Whitebox OEM vista home basic and XP home go for arround £50 inc VAT (our equivilent of sales tax) here in the uk. It is widely belived that the big vendors pay even less.
Home users will have little choice but to migrate as and when they buy thier next new PC, buisness users will be slower but some manufacturers are already putting out machines that are very difficult to find XP drivers for.
vista will replace XP just as XP replaced 2K, it will just take a bit of time.
I disagree, so does the German government, at present there is > 20,000 MW of installed wind power in Germany and the plans are to increase that substantially in the near future. It does sound like quite a bit but that figure is "installed capacity" which when reffering to the renewables other than dam based hydro is a eupherism for what it will generate under ideal conditions which may or may not line up with your peak load.
Such reneables make a nice supplement to an already stable grid and help to appease the greens but they will not replace fossil fuel based or nuclear power.
P.S. I concider dam based hydro seperately from other renewables because it is a much more practical source of power as it generates on the operators schedule not it's own but there are very few sites availible for new installations that aren't otherwise occupied.
you miss the critical thing, the BSA only tends to go after buisnesses.
So people get hooked on expensive software like photoshop through pirate copies then when they need to use it for work they buy it legit.
As for windows virtually everyone in the western world pays for it as part of thier computer purchase. Most PC vendors would not dare sell a machine with pirate windows preloaded (especially in theese days of WGA which invites users to rat on thier supplier in exchange for a free windows license) and most users expect windows installed and working when they buy the machine.
actually. I'm saying it wouldn't be entirely out of the question if it is an option for companies to become the "I'm Feeling Lucky" result for a keyword(s). I somewhat doubt they would do that. IIRC google has always been very hot on maintaining the seperation between search results and advertising. If people found the i'm feeling lucky result was different from the top search result because google were being paid do so I think it would be pretty bad PR.
Searching alone will only get you the data on the machine at the time you seize it. Some of that data may be useless because of encryption and other stuff may never have been stored on the hard drive in the first place. Placing spy software secretly will get them data that passes through that machine without being stored.
For a physiscal world analogy it is the difference between entering a house to search and seize and entering a house to play hidden survilence equipment and record conversations.
Since I hope that you neither omit Windows 2000 and Windows XP from the "desktop versions of 32 bit windows" set, nor are you ignorant about PAE I quote from that MS link
"Windows XP (all versions) 4 GB of physical RAM*"
"* Total physical address space is limited to 4 GB on these versions of Windows."
Since it is the total physical address space that is limited usable ram ends up somewhere between 3GB and 4GB depending on how much space the chipset reserves for IO.
2000 advanced server and all editions of server 2003 do support more ram but I do not consider those desktop operating systems
a sysadm at a client site shut me up by showing me a system with 6 GiB of memory running WinXP. It was a Dell blade server, can't remember which model though. System properties showed 6 GiB of memory installed. And yes, it felt and was sluggish-- but it had more than 4 GiB of usable memory." That finding contradicts the MS page you linked to and all experiance I have seen from people attempting to use XP32 on boxes with more than 4GB of ram.
It sounds like either it was running XP64 or someone had hacked out the limitation.
In the end, while Linux and Apple can afford to break things a little every now and then to consistently improve, Microsoft set themselves up for a massive fall because they spent so long trying to make everything work before that now they aren't trying as hard, everyone sees it as a failure. Yep, the most important thing to a customer is to keep thier software working.
Upgrading to a new version of windows eventually is practically forced by the older version getting it's support dropped by hardware and software vendors. There are already laptops that are a pain to get working correctly with XP.
Therefore software not working on a new version of the OS puts the customer in a difficult position. Upgrading the software can be expensive or impossible, not upgrading the OS becomes more and more untenable as the years pass and hardware has to be replaced. The opensource world has this problem far less because they can fix problems at source (forking the project if needed).
then your extra layer is largely irrelevant since the sequence only depends on a seed supplied by Microsoft's PRNG. There are plenty of other potential sources of randomness you can tap into, the least significant bits of timestamps for various external events, noise on analog inputs (such as the sound card) and many others.
child resistant catches/high level bolts sure but locks?!
