (and almost not at all understood). For those unfamiliar with neuro-anatomy the vagus nerve goes throughout the body and (primarily) picks up signals on heart rate etc. that are associate with an increased level of arousal. This nerve projects (again primarily) to the amygdala the brains "emotional center" so your brain knows your scared, excited, elated or any other state of high arousal (because often your body reacts before your brain does, altough this is a simplistic explanation) So all this device seems to do is cause a higher level of arousal. It doesn't seem to do much more than current drugs already do, except maybe that the level of control is more precise than just popping a pill.
McDonough's book "Cradle to Cradle" said pretty much the same thing, that is, current recycling methods don't work. He argues that the reason is that current products (aluminum cans etc.) weren't designed from the get go to be recyclable. Products need to be designed with an efficient life cycle in mind (thus the title "Cradle to Cradle"). Not only will this reduce "waste" but it will eliminate the concept of "waste" and in the long term it will be cheaper than current methods. The point is the traditional "reduce, reuse, recycle" only marginally slows down the invetitable approach to an ecological castastrophe but designing products from "cradle to cradle" stops the approach. Personally, I think he has a little too much faith in the ability of science and engineering to develop ecological life cycles but it's a nice idea.
It seems nobody on Slashdot has considered the other side of the coin . If the alternatives are (for example) Google offering a pitifully filtered search engine or the chinese government forcing it's own search engine on the people then what should google do? I realize that such censorship is unacceptable but change takes time so it would be better if the Chinese people could have at least some access to the outside world while people (i.e. everybody whining on Slashdot) actively fights for social reform in China. I'm sure I'll hear about how no accomadation is acceptable (most likely with many an invocation of Godwin's law) but Rome wasn't built in a day and it's apparent to me that nobody on Slashdot (myself included) understands the situation enough to say that American coporations should allow absolutely no censorship in China.
cat fishes you!
What many people have pointed out is that we use bad grammar simply because most of the prescriptive rules of grammar come from efforts to distinguish affluent speaker from "commoners" and have no real proven benefit as far as "effectively communicating." In fact there is more of a reason to use "poor" (in the prescriptive sense) grammar rather than "good" grammar because poor grammar can be written and understood more quickly good grammar. For example, in some circles "ur" is written and understood much more quickly "you're" or "your". Even if there is some cost due to confusing "you're" with "your" when typing "ur", "ur" might still end up having an overall higher net benefit than the other two words. In fact I wouldn't be suprised if "ur" some day becomes the dominant form of both words. My point is that language evolves organically and there is not, nor has there ever been, nor will there ever be a "correct" form of any language (just as the there will never be a "most highly evolved species") simply because there in no priviledged vantage point from which to view language. If we followed prescriptive we'd still be pronouncing make as "mach" and for no good reason (well asside from the spelling thing).
Just wanted to say that this is probably THE most intelligent psychology related post to slashdot. Kudos.
I hate these comparisons too. AND they're not even useful for predicting when we can simulate a "fully functioning brain." All of these predictions are based on equating neurons (which for one aren't the only "computations" going on in the brain) with simple transistor-like units (e.g.Perceptrons ). The truth is that when a neuron fires this leads to many possible different chemical cascades resulting in the production or destruction of neuro-transmitter, neuroreceptors and who knows what else. Talk to a neurscientist doing single cell research and they'll tell you that neuron is to perceptron as Boeing 747 is to paper airplane. Maybe you can learn something useful by using the modern computer as an analogy for the brain but it won't get you that far.
On a different note I think from the article it's unclear whether they mean to say that the brain is not like a modern digital computer with ram and hard disks etc. (which is most definitely correct) or whether they're trying to say something as silly as a brain couldn't be modeled by an ideal Turing machine (I think it's a fact that any given physical could be modeled by a Turing machine, though I could be wrong).
The real problem, of course, is that blacklists are needed in the first place. If ISPs did their jobs a little better (aol, hotmail and the likes), the amount of spam would already decrease significantly. And don't speak to me about chinese ISPs, since most spam comes from the US.
