Basically, the accuracy of a reference clock in principle determined by making two identical clocks, starting them synchronized, and measuring how long before they drift by an average of 1 second.
This has to do with how cold your atomic fountain is, and how well you isolate the particular magnetic sublevel you define the second in terms of.
Now, if you want to move to a mercury standard clock, you can do two things: first, calculate with QED the ratio of the freqencies between the transitions in cesium and mecury of interest. I don't know if we can do this well enough for these purposes or not. Second, you can redefine the second in terms of the oscillation frequency of some mercury transition at least within the accuracy of a current cesium clock.
The important thing to note is that mand physicists don't really care about how long a second is, as much as they care that two clocks run at the same rate, even if it is wrong.
I realize you probably understand what the article meant, but give me a break on the MS NBC conspiracy theory. Despite being a partnership between MS and NBC, they show a remarkable lack of favoratism towards MS.If anything, I would say it goes the other way: they go out of their way to disparage MS.
What I am saying, I guess, is never insult with style when you can insult with substance. MSNBC isn't exactly hampered by grade A reporting. This article is a prime example of their particular brand of News-lite.
Basically, it sounds like they called up a couple of people and asked their opinion on MS. Some of them didn't like it. There is not attempt to gather facts, or even a wide range of opinions, no attempt to delve into the reasons these people prefer Linux to Windows other than the simplistic "open source software lets us do more" and "Windows product activation is annoying", both of which are true, but hardly capture the reality of the situation. This is supposed to be exposing a trend, but provides only anecdotal evidence, nothign to indicate whether this is a real movement, or just the opinion of 3 or 4 guys.
At least the corporate PR-news I am used to seeing billed as "tech news" frequently contain facts, however slanted the tone may be.
Assuming he is guilty of this, it depends. If your goal is to cause the company to be devalues so seriously you can pick up their assets at a fraction of the cost, you want to disrupt their service in any way you can.
But more to the point, why is everyone here being so quick to condemn this guy. While I agree it is an odd situation, this is a pretty serious charge, and I think he deserves the benefit of the doubt.
Basically, I think that @Home were idiots. They made some really bad business decisions (Excite). Their cable partner contracts were poorly done, leaving them at a huge disadvantage when negotiating contract renewals. The "new economy" was crashing down around their ears, cutting into new subscriber rates, and DSL was aquiring a reputation for much better quality of service. AT&T gave them a not unreasonable buy out bid, but the @Home bondholders refused to accept that distressed assets sell at bargain prices. AT&T took their ball and went home, and @Home got nothing.
On the other hand, it is almost laughable to consider the possibility tht AT&T needed to spy on the fragile and unreliable @Home infrastructure to figure out how to make a nationwide network.
Not to put words in his mouth, but I believe he would say something like "It isn't where you are going, but how you get there"
And, if you want to get technical, no it isn't. The B5 call to arms/Crusade deal is about searching for a cure that is *going* to wipe out earth if it isn't found.
This is about rebuilding civilization from the ashes after a major catastrophe. Still, many other people have explored this. However, each is does it in a different way, from a different perspective. I hope jms will do it well (though I don't have showtime, so I won't be able to watch it the first time through)
IBM is largely targeting the markets where they number of instances > number of CPUs. Or, where you want peak loads of different applications load balanced dynamically (ie, without having to say "move this processor to that node").
As far as I know, Sun's dynamic domains are dynamic in the sense that they can be reconfigured on the fly, not that they automatically move resources where they are needed.
While it is true, I think it is not as stable as VM if one instance crashes. Also, while they tout the ability to dynamically reallocate processors, a friend of mine who tried it (on an E10K) said that if you tried to do it while running Sun Cluster (I think), the whole machine crashed. He claimed that this was not documented anywhere, though it is possible he just couldn't find it. In any case, having to choose between dynamic partitioning and automatic failover is kind of stupid.
Really? Out of the dozens of people I have heard bitch about AIX, not one has said it was unstable.
I have seen it used as a network server, a build machine, a graphics workstation (no joke!), a personal workstation used for scientific simulations, and a database server, all under relatively heavy load, and nobody has complained about it crashing.