If you haven't tought your children to behave by the time a full lock and key is the only thing that will stop them then either your child has serious mental problems (which does happen sometimes but should be fairly rare) or you were a very bad parent.
"By that time, the Earth and Moon will become caught up in what are called a "spin-orbit resonance" in which the Moon will circle the Earth in about 47 days (currently 29 days) and both Moon and Earth will rotate around their axes in the same time, always facing each other with the same side. Beyond this, it is hard to tell what will happen to the Earth-Moon system"
I would think that eventually friction (there isn't much friction in space but there is some) would make the system collapse if it was left to run long enough but that is likely to be a very long way off and the sun will probablly be long dead by then (of course the death of the sun will presumablly cause complications of it's own).
I never played the tape version of elite but the disk version on the BBC micro loaded data every time you moved system (hyperspace, galactic hyperspace and I belive escape capsule) or docked/undocked from a space station.
Afaict this varies a lot by location. I belive CD copy protection is more common in the UK than in the USA and more common in continental europe than in the uk.
also I notice in your selection of music types you don't mention pop, maybe that is significant.
I disagree with your assumption, when the common man puts his brand new purchased DVD is his brand new DVD player and finds it wont run, takes it back to the store and is told that it's an anti-piracy system that has stopped his legally purchased products from working but will he be told that or will the sales droid be as ignorant as he is? In any case I suspect that most people will not have just bought a new good quality DVD player and will thus assume that the problem is that thier DVD player is either old or cheap shit.
eg, Sony's Casino Royal not working on Sony's own current off the shelf players Now that is a monumental fuckup.
Some yes but far less than people on the ground have.
the higher up the user is the less bumps in the ground matter because the angle from the tower to the user for a given horizonal difference is steeper.
well one engine was a loss because of the fuel leak, mismanagement of the situation (the exact reasons for which have not been made public iirc but misdiagnoses is suspected) made them lose the other one too.
the other reason I can think of for an audible indication is that someone mandated that emergency calls should dial through keypad lock. Unfortunately the result of this combined with the short length of emergency numbers is to send a hell of a lot of accidental calls the emergency services way...
The real hit you take from software RAID is to the bus bandwidth, not the CPU. Most amateurs compare a hardware RAID controller to software RAID on a dinky little 32bit/33Mhz PCI bus (and remember that those onboard SATA ports are probably hanging off a regular 32bit/33Mhz PCI bus), that's why they frequently conclude software RAID is slower ("especially during rebuilds"). I think this is the key point, in high end server class hardware with fast busses the extra traffic from raid may be insignificant or it may not be depending on the array size and the busses in use. On an older desktop class board with modern drives, and a gigabit ethernet interface all on one 32bit 33mhz PCI bus it can be very significant especially when the array is degraded (a degraded parity based raid means that for every read that would have hit the dead drive you need to read the corresponding data from all the drives). The rebuild itself requires reading all data from all the remaining drives.
According to wikipedia modern desktop hard drives can do sustained transfers of 100MB/sec, one of theese will nearly saturate an ordinary 32/33 PCI bus, eight will saturate the fastest version of PCI-X 1.0 64/133.
To put it another way in a five disk array built from modern drives in a system based on basic PCI if you try and use more than 20% of the drives read capacity for the rebuild you will have saturated your PCI bus. With a hardware raid card none of the rebuild traffic will ever leave the raid card and you can probablly rebuild at that rate and barely notice it.
Looks like it comes down to a choice of use a hardware raid card or select your motherboard and controllers VERY carefully.
IMO the most important thing is when building a raid array DO NOT use drives from the same manufacturer. It may improve performance a bit to have an array of identical drives but it also seriously increases the chance of a second failure before replacement and rebuild are completed.
