Nonsense. Most models are wrong. They're still enormously useful compared to something that's more wrong. Newtonian mechanics is wrong, but it was -- and still is -- very useful for the overwhelming majority of situations.
It is very wrong to say the earth is flat. There are many, many ways of demonstrating its wrongness and assuming the earth is flat will lead you to wildly incorrect conclusions for many problems. It is less wrong to say the earth is a sphere. However, it's harder to demonstrate that it's wrong, and you can do many useful calculations assuming a sphere for simplicity. It's also wrong, but not very much, to say the earth is a slightly squashed sphere. It requires very careful measurement to demonstrate this, and it's such an accurate approximation to make that it's rare to see someone actually model the earth's correct shape.
Many of the image-processing solutions were very clever ways of hiding bugs in code. Some were more likely to pass a code review than others. Many of them would not really pass dedicated testing. The winning entry that you mention does character-substitution in an ASCII PPM file, replacing all digits with zeroes. If you look at the file in an image browser, it's actually redacted. If you look at the file in a text editor and have security in mind, you will immediately be very suspicious. The nice bit is that the code is very short and the error subtle enough that it's very easy to overlook the problem.
Besides being unnecessarily insulting, you're actually comparing two different things. (For one, don't claim "a massive amount" when you reference a single study and fail to link to that study. The paper is this.)
What you said is:
increases in the weight of passengers is also putting a drag on fuel economy. As we get fatter and fatter, it obviously takes more energy to move us around.
Which is simply not true. Car mass only affects fuel efficiency of acceleration and it affects it linearly in total car + contents mass.
What's supported by the article is your second point: that people buy more SUVs as a result of being heavier. However, they fail to separate out the other factors driving SUV purchase (a fact they freely admit but you don't).
Actually, many compact cars have excellent legroom. Some don't, just as some larger cars don't. I'm as tall as you, and my Prius has acceptable legroom -- better than some previous, much larger cars.
We can only construct pairs of files that have the same MD5 hash. There's no way to efficiently produce a file that has a particular, previously-published hash. However, in practice you wouldn't use MD5 anyway, but rather SHA-256 or better.
One reasonably good solution is to sign the hash using a key that is already stored in a difficult-to-modify part of the device.
An MD5 that you then store on disk along with the refresh image? Not so helpful. You could sign the MD5 hash, but then where do you put the public key to verify the signature. (At that point, at least, you've made it incrementally harder for malware to pull this off, since a fair bit of stuff has to be changed.)
Possibly, though it's also very likely they've developed a good model of the difference between an intentional and unintentional lane change. (To be fair, I've seen people on the road where it was hard to tell which it was. Maybe the car will teach them to drive better. Ha!)
Mercury has a pretty low vapor pressure. Once broken, the mercury vapor will recondense into liquid mercury.
Mercury vapor is more hazardous than liquid mercury, primarily because it enters through the lungs instead of through ingestion or skin contact. However, elemental mercury (both vapor and liquid) is fairly safe, particularly in the quantities we're talking about, because it absorbs so poorly into the bloodstream. Most legitimate mercury hazards are mercury salts or, particularly bad, organic mercury compounds, which absorb into the bloodstream more readily.
To be fair, they emit ionizing or near-ionizing radiation internally (ultraviolet). That's the part that makes them fluorescent -- the internal coating that absorbs UV radiation and reradiates it at a lower (visible) frequency. I think in the end the UV radiation from a CFL is lower than that from an incandescent.
Anything that's "instant on" or uses a transformer ("wall wart") is a vampire sucking off energy and wasting it. Cell phone chargers or any kind of charger, cordless house phones, computers, video game consoles, TVs, VCRs, DVD/BR players, stereos, laptop chargers, monitors, printers, microwaves... these are only a sample of the vampires in your house.
You mean electronics that are instant-on... and their efficiency varies dramatically. Plenty of non-electronic devices (and even simple electronic ones) are instant-on with zero power consumption in the interim. Incandescent bulbs, for example. (Also LED bulbs and many small electric motors, like a hand drill or vacuum cleaner.)
