The more important part, how the brain can sublimate operating complex machinery so that it doesn't require conscious thought to operate, isn't explained here.
The vast majority of your brain is dedicated to nonconscious processing, and it's easy to imagine that a neural network could do these sorts of tasks. The mystery is how conscious thought operates at all.
The rest of your argument, however, is correct. There's a very nice paper by Beilock et al. (2002) that shows that experienced soccer players dribble better when distracted than when consciously focusing on the skill.
Why deal with this by creating a confrontation with officers? Why not simply state, "It is my right not to show what's in this bag. If you want to see it, I'll go back to the register and return it." This seems a lot easier, doesn't get you in trouble with the cops, and still makes your point.
Actually if you look on PubMed you'll find a number of experiments that do demonstrate quantum aspects to microtubules and the brain; no, it doesn't prove consciousness is causal from them, but there's something to this; Well, then, this isn't evidence for quantum consciousness, is it? Recent research shows that quantum mechanical phenomena are involved in smell, but no one thinks the olfactory epithelium (or the system of the odorant and the olfactor epithelium) is independently conscious.
I think at the very least it partly explains memory in the brain. Very much partly; LTP is probably far more important. Microtubules may play some role in general neuroplasticity, in which case their importance goes far beyond memory, but that doesn't mean they play some role in consciousness.
And regarding Libet, well, I'm not sure the arising of the urge to mash a button is really indicative of anything on a grander scale. Maybe not, but most people would still say that they decided to pushed the button, rather than that they had the urge to push the button and so they did. If nothing else, it provides a good reason to question our interpretation of free will and our introspective analysis of it, while simultaneously demonstrating a sound way of combining objective and subjective analysis.
That's not true. Most of the attacks on free will I have seen coming from the neuroscience front take the form of experimental observations, rather than theoretical belief. The most celebrated of these is Benjamin Libet's EEG experiment, which I already mentioned in another post on this article. Basically, Libet attached participants to an EEG and asked them to perform a simple motor task. He devised a special clock that subjects could use to determine the time an event occurred with about a 50 ms margin of error, as previous experiments had shown (the mechanics aren't important). By asking subjects to report when they first decided to act according to this clock, Libet determined that there was a discrepancy of 200 ms before the first spike on the EEG indicating intent to act and when the subject became aware of his/her intent to act. Obviously, if a subject's decision happens before he/she is consciously aware, it's difficult to accept that the will could be entirely free.
Daniel Wegner's book The Illusion of Conscious Will covers this experiment and others in a more in-depth fashion.
There are ways of reconciling free will and determinism, in a way that permits the assumption of moral responsibility (among the most famous is Harry G. Frankfurt's theory of higher-order volitions, which I believe is mostly correct). But, it seems to me, there's little room anywhere for a non-deterministic view of the mind.
Does quantum physics have a hole? Yes. Does neuroscience have a hole? No. No major neuroscientists believe in a quantum mechanical model of the brain, because we have not yet observed any conclusive evidence that anything nondeterministic is going on in the neuron itself. We can predict neuronal behavior from observations at a level where only classical physics applies. We can build cochlear implants, which take the place of a large number of neurons, using neuronal simulation. Our model is not so far off that we can really postulate, as Roger Penrose, quantum consciousness's main apologist, does, that the Platonic values of good and truth reside in some strange hidden enclave of space-time.
Except for Roger Penrose's mathematical arguments and Stuart Hameroff's conclusions that microtubules could exhibit quantum mechanical behavior, I'm unaware of any research supporting quantum consciousness. And even Penrose and Hameroff have to make assumptions about a "non-computable influence" to postulate free will, since the Schrodinger Equation is basically deterministic. As much as I would like to believe in quantum consciousness, as much as that would simplify things, it turns out that "free will" is probably not what it seems. To understand ourselves, we have to do more than enumerate a list of things we think identify us and then find some quantum mechanical phenomenon with which to identify each.
On the other hand, Benjamin Libet's 1985 EEG experiment showed that true free will is basically impossible. While this evidence does not preclude that we may have C. A. Campbell's minimal amount of free will required for moral responsibility, and no more, it establishes to a reasonable scientific certainty that, when it comes to topics involving the self and consciousness, introspection is often false. By demonstrating that there are other ways of knowing the nature of the subjective beyond introspection, however, it gives us hope for the future.
