To become a business that succeeds over the long haul, you need to become difficult to replicate for some reason. Fundamentally there are only a few ways to stand out, and not all of these are necessarily viable strategies for a given market:
Low cost
High quality
Monopolistic control of a scarce resource
Privileged position in a market with network effects
Companies that undoubtedly have a defensible position include:
Walmart. They are the largest player in a market that is scale-dominated, so they have an automatic advantage re: costs.
Microsoft. Bill's genius was his early understanding of the importance of network effects, and then using them to catapault the OS and Office products to market dominance.
Mercedes-Benz. Quality is a difficult dimension to gain a defensible advantage along, since intellectual assets are easy to imitate (IP rights to too temporary and too specific). One way to do it is build a brand cache that maintains over time. I might also put Sony and Apple here.
Shell oil. They control scarce resources (oil rights) that make them hard to replicate.
eBay. The auction market is a network-effects market if there ever was one, and eBay has all of the business.
Companies that are iffier:
Dell. They are a low-cost producer, but are doing it through intelligence rather than scale. Until they get scale, intelligence is too easily copied.
Google. They stand out in terms of quality, but a smart competitor with equivalent technology could easily pull users away (remember Altavista?) so this isn't defensible in the long run. What they should try to do is convert their market into one where their scale gives them an advantage, but they haven't done this yet. For example their approach toward sponsored links doesn't give them a scale advantage, since they are compensated on a per-clickthru basis by their advertisers (so as an advertiser I would be just as willing to pay Startup X for this service). It's possible that their service will become sophisticated enough that only they will have the scale needed to build the infrastructure to deliver it (the situation the large chip fabs are in today).
Amazon. This is a tough one. While they don't have scale dominance in terms of volume relative to Border's etc., they certainly have the largest mail order retail distribution system in their target markets. So Border's etc. may not be able to compete with them effectively if it turns out that distributing to individuals is sufficiently different from distributing to stores that their scale in the latter can't advantage them in the former.
I don't know how you rate "most popular." Since computer use has been exploding exponentially, if you do it by user head count, no computer that's more that a couple of years old would count.
I doubt this is true. Commodore sold over 20 million C64s, and I would be very surprised if any specific model of Dell/IBM/whatever sold that many units today. The fact that there are so many nearly-identical models today means that none of them stand out enough to sell in large numbers.
I also doubt your statement that computer use is "expanding exponentially". In fact the growth rate of computer sales has been declining steadily, and it is quickly becoming a replacement business rather than one of selling to new users.
Even if there were indeed some intelligent life in the Gem 37 system, only in the first case (tens of light years) could we hope to ever communicate with them.
The goal of TPF is somewhat broader than SETI, which will only find life that is intelligent (i.e. capable and willing to send beamed radio transmissions to other stars). TPF will hunt for signs of ANY biological activity by looking for certain chemical signatures (e.g., an oxygen atmosphere).
The exciting thing about TPF is that it skips several (likely small) terms in the Drake equation. It is probably the likeliest near-term chance we have to find extraterrestrial life.
However, as a new parent, a scientist, and a PhD, I'll admit that I too have worries about WiFi and growing children. And I don't wear a tinfoil hat.
By coincidence I am also a new parent and a PhD (physics), and I understand firsthand your concern for your baby's safety. But consider the following:
Out of all the possible risks to your baby (SIDS, disease, falling, choking, etc.) where would you estimate WiFi exposure to be, on a ranked list?
The health impact of microwaves is one of the more intensively studied questions and yet there is still no statistically significant correlation. Doesn't this fact alone imply something about the answer to question 1?
Have you ever heard of a reasonable mechanism for low-frequency EM radiation to cause damage to biological tissue (beyond the heating effect, of course)?
How many more studies would have to be done for you to accept the no-correlation hypothesis?
The main reason is that the effect is so weak. A mission concept called LISA is being studied by ESA and NASA. The idea is to have 6 spacecraft orbiting the Sun, which together form a interferometer several million kilometers in size. The catch: Because the waves are so weak, the distances between these spacecraft would need to be controlled to within about a nanometer (!) to have any hope of detecting a signal. Needless to say a VERY challenging mission.
