The reason you need package signing in Internet Explorer is that extensions are provided by all sorts of different sites and you don't know which ones you can trust. Ditto for downloading ISO images and stuff. If all of the extensions come from a single trusted source, there is no need for package signing.
You think real engineers design things in their head in a padded room with no references? Being good at memorizing formulas is not the same as being a good engineer.
The problem is that it seems that a number of people using IE with the default security settings were never prompted that this thing was trying to install itself on their machines. Admittedly, it's possible these people were prompted and simply don't remember, but if there weren't, then there is a problem with IE's default configuration.
AFAIK, there is no record of any Mozilla extension installing itself without prompting the user first, and since most extensions are downloaded from centralized, trusted sources (basically just mozdev) there really isn't as much of a need to implement digital signatures. For that matter, this thing is apparently signed by Verisign, which means that IE's package signing system may be a security liability instead of a security benefit.
Yeah, but Mozilla, by default, prompts you before installing anything. It sounds like, for some default installations of IE, this gets installed without any prompting.
Lastly but obviously not least to the crowd which reads slashdot, thanks for correcting my grammar, sometimes my word usage is slightly off. I hope no one has trouble reading my posts.
Your word usage is slightly off? You left out a key phrase which made your comment self-contradictory. I hope to God that you're not an A.B. Duke; as it is, I'm embarrassed on behalf of the Duke academic community that someone receiving a scholarship at Duke is trying to pass off a major compositional mistake that resulted in a logical contradiction as a slight usage error...
Most of my friends are at public universities because they are great schools, just not quite as good as Duke, or some of those other private school-elitist ones you point out. Also, because they're giving me almost a full ride. So, I'm paying less for my degree than you, I got A's in calculus in high school, I'm getting A's here. Last semester I got an A+ in advanced physics.
There are exceptions to the rule, yes, but generally speaking, if it is possible to get a 100% on an exam given for an engineering course, then the exam really needs to be 4 times longer. As a general rule, engineering exams are designed to be taken over an 8-hour period with a computer and references, but are administered in a 2-hour period with a calculator and an equation sheet.:-p
Incidentally, your anecdote reminds me of a Spanish teacher at my high school who would not give a mark of 100%. The highest marks he gave were 99% plus some arbitrary fraction, like 5/8 or 27/32.
The scheme described in the article is not Bayesian at all. It's more like a very crude hash comparison. If two similar messages are concatenated, they should compress very well. If two dissimilar messages are concatenated, they will not compress as well.
An actual Bayesian filter would perform a statistical analysis of an existing body of spam and non-spam messages, identify key words or phrases that identify a message as spam or non-spam, and calculate the probability for every key word that a message containing that word is spam. Then every new message is classified as spam or non-spam by running a statistical analysis on its content, and the statistics of that message update and improve the probability model.
The interaction distance of EM forces associated with chemical bonds is at a microscopic scale, so even though the experiment occurs at a macroscopic scale, the interaction occurs at a microscopic scale. Also, the scale of the EM interaction between the nail and the earth is highly localized (to the contact area between the nail and the earth), whereas the gravitational interaction between the bowling ball and the earth is distributed over all the consituent particles of the bowling ball and the earth, some of which are interacting at very great distances. IIRC there is a theorem which allows us to analyze this gravitational interaction as if all the mass of the bowling ball and the earth were concentrated at the centers of those bodies, respectively, but under this analysis the scale of the gravitational interaction is still much larger than the scale of the EM interaction, so it's not really appropriate to compare them.
I wonder if a sufficient density of photons would collapse into a black hole.
I would speculate that one could not use photons to achieve critical mass for a black hole.
AFAIK, there is no gravitational interaction between photons, so they won't naturally collapse into a black hole.
Unless we're talking about gamma rays, most photons don't have enough energy to create virtual particles to provide rest mass to collapse into a black hole.
A point of confusion which seems to appear repeatedly in this thread is that, while the electromagnetic (EM) force seems to be stronger than gravity at microscopic scales,
the inverse square law implies that the ratio of these forces should remain constant with distance, but
everyday experience and astronomical evidence seems to suggest that gravity grows stronger than the EM force at macroscopic scales
I think the key to resolving this conundrum is to realize that the EM attraction is proportional to the relative charge difference between two bodies.