Most manufacturers seem to go through good and bad phases but you can't tell if drives are from a good or bad phase of a manufacturers life until some time after you bought the drive so diversification is your best defense.
I very much doubt anyone has a table of a corresponding input for every MD5 has.
there are 2^128 md5 hashes, if we assume we have to allow 32 bytes for each input (twice the length of the output seems like a reasonable starting point for space to allow) then that means such a table would be 2^128*32=2^133 bytes.
the largest readilly availible hard drives are about 2^40 bytes.
so to store your MD5 reversing table you would need about 2^93 hard drives.
using the approximation that 2^30 ~= 10^9 (yeah I know it's not very accurate but it will get us within an order of magnitude which is more than good enough to prove my point) that is about 8 billion billion billion hard drives.
UTC is not a "standard for converting", it is a definition of a way of representing time.
"unix time" as defined by all modern implementations (i'm not sure what posix specifies on the matter) is defined as the number of seconds since Jan 1, 1970 excluding leap seconds. This makes it easy to convert to and from civilian time (which is UTC with an offset configured based on the users timezone) but impossible to track accurately without external synchronisation and it also makes it suffer from time jumps.
Really the key problem is that leap seconds cannot be known in advance because we aren't that good at predicting the earths movements so when desgining a system you have to choose between either "maintaining clock accuracy needs external help" or "converting to civilian time correctly needs external help". Neither is a very desirable state of affairs but the former makes more sense since most clocks are bad enough that drift between resets and resetting inaccuracy for manually set clocks are probablly more than a second a year anyway and it means that a user updating manually doesn't have to enter a table of leap seconds so far (which trust me would really confuse most users trying to set thier clock manually).
Slashdot Mirror
Whitebox OEM vista home basic and XP home go for arround £50 inc VAT (our equivilent of sales tax) here in the uk. It is widely belived that the big vendors pay even less.
Home users will have little choice but to migrate as and when they buy thier next new PC, buisness users will be slower but some manufacturers are already putting out machines that are very difficult to find XP drivers for.
vista will replace XP just as XP replaced 2K, it will just take a bit of time.
I disagree, so does the German government, at present there is > 20,000 MW of installed
wind power in Germany and the plans are to increase that substantially in the near future.
It does sound like quite a bit but that figure is "installed capacity" which when reffering to the renewables other than dam based hydro is a eupherism for what it will generate under ideal conditions which may or may not line up with your peak load.
Such reneables make a nice supplement to an already stable grid and help to appease the greens but they will not replace fossil fuel based or nuclear power.
P.S. I concider dam based hydro seperately from other renewables because it is a much more practical source of power as it generates on the operators schedule not it's own but there are very few sites availible for new installations that aren't otherwise occupied.
you miss the critical thing, the BSA only tends to go after buisnesses.
So people get hooked on expensive software like photoshop through pirate copies then when they need to use it for work they buy it legit.
As for windows virtually everyone in the western world pays for it as part of thier computer purchase. Most PC vendors would not dare sell a machine with pirate windows preloaded (especially in theese days of WGA which invites users to rat on thier supplier in exchange for a free windows license) and most users expect windows installed and working when they buy the machine.
most survillance measures have corresponding countermeasures. That does not render them useless.
I agree though that a hardware approach would probablly be more effective.
actually. I'm saying it wouldn't be entirely out of the question if it is an option for companies to become the "I'm Feeling Lucky" result for a keyword(s).
I somewhat doubt they would do that. IIRC google has always been very hot on maintaining the seperation between search results and advertising. If people found the i'm feeling lucky result was different from the top search result because google were being paid do so I think it would be pretty bad PR.
Searching alone will only get you the data on the machine at the time you seize it. Some of that data may be useless because of encryption and other stuff may never have been stored on the hard drive in the first place. Placing spy software secretly will get them data that passes through that machine without being stored.
For a physiscal world analogy it is the difference between entering a house to search and seize and entering a house to play hidden survilence equipment and record conversations.