Terrorists would likely strike the elevator far below GEO - remember the elevator is almost 100,000 km long, and they'd be striking it within the bottom few km. This would do nothing. The operators would be like "Oh, jeez, those stupid terrorists tried to do something again, the elevator's drifting. OK, spool out another km of cable."
Remember however that the line is under very high tension, many thousands of tons equivalent. If it breaks and drifts it's not so easy to catch, hold on to, and reel back in. I suspect you'd want redundancy by having a line split into several pieces for the section in the atmosphere. This would be to protect against terrorist attack as well as accidental breakage (e.g., lightning strike)
Eventually, that information will likely come anyway, as a function of better theoretical models.
And where do these models come from, if we have no data? Science and technology cannot be advanced "on the cheap" -- a two-way interplay between theory (design) and experiment (test) is always required. This is the software developer's "test early, test often" mantra in a general form.
It's a good question, and I've often thought more scifi writers should address it. In most scifi AI is either not present (Dune), artificially rare (Star Wars, William Gibson), somehow deficient relative to the "real" intelligence of humans (Star Trek), or easily tamed into servitude (Asimov). Copouts one and all.
Ultimately, an economy is about control of resources. If the AIs never attain status of legal citizens, then humans still control the resources -- it's just a matter of who and what fraction. Most likely the wealthy get wealthier and the others lose their jobs, and a little bit of civil unrest arises to rebalance the situation (i.e., through government-mandated wealth redistribution).
The tricky case is if the AIs are given (or take) the right to control resources, e.g., by becoming legal citizens. It could be that humanity's slice of the resource pie would dwindle as AIs gain superintelligence. Humans then either: (a) merge with the AIs via neural interfaces, uploading, etc., or (b) become a secondary species, much as the non-human animals are on earth today (exploited or at best ignored).
This sounds like a fresh coat of paint on the old Fifth Generation Computing program. For you youngsters, this was a Japanese effort launched in the early 1980's to develop "intelligent" computers capable of natural language interaction, etc. At the time it was perceived as a major competitive threat to US and European technology companies.
Of course it was a colossal failure. Japan has gone from being a contender in the computing world to a nonentity. If Japan wants to get back on track, it would be well-advised to take smaller steps. Try making a robot with the intelligence of a cockroach first.
Of course the problem with the whole approach is that producing an AI is not yet an engineering exercise (as Apollo was in 1963).
> Further, the use of H1B holders is stupid for two unrelated reasons: you're shipping money to overseas economies,
And why exactly should a CEO of an American corporation care about this?
> and you're training a workforce to compete against you once it returns home (which most do).
Can you name for me a single example where an American company has suffered at the hands of competition from individuals it once employed on H1B visas?
My point: The H1B visa program may or may not be good for the US economy as a whole, but that is a completely separate issue from whether it's a good business decision for US Corporation X to employ H1B workers. If you dislike the H1B program, yell at your government officials, not at CEOs.
what goes on in the mind of the CEO. I just don't get the thinking a lot of times.
Maximize personal income
Maximize near-term corporate income or "shareholder value" (on a free cash flow basis if he/she is bright)
Treat employees well
Read these like Asimov's laws of robotics, i.e., order is significant. And like Asimov, what's really fun is when CEOs are in situations of inherent conflict between the laws. Great to see that smoke pouring out the ears.
An American company, Sage Systems, for example, has created a software program that allows utilities to "shed load instantly" if the system is at its peak and stressed to the limit, by "setting back a few thousand customers' thermostats by 2 degrees... [with] a single command over the internet"
This type of solution would manage long-term demand, but wouldn't address the typical cause of wide-scale outages -- demand spikes from sudden equipment failures that then propagate through the system.
One idea would be to have two grades of electricity available. Some outlets would be "high availability" and others would be "best effort" (I could see plugging a refrigerator and electronics into the former, and everything else into the latter). Each type would be metered separately, with the latter somewhat less expensive. When the power company needs to shed demand quickly, they cut the "best effort" power to a particular neighborhood.