Plenty of other complaints, though. Usually it goes something like "Would you believe this shit!!! XYZ is prototyped as returning a char instead of a long in the C libraries. At least it is stable, though"
Which they mention in the article. Basically Sun wants to use linux on low end servers like web servers that can be clustered to very high density to meed demand, and use their Enterprise line for "mainframe like" tasks.
To some extent he is right. I don't really think that buying a mainframe to replace 20 servers, even if they aren't the 1k machines Sun was talking about. However, if you get to 100 servers, or you have a mainframe with spare capacity used for other purposes, I think Linux on the s390 can be a good choice.
Sun and HP are both doing this now, as well, at least with CPUs (I don't think the do it with memory).
I know Sun had problems with the "hot add CPUs" idea -- people just didn't want to plug a cpu module into a million dollar machine while it was running and handling millions of dollars of transactions a day. Also, I know at least one case of someone trying to do that, and having the Sun technitian blow 3 CPU modules before giving up on it.
Given that the marginal cost of a CPU is low compared to the development costs, it really makes sense to ship disabled CPUs and turn them on as the customer needs them.
True. I know a lot of the early wind programs had pretty bad maintenence problems (I have seen several wind farms where at any given moment, half the windmills are offline), but I assume that has gotten better. I should not have excluded that -- wind power is excellent.
While ethanol produces greenhouse gasses, it is in direct proportion to the greenhouse gasses fixed by the growing of the corn.
That said, ethanol is a terribly inefficient fuel. The last dept. of agriculture study on it showed that the NEV (net energy value) of corn ethanol is somewhere around 1.25. Basically, this means that every 1.25 joules of ethanol energy you want, you need to spend 1 joule of energy on farm vehicle fuel, transportation, fertilizer, etc, all the while using up the most fertile land in the country on energy production, since corn needs much more fertile land than most other crops.
While I believe that we now have the technology and experience to build a relatively safe nuclear reactor (compared with any other kind of power plant), we have a limited amount of fissionable materials. I heard one calculation (I don't remember the source, so treat it skeptically) claim if the whole US converted to nuclear power, we would have about 20-30 years worth of power. Despite massive amounts of development money poured into them, nobody has demonstrated an ability to run breeder reactors cost effectivly, much less safely.
The best gasoline replacement I know of is methanol. Methanol can be generated from basically any plant, rather than only sucrose rich plants like corn. Some fast growing trees have a NEV as high as 25, can grow on land poorly suited to growing food products, has a multiple year harvest cycle, reducing errosion, and is all around a Good Thing(tm). these guys we could replace 1/2 of all gasoline consumtion with methanol without significantly affecting food prices. While that doesn't solve the electricity issue, it goes a long way towards reducing pollution and greenhouse emissions from cars. Plus methanol can either be burned in a traditional internal combustion engine, or used to power fuel cells. Thus, it could be implemented now, with existing technology while easy a transition to fuel cells if the power density issues are solved.
For electrical generation, I still want to hold out for solar, but it looks like it is going to be a while before the cost/kW is reasonable. I actually don't think it should be that hard to do so, it is just that most current research and demand for solar energy (ie, the space program) cares more about efficiency than cost. If the govt (or anyone with money) were to set a goal of solar cells with 1/2 the efficiency of current cells, but 1/10 the cost, I think we could acheive it in 10 years with moderate investment.
Well, assuming the energy liberated by annihilation doesn't destroy the nucleus, as you step down the periodic table, you will form progressivly more unstable isotopes. Eventually, the nucleus will undergo either nuclear fission, releasing several free neutrons, or beta decay, converting a neutron to a proton.
That said, I think the better way to accomplish this task is neutron capture. Since neutrons are neutral, you can use slow neutrons to bombard your target, drastically reducing the excess energy deposited.
Alcohol has a blood thinning property. This is currently believed to be the source of much of the aparent health benefits of moderate alcohol consumption -- thinning the blood reduces the risk of stroke and heart disease substantially.
Taking a half dose of asprin/day has a similar effect, and is a lot cheaper and safer than alcohol, but not nearly as much fun:)
First, Grovers algorithm does searching unordered lists, not sorting. In fact, I believe it has been "proved" that sorting cannot be done faster than O(N * ln N) with a quantum computer, though take thost proofs with a grain of salt, there is always the possibility of attacking the assumptions.
