Read as "We're too incopetent to implement a working DRM system ourselves, so we're asking the tech industry to join us in our crusade against our customers. I mean, they're all stealing, file-sharing criminals! They should be flushed out from their basements, full of computers with gigabytes of illegal high quality content like, uh, Britney Spears, with napalm! Your hear me!? NAPALM!!!! *spit flying* You know the smell of napalm and burned flesh? It's the smell of victory! And if they survive the napalm, they should be executed on the spot! *more ranting while drooling heavily*".
Nothing can move faster than the speed of light in vacuum, to be more precise. It's entirely possible to have particles traveling faster than the speed of light in a medium. For charged particles, this causes a radiation field known as Cherenkov radiation, first observed in 1934. It's somewhat similar to the sonic boom observed when something moves faster than the speed of sound.
... What this is probably going to mean, 10 or 20 years from now, is that HUGE numbers of people in retail and distributing are going to lose their jobs to these little tags....
This technology is probably inevitable though. As technology improves, more and more people can have their jobs replaced by computers. I know somebody is going to reply by saying "but it just shifts the jobs to somewhere else, e.g. the people who create these tags and create and maintain the software". Sure, to a degree, but if you follow the trends to their logical conclusion, you get to a point where millions of low-paying jobs are getting lost and being replaced by maybe a couple thousand higher paying jobs. At some point, something will have to give.. all those people who end up losing their jobs will be the retailers customers themselves, so their business drops.
Social progress (measured as GDP increase, or something like that) has largely been driven by technological innovations. RFID, if (or when) deployed large-scale, will decrease the workforce needed in retail and distribution. Of course, as you say, all those same people will not get jobs developing RFID hardware & software. If that were the case, there would be no point in RFID, since there would be no net efficiency gain! What will happen is that workforce will be freed to do new tasks, which noone has come up with yet. Look at things from a longer time scale, and to think of the economy has a whole rather than just the retail part.
As an example, say about 200 years ago a significant fraction of the population (about 80%) were directly involved in agriculture (i.e. they were farmers). Today, thanks to technological innovations farming productivity has increased about 100-fold, and as a consequence a quite small part of the population are farmers (4-5% in western europe and presumably smaller in the USA as farms are bigger there). This has allowed a lot of people to do other things than plowing the ground. We don't have a ~70% unemployment rate (i.e. 80%-5%-overhead of producing and developing farming machinery), as your line of reasoning suggests.
The same goes for RFID tags (or say, the introduction of automation, robots and such, in industry). So yes, in the short term and on a personal level, it might be unpleasant. I.e. aunt Tilly working as a cashier gets the boot and as she has little education she has a hard time getting a new job. But for society as a whole on a slightly longer timescale, the efficiency improvement made possible by RFID will be beneficial.
I think the time is ripe for some new kind of journal to appear. Published on the web, but still peer-reviewed. Look at arxiv, it's immensely popular even though it's just a preprint archive, no peer review.
The high cost associated with scientific publishing is caused by small circulation, not that the journals would have some big staff or things like that (the reviewers don't get paid). Why pay 'til your nose bleeds for distribution/printing when everybody has net access today?
No, the hard drive makers got it right. It was everyone else in the semiconductor industry who got it wrong by using SI prefixes (ie. the mega, giga etc. things) to mean various powers of two, when they really mean powers of ten. Nowadays there is a NIST standard which specifies prefixes for powers of two (the mebi, gibi, etc. things), but few people use those prefixes.
The point is that even though real applications typically achieve only 5-10% of the peak flops it's still a very fast machine. And better bang for the buck than other approaches to achieve the same level of performance.
some squid on a GSXR-1100 to zip by at some insane speed
Note to non-sportsbikers: Stupid bike riders are called 'squid' because that's what they most resemble after the accident they caused that cost their life.
Oh. I thought he meant SQUID = superconducting quantum interference device, which is a device for measuring extremely small magnetic fields. A SQUID might be of interrest in the sensor itself (if it would work at ambient temperature, that is), but I didn't really understand why he would want one on the bike...:)
Solid state lasers use electricity. The US is planning to deploy a 100 kW solid state laser on the JSF at some point. Scale this up to a few MW to punch through shell casings and similar armor and it could be useful on a ship. A few MW is nothing for a ships electrical system.