I know some large businesses have this kind of tiered service with power companies (here in California anyway).
Ok, so you propose to install a big rechargeable battery in my home, so that when the grid goes down I won't lose power. And if I think I can spare it, maybe sell a little back to the grid in times of crisis. (Note: The fact that the battery is a hydrogen fuel cell is totally irrelevant -- to me it's just a battery.)
This might have utility as a competitor to the current technology (gas-powered backup generators), although as a homeowner I like the fact that a generator can run indefinitely. However, it's just silly as a solution to global power problems:
I would be an idiot to sell any power back to the grid, because I don't know when the power is coming back on. Wouldn't I feel stupid if I sold power back and then the battery ran out?
It would be far more efficient to do system-wide power buffering at a few locations. Things like pumping water back to the top of the dam when the system is under-loaded, etc. This kind of distributed network only makes sense of power generation is distributed, as in solar.
Hydrogen advocates just can't build a compelling case for anything. As a favor, please don't bother us any more until you can make a usable replacement for these crappy cell phone and laptop batteries.
You should not despise that easily, just come by and visit and you will see what this so called "comunists" are doing.
Repeat after me three times: Capitalism and democracy are not the same thing.
One is a framework for organizing an economy, the other is a framework for organizing a government. A country may have both (US, Japan, European states), democracy only (early days of post-Soviet Russia), capitalism only (China), or neither (North Korea).
The Chinese leadership has been rather shrewd in its understanding here, recognizing that by freeing its economy it provides enough material success to the common man to quell any push toward greater personal freedom (the real threat to these same leaders). We fat Westerners tend to overemphasize the importance of personal freedom to a poor and oppressed people. To lowest order, people just want to eat. It remains to be seen what will happen in China when the common man's material needs are satisfied, or if economic growth stalls.
Of course, once the cognitive model becomes good enough, the temptation (economic imperative?) will be to offload some of the actual work onto it. This idea, taken to an extreme, is the topic of an excellent short story:
"I was six years old when my parents told me that there was a small, dark jewel inside my skull, learning to be me." (Greg Egan, "Learning to be me", Axiomatic collection)
Back on the topic of augmented intelligence, Kasparov has been advocating allowing mixed human/computer teams in "Advanced Chess" tournaments. It seems that the human/machine combination, with the right interface, yields far better chess play than either alone.
creating truly random numbers using inexpensive cameras
Fussy nitpick:
As a matter of principle we shouldn't talk about "true" random numbers. Even if you have a source of quantum randomness (the best you can do), in any practical scheme there will always be differences in how 0's and 1's are detected. The impact of these differences can be minimized through careful design, but they will always be there and will cause a departure from an exact 50/50 split of 0's and 1's.
For example, an attenuated laser passing through a simple 50/50 beamsplitter and into 2 single-photon detectors is far better than a lava lamp. But no beamsplitter is exactly 50/50, and this will cause an excess of 0's or 1's. (You could adjust the detection efficiency or optical alignment of the detectors to compensate, but the point is that this is a manual adjustment that will never result in exactly zero bias.) And no amount of algorithmic hashing or obfuscation can eliminate this bias (although it may reduce it significantly).
It will always be the case that any practical system for generating bit sequences of length N will never generate them with a frequency of exactly 2^(-N) each (in the limit of an infinite number of trials), and with zero correlation between successive bit sequences. "Truly" random numbers are an unattainable goal.
I've always wondered if the double-slit experiment, in a particle mode, would be a good way to generate random numbers.
These would certainly be random numbers in the strongest possible sense (quantum randomness). Couple of comments:
1) It will generally be biased to favor 0's or 1's, since each detector won't be equally illuminated. You minimize this with optical alignment, but it will always be there in some degree. This bias is a potential source of weakness, and no amount of hashing or other obfuscation really eliminates it. (Note: most other techniques, including lava lamps, share this weakness.)
2) There are easier ways of doing it. For example, the double slit isn't needed; a simple defocusing lens on a laser, illuminating two detectors would do the same thing. Or you could use a 50/50 beamsplitter and two detectors.