You're right that antioxidants in red wine might have some effect, but I would argue that the difference in the "buzz" one gets from red wine and the buzz one gets from vodka is minimal. Ethanol is a single molecule, C2H5OH. Remove the alcohol from red wine, and it is no longer psychoactive. If there's any objective difference between drinking wine and drinking beer, it's confined to effects on absorption and metabolism.
Subjective judgment of the "buzz" is likely highly dependent on social situation. Wine is a more refined drink, and people tend to think it's a more refined drink, and drink it in a more refined social situation. Does that mean it has a more refined buzz? Perhaps subjectively, but this more refined buzz is not something in the composition of the wine itself. Perhaps it's purely psychological, or perhaps it's related to the difference in consumption patterns between wine and vodka drinkers. There are too many variables involved to make any kind of accurate judgment.
Finally, it's far from a well-established scientific fact that mixing beer and liquor is a bad idea. One possibility is, when you start drinking liquor after you're already drunk from drinking too much beer, you can't really judge how much alcohol you're consuming and how fast you're consuming it, so you end up drinking too much and puking. There's also a school of thought that says that the difference is due to the carbonation, and another that says that it's due to the food or liquid content of the stomach. At the core, wine, beer, and hard liquor are all only alcohol delivery mechanisms.
I would argue that when you consume the alcohol is at least as important as how you consume te alcohol. The rate at which you consume alcohol determines how much alcohol is in your bloodstream at any given time, and different amounts of alcohol affect the brain in different ways (the neuropharamacology of alcohol is extremely complex, but Erowid is typically reliable when it comes to drugs). Also, there's the diurnal rhythm of alcohol metabolism. Getting drunk in the morning will leave you drunker longer than getting drunk at night, although, because acetaldehyde mediates many of the toxic effects of alcohol, if acetaldehyde metabolism doesn't follow the same rhythm, drinking in the morning could actually be better or worse for your brain. I'm sure there are some good papers on this topic, if you're interested.
No offense, but have you ever tried to find the Firefox beta on the website? Perhaps if the beta page were made more accessible, the Slashdot developers would. Also, the fact that this page exists before release isn't helping things.
The risers in most cases are actually unnecessary. I've built three computers with three different cases and never used them, because in each case, the board was offset enough from the back that they were completely unnecessary.
Also, sadly, usually 12V is not enough to create glorious displays of sparks.
"The bottom line is we still feel more comfortable using a Mac than a (Windows) PC," said Alan Paller, director of research for SANS.
But as Daines can attest, there are no guarantees.
"We're all sort of waiting with bated breath to see if any problem will happen and the jury is still out," said Thayer, the independent security consultant. "I don't think you'll find a consensus."
The article seems to be saying, "look, two people with Macs got infected with a virus! Now Windows is more secure than the Mac." For some reason, I trust the director of research at SANS more than this British chemical engineer or the "independent security consultant." Macs have never been immune to viruses, it's just that there are thousands of times fewer Mac viruses than PC viruses. And this is still the case.
But the reason OpenBSD is so secure is, really, that Theo is such a douche. He doesn't think that most people can code worth shit, and this attitude means that everything that gets into OpenBSD is analyzed and audited many, many times.
If the NSA could decrypt GPG-encrypted messages, it would have to have one of the following three things:
1. A miraculous mathematical advance that made the factoring of the product of two extremely large prime numbers much easier. (Unlikely.) 2. A quantum computer. (More unlikely.) 3. More conventional computer power than the rest of the world combined. (Extremely unlikely.)
All three are completely unrealistic. It is doubtful that the NSA can crack PGP, unless it's through a weakness in one of the symmetric ciphers and not the RSA/DH algorithim.
Well, yes, that's another hypothesis, but the function of sleep is still pretty ambiguous. While there are a few proven effects of sleep deprivation, little is known about the evolutionary purpose of sleep. You might want to look at this article from Nature for a more comprehensive and up-to-date review.
Provigil does seem to actually help stop you from crashing. There was a study (search PubMed if you're interested) where subjects stayed up for 64 hours straight. While those on some variation of amphetamine and those taking a placebo averaged 15 hours of recovery sleep, those on Provigil averaged 10.
It does appear that some stages of sleep (most notably, REM) are unnecessary, as certain drugs will nearly completely suppress them but people go on living their lives normally. One hypothesis suggests that REM merely exists to make waking easier, thus explaining why REM periods increase in duration as people are more rested. Really, no one knows much about sleep, so it does make sense to be cautious before embarking on some Great Sleep Deprivation Experiment.
