The Wall Street Journal and the New York Post have large international readerships. For many people, paying for an on-line subscription is both faster than waiting for the mail, and much cheaper, as fast international delivery of heavy newspapers costs a great deal of money.
The English language international "biggies" are probably the Wall Street Journal, the New York Post, the L.A. Times, the Jerusalem Post, and the Times (out of London, England). After this list is exhausted, there are not many papers with large international readerships. After this list is exhausted, there are not many other newspapers with a significant customer base willing to pay for delivery.
I just cannot see how a small market daily paper can charge for web viewing. My local daily has so few local articles that it isn't worth reading. I just skip the business section, because I have already read better articles on-line the previous day. The paper with better local content is my local weekly paper, and it is free.
The big papers need to figure out a way to pay for and finance local news. Local TV needs to find a way to finance local news. In general, a better business model is needed for local issues. Nevertheless, watching local readership nosedive by making local newspapers on-line subscription-only sites is not the correct model.
True, but there's also a lot more you can do to protect something from wear and tear when you're not concerned about its weight and cost to lift into orbit. It's actually much easier to make something on Earth that lasts that long than it is to make something for space that lasts that long. The reason we don't usually do so is it's even easier to make something that doesn't, and a lot less expensive to just service it as needed.
Snow, water and ice are really nasty. If you live near significant snow, you will have watched things just "move" around. Year after year, you can watch a fence move, or a big rock slowly move across a yard.
In many ways, water and ice are worse than space. As the water thaws and freezes, it picks up and moves considerable structures. In Southern Canada, you just put your footings down below the frost line. In the Canadian shield, most people don't have basements because it would mean blasting granite. By the time you hit the arctic, there is so much snow and ice, it becomes logistically difficult to put in proper footings.
The Russians are talking about building boats for the nuclear reactor. Sea can be more stable than land in some ways. But what do you do when a great big iceberg is coming your way? These reactors must be connected to something via a cable. They won't be easy to move. Essentially, if one of these reactors ever becomes ice-locked, it would be in danger of getting its hull crushed and sinking.
These reactors have to withstand ice, year after year, without fail. How is that going to work? We haven't built an ice-breaker that can survive rough service without on-going maintenance. How is a stationary boat going to do it without maintenance?
Additionally, if a space probe goes missing, it is largely without significant environmental consequences for planet earth. If one of these reactors fails, it could dump radioactive waste into the arctic ocean. Thanks to the jet stream, all the oceans are interconnected, and that radioactivity will go world wide.
Who isn't an outlaw on a social networking site? Isn't the whole point of them to show super cool or super something? Aren't social networking sites the everyman's new way of finding fame?
Yes, I am about to get flamed by every woman on slashdot for using the term "everyman". Except this is slashdot, so...
btrfs has several features that help prevent data loss, and in particular silent corruption of data on the disk. It is also handy to be able to take snapshots for backup purposes.
Ext3 and Ext4 are faster because they omit some of these features. There was recently some heated debate about ext4 and data loss, see the Slashdot discussion for more links.
With file systems, speed and data integrity are trade-offs.
Yes, and no. Not all computational theory uses Big O notation, because you can have algorithms are better with Big O notation, but in practice are so computationally awful that no one uses them. Certain array decomposition methods fit in this category. There are some methods that are essentially unused, because n needs to be so large that no one has a sufficiently large computationally tractable problem to get payback.
The other issue is that computational feasibility may not be O(2^n). It could be O(2^0.2n) or O(2^(log n)^2)). I would be betting on something similar to the later, particularly as when n is large the density of primes becomes thinner roughly to (log n)^2. Knocking out a divide by 2 constant doesn't change O(n) notation, and is only useful for a practical speedup. Changing the constant or formula in the exponent makes a big difference however.
It is quite possible that with small n, like 512 bits, that the computational speedups aren't yet apparent. Maybe some of these effects really come play when breaking 1,048,576 bit long code sequences. It is just that no one has got close enough to cracking 512 bits to worry about million bit problems.
The likelihood of breaking it is genuinely 1 in 2^n and can only be broken by brute force attack.
That's not strictly true. Although the discrete log problem is hard it is still a computational assumption. Proving that 2^n is a lower bound would be a significant achievement. This scheme is only "unbreakable" in the sense that RSA is - breaking it requires solving a problem that we suspect, but are unable to prove, is very hard.
Unless I am mistaken... [Proof that search space is much smaller than 2^n]... that can still leave a huge brute force search space of course.
Additionally, there are a few additional reasons why the lower bound must be smaller than 2^n.
1. The requirement of Xs, Ys, and Xd, Yd to be coprimes significantly reduces computational workload.
2. It is statistically possible to "get lucky", and randomly guess the right results, so the strict lower bound must be 1.