Anyway, I want to point out that searching is actually a very general algorithm. Basically, any algorithm that is much faster to validate than construct a solution can be done that way. For instace, one way to factor is to iterate through all numbers less than sqrt(N), testing each one by dividing N by it. This takes O(sqrt(N)) time. Grover's algorithm speeds this up to O(sqrt(sqrt(N)), which is still exponential in the number of bits in N. Not that this is NOT Shor's algorithm, which can indeed factor numbers in polynomic time in the number of bits.
Re:Your out of luck with Dish Network...
on
Comparing the DVRs?
·
· Score: 2
I belive the accessory port on the dish receiver is only an output for its own IR blaster, which can control a normal VCR to record off the dish. If not, I don't believe the specs are published that would allow the Tivo to control it.
Other DN receivers might have a serial port, but mine doesn't.
Re:Your out of luck with Dish Network...
on
Comparing the DVRs?
·
· Score: 2
Funny, my Tivo works fine with my Dish Network.
You need to use the IR blaster from the Tivo to change channels on the dish receiver, but it works fine.
Changing channels is really slow, since you have the combined latencies of the IR blaster, the dish receiver, and the MPEG encoder/decoder on the Tivo, but with the onscreen channel listing, I don't channel surf anymore.
The one thing that annoys me is that occasinally (I think when the dish receiver gets a firmware update), my receiver will turn off, and the Tivo has no way of knowing and doesn't turn it back on until I get home and notice that recordings are failing. I solved this by setting an auto-tune timer on my dish receiver, which wakes it up at 4:59, so at least the Simpsons and Babylon 5 get recorded.
Re:Portable MP3 Player to Plug In To Car Audio?
on
Review: SliMP3
·
· Score: 2
Mine does, though admitedly on the back, not the front. Many after market CD players and the like have a pair of RCA inputs on the back for hooking up to the original deck. It isn't so hard to bring this around front if you, say have a DIN head unit in a double DIN space.
Cool. I just looked at the back of my Sony Jukebox, and I also have the slink interface.
It looks like the only thing you are looking for is the audio connection, and yes, many new sound cards have both coax and fiber SPDIF IO. Check out the creative live/audigy, or the Soyo DRAGON motherboard. The digital audio should be lossless, and error corrected by the jukebox electronics, so you should be able to record with any standard recording software (wavr on my debian box). Throw together some scripts to synchronize the recording start with the signal from the jukebox and you should be set.
One thing you can do is build your own CD changer robot. If the only feature you care about is "remove disk from ejected tray and insert next disk", it shouldn't be that hard. But unless you have more than a couple hundred CDs, I suspect you are best buying 3 or 4 CD-ROMs and taking a day to feed CDs into them as fast as you can.
This would be way cool, but I forsee two major problems:
1) Speed. AFAIK, multi-disk CD changers only read at 1X. Even with the highest qualtiy settings, I can encode at 3-4 times that rate on my dual CPU PIII.
2) Access to TOC. This is the real killer: if you want all the nice freedb lookups to work right, you need to extract the TOC from the disk and compute a hash of it. I am almost positive this doesn't go down the SPDIF line.
The speed I could deal with (just leave it running when you go on vacation for a week or so), but unless you want a hard drive full of unnamed.mp3 files, you need to solve the TOC problem.
I have been using ResierFS for a few months now, and for the most part, it has worked well.
But the lack of a fsck is inexcusable. I have has a couple of "events" which might have corrupted a filesystem (my fault, not resierfs'), but had no way to check. Once, I went through my partitions one by one, copied them to a temp partition, mkfs'ed them, and copied them back. I put up with it because it was a pain in the butt to switch. Now that ext3 is in a Linus kernel, I am seriously considering running some tests, and converting back.
I used to be a skeptic about the possibility of quantum computing, but I have become a believer. Here is why:
1) It once was thought that gate precisions would need to be accurate to 1 in 10^12 or so, depending on the length of the computation. The problem was, unlike classical computers, quantum computers are not allowed to "latch" to a 1 or 0. However, there has been a substantial body of work that has shown how to implement error correction codes on quantum computers, without destroying the computation.