At microscopic scales, one is often dealing with individual EM charges, so the relative charge difference at that scale is large and the force is strong.
In macroscopic objects, it is difficult to separate macroscopic amounts of charge precisely because the EM force is quite strong, so macroscopic objects usually have relatively small charge differences and the macroscopic EM force seems relatively weak.
Compare this to gravity, which only has one type of charge--mass--which always increases as the size of the object increases.
At microscopic distances, you only have small amounts of charge associated with a weak force, so gravity seems weak
With macroscopic objects at macroscopic distances, you have lots and lots of charge associated with a weak force--enough to make gravity appear stronger than EM.
This is veering way off topic, but at the time the Nature paper was published, Risca was a high school student, and I believe that this research resulted in her winning the Intel Science Talent Search.
Hands-free sets do not make driving while talking on a cell-phone any safer. See this paper from the New England Journal of Medicine for details. Basically, they cross-correlated traffic accident reports with cell phone logs and found that talking on a cell phone while driving quadruples the risk of getting in an accident, regardless of whether or not the phone is hands-free. This increased risk of accident is comparable to the increased risk of accident while driving drunk.
The difference between talking on a cell phone and talking with a passenger is that the passenger is aware of the driving situation and can halt the conversation and/or call the driver's attention to the road in case of emergency.
Talking on a phone has been proven to be no more distracting than talking to a passenger.
That is not true. There was a paper published in the New England Journal of Medicine proving pretty conclusively that the increased risk of an accident while talking on a cell phone is about the same to the increased risk of getting in an accident while driving drunk. This risk increase is independent of whether or not the cell phone is hands-free. Intuitively, talking with a passenger should be safer than talking with someone on a cell phone because the passenger is also in the vehicle and is aware of the driver's situation, so he or she can halt the conversation or call the driver's attention to the road in case of emergency. A person on a cell phone is oblivious to the conditions the driver is facing and has no such safety factor.
I'm pretty sure this invention doesn't use memory metals. From skimming the patent, the power source for this watch movement seems to be a bimetallic coil spring. The different coefficients of thermal expansion of the metals cause the spring to coil or uncoil when it is heated or cooled. An escapement mechanism is probably used to connect this spring to the mechanical movement, so that regardless of whether the spring is coiling or uncoiling, it winds the watch.
All in all, I have to say its a pretty clever idea, but I'm surprised no one thought of it before. Bimetallic strips have been used to regulate mercury switches in thermostats for many years, and most mechanical clock and watches need an escapement so that the swinging of the pendulum or the oscillation of the spring only drives the movement forward and not backward.
The mechanism behind a thermal self-winding mechanical watch and a thermally-powered electrical watch are probably quite different. From skimming the patent, it seems that the thermal self-winding mechanical movement uses the thermal winding of a bimetallic spring to wind the mechanical movement of this watch (sort of like how a bimetallic spring is used to regulate a mercury switch in many thermostats). A thermally-powered electrical watch probably uses the voltage generated across a bimetallic junction to drive the quartz oscillator (which is more akin to a Peltier cooler run in reverse). I agree that, while the idea is clever, it does not deserve patent protection for the next twenty years. On the other hand, the market for this type of invention is pretty small, and other watch makers are free to continue manufacturing other types of self-powered movements (self-winding, kinetic generators, thermoelectric, etc.), so granting the patent does not especially harm the market.
I think it translates to something like "A good Christmas to you all." (buon = good; natale = birth, referring to Jesus' birth; etc.) But then again, I don't actually speak Italian, so I could be wrong...