A lot of the things you list now use trivial amounts of power when "off" (at least, the good ones do). Chargers in particular can cut out when they're not charging. Such things are easy to measure.
Increasing demand doesn't always mean increasing supply
It doesn't always, certainly, but sometimes it does. Original statement said (indirectly) that increased demand must necessarily result in increased prices, which is overly simplistic and often not true.
It's useful not to rely on intuition with quantum mechanics.
So, note that "quarkonium" isn't a particle, but rather a class of particles -- a quark bound to its antiquark. A collection of quarks held together by the strong force is a bound state. Bound states of quarks are particles.
See, what actually happened is that one person was arrested for videoing police as they were conducting field interviews. It's unreasonable to infer from that case that any person who videotapes a police officer would necessarily be arrested. It's quite unreasonable to infer that a person who videotapes a police officer committing a felony would be prosecuted.
Of course, if a policeman was actually raping someone, they're probably a dangerous enough individual that if they caught you videotaping them, arrest is just about the best thing that could happen to you.
The DoD actually stipulates 7-pass still. (However, physical destruction seems to be required for classified material.) 35-pass is the recommendation Gutmann made as so is often available in disk-wiping software.
The appeal of the multi-pass wipe is that it provides some degree of future-proofing (if people figure out a new technology for drive recovery, you may still be protected against it) and it's basically free if you're dealing with enough drives and have proper workflow.
The evidence suggests that the disk was partially zeroed, then that operation was cancelled and the disk was simply reformatted without first erasing it.
It's because writing zeroes takes time and is easy to screw up -- power loss, drive failures, etc. will stop the erasure process. Thermite is fast, reliable, and gives visual feedback that the operation has completed successfully.
It's not that simple. That's a reasonable description of an MFM disk, an old technology that isn't used any more. MFM disks were the topic of the Gutmann paper. Basically all claims that you can recover data from a zeroed drive are based on this paper. Gutmann has since repudiated it. Modern disks are substantially more complicated in terms of how a block of data gets turned into a collection of magnetizations, such that it's no longer reasonable to ever expect to get any useful information out of hysteresis (residual magnetization).
Nonetheless, the myth persists that somehow, magically, the government can read erased hard drives. What actually turns out to be the case is that people don't bother erasing hard drives.
(Also, it's not charge, it's moment. You can't add and remove magnetic charge because we haven't found any magnetic monopoles.)
The actual procedure as it was explained to me is that he used the OS X install-disk option to overwrite his disk and chose the Gutmann erasure option, which is a 35-pass wipe. It also takes forever and gives you a helpful progress bar indicating that it will take forever. Apparently he cancelled this and chose the zero-pass wipe -- also known as "just format the drive and install a new OS without actually erasing the disk".
Modern Mac OS X uses a single SHA-1 hash (salted) to store passwords. Older versions of OS X uses somewhat less-secure hashes, and if you've interacted with a Windows network you may have things like an NTLM hash to work with.
While the password is 11 characters, it's well within the set of passwords that a good dictionary attack generator will hit -- a word, a year, and some symbols. SHA-1 is cheap to crack.
This is a good example of why operating systems storing passwords should use key strengthening. A 1024-round HMAC is still trivially cheap to compute for a single password. Even if cracking this password took them only a month (a reasonable time for a long, guessable password), increasing the difficulty by 1024 would render it impossible to crack.
Well, that certainly sounds like a scientifically-sound assessment.
The Jensen study (1993) was brought on by purveyors of alcohol?
Animals are actually adapted to accept alcohol as part of their diet. Yeast (ever-present) produce alcohol normally as a byproduct of sugar digestion, potentially to kill off competitors. (Fungi are quite skilled at eliminating bacterial competitors, for example.) Animals as a result need to be able to tolerate some level of alcohols in their food source; hence (probably) the alcohol dehydrogenase enzyme.
Slashdot Mirror
Nonsense. Most models are wrong. They're still enormously useful compared to something that's more wrong. Newtonian mechanics is wrong, but it was -- and still is -- very useful for the overwhelming majority of situations.