I think the US was building a solid state laser (i.e. a thingy which uses electricity) for the JSF. It was 100 kW, and used capacitors charged by a generator coupled to the turbine axle. Something similar to this could easily be adapted to an aircraft carrier, they have many MW of electricity generation capability. Also, solid state lasers are a lot cheaper to shoot with (no chemicals consumed) and no toxic exhaust fumes.
i'd think it needs a hell of a accuracy to hit something that's miles away, like artillery shell. and having enough of them to destroy rocket bombardment is just a dream, light fast or not.
does this sound as starwars fud crap? yes.
There's nothing unphysical or so about laser weapons. They are entirely doable. I'd say that within 50 years industrialized countries will have deployed large number of various directed energy weapons, such as lasers.
Well, considering that you have been bashing OOP and evangelizing your "table-based" programming (or whatever you were calling it) for God knows how long, here's an excellent chance for you to demonstrate the superiority of your approach. Are you still going to implement it on DBASE with file-polling for "clustering"?:)
I'd say one reason scripting languages are popular for web development is that web apps spend most of their time waiting for IO (either file or DB) and pushing data to clients. Usually, they don't have to calculate very much. So it doesn't really matter that scripting languages are slow at tight inner loops etc.
The 9000 packages are one of the main reasons I use debian. Whatever open source program you need, you can be quite sure that it is in debian. Installing the program with all of its dependencies is a zillion times more convenient with apt-get than to hunt around the net for the packages (and possibly compile them).
But undoubtedly you are right, this huge number of packages slows down the release cycle. I don't know how this could be solved. Perhaps some kind of splitting into many sub-distributions, eg. debian-core debian-desktop, debian-server etc., would help somewhat. But then things get complicated when you inevitably get dependencies between the different sub-distros, and you have to integrate them for the release. And debian-core, which the other sub-distros would be built on, would inevitably be a bit older than the rest of the stuff.
Another thing is that everything in debian seems to be happening slow-motion. If you read the mailing-lists, there are endless nitpicking arguments about everything concievable. Democracy isn't about everybody taking part in every single decision. You elect leaders so they can make decision quickly, without everyone having to bother about it, as that doesn't scale. What would happen in society in general if we were to have referendums about every single thing the government decides?
..that in an article about debian there is this big ad for, yes you guessed it, Microsoft.
More seriously though, UPM looks very cool. Hopefully it will be a success. Although I find it hard to believe that debian would adopt it, maybe that's why he seems to be planning another distro.
The one with ~1 million cpu:s is the final Blue Gene. The one which this article talks about with 65000 cpu:s is Blue Gene/L, which is supposed to be a prototype of the final design.
Blue Gene is actually more like the architecture. The first machine in the family, Blue Gene/L which this announcement was about will be used for bomb simulation and have 65000 processors and about 200 teraflops performance. Later there will be another, simply called Blue Gene, with 1 million processors and an estimated 1 petaflops performance. You can think of Blue Gene/L has a prototype for the final one.
Here in Finland at least, we have a national supercomputing centre, which manages the supercomputers used by the universities and also commercial companies can buy computing time there. Certainly they get cost benefits of an arrangement like that compared to everyone buying their own supercomputer, which then sits underutilized most of the time.
Chrysler is a big company, so they certainly have many projects going on at the same time. By centralizing on one big cluster they can increase utilization compared to each department needing simulation capabilities having its own minicluster. So what the article probably means is that at the same time they can run 18 different simulations each using on average 12 cpus (108*2/18).
It's the same on (almost) all supercomputers. They have lots of users, but most users don't use that many cpus for their jobs. Take me, for example. On the supercomputer where I have an account there are 512 power4 cpus. Usually I use 8 or 16 cpus for my simulations, and so do almost all the other people using the same machine. About the only time the entire supercomputer is reserved for one job is when they're benchmarking it, which as you certainly can imagine isn't done so often.
Well, if it was well-known that TCO for a windows cluster would be lower than for a linux cluster on the same hardware, then they would certainly have bought a windows cluster instead. I don't think portability is a big problem here. Most numerical code is quite portable.
Looking at windowsclusters.org, it seems that most project there are using windows simply because MS agreed to supply both hardware and software, en exchange for using windows instead of linux.
Slashdot Mirror
Who still uses structs?
Yeah, I sure do. They bolted some syntactic sugar on top of good old structs and now they are called classes instead.