Re:Britney's Guide to Semiconductor Physics
on
Google's new toys
·
· Score: 1
Britney Spears's Guide to Semiconductor Physics does in fact have a wealth of pictures of Ms. Spears that have been marked up to illustrate concepts of, well, semiconductor physics.:-p I have to say, though, I think the band gap should be oriented vertically rather than horizontally--maybe that's just me...;-)
Even government employees (soldier, civil servants, etc.), who are directly compensated by taxpayer money, have free time--hours when they aren't working which they can spend sleeping or eating or shopping or watching TV or analyzing the population dynamics of Buffy or reading Slashdot. This graduate student is only indirectly compensated by taxpayer money, if at all--why can't he have free time and why can't he spend his free time doing whatever he wants (laws permitting)? You seem to think that by accepting governmental funding, graduate students should become academic slave labor. I find that notion offensive and hypocritical and, in protest, I will continue waste your tax money by sleeping and eating and doing whatever the hell I want (like reading and posting on Slashdot) on weekends.:-p
By your logic, military personnel, whose wages are fully subsidized by taxpayers, should not be allowed to sleep or eat, and certainly should not be granted leave or R&R, because every minute of that time could be spent hunting from Osama Bin Laden or invading Iraq. That's just a small subset of government personnel, all of whom are on taxpayer's tab, and all of whom are given sleeping time and eating time and vacation time and sick leave, which is an immeasurable waste of taxpayers money.
Also, by your logic, if your company were to receive a government contract, your logic says that I, a taxplaying citizen, get to rescind all your free time because all that free time is pissing away my tax money.
This guy did this in his free time. The fact that he had some free time and a creative outlet might indirectly (or maybe even directly) inspire him to develop a cure for cancer, in which case that free time pays for itself. Depriving him of all free time and creative outlets almost certainly would result in him going insane, in which case your imagined tax dollars are wasted anyway. Give him (and grad students and researchers everywhere) a break--they deserve it.
They spend all their time on serious, groundbreaking research, only.
YES, they should.
Apparently, you still don't get it. Even the academically elite need some free time and release from their research. What they do in that free time (laws permitting) is their own damn business.
Do you work for your employer 168 hours a week, 365 days a year? (I'll be a nice guy and let you take day February 29th off every four years, assuming you don't go crazy from sleep-deprivation or starve to death by then) Have you never used corporate resources for personal business (e.g. read Slashdot)? If you answered no to either of those questions, you are a hypocrite and have no right to bitch about graduate students doing things outside of their research.
You fail to take into account that death by vampire would not appear in murder statistics, since the body remains and continues to live a pretty normal life (besides the whole blood sucking thing). Even the new nocturnal sleep cycle might not stand out, seeing as how Sunnydale is a college town and all.
Incidentally, I think the best strategy for the Fastest Finger Game is not to play the game, but just to input a random sequence of A-B-C-D as quickly as possible. There are only 24 ways to arrange those letters, so you automatically have a 1 in 24 chance of winning. If you actually try to order the choices correctly, all bets are off. The only reason I put any thought into this at all is that I auditioned to be a contestant and made the cut. Sadly I never got a call to be on the show.:-(
I think Scientific American usually has a column for material like this, and I think that, at least in the past, Nature has a column or columns of this sort as well.
My point was that for a given population of a mechanical system, the phenomenon of wear and tear is a statistical process. For the population, it can be reasonably well predicted by statistics (mean, standard deviation, etc.), but for each individual case, it cannot be very well predicted.
Contrast this to computer code, which should be as deterministic as a system gets. Yet software failure still appears to be a random process--under identical test conditions on a single machine, some software will crash sometimes and run fine other times. And software can be written to identify bit errors and correct them, so the only reason software fails this way is because of human laziness.
Slashdot Mirror
The reason you need package signing in Internet Explorer is that extensions are provided by all sorts of different sites and you don't know which ones you can trust. Ditto for downloading ISO images and stuff. If all of the extensions come from a single trusted source, there is no need for package signing.
You think real engineers design things in their head in a padded room with no references? Being good at memorizing formulas is not the same as being a good engineer.
The problem is that it seems that a number of people using IE with the default security settings were never prompted that this thing was trying to install itself on their machines. Admittedly, it's possible these people were prompted and simply don't remember, but if there weren't, then there is a problem with IE's default configuration.
AFAIK, there is no record of any Mozilla extension installing itself without prompting the user first, and since most extensions are downloaded from centralized, trusted sources (basically just mozdev) there really isn't as much of a need to implement digital signatures. For that matter, this thing is apparently signed by Verisign, which means that IE's package signing system may be a security liability instead of a security benefit.