3. Even if using a bench mark like the "average time to break the code", the lower bound must still be much less than 2^n. One only needs to guess on average 1/2 of the possibilities to finish in the mean amount of time.
4. The requirement that A and B are primes must somehow limit the number of guesses considerably, especially if n is large, because the density of primes decreases with increasing n. It just might not be obvious how to make use of this information.
5. The algorithm requires random number generators for A, B, Xs, and Xd. It is very difficult to make "good" random number generators with computers. These algorithms are notorious for being easy to break. Bad seeds were the cause of the recent ssh bug.
Nearly all the places where I've seen problems have ultimately been down to the wrong cable type more often than a bad crimping job; the flexible cable is for patching, the stiff cable is for horizontal/vertical distribution!
My experience also. The solid cable into the flex cable connectors fails under vibration in an industrial environment. It also can fail in an office environment, when someone starts moving their computer around.
Sometimes, a candidate walks into an interview, and it is really obvious that they are a good hire. They are good communicators, get good grades, and in general, are good at everything. They tend to get promoted fast. Technically talented strong communicators are a rare commodity.
On the other hand, there is the "weed people out" strategy employed by many HR departments. This tends to result in the most vanilla decisions possible, and for many companies, is probably a disaster long-term. Also, many of the really stellar candidates, usually have major weaknesses. Bill Gates, Martha Stewart, whoever you pick, to succeed there will be compromises. Bill Gates wasn't a people person, and I don't think he was very good at programming either.
Sometimes, the stellar candidate is not the strongest one on paper. Sometimes, the "remove all negatives" strategy results in vanilla hires. I really think you have to look past the resume and meet some of the people to get an understanding of who you want to hire.
The immune system is poorly understood, and even today relatively unpredictable. Additionally, viruses tend to evolve, and this makes them unpredictable. Administering vaccines will cause permanent damage and possibly death to a very small percentage of the people given the vaccine. Taking a vaccine is like playing Russian Roulette. If you give enough different vaccines to a sufficiently large number of people, sooner or later something bad will happen. Adverse affects depends on what you have previously been exposed too, whether you currently have an infection and just don't know it yet, and if the immune system accidentally correlate this vaccine with something it really shouldn't attack. Currently, we can't predict that those complications for any given school kid.
The long-term effects of Gardasil are unknown. It only defends against a few specific common varieties of HPV. No one knows that if these are displaced, will deadlier cancer causing variations of HPV replace them? The way viruses spread, you really need to knock out the class, not just 3 sub-varieties. This is why the flu vaccine is given every year, and often does not work. The flu changes every year. It will be the same for HPV. It's just no one understands how fast it will happen. Will the virus mutate and the kids get cancer anyway?
Additionally, the current drug marketing process puts sales ahead of science. Even if Gardasil is safe, and ultimately saves lives, how much confidence do you have in vaccine testing? Sooner or later, some company will come up with a vaccine that really does have long-term side effects. Do you want every school aged girl in North America do be hit with a potentially serious "surprise" later in life? Make no mistake, we are giving these school kids Gardasil because of marketing, not science.
I just want my kids taking something because we understand the science, and the risks. Marketing isn't a good reason to be prescribing drugs and vaccines to kids.
If my friend and I each invest $100 into separate movies, and he gets back $200 versus my $125, it's obvious somebody made a better deal, and I want to know why my deal didn't stand up to his. That's the nature of the debate here.
With Hollywood accounting, if you as an actor put $100 into a film, your friend the accountant also puts in $100, and the movie makes $500 million, then you will get nothing, and your friend will get nothing, but somehow a studio executive in Hollywood will start driving around in fast cars.
Eventually, you and your friend will threaten to sue, and both of you will only get paid if your accountant friend was really wily, made copies of all the "cooked" books, and the studio doesn't want them released.
The PCIe bus is not the only problem. Even at 10Gb Ethernet, the load on the CPU of processing each interrupt corresponding to each packet arriving on the network becomes significant. At 1 Tb/s, your network speed is substantially higher than both the peak hard drive bandwidth (3 Gb/s) in your average desktop computer. At 1 Tb/s, you could fill a 1 TB hard drive in 8 seconds flat!
For the moment, these high speed technologies will be primarily used in the network and cabling closets, where the aggregate bandwidth of many computers (or at least many hard drives) can be gathered together.
The Shannon-Hartley theorem is not the relevant limit. The hard limit for copper is the cutt-off frequency, and for optical systems other technical challenges come into play.
Any given copper wire has an associated cutoff frequency. Passed this frequency, it is almost impossible to get significant amounts of energy to pass through the cable. The cutoff is very steep.