I can testify that Provigil (modafinil) works incredibly well. The only side effect I've experienced is a tendency to exacerbate headaches. If you stay up too long, you begin to feel some peripheral effects of sleep deprivation. (Staying up for 48 hours straight results in some astereognosis, whether you're on modafinil or not.) It doesn't appear to be addictive, since it's easy to stop and doesn't seem to produce real cravings. It is, however, vaguely habit-forming, as you realize you can just take a pill whenever you're tired and feel completely normal.
There's another major benefit of modafinil over amphetamine-type stimulants: you can go to sleep if you want to. It doesn't stop you from sleeping, just remove the fatigue. And apparently, you spend more time in deep sleep and less time in shallower stages. It's also much easier to get up in the morning. Overall, if you can find someplace to get it, I highly recommend it. One word of caution, however: while it doesn't seem to have too many adverse effects when combined with alcohol, experiences with other substances are mixed.
You could also just use a setuid CGI in an htaccess-protected directory. There'd have to be a pretty big security hole in Apache and a security hole in your script for anything bad to happen.
And I quote, from the summary:...a totally wireless, paperless checkout process, called EasyPay...the receipt will be in your inbox within an hour...
I would suspect that, in reality, if you don't want to give them your email, you just have to go to the register and wait longer and then you get a paper receipt. But it's certainly possible that a clerk with an EasyPay system would enter in your email as <bob@example.com> in order to speed things up, thus depriving you of any kind of receipt whatsoever...
Since I'm not already a registered Apple customer, any clerktrooper asking me for my email, snail address, or any other data not required to complete the transaction when I try to purchase products gets the old Jedi Mind Trick: you place an appropriate number of Federal Reserve Notes (or other bits of nicely-decorated paper) in your hand, wave your hand in front of the clerk, and you say "You don't need to see my identification".
In that case, the clerk matches your wave, says "You don't need a receipt," and sells you the product.
True, but it's not as if IBM didn't know what was going on. This Wikipedia article explains the circumstances, for anyone who's interested in further details.
Microsoft doesn't exactly own the rights on the components, but it's pretty close to. According to the Anandtech article, they have a license to manufacture the CPU. Likely (personally, I haven't confirmed it, but it seems probable), they have a license to manufacture the rest of the components as well. So you're correct that they can find another supplier if someone screws up.
"Full rights" to the CPU would mean they could also, for example, modify the design and license others to use their modified design. Suddenly, Microsoft becomes a powerful player in the embedded processor market by selling other people a chip that contains the results of decades of IBM R&D. IBM would, most likely, never let anyone do that, not even Microsoft.
It's also been shown that antidepressants induce neurogenesis. I don't remember the exact studies, but it appears that if you block neurogenesis the antidepressants lose their effects.
Of course, there are also a few studies suggesting antidepressants are little more effective than placebo to begin with, so who knows.
One could make the argument that it wasn't Steve Jobs' position that changed, but rather the feasibility of flash-based players.
Flash memory is now significantly cheaper than when Jobs made his announcement that Flash players sucked. Remember, the iPod nano has almost as much storage space as the original iPod, but uses flash and costs less. When Jobs made his announcement, a 512 MB player would set you back $250-$300, but the iPod shuffle costs $100.
There's little chance for a Web-based solution for working with documents to become more feasible, beyond a sudden standardization of Javascript across all browsers, which appears unlikely.
Well, yeah, for most people you can't trace the psychosis to cannabis smoking.
It's all related to how much you smoke early in your life (adult cannabis use appears to have no effect) and your genetic makeup.
It is not always evident that marijuana has caused the psychosis. The study I'm referring to traced the first 25 years of the lives of 1000 individuals. It only shows correlation and not causation, but if you read the study, it's quite difficult to provide an alternative explanation for the discrepency in rates of schizophreniform disorder.
It's called "medium-term" cognitive impairment. It disappears after at most 6 weeks.
It's much worse for some than others.
There's evidence saying there is none or a statistically insignificant effect of marijuana usage on IQ. There are also plenty of aspects of the thought process an IQ test doesn't measure, like, say, short-term memory, so this can't be taken as conclusive evidence that marijuana isn't harmful to the brain.
Now, on the other hand, psychosis is indeed a serious, proven, permanent risk of pot-smoking.
Because it's not common for a Java applet to require authorization to run, whereas every ActiveX control must.
You don't get used to pressing "Accept" to run the Java applet on Linux because it happens so rarely. It doesn't become a routine.