...for example, I think that Godel's incompleteness theorem hasn't changed society in the slightest.
Yes it has. It means that nobody can ever make a compiler that spots runtime errors. Software would be considerably less buggy were this not so, and I'm guessing that a majority of individuals in modern society have been directly or indirectly affected by software bugs.
Sorry, it's just that this info is old hat to anyone who keeps up on astro-phys stuff.
I agree, but this is mostly the fault of bad moderation. Some of the questions were actually quite provoking if you read them. The fact that only the softball RTFB ones got modded up is a punishment we inflict on ourselves.
...there are a large number of fundamental constants which are balanced just perfectly to allow complex systems to arise in the Universe. This creates a question: how did this perfect balance come to be?
You know, I flipped a coin 30 times the other day and came out with the following result: HTHHTTTHHTHHTHTTHHTHHTTTHHTHTT
Now I ask you: what is the likelihood of my getting exactly this sequence of heads and tails? And to think I got exactly this, without a single mistake! I can't believe my luck! Clearly the result of divine intervention!
(Message: Be careful trying to apply probability theory when the result is a priori known.)
How can science have anything truly definitive to say about the distant future of the universe?
It seems to me that the action of life itself generates unpredictability, and that we likely have no more conception of what life will be like in 10 million years than a chimp does of us. Rather than talk about the distant future of the universe, how about this much simpler question:
What will the surface of the earth be like in one million years?
(If you don't think this question is simpler, then you are making assumptions about the scope of life's potential impact on the physical world.)
If you want to instead think of life as just another physical process, then the question boils down to:
Are the laws of the universe such that the future can be predicted (or at least broadly characterized)? Or like many cellular automata (e.g., Conway's Life), is the only way to predict the future to in effect run a simulation?
given how the US/Canadian networks are intermingled, you can't be sure packets won't go through the US. As a matter of fact, to go to my library, eight blocks from my home, packets go through New-York City
This rerouting through the US is an intentional feature of every router we sell you guys. It's so we can keep tabs on what you're up to.
This idea amounts to statistical multiplexing, where you get the most efficient use of a fixed resource (in this case CDs) by pooling consumer demand for that resource. An IP router works this way to manage demand for bandwidth.
Statistical multiplexing has real value when the average consumer demands far less than the entire resource. If you play a typical CD in your collection for an hour per week, then that CD is being wasted for the other 167 hours it's sitting on the shelf unused. The idea is to share this single CD with 167 other friends and thereby distribute the costs. (The real answer is somewhat less than 167 in order to decrease the likelihood of collisions when trying to access the resource -- "buffer overflow" in the router analogy.) Interestingly the method is most valuable when applied to the music that you play the least.
When applied to a large number of subscribers (large meaning much larger than the number of distinct CDs in release), the economics are very attractive. If the average subscriber listens to music 10% of the time, then the per-subscriber cost of purchasing the entire central CD repository is only 10% the cost of a single CD!
One challenge would be convincing the court that the technology really does limit the number of simultaneous listeners, and that there's no way to spoof it (this killed mp3.com). At a minimum you'd need a persistent net connection to request/release control of CDs as you play them. I could see a scenario where you purchase the music that you don't want to be chained to the net to listen to (e.g., in your iPod), and you listen to the rest through the service.
Slashdot Mirror
To become a business that succeeds over the long haul, you need to become difficult to replicate for some reason. Fundamentally there are only a few ways to stand out, and not all of these are necessarily viable strategies for a given market:
Companies that undoubtedly have a defensible position include:
Companies that are iffier:
I don't know how you rate "most popular." Since computer use has been exploding exponentially, if you do it by user head count, no computer that's more that a couple of years old would count.
I doubt this is true. Commodore sold over 20 million C64s, and I would be very surprised if any specific model of Dell/IBM/whatever sold that many units today. The fact that there are so many nearly-identical models today means that none of them stand out enough to sell in large numbers.
I also doubt your statement that computer use is "expanding exponentially". In fact the growth rate of computer sales has been declining steadily, and it is quickly becoming a replacement business rather than one of selling to new users.