2) Decoherence, or the problem with entropic contamination by the environment, is a serious one, and the one that was always thought to be the nail in the coffin. However, now that people have actually started trying to make systems resilient to decoherence, they have measured systems that decohere on time scales from ms to hours -- giving considerable time to do computation.
3) One particular QC proposal, germanium quantum dots on silicon, has made incredible progress recently. A couple of years ago, someone wrote down the "six impossible tasks to make Si quantum computers work", three of which have already been accomplished.
We have already accomplished so much in QC that people thought would be impossible, it hardly seems logical to give up now. I can't say for certain, obviously, that QC will ever be a reality, but I would give it better than 50-50 odds at the moment.
Orthogonality in this case refers to modes of the electric field. A mode is labeled by a frequency and a polarization state. So, yes, two beams of different wavelength are orthogonal modes, and can be resolved by using a grating to diffract each component to a seperate detector. This is called WDM -- wavelength division multiplexing.
States with opposite polarization (horizontal and vertical, right and left circular, etc) are also orthogonal modes, and can be seperated, for instance, by a polarizing beam splitter. This is PDM.
The total bandwidth of a communication link is bps/mode * useful modes, so either increasing the number of frequencies or polarizations, or both, can improve bandwidth.
WDM is limited because each pulse actually covers a range of frequencies, and you need to choose them far enough apart that they don't overlap, or you won't be able to discriminate them well. PDM is limited because it is hard to get fibers to not fs*k with the polarization of light, plus there are only two orthogonal states, so you can only easily get a factor of 2 improvment in bandwidth.
While there is certainly a risk that porn site owners will abuse credit card data, and it is certainly easy to villianize the entire porn industry, but I don't think that is the real issue. I should have to, or even be asked, to give someone my credit card number unless I am buying something from them. Using credit cards as age verification is convenient but dumb.
People will say "If you don't like it, don't surf for porn", but they are implicitly making a moral judgment about viewing porn... They really mean "You dirty person, you shouldn't do that anyway, so don't complain about giving out your credit card."
Slashdot Mirror
Basically, the accuracy of a reference clock in principle determined by making two identical clocks, starting them synchronized, and measuring how long before they drift by an average of 1 second.
This has to do with how cold your atomic fountain is, and how well you isolate the particular magnetic sublevel you define the second in terms of.
Now, if you want to move to a mercury standard clock, you can do two things: first, calculate with QED the ratio of the freqencies between the transitions in cesium and mecury of interest. I don't know if we can do this well enough for these purposes or not. Second, you can redefine the second in terms of the oscillation frequency of some mercury transition at least within the accuracy of a current cesium clock.
The important thing to note is that mand physicists don't really care about how long a second is, as much as they care that two clocks run at the same rate, even if it is wrong.
I realize you probably understand what the article meant, but give me a break on the MS NBC conspiracy theory. Despite being a partnership between MS and NBC, they show a remarkable lack of favoratism towards MS.If anything, I would say it goes the other way: they go out of their way to disparage MS.
What I am saying, I guess, is never insult with style when you can insult with substance. MSNBC isn't exactly hampered by grade A reporting. This article is a prime example of their particular brand of News-lite.
Basically, it sounds like they called up a couple of people and asked their opinion on MS. Some of them didn't like it. There is not attempt to gather facts, or even a wide range of opinions, no attempt to delve into the reasons these people prefer Linux to Windows other than the simplistic "open source software lets us do more" and "Windows product activation is annoying", both of which are true, but hardly capture the reality of the situation. This is supposed to be exposing a trend, but provides only anecdotal evidence, nothign to indicate whether this is a real movement, or just the opinion of 3 or 4 guys.
At least the corporate PR-news I am used to seeing billed as "tech news" frequently contain facts, however slanted the tone may be.
Assuming he is guilty of this, it depends. If your goal is to cause the company to be devalues so seriously you can pick up their assets at a fraction of the cost, you want to disrupt their service in any way you can.