Yeah, but Mozilla, by default, prompts you before installing anything. It sounds like, for some default installations of IE, this gets installed without any prompting.
And you're not elitist?
There are exceptions to the rule, yes, but generally speaking, if it is possible to get a 100% on an exam given for an engineering course, then the exam really needs to be 4 times longer. As a general rule, engineering exams are designed to be taken over an 8-hour period with a computer and references, but are administered in a 2-hour period with a calculator and an equation sheet. :-p
Incidentally, your anecdote reminds me of a Spanish teacher at my high school who would not give a mark of 100%. The highest marks he gave were 99% plus some arbitrary fraction, like 5/8 or 27/32.
The scheme described in the article is not Bayesian at all. It's more like a very crude hash comparison. If two similar messages are concatenated, they should compress very well. If two dissimilar messages are concatenated, they will not compress as well.
An actual Bayesian filter would perform a statistical analysis of an existing body of spam and non-spam messages, identify key words or phrases that identify a message as spam or non-spam, and calculate the probability for every key word that a message containing that word is spam. Then every new message is classified as spam or non-spam by running a statistical analysis on its content, and the statistics of that message update and improve the probability model.
The interaction distance of EM forces associated with chemical bonds is at a microscopic scale, so even though the experiment occurs at a macroscopic scale, the interaction occurs at a microscopic scale. Also, the scale of the EM interaction between the nail and the earth is highly localized (to the contact area between the nail and the earth), whereas the gravitational interaction between the bowling ball and the earth is distributed over all the consituent particles of the bowling ball and the earth, some of which are interacting at very great distances. IIRC there is a theorem which allows us to analyze this gravitational interaction as if all the mass of the bowling ball and the earth were concentrated at the centers of those bodies, respectively, but under this analysis the scale of the gravitational interaction is still much larger than the scale of the EM interaction, so it's not really appropriate to compare them.
- the inverse square law implies that the ratio of these forces should remain constant with distance, but
- everyday experience and astronomical evidence seems to suggest that gravity grows stronger than the EM force at macroscopic scales
I think the key to resolving this conundrum is to realize that the EM attraction is proportional to the relative charge difference between two bodies.- At microscopic scales, one is often dealing with individual EM charges, so the relative charge difference at that scale is large and the force is strong.
- In macroscopic objects, it is difficult to separate macroscopic amounts of charge precisely because the EM force is quite strong, so macroscopic objects usually have relatively small charge differences and the macroscopic EM force seems relatively weak.
Compare this to gravity, which only has one type of charge--mass--which always increases as the size of the object increases.This is veering way off topic, but at the time the Nature paper was published, Risca was a high school student, and I believe that this research resulted in her winning the Intel Science Talent Search.
Hands-free sets do not make driving while talking on a cell-phone any safer. See this paper from the New England Journal of Medicine for details. Basically, they cross-correlated traffic accident reports with cell phone logs and found that talking on a cell phone while driving quadruples the risk of getting in an accident, regardless of whether or not the phone is hands-free. This increased risk of accident is comparable to the increased risk of accident while driving drunk.
The difference between talking on a cell phone and talking with a passenger is that the passenger is aware of the driving situation and can halt the conversation and/or call the driver's attention to the road in case of emergency.
I'm pretty sure this invention doesn't use memory metals. From skimming the patent, the power source for this watch movement seems to be a bimetallic coil spring. The different coefficients of thermal expansion of the metals cause the spring to coil or uncoil when it is heated or cooled. An escapement mechanism is probably used to connect this spring to the mechanical movement, so that regardless of whether the spring is coiling or uncoiling, it winds the watch.
All in all, I have to say its a pretty clever idea, but I'm surprised no one thought of it before. Bimetallic strips have been used to regulate mercury switches in thermostats for many years, and most mechanical clock and watches need an escapement so that the swinging of the pendulum or the oscillation of the spring only drives the movement forward and not backward.