Read as "We're too incopetent to implement a working DRM system ourselves, so we're asking the tech industry to join us in our crusade against our customers. I mean, they're all stealing, file-sharing criminals! They should be flushed out from their basements, full of computers with gigabytes of illegal high quality content like, uh, Britney Spears, with napalm! Your hear me!? NAPALM!!!! *spit flying* You know the smell of napalm and burned flesh? It's the smell of victory! And if they survive the napalm, they should be executed on the spot! *more ranting while drooling heavily*".
Makes one wonder if they really are planning to switch, or if it's yet another scheme to extort free MS-licences from Billy-boy..
Nothing can move faster than the speed of light in vacuum, to be more precise. It's entirely possible to have particles traveling faster than the speed of light in a medium. For charged particles, this causes a radiation field known as Cherenkov radiation, first observed in 1934. It's somewhat similar to the sonic boom observed when something moves faster than the speed of sound.
It can't be cheap:
MBE = Mega-buck evaporator
Well, you dont' have dust filters on jet engines equipped with afterburners either.
This technology is probably inevitable though. As technology improves, more and more people can have their jobs replaced by computers. I know somebody is going to reply by saying "but it just shifts the jobs to somewhere else, e.g. the people who create these tags and create and maintain the software". Sure, to a degree, but if you follow the trends to their logical conclusion, you get to a point where millions of low-paying jobs are getting lost and being replaced by maybe a couple thousand higher paying jobs. At some point, something will have to give
Social progress (measured as GDP increase, or something like that) has largely been driven by technological innovations. RFID, if (or when) deployed large-scale, will decrease the workforce needed in retail and distribution. Of course, as you say, all those same people will not get jobs developing RFID hardware & software. If that were the case, there would be no point in RFID, since there would be no net efficiency gain! What will happen is that workforce will be freed to do new tasks, which noone has come up with yet. Look at things from a longer time scale, and to think of the economy has a whole rather than just the retail part.
As an example, say about 200 years ago a significant fraction of the population (about 80%) were directly involved in agriculture (i.e. they were farmers). Today, thanks to technological innovations farming productivity has increased about 100-fold, and as a consequence a quite small part of the population are farmers (4-5% in western europe and presumably smaller in the USA as farms are bigger there). This has allowed a lot of people to do other things than plowing the ground. We don't have a ~70% unemployment rate (i.e. 80%-5%-overhead of producing and developing farming machinery), as your line of reasoning suggests.
The same goes for RFID tags (or say, the introduction of automation, robots and such, in industry). So yes, in the short term and on a personal level, it might be unpleasant. I.e. aunt Tilly working as a cashier gets the boot and as she has little education she has a hard time getting a new job. But for society as a whole on a slightly longer timescale, the efficiency improvement made possible by RFID will be beneficial.
I think the time is ripe for some new kind of journal to appear. Published on the web, but still peer-reviewed. Look at arxiv, it's immensely popular even though it's just a preprint archive, no peer review.
The high cost associated with scientific publishing is caused by small circulation, not that the journals would have some big staff or things like that (the reviewers don't get paid). Why pay 'til your nose bleeds for distribution/printing when everybody has net access today?
No, the hard drive makers got it right. It was everyone else in the semiconductor industry who got it wrong by using SI prefixes (ie. the mega, giga etc. things) to mean various powers of two, when they really mean powers of ten. Nowadays there is a NIST standard which specifies prefixes for powers of two (the mebi, gibi, etc. things), but few people use those prefixes.
Yeah? Really? So what?
The point is that even though real applications typically achieve only 5-10% of the peak flops it's still a very fast machine. And better bang for the buck than other approaches to achieve the same level of performance.
some squid on a GSXR-1100 to zip by at some insane speed
Note to non-sportsbikers: Stupid bike riders are called 'squid' because that's what they most resemble after the accident they caused that cost their life.
Oh. I thought he meant SQUID = superconducting quantum interference device, which is a device for measuring extremely small magnetic fields. A SQUID might be of interrest in the sensor itself (if it would work at ambient temperature, that is), but I didn't really understand why he would want one on the bike...
Solid state lasers use electricity. The US is planning to deploy a 100 kW solid state laser on the JSF at some point. Scale this up to a few MW to punch through shell casings and similar armor and it could be useful on a ship. A few MW is nothing for a ships electrical system.