For most types of coaxial cable, the cutoff frequency is on the order of 1 GHz to 8 GHz. Since the bandwidth required for a working communications link is generally higher than the bandwidth of the cable, copper wiring will top out at something on the order of a few GHz for most practical applications. UTP cable, as used in existing Ethernet, will perform worse than coaxial cable. For practical purposes, we have probably used all of its available bandwidth for 1 Gb Ethernet. UTP has a cutoff frequency on the order of 300 to 500 MHz, if memory serves. As such, the 1 Gb Ethernet specification resorts to uses all four pairs to achieve the 1Gb rated speed.
To increase bandwidth further, either microwave or optical waveguides can be used. Microwave waveguides are not practical for personal computer use. This leaves optical fiber, which is an optical waveguide.
Optical fiber has an essentially unlimited bandwidth, on the order of 500 Tb/s. Its performance is primarily limited by cost reasons and technical reasons relating to receivers and transmitters. It is difficult to generate the variable frequency light sources required to make use of the vast amounts of light spectrum. Separation of the light sources at the receiver is also a major issue. There are optical dispersion problems relating to the cable, but these are easier to deal with than the problems of creating a precision wide-band variable frequency laser.
In general, the technologies at optical speeds are not as well developed as the electrical technologies for microwave, broadcast, and copper communications transmission. It is much more difficult to use all available bandwidth at optical speeds, than at copper speeds. However, since the theoretical bandwidth at optical speeds is huge, much higher communication speeds are possible with optical.
Can any rational number be expressed as a sum the powers of irrational numbers like pi or e?
Yes. Irrational number systems can exist, and can represent rational numbers. The rational numbers will appear as an infinite series, much like as the expansion for pi ended with the lengthy discussion with the GP poster.
So far, I have not found a use for such a numbering system. pi occurs regularly. pi^2 and pi^3 and higher orders of pi usually correspond to calculation errors for most practical systems. Thus a base pi numbering system appears useless. However, e and pi occur so frequently that base-pi and base-e units do exist.
The base-pi unit (as opposed to a full numbering system) is radians, and radians are widely used in math and engineering. Similarly, exponentials (e) are often expressed as phasors and are widely used in electrical engineering. (Yes, phasors do exist, and they aren't something off Star Trek.)
And thank you for finally answering my question. Now let's see you do it with -1.;)
Thanks for the feedback. I think you really need to do some university level math courses. Your mind is obviously thinking in that direction. Some of this stuff was hinted at in high-school, but not properly covered.
-1 can be represented as a power of two as well:
-1 = 2^(i*pi*ln(e)/ln(2)) where i=sqrt(-1), ln(2) is the natural logarithm of 2, and ln(e)=1. As -1 is negative, but not irrational, the power series expansion has only one element, with an imaginary/complex exponent.
Finally, there are non-trivial methods to find the components of pi. That is why I suggested working it out from a power series expansion of 3*acos(x), evaluated at x=1/2. Computer algorithms compute pi regularly, but they are non-trivial operations.
Normally, this would be the time I would march a student over to the nearest blackboard. All I can say is this, type the following expression into a calculator:
2^1+2^0+2^-3+2^-6+2^-11+2^-12+2^-13+2^-14+2^-15+2^-16+2^-18+2^-19+2^-21+2^-23+2^-25+2^-29
It should give you 3.141592653, which is pi to 10 digits. The only thing special about irrational numbers is that they require an infinite non-repeating series of numbers to represent them. In base 2, this means an infinite series of powers of two, and negative exponents are required.
I suspect your definition of coefficient, power series and/or power series expansion differs fundamentally from mine. I'm also having difficulty following your notation. As I teach a course in first year Calculus, and being able to represent pi to an arbitrary precision is an important piece of computer science, I can assure you that solving this problem is possible.
Additionally, if I do understand your notation properly, where it looks like your solving for exponents, the relevant exponents are:
p1 = 2, p2 = 1, p3 = -3, p4 = -6, p5 = -11, p6 = -12, and so on.
Any number that can be represented in Base-10 decimal notation, can be represented in Base-2 notation. Pi is 3.14159... in decimal notation. There exists a corresponding binary sequence of numbers 11.001001... that represent pi in Base-2 notation. In base 10 notation, each digit corresponds to a power of 10. Left hand digits go from 10^0, 10^1, 10^2, and increase going left. Right hand digits go from 10^-1, 10^-2, 10^-3 and decrease going right. Similarly, in Base-2, the same thing happens. For pi, the weightings are 1*2^1+1*2^0+0*2^-1+0*2^-2+1*2^-3+... and so on.
The only complexity to pi is that it is an infinite, non-repeating sequence of digits. There is are multiple methods of finding them, and super-computer people spend time experimenting with them. For most uses, presenting the bits for a binary expansion of pi is pointless, because the binary expansion for pi can be trivially calculated from the binary bits of any appropriate infinite series expansion. Evaluating 3*arccos(x) at x=1/2 is just one example. The advantage of the arccos expansion is that it makes some of the resulting calculations simpler, as it already has lots of powers of 2 in it already. As such, all one needs to do is express the co-efficients in binary, add them up, and the result is the expansion for pi to any arbitrary level of accuracy. Additionally, the infinite series has the important property that it represents the exact value of pi.
