It looks to me like he simply missed out the word "aren't". It's pretty easy to skip words accidentally if you are posting in a rush and your typing speed is slower than your thinking speed.
To me a world war implies two world superpowers (either individual countries or allegiences) directly fighting each other (not merely fighting proxy wars) with most of the world aligning to one or the other of them. Currently all the worlds superpowers are pretty friendly with each other.
No they will run to get what is probablly a much better connection at a much lower price. Sure the customer service may be better from a local wireless based provider but it's hard to beat fixed connections for reliable high performance service and since the telephone charges are already paying for the wiring the extra charge for DSL can be relatively low and still make decent money for the telco.
At the simplest level you buy a suitable internet connection (one that allows resale, large numbers of IPs etc) and then resell to customers over some kind of connection.
Beyond this you get into multihoming and requesting your IP blocks directly from the RIR.
AIUI deviations between the earths rotation and the SI second are grouped up and presented in the form of leap seconds (which may be either positive or negative) at midnight UTC. Whether these come during the working day or not will vary depending on where you live.
2 microseconds per day is a pretty small change though. Other changes such as the gradual slowing of the earths rotation will have a far more significant impact on the frequency of leap seconds (assuming that is that the proposal to make civil time entirely tied to atom time doesn't go through...)
1: If the new upstart is succesfull they (and/or copycats) may spread into areas where the incumbent telcos do offer broadband. 2: Just because a telco doesn't offer broadband in an area doesn't mean they aren't offering service at all. In the absence of broadband there is presumably money to be made from dialup.
It depends what you mean by self correcting. Ultimately the old people will die but not without potentially a lot of suffering happening first and if the birth rate stays low the problem won't go away because there will remain more old people thand young people.
I see a few outcomes of a sustained very low birthrate.
1: the old people have sufficiant money invested in foreign investments to pay for their retirement despite the inevitable shrink in exports and the number of younger citizens remains just about sufficiant to care for the old people, manage the imports and possiblly maintain some small ammount of other acticity. 2: the people work until they drop and people who can no longer look after themselves are simply left to die and/or institutionalised with a very minimal level of care because there simply isn't the resources to care for them properly. 3: the goverment allows massive immigration of young people to try and produce a population with a greater weight towards the productive (and hence taxpaying) age range. This solves the immediate problem but unless the immigrants take on the native culture it is likely to lead to the native culture being becoming within a few generations..
The problem is that low birthrates don't just mean a reduction in overall population (which I tend to agree would be a good thing for the world in general) it also means a greater proportion of the population are too old to work and in some cases too old to look after themselves.
One way to work arround this is to allow lots of immigration but that can mean the natives becoming a minority in their own country within a few generations which many people find rather unpalatable. It also only works as long as there are high bithrate countries to act as a source of immigrants.
My first guess would be a worker being lax about actually using his safety harness* (possiblly caused by time or other pressures, possiblly just as a result of overconfidence) on combined with a sudden surprise making him lose his footing.
*note that wearing a safety harness is not the same as using it. For a safety harness to protect the wearer it has to be actually attatched to something solid, prefferablly with a shock absorbing section in between the harness and attatchment point.
I've certainly seen reports of SSDs simply refusing to talk to their host computers. It's certainly possible every chip on the drive was fried but IMO it's just as likely that the controller chips or the voltage regulators (I don't know what modern flash runs at but I'm pretty sure it's less than the 5V that standard 2.5 inch SSDs run off*) have died leaving the flash intact.
*Afaict 5V is the only voltage you can rely on being present for a 2.5 inch drive.
Media and documents I like to put on a secondary partition anyway so I don't lose them if/when I have to blow windows away and reinstall so putting them on a seperate drive would be no big deal to me..
Bulky applications can indeed be a pain in multi drive setups (haven't gone SSD myself yet though I intend to do so on my next machine but I do have machines with multiple drives because their storage needs have grown over the years). IIRC steam is a particular pain because all steam applications have to be installed in subdirectories (with names chosen by steam) of the steam directory. So you either have every steam game on your HDD or every steam game on your SSD.