I think the US was building a solid state laser (i.e. a thingy which uses electricity) for the JSF. It was 100 kW, and used capacitors charged by a generator coupled to the turbine axle. Something similar to this could easily be adapted to an aircraft carrier, they have many MW of electricity generation capability. Also, solid state lasers are a lot cheaper to shoot with (no chemicals consumed) and no toxic exhaust fumes.
i'd think it needs a hell of a accuracy to hit something that's miles away, like artillery shell. and having enough of them to destroy rocket bombardment is just a dream, light fast or not.
does this sound as starwars fud crap? yes.
There's nothing unphysical or so about laser weapons. They are entirely doable. I'd say that within 50 years industrialized countries will have deployed large number of various directed energy weapons, such as lasers.
Well, considering that you have been bashing OOP and evangelizing your "table-based" programming (or whatever you were calling it) for God knows how long, here's an excellent chance for you to demonstrate the superiority of your approach. Are you still going to implement it on DBASE with file-polling for "clustering"? :)
I'd say one reason scripting languages are popular for web development is that web apps spend most of their time waiting for IO (either file or DB) and pushing data to clients. Usually, they don't have to calculate very much. So it doesn't really matter that scripting languages are slow at tight inner loops etc.
The 9000 packages are one of the main reasons I use debian. Whatever open source program you need, you can be quite sure that it is in debian. Installing the program with all of its dependencies is a zillion times more convenient with apt-get than to hunt around the net for the packages (and possibly compile them).
But undoubtedly you are right, this huge number of packages slows down the release cycle. I don't know how this could be solved. Perhaps some kind of splitting into many sub-distributions, eg. debian-core debian-desktop, debian-server etc., would help somewhat. But then things get complicated when you inevitably get dependencies between the different sub-distros, and you have to integrate them for the release. And debian-core, which the other sub-distros would be built on, would inevitably be a bit older than the rest of the stuff.
Another thing is that everything in debian seems to be happening slow-motion. If you read the mailing-lists, there are endless nitpicking arguments about everything concievable. Democracy isn't about everybody taking part in every single decision. You elect leaders so they can make decision quickly, without everyone having to bother about it, as that doesn't scale. What would happen in society in general if we were to have referendums about every single thing the government decides?
..that in an article about debian there is this big ad for, yes you guessed it, Microsoft.
More seriously though, UPM looks very cool. Hopefully it will be a success. Although I find it hard to believe that debian would adopt it, maybe that's why he seems to be planning another distro.
The one with ~1 million cpu:s is the final Blue Gene. The one which this article talks about with 65000 cpu:s is Blue Gene/L, which is supposed to be a prototype of the final design.
Blue Gene is actually more like the architecture. The first machine in the family, Blue Gene/L which this announcement was about will be used for bomb simulation and have 65000 processors and about 200 teraflops performance. Later there will be another, simply called Blue Gene, with 1 million processors and an estimated 1 petaflops performance. You can think of Blue Gene/L has a prototype for the final one.
But luckily they don't have to do that as they aren't so blindlingly stupid that they would try to run one kernel for all 65000 cpus.
Here in Finland at least, we have a national supercomputing centre, which manages the supercomputers used by the universities and also commercial companies can buy computing time there. Certainly they get cost benefits of an arrangement like that compared to everyone buying their own supercomputer, which then sits underutilized most of the time.
Chrysler is a big company, so they certainly have many projects going on at the same time. By centralizing on one big cluster they can increase utilization compared to each department needing simulation capabilities having its own minicluster. So what the article probably means is that at the same time they can run 18 different simulations each using on average 12 cpus (108*2/18).
It's the same on (almost) all supercomputers. They have lots of users, but most users don't use that many cpus for their jobs. Take me, for example. On the supercomputer where I have an account there are 512 power4 cpus. Usually I use 8 or 16 cpus for my simulations, and so do almost all the other people using the same machine. About the only time the entire supercomputer is reserved for one job is when they're benchmarking it, which as you certainly can imagine isn't done so often.
Well, if it was well-known that TCO for a windows cluster would be lower than for a linux cluster on the same hardware, then they would certainly have bought a windows cluster instead. I don't think portability is a big problem here. Most numerical code is quite portable.
Looking at windowsclusters.org, it seems that most project there are using windows simply because MS agreed to supply both hardware and software, en exchange for using windows instead of linux.
Last I heard they ran one kernel per compute node (2 cpus ?). Their shmem library creates the impression of a shared memory.