Also, many browsers won't run unsigned Java applets outside of a sandbox at all, whereas Internet Explorer will (or at least used to) simply say "This applet is unsigned. Do you want to run it?" or something to that effect.
It's true that the combustion of gasoline isn't complete; if it were, you wouldn't get any pollutants. The combustion of any hydrocarbon produces (ideally) only H2O and CO2. All of the additional pollutants are products of incomplete combustion.
Slashdot Mirror
The more important part, how the brain can sublimate operating complex machinery so that it doesn't require conscious thought to operate, isn't explained here.
The vast majority of your brain is dedicated to nonconscious processing, and it's easy to imagine that a neural network could do these sorts of tasks. The mystery is how conscious thought operates at all.
The rest of your argument, however, is correct. There's a very nice paper by Beilock et al. (2002) that shows that experienced soccer players dribble better when distracted than when consciously focusing on the skill.
Why deal with this by creating a confrontation with officers? Why not simply state, "It is my right not to show what's in this bag. If you want to see it, I'll go back to the register and return it." This seems a lot easier, doesn't get you in trouble with the cops, and still makes your point.
That's not true. Most of the attacks on free will I have seen coming from the neuroscience front take the form of experimental observations, rather than theoretical belief. The most celebrated of these is Benjamin Libet's EEG experiment, which I already mentioned in another post on this article. Basically, Libet attached participants to an EEG and asked them to perform a simple motor task. He devised a special clock that subjects could use to determine the time an event occurred with about a 50 ms margin of error, as previous experiments had shown (the mechanics aren't important). By asking subjects to report when they first decided to act according to this clock, Libet determined that there was a discrepancy of 200 ms before the first spike on the EEG indicating intent to act and when the subject became aware of his/her intent to act. Obviously, if a subject's decision happens before he/she is consciously aware, it's difficult to accept that the will could be entirely free.
Daniel Wegner's book The Illusion of Conscious Will covers this experiment and others in a more in-depth fashion.
There are ways of reconciling free will and determinism, in a way that permits the assumption of moral responsibility (among the most famous is Harry G. Frankfurt's theory of higher-order volitions, which I believe is mostly correct). But, it seems to me, there's little room anywhere for a non-deterministic view of the mind.
Does quantum physics have a hole? Yes. Does neuroscience have a hole? No. No major neuroscientists believe in a quantum mechanical model of the brain, because we have not yet observed any conclusive evidence that anything nondeterministic is going on in the neuron itself. We can predict neuronal behavior from observations at a level where only classical physics applies. We can build cochlear implants, which take the place of a large number of neurons, using neuronal simulation. Our model is not so far off that we can really postulate, as Roger Penrose, quantum consciousness's main apologist, does, that the Platonic values of good and truth reside in some strange hidden enclave of space-time.
Except for Roger Penrose's mathematical arguments and Stuart Hameroff's conclusions that microtubules could exhibit quantum mechanical behavior, I'm unaware of any research supporting quantum consciousness. And even Penrose and Hameroff have to make assumptions about a "non-computable influence" to postulate free will, since the Schrodinger Equation is basically deterministic. As much as I would like to believe in quantum consciousness, as much as that would simplify things, it turns out that "free will" is probably not what it seems. To understand ourselves, we have to do more than enumerate a list of things we think identify us and then find some quantum mechanical phenomenon with which to identify each.
On the other hand, Benjamin Libet's 1985 EEG experiment showed that true free will is basically impossible. While this evidence does not preclude that we may have C. A. Campbell's minimal amount of free will required for moral responsibility, and no more, it establishes to a reasonable scientific certainty that, when it comes to topics involving the self and consciousness, introspection is often false. By demonstrating that there are other ways of knowing the nature of the subjective beyond introspection, however, it gives us hope for the future.
You're right that antioxidants in red wine might have some effect, but I would argue that the difference in the "buzz" one gets from red wine and the buzz one gets from vodka is minimal. Ethanol is a single molecule, C2H5OH. Remove the alcohol from red wine, and it is no longer psychoactive. If there's any objective difference between drinking wine and drinking beer, it's confined to effects on absorption and metabolism.
Subjective judgment of the "buzz" is likely highly dependent on social situation. Wine is a more refined drink, and people tend to think it's a more refined drink, and drink it in a more refined social situation. Does that mean it has a more refined buzz? Perhaps subjectively, but this more refined buzz is not something in the composition of the wine itself. Perhaps it's purely psychological, or perhaps it's related to the difference in consumption patterns between wine and vodka drinkers. There are too many variables involved to make any kind of accurate judgment.