In short, in base 10, decimal:
pi = 3.141... = 3 * 10^0 + 1 * 10^-1 + 4 * 10^-2 + 1 * 10^-3 +... and in base 2, binary:
pi = 11.001... = 1 * 2^1 + 1 * 2^0 + 0 * 2^-1 + 0 * 2^-2 + 1 * 2^-3 +...
Okay, smart guy, give me an exact expression for pi expressed as a sum of powers of two.
It can be done. Irrational numbers expand to an infinite series.
Try evaluating the power series expansion of 3*arccos(x), at x=1/2. The details should be in most first year calculus texts. It should contain lots of powers of two and have the property that 3*arccos(1/2)=pi.
You can even express pi() as an infinite series of powers of two where all the coefficients are zero or one, and the proof for that will also be in some many math texts dealing with irrational numbers.
For some reason a significant amount say yes... I can't understand why one would if they have already committed some terrible crimes, but they seem to.
The trouble is trolling through a large enough database will randomly create some false positives. Even if you assume the DNA data is reliable, which it isn't, then you still have the problem of "I bumped into this person, and through some strange series of events my skin cells contaminated the sample." There was already a case of the police desperately searching for a serial killer for 6 murders, only to later realize that the suspect was a technician who was accidentally contaminating the samples. Check out the story here, here or here.
The problem with modern DNA techniques is they can be too sensitive. They light up anyone who ever came in contact with the sample. Even if this is through accidental contact.
Good Luck if you actually had sex with the girl, and the rapist used a condom. Your going to be a suspect no matter what you do. Hope you have a really good alibi, a really good lawyer, and that the girl swears that you are a nice guy. Your going to need that alibi for the next 100 years.
The pirate bay case means google may have to pay when people use google to download copyrighted material.
That is exactly what is happening. The newspapers are going after Google for money. They want to charge Google for the headlines and short blurbs on the Google News page.
Essentially, everything on the web is copyrighted, Google has money, so people are going to start going after Google. Precedents like this are going to make things worse.
For the moment, the attacks will be fairly non-obvious, but the *AA will push for increasingly draconian copyright laws. Eventually, if they are not stopped, excessive rules will kill the Internet. For those with a sense of history, one can have either a small network with lots of control or a big network with lots of users and lax controls. What made the internet successful was the relative lack of control. Let's hope this doesn't change.
Every series chain needs its own constant current source. Otherwise, you get into a load-leveling problem between the chains, where if one LED has a negative temperature coefficient, it can overhead, reduce it's resistance, and draw even more current and heat into that particular chain until device failure occurs.
For LEDs, a simple resistor is often used as a constant current source. For efficiency reasons, a constant current switching power supply may be used in high-power designs.
The resistor / switching power supply design can also be combined. Use a resistor to regulate current in the individual chains, then regulate the overall voltage to the module such that the LED chain experiencing the highest current draw is kept within parameters. Depending on the power levels involved, even simple things like wire resistance, inductance and the equivalent series resistance of the LEDs can also serve as current limiting resistances. As such, the "current limiting resistors" might not be physical components, and instead could be the intrinsic properties of some of the devices already in the circuit. The detailed design of a high-power LED array can be quite subtle.
I think the mental anguish was caused by being told to fix other people's bugs, then flunking your performance review because you succeeded. This is an example of performance incentives gone horribly wrong. Obviously, someone felt that fixing fewer bugs indicated greater programming prowess, better software, and the resulting early ship date! I'm not surprised the GP post talked about Microsoft.
The problem is not only the data set could be huge, it doesn't prove anything either. If I click the link to slashdot, then it could reasonably be argued that I went to slashdot. If you start collecting every link contained in the page, then you immediately invoke the "I didn't go there" defense. If your web browser precaches all the links on all the links in a page, then things really start getting complex. Eventually, all the defense lawyer has to do is show the database is hopelessly corrupted with all sorts of information having nothing to do with the client on trial.
The other complication is that some companies will be running automated web spidering programs, and other companies will be forwarding data from other companies (like other ISPs). The data collection directive may result in simply having to store a significant percentage of your total outgoing bandwidth. Even if that percentage is a few percent, a few percent of a big pipe adds up quickly. 100 Mbps / 8 (bits / byte) * 5% * 3600 (seconds/hour) * 24 (hours/day) * 365 (days/year) = 20 (TB/yr). 20 Terabytes per year is a large amount of data, especially if you want to make it searchable. I'm sure there are lots of ISPs out there using connections bigger than 100 Mbps too.