It is very wrong to say the earth is flat. There are many, many ways of demonstrating its wrongness and assuming the earth is flat will lead you to wildly incorrect conclusions for many problems.
It is less wrong to say the earth is a sphere. However, it's harder to demonstrate that it's wrong, and you can do many useful calculations assuming a sphere for simplicity.
It's also wrong, but not very much, to say the earth is a slightly squashed sphere. It requires very careful measurement to demonstrate this, and it's such an accurate approximation to make that it's rare to see someone actually model the earth's correct shape.
It's the underhanded C contest.
Many of the image-processing solutions were very clever ways of hiding bugs in code. Some were more likely to pass a code review than others. Many of them would not really pass dedicated testing. The winning entry that you mention does character-substitution in an ASCII PPM file, replacing all digits with zeroes. If you look at the file in an image browser, it's actually redacted. If you look at the file in a text editor and have security in mind, you will immediately be very suspicious. The nice bit is that the code is very short and the error subtle enough that it's very easy to overlook the problem.
Besides being unnecessarily insulting, you're actually comparing two different things. (For one, don't claim "a massive amount" when you reference a single study and fail to link to that study. The paper is this.)
What you said is:
increases in the weight of passengers is also putting a drag on fuel economy. As we get fatter and fatter, it obviously takes more energy to move us around.
Which is simply not true. Car mass only affects fuel efficiency of acceleration and it affects it linearly in total car + contents mass.
What's supported by the article is your second point: that people buy more SUVs as a result of being heavier. However, they fail to separate out the other factors driving SUV purchase (a fact they freely admit but you don't).
It's not significant at all. Do a little quantitative analysis.
Actually, many compact cars have excellent legroom. Some don't, just as some larger cars don't. I'm as tall as you, and my Prius has acceptable legroom -- better than some previous, much larger cars.
We can only construct pairs of files that have the same MD5 hash. There's no way to efficiently produce a file that has a particular, previously-published hash. However, in practice you wouldn't use MD5 anyway, but rather SHA-256 or better.
One reasonably good solution is to sign the hash using a key that is already stored in a difficult-to-modify part of the device.
An MD5 that you then store on disk along with the refresh image? Not so helpful. You could sign the MD5 hash, but then where do you put the public key to verify the signature. (At that point, at least, you've made it incrementally harder for malware to pull this off, since a fair bit of stuff has to be changed.)
Possibly, though it's also very likely they've developed a good model of the difference between an intentional and unintentional lane change. (To be fair, I've seen people on the road where it was hard to tell which it was. Maybe the car will teach them to drive better. Ha!)
Mercury has a pretty low vapor pressure. Once broken, the mercury vapor will recondense into liquid mercury.
Mercury vapor is more hazardous than liquid mercury, primarily because it enters through the lungs instead of through ingestion or skin contact. However, elemental mercury (both vapor and liquid) is fairly safe, particularly in the quantities we're talking about, because it absorbs so poorly into the bloodstream. Most legitimate mercury hazards are mercury salts or, particularly bad, organic mercury compounds, which absorb into the bloodstream more readily.
To be fair, they emit ionizing or near-ionizing radiation internally (ultraviolet). That's the part that makes them fluorescent -- the internal coating that absorbs UV radiation and reradiates it at a lower (visible) frequency. I think in the end the UV radiation from a CFL is lower than that from an incandescent.
You should try fancier fluorescents with a frequency higher than 60 Hz.
When death taxes are paid
It's called inheritance tax. Death is free, but inheritance of wealth is taxable (for amounts greater than many millions of dollars).
The distribution of income is only slightly less skewed.
Anything that's "instant on" or uses a transformer ("wall wart") is a vampire sucking off energy and wasting it. Cell phone chargers or any kind of charger, cordless house phones, computers, video game consoles, TVs, VCRs, DVD/BR players, stereos, laptop chargers, monitors, printers, microwaves... these are only a sample of the vampires in your house.
You mean electronics that are instant-on... and their efficiency varies dramatically. Plenty of non-electronic devices (and even simple electronic ones) are instant-on with zero power consumption in the interim. Incandescent bulbs, for example. (Also LED bulbs and many small electric motors, like a hand drill or vacuum cleaner.)