Be aware that when you dim an incandescent bulb (whether regular or halogen) the lumen output goes down WAY quicker than the power consumption. It would almost certainly be far more efficient to have a low power LED or flourescent bulb for constant illumination and only turn on the halogen when you need bright light.
Hopefully LED lighting will mature in the not too distant future. LEDs are efficient and can be dimmed and/or switched in groups within a fixture to provide variable light output. The problem is that good powerful LEDs are still rare and expensive and noone seems to be really making high quality LED specific fixtures yet.
I'm relatively ignorant, but AFAIK, it is common for "random number generators" with computers to not really be random at least via software because there will be underlying code based on an algorithm, since they are logical devices. So if you figure out the algorithm you can predict the "randomness".
Generally to predict the "random numbers" you need to know.
1: what algorithm is in use 2: either the internal state of the random number generator or the combination of seed value(s) and how many random numbers have been generated.
The difficulty of figuring these out varies hugely. If you can get hold of a copy of the software you can obviously see what algorithm it's using (a little harder with binaries than with source but far from impossible). If you can't then depending on the algorithm you may be able to identify it from what the output looks like or by trial and error in trying attacks for various algorithms. You can try and guess the seed data but provided it is sufficient in quantity and competently sourced this is also highly unlikely. Really the OS needs to be involved in this collection as it is difficult for many user level applications to collect sufficient seed data on their own.
Once you know the algorithm then things depend hugely on what sort of algorithm you are dealing with. Some algorithms either have so little internal state that the internal state can simply be brute forced or have mathematical weaknesses that allow calculation of the state from a sample of this output.
Unfortunately most programming languages come with a default random number generators that is insecure. Sometimes there is a secure one as well but often users who require secure random number generators are forced to turn to third party libraries and/or platform specific interfaces. The result is that often coders end up using insecure random number generators in situations where a secure one should be used.
IMO the default random number generator in a language should be one designed for security. IMO it's better that people use a secure random number generator when an insecure one would do than they use an ordinary random number generator where a secure one is needed.
Do you arrest the CEO of Smith&Wesson for a bank robbery?
There is a line between making a device that can potentially be used in a crime and making a device whose only substantial use is to commit a crime. The only substantial use for a points code generator is to fraudulently obtain points.
Guns do have legal uses. Nevertheless most civilised countries regulate them heavily (afaict even the US regulates them though not as strictly as other places) because the danger from their illegal use is perceived (rightly or wrongly) to outweigh the legitimate uses.
IANAL but as I understand it making something with both legal and illegal uses is generally OK unless explicity banned/regulated. Making something with only illegal uses or actively promoting illegal use of your device/service is not (this is what has got many torrent sites into trouble).
A frequency called the Nyquist frequency is defined to be half the sampling frequency of the digitally processed signal. It can be mathematically proven that aliasing can be avoided if the Nyquist frequency is greater than the maximum component frequency of the signal being sampled..
True but that is a very big IF. Real signals are almost certain to have components beyond the audible. These must be filtered out before sampling or they will be aliased down into the audible range. Afaict it's pretty much impossible to make an analog filter that passes 20KH\ perfectly yet applies a massive cut (big enough to put the aliased signals below the noise floor) cut at 22.05KHz
It's possible to get around this by sampling at a higher rate then filtering and downsampling (typcially you would combine the two operations for efficiency) digitally but digital filters have their own set of issues (for example a basic FIR design will produce "pre-echo")
IMO the difference between 58.7% and 58.0% compression is negligible and for a store wider support is more important than faster compression.
Yes we geeks have no trouble rigging up a script to convert anything to anything (though preserving tagging can be trickier) but we are a tiny minority of the market.
Commercial pressed CDs are very different from CD-Rs
A commercial pressed CD is made by pressing a pattern of pits into a disc of plastic and then coating the surface with aluminium. The metal layer is then coated with a protective layer and labeling is applied on top of that. Generally provided it is not badly scratched and the lacquering was done properly this is a very stable construction.