That's evil. Note to self: automatic data deletion devices must include accelerometers.
You seem to know what you are talking about. Just how hard is it to build a secure computing resource at a remote site, that isn't (easily) vulnerable to data loss if someone steals the equipment? Something that can stop the common criminal?
Wouldn't it be simpler to create an encrypted file system with a self-destructing key?
That way, when the FBI seized the servers, they could automatically delete all the data for you. Then when it hit court, it would be "well your honour, if the FBI told me what they were up to in advance, then I would have cooperated with them. As it is, this device prevents thieves from accessing sensitive company data. It prevents data thefts like the ones that happened at the department of defense, the CIA, the IRS, and the FBI."
The cops might be seriously annoyed with you, but you are going to be a criminal anyway...
Slashdot Mirror
The Wall Street Journal and the New York Post have large international readerships. For many people, paying for an on-line subscription is both faster than waiting for the mail, and much cheaper, as fast international delivery of heavy newspapers costs a great deal of money.
The English language international "biggies" are probably the Wall Street Journal, the New York Post, the L.A. Times, the Jerusalem Post, and the Times (out of London, England). After this list is exhausted, there are not many papers with large international readerships. After this list is exhausted, there are not many other newspapers with a significant customer base willing to pay for delivery.
I just cannot see how a small market daily paper can charge for web viewing. My local daily has so few local articles that it isn't worth reading. I just skip the business section, because I have already read better articles on-line the previous day. The paper with better local content is my local weekly paper, and it is free.
The big papers need to figure out a way to pay for and finance local news. Local TV needs to find a way to finance local news. In general, a better business model is needed for local issues. Nevertheless, watching local readership nosedive by making local newspapers on-line subscription-only sites is not the correct model.
Snow, water and ice are really nasty. If you live near significant snow, you will have watched things just "move" around. Year after year, you can watch a fence move, or a big rock slowly move across a yard.
In many ways, water and ice are worse than space. As the water thaws and freezes, it picks up and moves considerable structures. In Southern Canada, you just put your footings down below the frost line. In the Canadian shield, most people don't have basements because it would mean blasting granite. By the time you hit the arctic, there is so much snow and ice, it becomes logistically difficult to put in proper footings.
The Russians are talking about building boats for the nuclear reactor. Sea can be more stable than land in some ways. But what do you do when a great big iceberg is coming your way? These reactors must be connected to something via a cable. They won't be easy to move. Essentially, if one of these reactors ever becomes ice-locked, it would be in danger of getting its hull crushed and sinking.
These reactors have to withstand ice, year after year, without fail. How is that going to work? We haven't built an ice-breaker that can survive rough service without on-going maintenance. How is a stationary boat going to do it without maintenance?
Additionally, if a space probe goes missing, it is largely without significant environmental consequences for planet earth. If one of these reactors fails, it could dump radioactive waste into the arctic ocean. Thanks to the jet stream, all the oceans are interconnected, and that radioactivity will go world wide.
Who isn't an outlaw on a social networking site? Isn't the whole point of them to show super cool or super something? Aren't social networking sites the everyman's new way of finding fame?
Yes, I am about to get flamed by every woman on slashdot for using the term "everyman". Except this is slashdot, so ...
btrfs has several features that help prevent data loss, and in particular silent corruption of data on the disk. It is also handy to be able to take snapshots for backup purposes.
Ext3 and Ext4 are faster because they omit some of these features. There was recently some heated debate about ext4 and data loss, see the Slashdot discussion for more links.
With file systems, speed and data integrity are trade-offs.
Yes, and no. Not all computational theory uses Big O notation, because you can have algorithms are better with Big O notation, but in practice are so computationally awful that no one uses them. Certain array decomposition methods fit in this category. There are some methods that are essentially unused, because n needs to be so large that no one has a sufficiently large computationally tractable problem to get payback.
The other issue is that computational feasibility may not be O(2^n). It could be O(2^0.2n) or O(2^(log n)^2)). I would be betting on something similar to the later, particularly as when n is large the density of primes becomes thinner roughly to (log n)^2. Knocking out a divide by 2 constant doesn't change O(n) notation, and is only useful for a practical speedup. Changing the constant or formula in the exponent makes a big difference however.
It is quite possible that with small n, like 512 bits, that the computational speedups aren't yet apparent. Maybe some of these effects really come play when breaking 1,048,576 bit long code sequences. It is just that no one has got close enough to cracking 512 bits to worry about million bit problems.
Additionally, there are a few additional reasons why the lower bound must be smaller than 2^n.
1. The requirement of Xs, Ys, and Xd, Yd to be coprimes significantly reduces computational workload.
2. It is statistically possible to "get lucky", and randomly guess the right results, so the strict lower bound must be 1.