A lot of the things you list now use trivial amounts of power when "off" (at least, the good ones do). Chargers in particular can cut out when they're not charging. Such things are easy to measure.
Increasing demand doesn't always mean increasing supply
It doesn't always, certainly, but sometimes it does. Original statement said (indirectly) that increased demand must necessarily result in increased prices, which is overly simplistic and often not true.
That's not math, that's economics. It's not even very good economics.
It's useful not to rely on intuition with quantum mechanics.
So, note that "quarkonium" isn't a particle, but rather a class of particles -- a quark bound to its antiquark. A collection of quarks held together by the strong force is a bound state. Bound states of quarks are particles.
This is an unreasonable generalization.
See, what actually happened is that one person was arrested for videoing police as they were conducting field interviews. It's unreasonable to infer from that case that any person who videotapes a police officer would necessarily be arrested. It's quite unreasonable to infer that a person who videotapes a police officer committing a felony would be prosecuted.
Of course, if a policeman was actually raping someone, they're probably a dangerous enough individual that if they caught you videotaping them, arrest is just about the best thing that could happen to you.
The DoD actually stipulates 7-pass still. (However, physical destruction seems to be required for classified material.) 35-pass is the recommendation Gutmann made as so is often available in disk-wiping software.
The appeal of the multi-pass wipe is that it provides some degree of future-proofing (if people figure out a new technology for drive recovery, you may still be protected against it) and it's basically free if you're dealing with enough drives and have proper workflow.
The evidence suggests that the disk was partially zeroed, then that operation was cancelled and the disk was simply reformatted without first erasing it.
It's because writing zeroes takes time and is easy to screw up -- power loss, drive failures, etc. will stop the erasure process. Thermite is fast, reliable, and gives visual feedback that the operation has completed successfully.
It's not that simple. That's a reasonable description of an MFM disk, an old technology that isn't used any more. MFM disks were the topic of the Gutmann paper. Basically all claims that you can recover data from a zeroed drive are based on this paper. Gutmann has since repudiated it. Modern disks are substantially more complicated in terms of how a block of data gets turned into a collection of magnetizations, such that it's no longer reasonable to ever expect to get any useful information out of hysteresis (residual magnetization).
Nonetheless, the myth persists that somehow, magically, the government can read erased hard drives. What actually turns out to be the case is that people don't bother erasing hard drives.
(Also, it's not charge, it's moment. You can't add and remove magnetic charge because we haven't found any magnetic monopoles.)
The actual procedure as it was explained to me is that he used the OS X install-disk option to overwrite his disk and chose the Gutmann erasure option, which is a 35-pass wipe. It also takes forever and gives you a helpful progress bar indicating that it will take forever. Apparently he cancelled this and chose the zero-pass wipe -- also known as "just format the drive and install a new OS without actually erasing the disk".
Pro tip: zero-pass wipe is not secure.
Modern Mac OS X uses a single SHA-1 hash (salted) to store passwords. Older versions of OS X uses somewhat less-secure hashes, and if you've interacted with a Windows network you may have things like an NTLM hash to work with.
While the password is 11 characters, it's well within the set of passwords that a good dictionary attack generator will hit -- a word, a year, and some symbols. SHA-1 is cheap to crack.
This is a good example of why operating systems storing passwords should use key strengthening. A 1024-round HMAC is still trivially cheap to compute for a single password. Even if cracking this password took them only a month (a reasonable time for a long, guessable password), increasing the difficulty by 1024 would render it impossible to crack.
Well, that certainly sounds like a scientifically-sound assessment.
The Jensen study (1993) was brought on by purveyors of alcohol?
Animals are actually adapted to accept alcohol as part of their diet. Yeast (ever-present) produce alcohol normally as a byproduct of sugar digestion, potentially to kill off competitors. (Fungi are quite skilled at eliminating bacterial competitors, for example.) Animals as a result need to be able to tolerate some level of alcohols in their food source; hence (probably) the alcohol dehydrogenase enzyme.