CD-Rs OTOH use a heat sensitive dye. Patterns burnt into the dye serve as a substitute for the pits. There are a number of dyes in use and many of the cheaper ones are rather unstable. Even the best ones almost certainly aren't as stable as the metal on pits of a commercial CD.
Randomness won't lose but it won't win either. Overall it will draw against any non-cheating* strategy. Therefore it would be the best strategy under the following conditions
1: everyone had a true rng or a a csprng in their head. 2: noone was cheating* and/or the player had the ability to throw consistenly in a way that defeated cheats 3: everyone was rational 4: everyone assumed everyone else was rational.
However people are neither rational or good at coming up with random numbers....
*A "cheating" strategy would be one where you try and determine what your opponent is in the process of throwing before you make you mind up on what you will throw...
It's easy to make incandescent bulbs last a very long time by running them at lower temperatures but it comes at a high price in efficiency and also results in a somewhat orange light.
For features on specific intel processors generally googling the model number will brink up a link to a page on ark.intel.com (never had much luck finding these pages with search tools on intel's site) within the first few results. For performance comparisons I look at a variety of review sites, anandtech bench is especially good when you want a quick comparision between two CPUs. Not sure about features on AMD processors but I bet it's on AMDs site somewhere
What is your budget for CPU/MB? If it's very low you are probablly better off with AMD, if it's $300 or more you are probablly better off with intel. Do you intend to replace this CPU while keeping the motherboard you bought with it? If so then AM3 and LGA1366 are reaching the end of their life while LGA1155 is at the start of it's life. Do you need virtualisation with support for dedicating hardware to particular VMs (intel call this VT-D,AMD call it AMD-Vi) or just ordinary hardware assisted virtualisation (intel call this VT-x, AMD call it AMD-V) Do you want to overclock? Can you live with a mainstream platform (less ram channels and less PCIe) and are you prepared to wait for the fixed P67/H67 boards? If so then LGA1155 platform gives far better bang per buck than LGA1366.
One annoyance to be aware of with LGA1155 is that you can't really overclock by base clock anymore and for some reason you can get either unlocked multiplier or VT-d but not both. Also only P67 allows multiplier based overclocking, H67 doesn't for some reason
Extreme edition processors rarely make sense financially (there are exceptions such as the 980x prior to the release of sandy bridge and the recent price drop on the 970).
Panel growth was outstripping the grids ability to handle the regional ups and downs of so many panels.
And therin lies the real problem with solar and wind, they generate when the weather is right which may or may not match up with when you need the most power.
Pumped storage and demand side management can help but pumped storage is very expensive and encouraging demand side management requires exposing end customers to tarrifs that vary with time of day which opens up huge cans of worms both form a point of view of privacy and from a point of view of keeping tarrifs understandable to consumers. Further both pumped storage and demand side management are only likely to smooth out short term peaks/troughs not weeks of poor weather.
For this reason I would be very surprised if solar and wind ever become more than minor players in the electricty market.
As can be seen from that chart in the USA only a tiny amount of oil goes to electricty generation (I believe the situation is similar here in europe but since this is a US site I used a US example). Over half of oil used in the US goes to transportation (with another large chunk going to insturial uses). Therefore nuclear power stations generating electricty won't directly do anything about the oil problem they will just displace natural gas and/or coal generation. Solving the oil problem will require at least one of the follwing
1: massive develpment of fischer tropsh and similar processes that can turn other energy sources into liquid fuel 2: a massive switch to electric transportation (not just cars but lorries, trains etc as well) 3: a massive reduction in the ammount of transportation services used (both of people and goods)
Slashdot Mirror
It looks to me like he simply missed out the word "aren't". It's pretty easy to skip words accidentally if you are posting in a rush and your typing speed is slower than your thinking speed.
To me a world war implies two world superpowers (either individual countries or allegiences) directly fighting each other (not merely fighting proxy wars) with most of the world aligning to one or the other of them. Currently all the worlds superpowers are pretty friendly with each other.