3. Even if using a bench mark like the "average time to break the code", the lower bound must still be much less than 2^n. One only needs to guess on average 1/2 of the possibilities to finish in the mean amount of time.
4. The requirement that A and B are primes must somehow limit the number of guesses considerably, especially if n is large, because the density of primes decreases with increasing n. It just might not be obvious how to make use of this information.
5. The algorithm requires random number generators for A, B, Xs, and Xd. It is very difficult to make "good" random number generators with computers. These algorithms are notorious for being easy to break. Bad seeds were the cause of the recent ssh bug.
My experience also. The solid cable into the flex cable connectors fails under vibration in an industrial environment. It also can fail in an office environment, when someone starts moving their computer around.
Sometimes, a candidate walks into an interview, and it is really obvious that they are a good hire. They are good communicators, get good grades, and in general, are good at everything. They tend to get promoted fast. Technically talented strong communicators are a rare commodity.
On the other hand, there is the "weed people out" strategy employed by many HR departments. This tends to result in the most vanilla decisions possible, and for many companies, is probably a disaster long-term. Also, many of the really stellar candidates, usually have major weaknesses. Bill Gates, Martha Stewart, whoever you pick, to succeed there will be compromises. Bill Gates wasn't a people person, and I don't think he was very good at programming either.
Sometimes, the stellar candidate is not the strongest one on paper. Sometimes, the "remove all negatives" strategy results in vanilla hires. I really think you have to look past the resume and meet some of the people to get an understanding of who you want to hire.
The immune system is poorly understood, and even today relatively unpredictable. Additionally, viruses tend to evolve, and this makes them unpredictable. Administering vaccines will cause permanent damage and possibly death to a very small percentage of the people given the vaccine. Taking a vaccine is like playing Russian Roulette. If you give enough different vaccines to a sufficiently large number of people, sooner or later something bad will happen. Adverse affects depends on what you have previously been exposed too, whether you currently have an infection and just don't know it yet, and if the immune system accidentally correlate this vaccine with something it really shouldn't attack. Currently, we can't predict that those complications for any given school kid.
The long-term effects of Gardasil are unknown. It only defends against a few specific common varieties of HPV. No one knows that if these are displaced, will deadlier cancer causing variations of HPV replace them? The way viruses spread, you really need to knock out the class, not just 3 sub-varieties. This is why the flu vaccine is given every year, and often does not work. The flu changes every year. It will be the same for HPV. It's just no one understands how fast it will happen. Will the virus mutate and the kids get cancer anyway?
Additionally, the current drug marketing process puts sales ahead of science. Even if Gardasil is safe, and ultimately saves lives, how much confidence do you have in vaccine testing? Sooner or later, some company will come up with a vaccine that really does have long-term side effects. Do you want every school aged girl in North America do be hit with a potentially serious "surprise" later in life? Make no mistake, we are giving these school kids Gardasil because of marketing, not science.
I just want my kids taking something because we understand the science, and the risks. Marketing isn't a good reason to be prescribing drugs and vaccines to kids.
With Hollywood accounting, if you as an actor put $100 into a film, your friend the accountant also puts in $100, and the movie makes $500 million, then you will get nothing, and your friend will get nothing, but somehow a studio executive in Hollywood will start driving around in fast cars.
Eventually, you and your friend will threaten to sue, and both of you will only get paid if your accountant friend was really wily, made copies of all the "cooked" books, and the studio doesn't want them released.
The PCIe bus is not the only problem. Even at 10Gb Ethernet, the load on the CPU of processing each interrupt corresponding to each packet arriving on the network becomes significant. At 1 Tb/s, your network speed is substantially higher than both the peak hard drive bandwidth (3 Gb/s) in your average desktop computer. At 1 Tb/s, you could fill a 1 TB hard drive in 8 seconds flat!
For the moment, these high speed technologies will be primarily used in the network and cabling closets, where the aggregate bandwidth of many computers (or at least many hard drives) can be gathered together.
The Shannon-Hartley theorem is not the relevant limit. The hard limit for copper is the cutt-off frequency, and for optical systems other technical challenges come into play.
Any given copper wire has an associated cutoff frequency. Passed this frequency, it is almost impossible to get significant amounts of energy to pass through the cable. The cutoff is very steep.
For most types of coaxial cable, the cutoff frequency is on the order of 1 GHz to 8 GHz. Since the bandwidth required for a working communications link is generally higher than the bandwidth of the cable, copper wiring will top out at something on the order of a few GHz for most practical applications. UTP cable, as used in existing Ethernet, will perform worse than coaxial cable. For practical purposes, we have probably used all of its available bandwidth for 1 Gb Ethernet. UTP has a cutoff frequency on the order of 300 to 500 MHz, if memory serves. As such, the 1 Gb Ethernet specification resorts to uses all four pairs to achieve the 1Gb rated speed.