Even if there were indeed some intelligent life in the Gem 37 system, only in the first case (tens of light years) could we hope to ever communicate with them.
The goal of TPF is somewhat broader than SETI, which will only find life that is intelligent (i.e. capable and willing to send beamed radio transmissions to other stars). TPF will hunt for signs of ANY biological activity by looking for certain chemical signatures (e.g., an oxygen atmosphere).
The exciting thing about TPF is that it skips several (likely small) terms in the Drake equation. It is probably the likeliest near-term chance we have to find extraterrestrial life.
However, as a new parent, a scientist, and a PhD, I'll admit that I too have worries about WiFi and growing children. And I don't wear a tinfoil hat.
By coincidence I am also a new parent and a PhD (physics), and I understand firsthand your concern for your baby's safety. But consider the following:
Why couldn't we put this lab in orbit?
The main reason is that the effect is so weak. A mission concept called LISA is being studied by ESA and NASA. The idea is to have 6 spacecraft orbiting the Sun, which together form a interferometer several million kilometers in size. The catch: Because the waves are so weak, the distances between these spacecraft would need to be controlled to within about a nanometer (!) to have any hope of detecting a signal. Needless to say a VERY challenging mission.
A lot of other interesting missions would be enabled by good formation flight technology. Look at NASA's Terrestrial Planet Finder mission, or the ESA's similar Darwin mission.
Terrorists would likely strike the elevator far below GEO - remember the elevator is almost 100,000 km long, and they'd be striking it within the bottom few km. This would do nothing. The operators would be like "Oh, jeez, those stupid terrorists tried to do something again, the elevator's drifting. OK, spool out another km of cable."
Remember however that the line is under very high tension, many thousands of tons equivalent. If it breaks and drifts it's not so easy to catch, hold on to, and reel back in. I suspect you'd want redundancy by having a line split into several pieces for the section in the atmosphere. This would be to protect against terrorist attack as well as accidental breakage (e.g., lightning strike)
Eventually, that information will likely come anyway, as a function of better theoretical models.
And where do these models come from, if we have no data? Science and technology cannot be advanced "on the cheap" -- a two-way interplay between theory (design) and experiment (test) is always required. This is the software developer's "test early, test often" mantra in a general form.
do not welcome our Microsoft overlords.
What do the people do?
It's a good question, and I've often thought more scifi writers should address it. In most scifi AI is either not present (Dune), artificially rare (Star Wars, William Gibson), somehow deficient relative to the "real" intelligence of humans (Star Trek), or easily tamed into servitude (Asimov). Copouts one and all.
Ultimately, an economy is about control of resources. If the AIs never attain status of legal citizens, then humans still control the resources -- it's just a matter of who and what fraction. Most likely the wealthy get wealthier and the others lose their jobs, and a little bit of civil unrest arises to rebalance the situation (i.e., through government-mandated wealth redistribution).
The tricky case is if the AIs are given (or take) the right to control resources, e.g., by becoming legal citizens. It could be that humanity's slice of the resource pie would dwindle as AIs gain superintelligence. Humans then either: (a) merge with the AIs via neural interfaces, uploading, etc., or (b) become a secondary species, much as the non-human animals are on earth today (exploited or at best ignored).
This sounds like a fresh coat of paint on the old Fifth Generation Computing program. For you youngsters, this was a Japanese effort launched in the early 1980's to develop "intelligent" computers capable of natural language interaction, etc. At the time it was perceived as a major competitive threat to US and European technology companies.
Of course it was a colossal failure. Japan has gone from being a contender in the computing world to a nonentity. If Japan wants to get back on track, it would be well-advised to take smaller steps. Try making a robot with the intelligence of a cockroach first.
Of course the problem with the whole approach is that producing an AI is not yet an engineering exercise (as Apollo was in 1963).
> Further, the use of H1B holders is stupid for two unrelated reasons: you're shipping money to overseas economies,
And why exactly should a CEO of an American corporation care about this?
> and you're training a workforce to compete against you once it returns home (which most do).