Since I hope that you neither omit Windows 2000 and Windows XP from the "desktop versions of 32 bit windows" set, nor are you ignorant about PAE
I quote from that MS link
"Windows XP (all versions) 4 GB of physical RAM*"
"* Total physical address space is limited to 4 GB on these versions of Windows."
Since it is the total physical address space that is limited usable ram ends up somewhere between 3GB and 4GB depending on how much space the chipset reserves for IO.
2000 advanced server and all editions of server 2003 do support more ram but I do not consider those desktop operating systems
a sysadm at a client site shut me up by showing me a system with 6 GiB of memory running WinXP. It was a Dell blade server, can't remember which model though. System properties showed 6 GiB of memory installed. And yes, it felt and was sluggish-- but it had more than 4 GiB of usable memory."
That finding contradicts the MS page you linked to and all experiance I have seen from people attempting to use XP32 on boxes with more than 4GB of ram.
It sounds like either it was running XP64 or someone had hacked out the limitation.
In the end, while Linux and Apple can afford to break things a little every now and then to consistently improve, Microsoft set themselves up for a massive fall because they spent so long trying to make everything work before that now they aren't trying as hard, everyone sees it as a failure.
Yep, the most important thing to a customer is to keep thier software working.
Upgrading to a new version of windows eventually is practically forced by the older version getting it's support dropped by hardware and software vendors. There are already laptops that are a pain to get working correctly with XP.
Therefore software not working on a new version of the OS puts the customer in a difficult position. Upgrading the software can be expensive or impossible, not upgrading the OS becomes more and more untenable as the years pass and hardware has to be replaced. The opensource world has this problem far less because they can fix problems at source (forking the project if needed).
I think the NT line puts limits on what processes can debug each other though I don't know exactly what they are.
then your extra layer is largely irrelevant since the sequence only depends on a seed supplied by Microsoft's PRNG.
There are plenty of other potential sources of randomness you can tap into, the least significant bits of timestamps for various external events, noise on analog inputs (such as the sound card) and many others.
child resistant catches/high level bolts sure but locks?!
If you haven't tought your children to behave by the time a full lock and key is the only thing that will stop them then either your child has serious mental problems (which does happen sometimes but should be fairly rare) or you were a very bad parent.
well the wikipedia article says
"By that time, the Earth and Moon will become caught up in what are called a "spin-orbit resonance" in which the Moon will circle the Earth in about 47 days (currently 29 days) and both Moon and Earth will rotate around their axes in the same time, always facing each other with the same side. Beyond this, it is hard to tell what will happen to the Earth-Moon system"
I would think that eventually friction (there isn't much friction in space but there is some) would make the system collapse if it was left to run long enough but that is likely to be a very long way off and the sun will probablly be long dead by then (of course the death of the sun will presumablly cause complications of it's own).
I never played the tape version of elite but the disk version on the BBC micro loaded data every time you moved system (hyperspace, galactic hyperspace and I belive escape capsule) or docked/undocked from a space station.
Afaict this varies a lot by location. I belive CD copy protection is more common in the UK than in the USA and more common in continental europe than in the uk.
also I notice in your selection of music types you don't mention pop, maybe that is significant.
I disagree with your assumption, when the common man puts his brand new purchased DVD is his brand new DVD player and finds it wont run, takes it back to the store and is told that it's an anti-piracy system that has stopped his legally purchased products from working
but will he be told that or will the sales droid be as ignorant as he is? In any case I suspect that most people will not have just bought a new good quality DVD player and will thus assume that the problem is that thier DVD player is either old or cheap shit.
eg, Sony's Casino Royal not working on Sony's own current off the shelf players
Now that is a monumental fuckup.