No they will run to get what is probablly a much better connection at a much lower price. Sure the customer service may be better from a local wireless based provider but it's hard to beat fixed connections for reliable high performance service and since the telephone charges are already paying for the wiring the extra charge for DSL can be relatively low and still make decent money for the telco.
How does one become an ISP?
At the simplest level you buy a suitable internet connection (one that allows resale, large numbers of IPs etc) and then resell to customers over some kind of connection.
Beyond this you get into multihoming and requesting your IP blocks directly from the RIR.
AIUI deviations between the earths rotation and the SI second are grouped up and presented in the form of leap seconds (which may be either positive or negative) at midnight UTC. Whether these come during the working day or not will vary depending on where you live.
2 microseconds per day is a pretty small change though. Other changes such as the gradual slowing of the earths rotation will have a far more significant impact on the frequency of leap seconds (assuming that is that the proposal to make civil time entirely tied to atom time doesn't go through...)
I can think of a couple of reasons
1: If the new upstart is succesfull they (and/or copycats) may spread into areas where the incumbent telcos do offer broadband.
2: Just because a telco doesn't offer broadband in an area doesn't mean they aren't offering service at all. In the absence of broadband there is presumably money to be made from dialup.
being becoming
That should have said becoming a minority.
It depends what you mean by self correcting. Ultimately the old people will die but not without potentially a lot of suffering happening first and if the birth rate stays low the problem won't go away because there will remain more old people thand young people.
I see a few outcomes of a sustained very low birthrate.
1: the old people have sufficiant money invested in foreign investments to pay for their retirement despite the inevitable shrink in exports and the number of younger citizens remains just about sufficiant to care for the old people, manage the imports and possiblly maintain some small ammount of other acticity.
2: the people work until they drop and people who can no longer look after themselves are simply left to die and/or institutionalised with a very minimal level of care because there simply isn't the resources to care for them properly.
3: the goverment allows massive immigration of young people to try and produce a population with a greater weight towards the productive (and hence taxpaying) age range. This solves the immediate problem but unless the immigrants take on the native culture it is likely to lead to the native culture being becoming within a few generations..
The problem is that low birthrates don't just mean a reduction in overall population (which I tend to agree would be a good thing for the world in general) it also means a greater proportion of the population are too old to work and in some cases too old to look after themselves.
One way to work arround this is to allow lots of immigration but that can mean the natives becoming a minority in their own country within a few generations which many people find rather unpalatable. It also only works as long as there are high bithrate countries to act as a source of immigrants.
My first guess would be a worker being lax about actually using his safety harness* (possiblly caused by time or other pressures, possiblly just as a result of overconfidence) on combined with a sudden surprise making him lose his footing.
*note that wearing a safety harness is not the same as using it. For a safety harness to protect the wearer it has to be actually attatched to something solid, prefferablly with a shock absorbing section in between the harness and attatchment point.
I've certainly seen reports of SSDs simply refusing to talk to their host computers. It's certainly possible every chip on the drive was fried but IMO it's just as likely that the controller chips or the voltage regulators (I don't know what modern flash runs at but I'm pretty sure it's less than the 5V that standard 2.5 inch SSDs run off*) have died leaving the flash intact.
*Afaict 5V is the only voltage you can rely on being present for a 2.5 inch drive.
Media and documents I like to put on a secondary partition anyway so I don't lose them if/when I have to blow windows away and reinstall so putting them on a seperate drive would be no big deal to me..
Bulky applications can indeed be a pain in multi drive setups (haven't gone SSD myself yet though I intend to do so on my next machine but I do have machines with multiple drives because their storage needs have grown over the years). IIRC steam is a particular pain because all steam applications have to be installed in subdirectories (with names chosen by steam) of the steam directory. So you either have every steam game on your HDD or every steam game on your SSD.
Be aware that when you dim an incandescent bulb (whether regular or halogen) the lumen output goes down WAY quicker than the power consumption. It would almost certainly be far more efficient to have a low power LED or flourescent bulb for constant illumination and only turn on the halogen when you need bright light.