I think that the most effective way to filter spam (in terms of maximum hits and least false postives) would be to have individual messages marked as spam. For example when a message is received a user can mark it as spam. The message would then be sent to some central server that would find statistical regularities (so that slight variations of the same message are also marked as spam) in the messages and block the message from being received by other users. (Doesn't Gmail already do this?) Heck individuals you could could SET the hit/false alarm ratio (based on how the algorithm has worked in the past) to whatever they think is optimal. Of course this means that some people will have to receive the message first but after spammers see a dastric decrease in the amount of return they may stop spamming as much. Of course the spammers might try to counter the filter by having clever algorithms that generate unique messages or by hiring super poor people to compose a variety of different spams but I seriously doubt they'll be able to keep up with the awesome power of Google!!!
true enough for this article. I guess I'm just commenting on a more general trend I see of bad neuroscience and psychology on slashdot and the mainstream media (as well as in the field). These studies aren't neccessarily bad but the way they are discussed often is.
http://www.jsmf.org/about/s/badneuro/archive.htm
as a graduate student in psychology I am upset by the frequent unscientific articles posted on slashdot relating to my field. Here are a couple of points that I think need to be made:
.000001% increase in IQ but in 10,0000) people. I can assure most of the results reported in this article are based on studies that are closer to the later method than the former. Furthermore this .00001% increase (is in the previous example) is an AVERAGE. That it is, for you in particular eating breakfast may even decrease your IQ. That is, infering something is true about an individual from the group that individual belongs to is a fallacy. There's even a name for this fallacy (the ecological fallacy).
1. All of you probably know this but I'll just emphasize the point: "correlation is not causation."
2. All these "facts" are based on very questionable statistical techniques. I won't go into these techniques here but the idea in these experiments is that you want to make sure the effect of some manipulated variable is above chance (statistical significance). There are two ways to do this: have a large effect (e.g. eating breakfast causes a 10% increase in IQ) or just get lots of people so that even a small effect is not very likely by chance (e.g. eating breakfast causes a
3. Finding a brain area that is "activated" (fMRI) during a certain cognitive task is about as unexciting as learning that people use their hands to perform a physical task. OF COURSE, there is some area (or mostly likely areas) of the brain that is used in cognitive task just as of course there are areas of the body that perform "physical" tasks. Furthermore, it's unclear if "activation" should mean anything at all. Activation is defined as some small change in blood flow (which is correlated with neural activity) as measured via the oxygen levels picked up by a fMRI machine. Again, these difference results just need to be statistically significant to get published so that who knows what that meeds for individuals.
Little offtopic but shoddy science is the bane of all true geeks damn it.
....you are all wrong. People are notoriously bad http://www.analytictech.com/mb021/notes_on_gilovic h.htm at finding patterns in certain things and the economy happens to be one of those things. The problem is two fold: One) We're likely two see random occurences as evidence for our position. For example, if all our firends lose their jobs at once this is a rare occurence but not unlikely to happen to at least a couple people given that there are some many people in the US. If you flipped a coin a million times you might end up with a run of 100 heads in a row. The second problem is that even if people do use real evidence (ie unemployement rates) they are likely to associate it with the wrong cause. For example, Andrew Jackson ruined the economy by devaluing paper money by making gold the only way to buy land. He was long gone before the the disastrous effects of his decision were apparent. So why don't we blame clinto, bush sr., or even FDR instead of Bush jr.? It's sad to say but human beings are cognitively unable to associate fluctuations in the economy with the current present. (at least without a huge amout of studying or in rare a situation where it is completely obvious to everyone that the current president screwed up).
Just as a side note on blaming presidents for things they had not much to do with (ie the economy), incubent presidents are way more likely to lose in areas were there has been natural disasters. apparently they get blamed for not fixing (and sometimes not preventing) natural disasters. Looks like bush might not win florida this year.....