Some yes but far less than people on the ground have.
the higher up the user is the less bumps in the ground matter because the angle from the tower to the user for a given horizonal difference is steeper.
well one engine was a loss because of the fuel leak, mismanagement of the situation (the exact reasons for which have not been made public iirc but misdiagnoses is suspected) made them lose the other one too.
the other reason I can think of for an audible indication is that someone mandated that emergency calls should dial through keypad lock. Unfortunately the result of this combined with the short length of emergency numbers is to send a hell of a lot of accidental calls the emergency services way...
there are a variety of methods, the easiest is to link the incoming 5V to the USB 5V.
Various possible methods are listed at http://www.nslu2-linux.org/wiki/HowTo/ForcePowerAlwaysOn
The real hit you take from software RAID is to the bus bandwidth, not the CPU. Most amateurs compare a hardware RAID controller to software RAID on a dinky little 32bit/33Mhz PCI bus (and remember that those onboard SATA ports are probably hanging off a regular 32bit/33Mhz PCI bus), that's why they frequently conclude software RAID is slower ("especially during rebuilds").
I think this is the key point, in high end server class hardware with fast busses the extra traffic from raid may be insignificant or it may not be depending on the array size and the busses in use. On an older desktop class board with modern drives, and a gigabit ethernet interface all on one 32bit 33mhz PCI bus it can be very significant especially when the array is degraded (a degraded parity based raid means that for every read that would have hit the dead drive you need to read the corresponding data from all the drives). The rebuild itself requires reading all data from all the remaining drives.
According to wikipedia modern desktop hard drives can do sustained transfers of 100MB/sec, one of theese will nearly saturate an ordinary 32/33 PCI bus, eight will saturate the fastest version of PCI-X 1.0 64/133.
To put it another way in a five disk array built from modern drives in a system based on basic PCI if you try and use more than 20% of the drives read capacity for the rebuild you will have saturated your PCI bus. With a hardware raid card none of the rebuild traffic will ever leave the raid card and you can probablly rebuild at that rate and barely notice it.
Looks like it comes down to a choice of use a hardware raid card or select your motherboard and controllers VERY carefully.
mmm, last I heared those things tended to perform considerablly worse than linux software raid.
IMO the most important thing is when building a raid array DO NOT use drives from the same manufacturer. It may improve performance a bit to have an array of identical drives but it also seriously increases the chance of a second failure before replacement and rebuild are completed.
Most manufacturers seem to go through good and bad phases but you can't tell if drives are from a good or bad phase of a manufacturers life until some time after you bought the drive so diversification is your best defense.
I very much doubt anyone has a table of a corresponding input for every MD5 has.
there are 2^128 md5 hashes, if we assume we have to allow 32 bytes for each input (twice the length of the output seems like a reasonable starting point for space to allow) then that means such a table would be 2^128*32=2^133 bytes.
the largest readilly availible hard drives are about 2^40 bytes.
so to store your MD5 reversing table you would need about 2^93 hard drives.
using the approximation that 2^30 ~= 10^9 (yeah I know it's not very accurate but it will get us within an order of magnitude which is more than good enough to prove my point) that is about 8 billion billion billion hard drives.
Your post contains some factual mistakes.
UTC is not a "standard for converting", it is a definition of a way of representing time.
"unix time" as defined by all modern implementations (i'm not sure what posix specifies on the matter) is defined as the number of seconds since Jan 1, 1970 excluding leap seconds. This makes it easy to convert to and from civilian time (which is UTC with an offset configured based on the users timezone) but impossible to track accurately without external synchronisation and it also makes it suffer from time jumps.
Really the key problem is that leap seconds cannot be known in advance because we aren't that good at predicting the earths movements so when desgining a system you have to choose between either "maintaining clock accuracy needs external help" or "converting to civilian time correctly needs external help". Neither is a very desirable state of affairs but the former makes more sense since most clocks are bad enough that drift between resets and resetting inaccuracy for manually set clocks are probablly more than a second a year anyway and it means that a user updating manually doesn't have to enter a table of leap seconds so far (which trust me would really confuse most users trying to set thier clock manually).