Hopefully LED lighting will mature in the not too distant future. LEDs are efficient and can be dimmed and/or switched in groups within a fixture to provide variable light output. The problem is that good powerful LEDs are still rare and expensive and noone seems to be really making high quality LED specific fixtures yet.
I've seen all drive manufacturers cycle from the worst on the market to the best and back again several times over.
And the trouble is by the time a drive has been out long enough to know if it's a lemon or not it's obsolete anyway :(.
I'm relatively ignorant, but AFAIK, it is common for "random number generators" with computers to not really be random at least via software because there will be underlying code based on an algorithm, since they are logical devices. So if you figure out the algorithm you can predict the "randomness".
Generally to predict the "random numbers" you need to know.
1: what algorithm is in use
2: either the internal state of the random number generator or the combination of seed value(s) and how many random numbers have been generated.
The difficulty of figuring these out varies hugely. If you can get hold of a copy of the software you can obviously see what algorithm it's using (a little harder with binaries than with source but far from impossible). If you can't then depending on the algorithm you may be able to identify it from what the output looks like or by trial and error in trying attacks for various algorithms. You can try and guess the seed data but provided it is sufficient in quantity and competently sourced this is also highly unlikely. Really the OS needs to be involved in this collection as it is difficult for many user level applications to collect sufficient seed data on their own.
Once you know the algorithm then things depend hugely on what sort of algorithm you are dealing with. Some algorithms either have so little internal state that the internal state can simply be brute forced or have mathematical weaknesses that allow calculation of the state from a sample of this output.
Unfortunately most programming languages come with a default random number generators that is insecure. Sometimes there is a secure one as well but often users who require secure random number generators are forced to turn to third party libraries and/or platform specific interfaces. The result is that often coders end up using insecure random number generators in situations where a secure one should be used.
IMO the default random number generator in a language should be one designed for security. IMO it's better that people use a secure random number generator when an insecure one would do than they use an ordinary random number generator where a secure one is needed.
Do you arrest the CEO of Smith&Wesson for a bank robbery?
There is a line between making a device that can potentially be used in a crime and making a device whose only substantial use is to commit a crime. The only substantial use for a points code generator is to fraudulently obtain points.
Guns do have legal uses. Nevertheless most civilised countries regulate them heavily (afaict even the US regulates them though not as strictly as other places) because the danger from their illegal use is perceived (rightly or wrongly) to outweigh the legitimate uses.
IANAL but as I understand it making something with both legal and illegal uses is generally OK unless explicity banned/regulated. Making something with only illegal uses or actively promoting illegal use of your device/service is not (this is what has got many torrent sites into trouble).
A frequency called the Nyquist frequency is defined to be half the sampling frequency of the digitally processed signal. It can be mathematically proven that aliasing can be avoided if the Nyquist frequency is greater than the maximum component frequency of the signal being sampled..
True but that is a very big IF. Real signals are almost certain to have components beyond the audible. These must be filtered out before sampling or they will be aliased down into the audible range. Afaict it's pretty much impossible to make an analog filter that passes 20KH\ perfectly yet applies a massive cut (big enough to put the aliased signals below the noise floor) cut at 22.05KHz
It's possible to get around this by sampling at a higher rate then filtering and downsampling (typcially you would combine the two operations for efficiency) digitally but digital filters have their own set of issues (for example a basic FIR design will produce "pre-echo")
The music industry a while back decided that breaking apples lockin was more important to them than having DRM on downloadable music.
according to http://wiki.hydrogenaudio.org/index.php?title=Lossless_comparison#Comparison_Table wavpack is faster to compress and offers marginally better compression but flac has wider support from both hardware and software.
IMO the difference between 58.7% and 58.0% compression is negligible and for a store wider support is more important than faster compression.
Yes we geeks have no trouble rigging up a script to convert anything to anything (though preserving tagging can be trickier) but we are a tiny minority of the market.