The mechanism behind a thermal self-winding mechanical watch and a thermally-powered electrical watch are probably quite different. From skimming the patent, it seems that the thermal self-winding mechanical movement uses the thermal winding of a bimetallic spring to wind the mechanical movement of this watch (sort of like how a bimetallic spring is used to regulate a mercury switch in many thermostats). A thermally-powered electrical watch probably uses the voltage generated across a bimetallic junction to drive the quartz oscillator (which is more akin to a Peltier cooler run in reverse). I agree that, while the idea is clever, it does not deserve patent protection for the next twenty years. On the other hand, the market for this type of invention is pretty small, and other watch makers are free to continue manufacturing other types of self-powered movements (self-winding, kinetic generators, thermoelectric, etc.), so granting the patent does not especially harm the market.
I think it translates to something like "A good Christmas to you all." (buon = good; natale = birth, referring to Jesus' birth; etc.) But then again, I don't actually speak Italian, so I could be wrong...
Britney Spears's Guide to Semiconductor Physics does in fact have a wealth of pictures of Ms. Spears that have been marked up to illustrate concepts of, well, semiconductor physics. :-p I have to say, though, I think the band gap should be oriented vertically rather than horizontally--maybe that's just me... ;-)
Even government employees (soldier, civil servants, etc.), who are directly compensated by taxpayer money, have free time--hours when they aren't working which they can spend sleeping or eating or shopping or watching TV or analyzing the population dynamics of Buffy or reading Slashdot. This graduate student is only indirectly compensated by taxpayer money, if at all--why can't he have free time and why can't he spend his free time doing whatever he wants (laws permitting)? You seem to think that by accepting governmental funding, graduate students should become academic slave labor. I find that notion offensive and hypocritical and, in protest, I will continue waste your tax money by sleeping and eating and doing whatever the hell I want (like reading and posting on Slashdot) on weekends. :-p
By your logic, military personnel, whose wages are fully subsidized by taxpayers, should not be allowed to sleep or eat, and certainly should not be granted leave or R&R, because every minute of that time could be spent hunting from Osama Bin Laden or invading Iraq. That's just a small subset of government personnel, all of whom are on taxpayer's tab, and all of whom are given sleeping time and eating time and vacation time and sick leave, which is an immeasurable waste of taxpayers money.
Also, by your logic, if your company were to receive a government contract, your logic says that I, a taxplaying citizen, get to rescind all your free time because all that free time is pissing away my tax money.
This guy did this in his free time. The fact that he had some free time and a creative outlet might indirectly (or maybe even directly) inspire him to develop a cure for cancer, in which case that free time pays for itself. Depriving him of all free time and creative outlets almost certainly would result in him going insane, in which case your imagined tax dollars are wasted anyway. Give him (and grad students and researchers everywhere) a break--they deserve it.
Do you work for your employer 168 hours a week, 365 days a year? (I'll be a nice guy and let you take day February 29th off every four years, assuming you don't go crazy from sleep-deprivation or starve to death by then) Have you never used corporate resources for personal business (e.g. read Slashdot)? If you answered no to either of those questions, you are a hypocrite and have no right to bitch about graduate students doing things outside of their research.
Hal-9001
Electrical Engineering Graduate Student
You fail to take into account that death by vampire would not appear in murder statistics, since the body remains and continues to live a pretty normal life (besides the whole blood sucking thing). Even the new nocturnal sleep cycle might not stand out, seeing as how Sunnydale is a college town and all.
Incidentally, I think the best strategy for the Fastest Finger Game is not to play the game, but just to input a random sequence of A-B-C-D as quickly as possible. There are only 24 ways to arrange those letters, so you automatically have a 1 in 24 chance of winning. If you actually try to order the choices correctly, all bets are off. The only reason I put any thought into this at all is that I auditioned to be a contestant and made the cut. Sadly I never got a call to be on the show. :-(
I think Scientific American usually has a column for material like this, and I think that, at least in the past, Nature has a column or columns of this sort as well.
My point was that for a given population of a mechanical system, the phenomenon of wear and tear is a statistical process. For the population, it can be reasonably well predicted by statistics (mean, standard deviation, etc.), but for each individual case, it cannot be very well predicted.
Contrast this to computer code, which should be as deterministic as a system gets. Yet software failure still appears to be a random process--under identical test conditions on a single machine, some software will crash sometimes and run fine other times. And software can be written to identify bit errors and correct them, so the only reason software fails this way is because of human laziness.