But more to the point, why is everyone here being so quick to condemn this guy. While I agree it is an odd situation, this is a pretty serious charge, and I think he deserves the benefit of the doubt.
Basically, I think that @Home were idiots. They made some really bad business decisions (Excite). Their cable partner contracts were poorly done, leaving them at a huge disadvantage when negotiating contract renewals. The "new economy" was crashing down around their ears, cutting into new subscriber rates, and DSL was aquiring a reputation for much better quality of service. AT&T gave them a not unreasonable buy out bid, but the @Home bondholders refused to accept that distressed assets sell at bargain prices. AT&T took their ball and went home, and @Home got nothing.
On the other hand, it is almost laughable to consider the possibility tht AT&T needed to spy on the fragile and unreliable @Home infrastructure to figure out how to make a nationwide network.
Not to put words in his mouth, but I believe he would say something like "It isn't where you are going, but how you get there"
And, if you want to get technical, no it isn't. The B5 call to arms/Crusade deal is about searching for a cure that is *going* to wipe out earth if it isn't found.
This is about rebuilding civilization from the ashes after a major catastrophe. Still, many other people have explored this. However, each is does it in a different way, from a different perspective. I hope jms will do it well (though I don't have showtime, so I won't be able to watch it the first time through)
IBM is largely targeting the markets where they number of instances > number of CPUs. Or, where you want peak loads of different applications load balanced dynamically (ie, without having to say "move this processor to that node").
As far as I know, Sun's dynamic domains are dynamic in the sense that they can be reconfigured on the fly, not that they automatically move resources where they are needed.
While it is true, I think it is not as stable as VM if one instance crashes. Also, while they tout the ability to dynamically reallocate processors, a friend of mine who tried it (on an E10K) said that if you tried to do it while running Sun Cluster (I think), the whole machine crashed. He claimed that this was not documented anywhere, though it is possible he just couldn't find it. In any case, having to choose between dynamic partitioning and automatic failover is kind of stupid.
Really? Out of the dozens of people I have heard bitch about AIX, not one has said it was unstable.
I have seen it used as a network server, a build machine, a graphics workstation (no joke!), a personal workstation used for scientific simulations, and a database server, all under relatively heavy load, and nobody has complained about it crashing.
Plenty of other complaints, though. Usually it goes something like "Would you believe this shit!!! XYZ is prototyped as returning a char instead of a long in the C libraries. At least it is stable, though"
Which they mention in the article. Basically Sun wants to use linux on low end servers like web servers that can be clustered to very high density to meed demand, and use their Enterprise line for "mainframe like" tasks.
To some extent he is right. I don't really think that buying a mainframe to replace 20 servers, even if they aren't the 1k machines Sun was talking about. However, if you get to 100 servers, or you have a mainframe with spare capacity used for other purposes, I think Linux on the s390 can be a good choice.
Sun and HP are both doing this now, as well, at least with CPUs (I don't think the do it with memory).
I know Sun had problems with the "hot add CPUs" idea -- people just didn't want to plug a cpu module into a million dollar machine while it was running and handling millions of dollars of transactions a day. Also, I know at least one case of someone trying to do that, and having the Sun technitian blow 3 CPU modules before giving up on it.
Given that the marginal cost of a CPU is low compared to the development costs, it really makes sense to ship disabled CPUs and turn them on as the customer needs them.
True. I know a lot of the early wind programs had pretty bad maintenence problems (I have seen several wind farms where at any given moment, half the windmills are offline), but I assume that has gotten better. I should not have excluded that -- wind power is excellent.
While ethanol produces greenhouse gasses, it is in direct proportion to the greenhouse gasses fixed by the growing of the corn.
That said, ethanol is a terribly inefficient fuel. The last dept. of agriculture study on it showed that the NEV (net energy value) of corn ethanol is somewhere around 1.25. Basically, this means that every 1.25 joules of ethanol energy you want, you need to spend 1 joule of energy on farm vehicle fuel, transportation, fertilizer, etc, all the while using up the most fertile land in the country on energy production, since corn needs much more fertile land than most other crops.