To increase bandwidth further, either microwave or optical waveguides can be used. Microwave waveguides are not practical for personal computer use. This leaves optical fiber, which is an optical waveguide.
Optical fiber has an essentially unlimited bandwidth, on the order of 500 Tb/s. Its performance is primarily limited by cost reasons and technical reasons relating to receivers and transmitters. It is difficult to generate the variable frequency light sources required to make use of the vast amounts of light spectrum. Separation of the light sources at the receiver is also a major issue. There are optical dispersion problems relating to the cable, but these are easier to deal with than the problems of creating a precision wide-band variable frequency laser.
In general, the technologies at optical speeds are not as well developed as the electrical technologies for microwave, broadcast, and copper communications transmission. It is much more difficult to use all available bandwidth at optical speeds, than at copper speeds. However, since the theoretical bandwidth at optical speeds is huge, much higher communication speeds are possible with optical.
Yes. Irrational number systems can exist, and can represent rational numbers. The rational numbers will appear as an infinite series, much like as the expansion for pi ended with the lengthy discussion with the GP poster.
So far, I have not found a use for such a numbering system. pi occurs regularly. pi^2 and pi^3 and higher orders of pi usually correspond to calculation errors for most practical systems. Thus a base pi numbering system appears useless. However, e and pi occur so frequently that base-pi and base-e units do exist.
The base-pi unit (as opposed to a full numbering system) is radians, and radians are widely used in math and engineering. Similarly, exponentials (e) are often expressed as phasors and are widely used in electrical engineering. (Yes, phasors do exist, and they aren't something off Star Trek.)
Thanks for the feedback. I think you really need to do some university level math courses. Your mind is obviously thinking in that direction. Some of this stuff was hinted at in high-school, but not properly covered.
-1 can be represented as a power of two as well:
-1 = 2^(i*pi*ln(e)/ln(2))
where i=sqrt(-1), ln(2) is the natural logarithm of 2, and ln(e)=1. As -1 is negative, but not irrational, the power series expansion has only one element, with an imaginary/complex exponent.
Finally, there are non-trivial methods to find the components of pi. That is why I suggested working it out from a power series expansion of 3*acos(x), evaluated at x=1/2. Computer algorithms compute pi regularly, but they are non-trivial operations.
Normally, this would be the time I would march a student over to the nearest blackboard. All I can say is this, type the following expression into a calculator:
2^1+2^0+2^-3+2^-6+2^-11+2^-12+2^-13+2^-14+2^-15+2^-16+2^-18+2^-19+2^-21+2^-23+2^-25+2^-29
It should give you 3.141592653, which is pi to 10 digits. The only thing special about irrational numbers is that they require an infinite non-repeating series of numbers to represent them. In base 2, this means an infinite series of powers of two, and negative exponents are required.
I suspect your definition of coefficient, power series and/or power series expansion differs fundamentally from mine. I'm also having difficulty following your notation. As I teach a course in first year Calculus, and being able to represent pi to an arbitrary precision is an important piece of computer science, I can assure you that solving this problem is possible.
Additionally, if I do understand your notation properly, where it looks like your solving for exponents, the relevant exponents are:
p1 = 2, p2 = 1, p3 = -3, p4 = -6, p5 = -11, p6 = -12, and so on.
Any number that can be represented in Base-10 decimal notation, can be represented in Base-2 notation. Pi is 3.14159 ... in decimal notation. There exists a corresponding binary sequence of numbers 11.001001 ... that represent pi in Base-2 notation. In base 10 notation, each digit corresponds to a power of 10. Left hand digits go from 10^0, 10^1, 10^2, and increase going left. Right hand digits go from 10^-1, 10^-2, 10^-3 and decrease going right. Similarly, in Base-2, the same thing happens. For pi, the weightings are 1*2^1+1*2^0+0*2^-1+0*2^-2+1*2^-3+... and so on.
The only complexity to pi is that it is an infinite, non-repeating sequence of digits. There is are multiple methods of finding them, and super-computer people spend time experimenting with them. For most uses, presenting the bits for a binary expansion of pi is pointless, because the binary expansion for pi can be trivially calculated from the binary bits of any appropriate infinite series expansion. Evaluating 3*arccos(x) at x=1/2 is just one example. The advantage of the arccos expansion is that it makes some of the resulting calculations simpler, as it already has lots of powers of 2 in it already. As such, all one needs to do is express the co-efficients in binary, add them up, and the result is the expansion for pi to any arbitrary level of accuracy. Additionally, the infinite series has the important property that it represents the exact value of pi.
In short, in base 10, decimal: ...