Can you name for me a single example where an American company has suffered at the hands of competition from individuals it once employed on H1B visas?
My point: The H1B visa program may or may not be good for the US economy as a whole, but that is a completely separate issue from whether it's a good business decision for US Corporation X to employ H1B workers. If you dislike the H1B program, yell at your government officials, not at CEOs.
what goes on in the mind of the CEO. I just don't get the thinking a lot of times.
Read these like Asimov's laws of robotics, i.e., order is significant. And like Asimov, what's really fun is when CEOs are in situations of inherent conflict between the laws. Great to see that smoke pouring out the ears.
An American company, Sage Systems, for example, has created a software program that allows utilities to "shed load instantly" if the system is at its peak and stressed to the limit, by "setting back a few thousand customers' thermostats by 2 degrees ... [with] a single command over the internet"
This type of solution would manage long-term demand, but wouldn't address the typical cause of wide-scale outages -- demand spikes from sudden equipment failures that then propagate through the system.
One idea would be to have two grades of electricity available. Some outlets would be "high availability" and others would be "best effort" (I could see plugging a refrigerator and electronics into the former, and everything else into the latter). Each type would be metered separately, with the latter somewhat less expensive. When the power company needs to shed demand quickly, they cut the "best effort" power to a particular neighborhood.
I know some large businesses have this kind of tiered service with power companies (here in California anyway).
Ok, so you propose to install a big rechargeable battery in my home, so that when the grid goes down I won't lose power. And if I think I can spare it, maybe sell a little back to the grid in times of crisis. (Note: The fact that the battery is a hydrogen fuel cell is totally irrelevant -- to me it's just a battery.)
This might have utility as a competitor to the current technology (gas-powered backup generators), although as a homeowner I like the fact that a generator can run indefinitely. However, it's just silly as a solution to global power problems:
Hydrogen advocates just can't build a compelling case for anything. As a favor, please don't bother us any more until you can make a usable replacement for these crappy cell phone and laptop batteries.
You should not despise that easily, just come by and visit and you will see what this so called "comunists" are doing.
Repeat after me three times: Capitalism and democracy are not the same thing.
One is a framework for organizing an economy, the other is a framework for organizing a government. A country may have both (US, Japan, European states), democracy only (early days of post-Soviet Russia), capitalism only (China), or neither (North Korea).
The Chinese leadership has been rather shrewd in its understanding here, recognizing that by freeing its economy it provides enough material success to the common man to quell any push toward greater personal freedom (the real threat to these same leaders). We fat Westerners tend to overemphasize the importance of personal freedom to a poor and oppressed people. To lowest order, people just want to eat. It remains to be seen what will happen in China when the common man's material needs are satisfied, or if economic growth stalls.
Of course, once the cognitive model becomes good enough, the temptation (economic imperative?) will be to offload some of the actual work onto it. This idea, taken to an extreme, is the topic of an excellent short story:
"I was six years old when my parents told me that there was a small, dark jewel inside my skull, learning to be me." (Greg Egan, "Learning to be me", Axiomatic collection)
Back on the topic of augmented intelligence, Kasparov has been advocating allowing mixed human/computer teams in "Advanced Chess" tournaments. It seems that the human/machine combination, with the right interface, yields far better chess play than either alone.
creating truly random numbers using inexpensive cameras
Fussy nitpick:
As a matter of principle we shouldn't talk about "true" random numbers. Even if you have a source of quantum randomness (the best you can do), in any practical scheme there will always be differences in how 0's and 1's are detected. The impact of these differences can be minimized through careful design, but they will always be there and will cause a departure from an exact 50/50 split of 0's and 1's.
For example, an attenuated laser passing through a simple 50/50 beamsplitter and into 2 single-photon detectors is far better than a lava lamp. But no beamsplitter is exactly 50/50, and this will cause an excess of 0's or 1's. (You could adjust the detection efficiency or optical alignment of the detectors to compensate, but the point is that this is a manual adjustment that will never result in exactly zero bias.) And no amount of algorithmic hashing or obfuscation can eliminate this bias (although it may reduce it significantly).