While I believe that we now have the technology and experience to build a relatively safe nuclear reactor (compared with any other kind of power plant), we have a limited amount of fissionable materials. I heard one calculation (I don't remember the source, so treat it skeptically) claim if the whole US converted to nuclear power, we would have about 20-30 years worth of power. Despite massive amounts of development money poured into them, nobody has demonstrated an ability to run breeder reactors cost effectivly, much less safely.
The best gasoline replacement I know of is methanol. Methanol can be generated from basically any plant, rather than only sucrose rich plants like corn. Some fast growing trees have a NEV as high as 25, can grow on land poorly suited to growing food products, has a multiple year harvest cycle, reducing errosion, and is all around a Good Thing(tm). these guys we could replace 1/2 of all gasoline consumtion with methanol without significantly affecting food prices. While that doesn't solve the electricity issue, it goes a long way towards reducing pollution and greenhouse emissions from cars. Plus methanol can either be burned in a traditional internal combustion engine, or used to power fuel cells. Thus, it could be implemented now, with existing technology while easy a transition to fuel cells if the power density issues are solved.
For electrical generation, I still want to hold out for solar, but it looks like it is going to be a while before the cost/kW is reasonable. I actually don't think it should be that hard to do so, it is just that most current research and demand for solar energy (ie, the space program) cares more about efficiency than cost. If the govt (or anyone with money) were to set a goal of solar cells with 1/2 the efficiency of current cells, but 1/10 the cost, I think we could acheive it in 10 years with moderate investment.
Just as a note, positrons and protons are not the same thing.
Well, assuming the energy liberated by annihilation doesn't destroy the nucleus, as you step down the periodic table, you will form progressivly more unstable isotopes. Eventually, the nucleus will undergo either nuclear fission, releasing several free neutrons, or beta decay, converting a neutron to a proton.
That said, I think the better way to accomplish this task is neutron capture. Since neutrons are neutral, you can use slow neutrons to bombard your target, drastically reducing the excess energy deposited.
Alcohol has a blood thinning property. This is currently believed to be the source of much of the aparent health benefits of moderate alcohol consumption -- thinning the blood reduces the risk of stroke and heart disease substantially.
:)
Taking a half dose of asprin/day has a similar effect, and is a lot cheaper and safer than alcohol, but not nearly as much fun
First, Grovers algorithm does searching unordered lists, not sorting. In fact, I believe it has been "proved" that sorting cannot be done faster than O(N * ln N) with a quantum computer, though take thost proofs with a grain of salt, there is always the possibility of attacking the assumptions.
Anyway, I want to point out that searching is actually a very general algorithm. Basically, any algorithm that is much faster to validate than construct a solution can be done that way. For instace, one way to factor is to iterate through all numbers less than sqrt(N), testing each one by dividing N by it. This takes O(sqrt(N)) time. Grover's algorithm speeds this up to O(sqrt(sqrt(N)), which is still exponential in the number of bits in N. Not that this is NOT Shor's algorithm, which can indeed factor numbers in polynomic time in the number of bits.
I belive the accessory port on the dish receiver is only an output for its own IR blaster, which can control a normal VCR to record off the dish. If not, I don't believe the specs are published that would allow the Tivo to control it.
Other DN receivers might have a serial port, but mine doesn't.
Funny, my Tivo works fine with my Dish Network.
You need to use the IR blaster from the Tivo to change channels on the dish receiver, but it works fine.
Changing channels is really slow, since you have the combined latencies of the IR blaster, the dish receiver, and the MPEG encoder/decoder on the Tivo, but with the onscreen channel listing, I don't channel surf anymore.
The one thing that annoys me is that occasinally (I think when the dish receiver gets a firmware update), my receiver will turn off, and the Tivo has no way of knowing and doesn't turn it back on until I get home and notice that recordings are failing. I solved this by setting an auto-tune timer on my dish receiver, which wakes it up at 4:59, so at least the Simpsons and Babylon 5 get recorded.
Mine does, though admitedly on the back, not the front. Many after market CD players and the like have a pair of RCA inputs on the back for hooking up to the original deck. It isn't so hard to bring this around front if you, say have a DIN head unit in a double DIN space.
Cool. I just looked at the back of my Sony Jukebox, and I also have the slink interface.