Finally, it's far from a well-established scientific fact that mixing beer and liquor is a bad idea. One possibility is, when you start drinking liquor after you're already drunk from drinking too much beer, you can't really judge how much alcohol you're consuming and how fast you're consuming it, so you end up drinking too much and puking. There's also a school of thought that says that the difference is due to the carbonation, and another that says that it's due to the food or liquid content of the stomach. At the core, wine, beer, and hard liquor are all only alcohol delivery mechanisms.
I would argue that when you consume the alcohol is at least as important as how you consume te alcohol. The rate at which you consume alcohol determines how much alcohol is in your bloodstream at any given time, and different amounts of alcohol affect the brain in different ways (the neuropharamacology of alcohol is extremely complex, but Erowid is typically reliable when it comes to drugs). Also, there's the diurnal rhythm of alcohol metabolism. Getting drunk in the morning will leave you drunker longer than getting drunk at night, although, because acetaldehyde mediates many of the toxic effects of alcohol, if acetaldehyde metabolism doesn't follow the same rhythm, drinking in the morning could actually be better or worse for your brain. I'm sure there are some good papers on this topic, if you're interested.
No offense, but have you ever tried to find the Firefox beta on the website? Perhaps if the beta page were made more accessible, the Slashdot developers would. Also, the fact that this page exists before release isn't helping things.
The risers in most cases are actually unnecessary. I've built three computers with three different cases and never used them, because in each case, the board was offset enough from the back that they were completely unnecessary.
Also, sadly, usually 12V is not enough to create glorious displays of sparks.
"The bottom line is we still feel more comfortable using a Mac than a (Windows) PC," said Alan Paller, director of research for SANS.
But as Daines can attest, there are no guarantees.
"We're all sort of waiting with bated breath to see if any problem will happen and the jury is still out," said Thayer, the independent security consultant. "I don't think you'll find a consensus."
The article seems to be saying, "look, two people with Macs got infected with a virus! Now Windows is more secure than the Mac." For some reason, I trust the director of research at SANS more than this British chemical engineer or the "independent security consultant." Macs have never been immune to viruses, it's just that there are thousands of times fewer Mac viruses than PC viruses. And this is still the case.
But the reason OpenBSD is so secure is, really, that Theo is such a douche. He doesn't think that most people can code worth shit, and this attitude means that everything that gets into OpenBSD is analyzed and audited many, many times.
If the NSA could decrypt GPG-encrypted messages, it would have to have one of the following three things:
1. A miraculous mathematical advance that made the factoring of the product of two extremely large prime numbers much easier. (Unlikely.)
2. A quantum computer. (More unlikely.)
3. More conventional computer power than the rest of the world combined. (Extremely unlikely.)
All three are completely unrealistic. It is doubtful that the NSA can crack PGP, unless it's through a weakness in one of the symmetric ciphers and not the RSA/DH algorithim.
Well, yes, that's another hypothesis, but the function of sleep is still pretty ambiguous. While there are a few proven effects of sleep deprivation, little is known about the evolutionary purpose of sleep. You might want to look at this article from Nature for a more comprehensive and up-to-date review.
Provigil does seem to actually help stop you from crashing. There was a study (search PubMed if you're interested) where subjects stayed up for 64 hours straight. While those on some variation of amphetamine and those taking a placebo averaged 15 hours of recovery sleep, those on Provigil averaged 10.
It does appear that some stages of sleep (most notably, REM) are unnecessary, as certain drugs will nearly completely suppress them but people go on living their lives normally. One hypothesis suggests that REM merely exists to make waking easier, thus explaining why REM periods increase in duration as people are more rested. Really, no one knows much about sleep, so it does make sense to be cautious before embarking on some Great Sleep Deprivation Experiment.
I can testify that Provigil (modafinil) works incredibly well. The only side effect I've experienced is a tendency to exacerbate headaches. If you stay up too long, you begin to feel some peripheral effects of sleep deprivation. (Staying up for 48 hours straight results in some astereognosis, whether you're on modafinil or not.) It doesn't appear to be addictive, since it's easy to stop and doesn't seem to produce real cravings. It is, however, vaguely habit-forming, as you realize you can just take a pill whenever you're tired and feel completely normal.