It will always be the case that any practical system for generating bit sequences of length N will never generate them with a frequency of exactly 2^(-N) each (in the limit of an infinite number of trials), and with zero correlation between successive bit sequences. "Truly" random numbers are an unattainable goal.
I've always wondered if the double-slit experiment, in a particle mode, would be a good way to generate random numbers.
These would certainly be random numbers in the strongest possible sense (quantum randomness). Couple of comments:
1) It will generally be biased to favor 0's or 1's, since each detector won't be equally illuminated. You minimize this with optical alignment, but it will always be there in some degree. This bias is a potential source of weakness, and no amount of hashing or other obfuscation really eliminates it. (Note: most other techniques, including lava lamps, share this weakness.)
2) There are easier ways of doing it. For example, the double slit isn't needed; a simple defocusing lens on a laser, illuminating two detectors would do the same thing. Or you could use a 50/50 beamsplitter and two detectors.
Now if only there was .wmv for Mac, for pr0n...
This here will play most wmv files on OS X.
Yes it has. It means that nobody can ever make a compiler that spots runtime errors. Software would be considerably less buggy were this not so, and I'm guessing that a majority of individuals in modern society have been directly or indirectly affected by software bugs.
Sorry, it's just that this info is old hat to anyone who keeps up on astro-phys stuff.
I agree, but this is mostly the fault of bad moderation. Some of the questions were actually quite provoking if you read them. The fact that only the softball RTFB ones got modded up is a punishment we inflict on ourselves.
You know, I flipped a coin 30 times the other day and came out with the following result: HTHHTTTHHTHHTHTTHHTHHTTTHHTHTT
Now I ask you: what is the likelihood of my getting exactly this sequence of heads and tails? And to think I got exactly this, without a single mistake! I can't believe my luck! Clearly the result of divine intervention!
(Message: Be careful trying to apply probability theory when the result is a priori known.)
How can science have anything truly definitive to say about the distant future of the universe?
It seems to me that the action of life itself generates unpredictability, and that we likely have no more conception of what life will be like in 10 million years than a chimp does of us. Rather than talk about the distant future of the universe, how about this much simpler question:
What will the surface of the earth be like in one million years?
(If you don't think this question is simpler, then you are making assumptions about the scope of life's potential impact on the physical world.)
If you want to instead think of life as just another physical process, then the question boils down to:
Are the laws of the universe such that the future can be predicted (or at least broadly characterized)? Or like many cellular automata (e.g., Conway's Life), is the only way to predict the future to in effect run a simulation?
given how the US/Canadian networks are intermingled, you can't be sure packets won't go through the US. As a matter of fact, to go to my library, eight blocks from my home, packets go through New-York City
This rerouting through the US is an intentional feature of every router we sell you guys. It's so we can keep tabs on what you're up to.
A safe Canadian is a surveiled Canadian.
This idea amounts to statistical multiplexing, where you get the most efficient use of a fixed resource (in this case CDs) by pooling consumer demand for that resource. An IP router works this way to manage demand for bandwidth.
Statistical multiplexing has real value when the average consumer demands far less than the entire resource. If you play a typical CD in your collection for an hour per week, then that CD is being wasted for the other 167 hours it's sitting on the shelf unused. The idea is to share this single CD with 167 other friends and thereby distribute the costs. (The real answer is somewhat less than 167 in order to decrease the likelihood of collisions when trying to access the resource -- "buffer overflow" in the router analogy.) Interestingly the method is most valuable when applied to the music that you play the least.
When applied to a large number of subscribers (large meaning much larger than the number of distinct CDs in release), the economics are very attractive. If the average subscriber listens to music 10% of the time, then the per-subscriber cost of purchasing the entire central CD repository is only 10% the cost of a single CD!
One challenge would be convincing the court that the technology really does limit the number of simultaneous listeners, and that there's no way to spoof it (this killed mp3.com). At a minimum you'd need a persistent net connection to request/release control of CDs as you play them. I could see a scenario where you purchase the music that you don't want to be chained to the net to listen to (e.g., in your iPod), and you listen to the rest through the service.