Commercial pressed CDs are very different from CD-Rs
A commercial pressed CD is made by pressing a pattern of pits into a disc of plastic and then coating the surface with aluminium. The metal layer is then coated with a protective layer and labeling is applied on top of that. Generally provided it is not badly scratched and the lacquering was done properly this is a very stable construction.
CD-Rs OTOH use a heat sensitive dye. Patterns burnt into the dye serve as a substitute for the pits. There are a number of dyes in use and many of the cheaper ones are rather unstable. Even the best ones almost certainly aren't as stable as the metal on pits of a commercial CD.
Randomness won't lose but it won't win either. Overall it will draw against any non-cheating* strategy. Therefore it would be the best strategy under the following conditions
1: everyone had a true rng or a a csprng in their head.
2: noone was cheating* and/or the player had the ability to throw consistenly in a way that defeated cheats
3: everyone was rational
4: everyone assumed everyone else was rational.
However people are neither rational or good at coming up with random numbers....
*A "cheating" strategy would be one where you try and determine what your opponent is in the process of throwing before you make you mind up on what you will throw...
It's easy to make incandescent bulbs last a very long time by running them at lower temperatures but it comes at a high price in efficiency and also results in a somewhat orange light.
For features on specific intel processors generally googling the model number will brink up a link to a page on ark.intel.com (never had much luck finding these pages with search tools on intel's site) within the first few results. For performance comparisons I look at a variety of review sites, anandtech bench is especially good when you want a quick comparision between two CPUs. Not sure about features on AMD processors but I bet it's on AMDs site somewhere
What is your budget for CPU/MB? If it's very low you are probablly better off with AMD, if it's $300 or more you are probablly better off with intel.
Do you intend to replace this CPU while keeping the motherboard you bought with it? If so then AM3 and LGA1366 are reaching the end of their life while LGA1155 is at the start of it's life.
Do you need virtualisation with support for dedicating hardware to particular VMs (intel call this VT-D,AMD call it AMD-Vi) or just ordinary hardware assisted virtualisation (intel call this VT-x, AMD call it AMD-V)
Do you want to overclock?
Can you live with a mainstream platform (less ram channels and less PCIe) and are you prepared to wait for the fixed P67/H67 boards? If so then LGA1155 platform gives far better bang per buck than LGA1366.
One annoyance to be aware of with LGA1155 is that you can't really overclock by base clock anymore and for some reason you can get either unlocked multiplier or VT-d but not both. Also only P67 allows multiplier based overclocking, H67 doesn't for some reason
Extreme edition processors rarely make sense financially (there are exceptions such as the 980x prior to the release of sandy bridge and the recent price drop on the 970).
Panel growth was outstripping the grids ability to handle the regional ups and downs of so many panels.
And therin lies the real problem with solar and wind, they generate when the weather is right which may or may not match up with when you need the most power.
Pumped storage and demand side management can help but pumped storage is very expensive and encouraging demand side management requires exposing end customers to tarrifs that vary with time of day which opens up huge cans of worms both form a point of view of privacy and from a point of view of keeping tarrifs understandable to consumers. Further both pumped storage and demand side management are only likely to smooth out short term peaks/troughs not weeks of poor weather.
For this reason I would be very surprised if solar and wind ever become more than minor players in the electricty market.
While I agree we should be moving towards non-fossil fuel energy sources I suggest taking a look at the chart below.
http://upload.wikimedia.org/wikipedia/commons/5/54/LLNL_US_Energy_Flow_2009.png
As can be seen from that chart in the USA only a tiny amount of oil goes to electricty generation (I believe the situation is similar here in europe but since this is a US site I used a US example). Over half of oil used in the US goes to transportation (with another large chunk going to insturial uses). Therefore nuclear power stations generating electricty won't directly do anything about the oil problem they will just displace natural gas and/or coal generation. Solving the oil problem will require at least one of the follwing
1: massive develpment of fischer tropsh and similar processes that can turn other energy sources into liquid fuel
2: a massive switch to electric transportation (not just cars but lorries, trains etc as well)
3: a massive reduction in the ammount of transportation services used (both of people and goods)