There's another major benefit of modafinil over amphetamine-type stimulants: you can go to sleep if you want to. It doesn't stop you from sleeping, just remove the fatigue. And apparently, you spend more time in deep sleep and less time in shallower stages. It's also much easier to get up in the morning. Overall, if you can find someplace to get it, I highly recommend it. One word of caution, however: while it doesn't seem to have too many adverse effects when combined with alcohol, experiences with other substances are mixed.
You could also just use a setuid CGI in an htaccess-protected directory. There'd have to be a pretty big security hole in Apache and a security hole in your script for anything bad to happen.
Errr, an Apple Store.
...a totally wireless, paperless checkout process, called EasyPay...the receipt will be in your inbox within an hour...
And I quote, from the summary:
I would suspect that, in reality, if you don't want to give them your email, you just have to go to the register and wait longer and then you get a paper receipt. But it's certainly possible that a clerk with an EasyPay system would enter in your email as <bob@example.com> in order to speed things up, thus depriving you of any kind of receipt whatsoever...
Since I'm not already a registered Apple customer, any clerktrooper asking me for my email, snail address, or any other data not required to complete the transaction when I try to purchase products gets the old Jedi Mind Trick: you place an appropriate number of Federal Reserve Notes (or other bits of nicely-decorated paper) in your hand, wave your hand in front of the clerk, and you say "You don't need to see my identification".
In that case, the clerk matches your wave, says "You don't need a receipt," and sells you the product.
But when you need to return it, you're screwed.
True, but it's not as if IBM didn't know what was going on. This Wikipedia article explains the circumstances, for anyone who's interested in further details.
Microsoft doesn't exactly own the rights on the components, but it's pretty close to. According to the Anandtech article, they have a license to manufacture the CPU. Likely (personally, I haven't confirmed it, but it seems probable), they have a license to manufacture the rest of the components as well. So you're correct that they can find another supplier if someone screws up.
"Full rights" to the CPU would mean they could also, for example, modify the design and license others to use their modified design. Suddenly, Microsoft becomes a powerful player in the embedded processor market by selling other people a chip that contains the results of decades of IBM R&D. IBM would, most likely, never let anyone do that, not even Microsoft.
It's also been shown that antidepressants induce neurogenesis. I don't remember the exact studies, but it appears that if you block neurogenesis the antidepressants lose their effects.
Of course, there are also a few studies suggesting antidepressants are little more effective than placebo to begin with, so who knows.
One could make the argument that it wasn't Steve Jobs' position that changed, but rather the feasibility of flash-based players.
Flash memory is now significantly cheaper than when Jobs made his announcement that Flash players sucked. Remember, the iPod nano has almost as much storage space as the original iPod, but uses flash and costs less. When Jobs made his announcement, a 512 MB player would set you back $250-$300, but the iPod shuffle costs $100.
There's little chance for a Web-based solution for working with documents to become more feasible, beyond a sudden standardization of Javascript across all browsers, which appears unlikely.
Well, yeah, for most people you can't trace the psychosis to cannabis smoking.
It's all related to how much you smoke early in your life (adult cannabis use appears to have no effect) and your genetic makeup.
It is not always evident that marijuana has caused the psychosis. The study I'm referring to traced the first 25 years of the lives of 1000 individuals. It only shows correlation and not causation, but if you read the study, it's quite difficult to provide an alternative explanation for the discrepency in rates of schizophreniform disorder.
It's called "medium-term" cognitive impairment. It disappears after at most 6 weeks.
It's much worse for some than others.
There's evidence saying there is none or a statistically insignificant effect of marijuana usage on IQ. There are also plenty of aspects of the thought process an IQ test doesn't measure, like, say, short-term memory, so this can't be taken as conclusive evidence that marijuana isn't harmful to the brain.
Now, on the other hand, psychosis is indeed a serious, proven, permanent risk of pot-smoking.
Because it's not common for a Java applet to require authorization to run, whereas every ActiveX control must.
You don't get used to pressing "Accept" to run the Java applet on Linux because it happens so rarely. It doesn't become a routine.
Also, many browsers won't run unsigned Java applets outside of a sandbox at all, whereas Internet Explorer will (or at least used to) simply say "This applet is unsigned. Do you want to run it?" or something to that effect.
It's true that the combustion of gasoline isn't complete; if it were, you wouldn't get any pollutants. The combustion of any hydrocarbon produces (ideally) only H2O and CO2. All of the additional pollutants are products of incomplete combustion.
It is also true that a good deal of energy in an internal combustion engine is lost as heat, but there are multiple sources of inefficiency. See Wikipedia's page on the operation of an internal combustion engine for more.