It looks like the only thing you are looking for is the audio connection, and yes, many new sound cards have both coax and fiber SPDIF IO. Check out the creative live/audigy, or the Soyo DRAGON motherboard. The digital audio should be lossless, and error corrected by the jukebox electronics, so you should be able to record with any standard recording software (wavr on my debian box). Throw together some scripts to synchronize the recording start with the signal from the jukebox and you should be set.
One thing you can do is build your own CD changer robot. If the only feature you care about is "remove disk from ejected tray and insert next disk", it shouldn't be that hard. But unless you have more than a couple hundred CDs, I suspect you are best buying 3 or 4 CD-ROMs and taking a day to feed CDs into them as fast as you can.
This would be way cool, but I forsee two major problems:
.mp3 files, you need to solve the TOC problem.
1) Speed. AFAIK, multi-disk CD changers only read at 1X. Even with the highest qualtiy settings, I can encode at 3-4 times that rate on my dual CPU PIII.
2) Access to TOC. This is the real killer: if you want all the nice freedb lookups to work right, you need to extract the TOC from the disk and compute a hash of it. I am almost positive this doesn't go down the SPDIF line.
The speed I could deal with (just leave it running when you go on vacation for a week or so), but unless you want a hard drive full of unnamed
I have been using ResierFS for a few months now, and for the most part, it has worked well.
But the lack of a fsck is inexcusable. I have has a couple of "events" which might have corrupted a filesystem (my fault, not resierfs'), but had no way to check. Once, I went through my partitions one by one, copied them to a temp partition, mkfs'ed them, and copied them back. I put up with it because it was a pain in the butt to switch. Now that ext3 is in a Linus kernel, I am seriously considering running some tests, and converting back.
I used to be a skeptic about the possibility of quantum computing, but I have become a believer. Here is why:
1) It once was thought that gate precisions would need to be accurate to 1 in 10^12 or so, depending on the length of the computation. The problem was, unlike classical computers, quantum computers are not allowed to "latch" to a 1 or 0. However, there has been a substantial body of work that has shown how to implement error correction codes on quantum computers, without destroying the computation.
2) Decoherence, or the problem with entropic contamination by the environment, is a serious one, and the one that was always thought to be the nail in the coffin. However, now that people have actually started trying to make systems resilient to decoherence, they have measured systems that decohere on time scales from ms to hours -- giving considerable time to do computation.
3) One particular QC proposal, germanium quantum dots on silicon, has made incredible progress recently. A couple of years ago, someone wrote down the "six impossible tasks to make Si quantum computers work", three of which have already been accomplished.
We have already accomplished so much in QC that people thought would be impossible, it hardly seems logical to give up now. I can't say for certain, obviously, that QC will ever be a reality, but I would give it better than 50-50 odds at the moment.
Orthogonality in this case refers to modes of the electric field. A mode is labeled by a frequency and a polarization state. So, yes, two beams of different wavelength are orthogonal modes, and can be resolved by using a grating to diffract each component to a seperate detector. This is called WDM -- wavelength division multiplexing.
States with opposite polarization (horizontal and vertical, right and left circular, etc) are also orthogonal modes, and can be seperated, for instance, by a polarizing beam splitter. This is PDM.
The total bandwidth of a communication link is bps/mode * useful modes, so either increasing the number of frequencies or polarizations, or both, can improve bandwidth.
WDM is limited because each pulse actually covers a range of frequencies, and you need to choose them far enough apart that they don't overlap, or you won't be able to discriminate them well. PDM is limited because it is hard to get fibers to not fs*k with the polarization of light, plus there are only two orthogonal states, so you can only easily get a factor of 2 improvment in bandwidth.
While there is certainly a risk that porn site owners will abuse credit card data, and it is certainly easy to villianize the entire porn industry, but I don't think that is the real issue. I should have to, or even be asked, to give someone my credit card number unless I am buying something from them. Using credit cards as age verification is convenient but dumb.
People will say "If you don't like it, don't surf for porn", but they are implicitly making a moral judgment about viewing porn... They really mean "You dirty person, you shouldn't do that anyway, so don't complain about giving out your credit card."