...
pi = 3.141... = 3 * 10^0 + 1 * 10^-1 + 4 * 10^-2 + 1 * 10^-3 +
and in base 2, binary:
pi = 11.001... = 1 * 2^1 + 1 * 2^0 + 0 * 2^-1 + 0 * 2^-2 + 1 * 2^-3 +
It can be done. Irrational numbers expand to an infinite series.
Try evaluating the power series expansion of 3*arccos(x), at x=1/2. The details should be in most first year calculus texts. It should contain lots of powers of two and have the property that 3*arccos(1/2)=pi.
You can even express pi() as an infinite series of powers of two where all the coefficients are zero or one, and the proof for that will also be in some many math texts dealing with irrational numbers.
The trouble is trolling through a large enough database will randomly create some false positives. Even if you assume the DNA data is reliable, which it isn't, then you still have the problem of "I bumped into this person, and through some strange series of events my skin cells contaminated the sample." There was already a case of the police desperately searching for a serial killer for 6 murders, only to later realize that the suspect was a technician who was accidentally contaminating the samples. Check out the story here, here or here.
The problem with modern DNA techniques is they can be too sensitive. They light up anyone who ever came in contact with the sample. Even if this is through accidental contact.
Good Luck if you actually had sex with the girl, and the rapist used a condom. Your going to be a suspect no matter what you do. Hope you have a really good alibi, a really good lawyer, and that the girl swears that you are a nice guy. Your going to need that alibi for the next 100 years.
That is exactly what is happening. The newspapers are going after Google for money. They want to charge Google for the headlines and short blurbs on the Google News page.
Essentially, everything on the web is copyrighted, Google has money, so people are going to start going after Google. Precedents like this are going to make things worse.
For the moment, the attacks will be fairly non-obvious, but the *AA will push for increasingly draconian copyright laws. Eventually, if they are not stopped, excessive rules will kill the Internet. For those with a sense of history, one can have either a small network with lots of control or a big network with lots of users and lax controls. What made the internet successful was the relative lack of control. Let's hope this doesn't change.
Every series chain needs its own constant current source. Otherwise, you get into a load-leveling problem between the chains, where if one LED has a negative temperature coefficient, it can overhead, reduce it's resistance, and draw even more current and heat into that particular chain until device failure occurs.
For LEDs, a simple resistor is often used as a constant current source. For efficiency reasons, a constant current switching power supply may be used in high-power designs.
The resistor / switching power supply design can also be combined. Use a resistor to regulate current in the individual chains, then regulate the overall voltage to the module such that the LED chain experiencing the highest current draw is kept within parameters. Depending on the power levels involved, even simple things like wire resistance, inductance and the equivalent series resistance of the LEDs can also serve as current limiting resistances. As such, the "current limiting resistors" might not be physical components, and instead could be the intrinsic properties of some of the devices already in the circuit. The detailed design of a high-power LED array can be quite subtle.
I think the mental anguish was caused by being told to fix other people's bugs, then flunking your performance review because you succeeded. This is an example of performance incentives gone horribly wrong. Obviously, someone felt that fixing fewer bugs indicated greater programming prowess, better software, and the resulting early ship date! I'm not surprised the GP post talked about Microsoft.
The problem is not only the data set could be huge, it doesn't prove anything either. If I click the link to slashdot, then it could reasonably be argued that I went to slashdot. If you start collecting every link contained in the page, then you immediately invoke the "I didn't go there" defense. If your web browser precaches all the links on all the links in a page, then things really start getting complex. Eventually, all the defense lawyer has to do is show the database is hopelessly corrupted with all sorts of information having nothing to do with the client on trial.
The other complication is that some companies will be running automated web spidering programs, and other companies will be forwarding data from other companies (like other ISPs). The data collection directive may result in simply having to store a significant percentage of your total outgoing bandwidth. Even if that percentage is a few percent, a few percent of a big pipe adds up quickly. 100 Mbps / 8 (bits / byte) * 5% * 3600 (seconds/hour) * 24 (hours/day) * 365 (days/year) = 20 (TB/yr). 20 Terabytes per year is a large amount of data, especially if you want to make it searchable. I'm sure there are lots of ISPs out there using connections bigger than 100 Mbps too.
That's evil. Note to self: automatic data deletion devices must include accelerometers.
You seem to know what you are talking about. Just how hard is it to build a secure computing resource at a remote site, that isn't (easily) vulnerable to data loss if someone steals the equipment? Something that can stop the common criminal?
Wouldn't it be simpler to create an encrypted file system with a self-destructing key?
That way, when the FBI seized the servers, they could automatically delete all the data for you. Then when it hit court, it would be "well your honour, if the FBI told me what they were up to in advance, then I would have cooperated with them. As it is, this device prevents thieves from accessing sensitive company data. It prevents data thefts like the ones that happened at the department of defense, the CIA, the IRS, and the FBI."
The cops might be seriously annoyed with you, but you are going to be a criminal anyway ...