Everyone's missed the mark slightly, it seems. Boy am I glad I paid attention in my EE class right now...
The most widely-used building block of modern computer circuits is the transistor. They come in 2 basic flavors, n-type and p-type. The difference, in layman's terms, is that n-types switch current on, and p-types switch current off. This was covered pretty well.
The computer world works in terms of binary logic functions, so we use transistors to build logic circuits. The simplest logic circuit that can be built is an inverter (NOT gate, etc). This takes 2 transistors (one of each type). Doing this on a molecule level is what IBM has just accomplished.
By itself, the NOT function is hardly useful, since it has only one input. To perform complex calculations, we use functions like OR, AND, NOR, NAND, XOR, etc, which have 2 inputs. As it turns out, the ones that human beings comprehend easily (AND and OR) are not the easiest to build, transistor-wise. NAND and NOR each require 4 transistors to build (two of each type), and have similar layouts. This is convenient because both functions (as mentioned) can be used exclusively to generate any possible binary function.
In practice, NAND gates are used most often in computer logic. Why? Although they are equally easy to build, the NOR gate requires you to stack two p-type transistors on top of each other. Due to their electrical characteristics, p-type transistors are significantly slower than n-type ones. Stacking them together only exacerbates the performance problem. Hence, we use NANDs whenever possible.
However, we still see a significant number of NOR gates in most computer architectures. The reason? NOR gates are the fundamental building block of flip-flops (a circuit that can "remember" a value), which are combined to make registers. Since RISC architectures tend to have a lot of registers (both visible and hidden in the pipeline system), we see plenty of NOR gates as well.
I wouldn't call it policing, we're just trying to comply with the law
But it is policing! What they're doing is no different from a cop sitting by the side of the road catching speeders. The only difference is that they're also bypassing the rest of the legal process by deciding your guilt and then punishing you by shutting down your account.
Just because some dippy spokesperson "wouldn't call it policing" doesn't mean it's not!
After reading the article, it seems that the reason this isn't "catching on" is that there's really nothing to it. There seems to be this vague idea that peer-to-peer is some holy grail of networking, and that any day now it will lead us away from the "limitations" of the client-server model.
Unfortunately, no one seems to be able to come up with many concrete applications of this idea (other than a few mentioned, like music sharing). The reason is that most of the functions we expect from a network are based on client-server technology. There seems to be this hazy notion of "conversations" between machines, in the same way that two friends could have a chat, and this is not the kind of fluffy computing model that I'm looking for when I visit my bank online to pay my bills.
I definitely agree that the HTTP protocol is ill-equipped to deal with the demands of today's network applications, but the solution is to standardize a better protocol, not throw out the entire model.
The GeForce3 is listed for $290 on pricewatch, which is certainly better than the $600 it was originally.
Re:Ah, but it DOES have a dollar sign in front of
on
Dynamix Closed Down?
·
· Score: 1
You're right about customer loyalty, but I'm not sure it works the same way in the game industry (especially with the game publishers).
If I buy a car and have a miserable experience with the dealership or company, I can always get my next car somewhere else.
Games, unlike cars, are one of a kind. When a great new game comes out, you don't have the option of choosing which publisher to get it from; you take it or leave it.
Suppose Tribes 3 was published by Sierra in a few years. How many people here would honestly not get it because of what Sierra did to Dynamix?
True, mobile machines do pose a problem with routing. Right now there are two methods in development for handling mobile routing:
1. Every mobile machine must be associated with a "home agent" machine, which cannot move (and therefore has a fixed IP adddress). All outgoing packets are tagged with the home agent's IP address, and return packets are sent there first using traditional routing methods. The home agent then forwards the packets to the mobile machine. To be able to do this, the mobile machine must continaully update the home agent on its current location. As the mobile machine changes location, it also changes IP addresses, and sends notifications to the home agent containing the new mobile IP address.
The advantage of this approach is that it is transparent to the upper application layers AND to the other end of the connection (it appears that the home agent is the final destination).
2. The other approach is more advanced, and requires the participation of both ends of the connection. The mobile machine, when it changes locations, sends a notification directly to the remote machine with its new IP address. This way, further packets can be sent directly to the new location, without the need of a home agent machine.
In short, mobile IP will always require somewhat of a "hack", and you already do change IP addresses when you change locations. Having those addresses correspond directly to the geographic location would simply eliminate some of the complications of routing.
I've been contemplating an idea for a routing protocol based on geographic location.
Right now, our notion of the "destination" of a packet is based on IP addresses, which are somewhat arbitrarily chosen and have little relationship with the physical location of the target machine. To make this work, we've needed to employ complex routing tables and algortihms with relatively large upkeep and administration requirements.
As the 'net becomes more strongly connected, there will be even more paths for packets to take, and it seems logical to try and simplify routing. If the "address" of a machine were derived from its physical geographic location, then packets could be routed simply by "sending them in that general direction". Instead of complex routing tables, routers would only need to know their relative geographic location in order to send packets toward the target. Conventional routing methods could be used on a local scale to calculate the final hop or two for the packet.
Needless to say, this method would trivialize the problems posed in the article as well.
Sierra "management" knows exactly what they're doing here. Tribes 2, like its predecessor, is a game with a very strong cult following, and less of a general appeal to the average gamer. Most people who really enjoy T2 were also avid Tribes fans, and were watching the whole development process very closely.
What this means is that most of the target audience for Tribes 2 has already bought the game. Management knows this, and so they will feel completely justified in cutting off T2 support, since most of the profits from the game have already been made.
And from upper management's point of view, things like "customer satisfaction" mean nothing in this case, because it doesn't have a dollar sign in front of it.
Why bother with the speakers and mic. Once the MP3 is converted back to analog couldn't you just pipe the analog signal back in and convert it back to digital then store it as the new file?
Although this approach will not remove the watermark, it WILL allow you to play the audio on a different player (one that does not check for watermarks).
No matter what the regulators do, there will always be a software hack (non-standard audio player, etc) to ignore the watermark. The only way to ensure that the watermark can be detected is with a special sound card. Even if the regulators come up with such a thing, a discerning customer (aka geek) will always opt for the non-restrictive hardware. Therefore, the only way this could become truly ugly is if the government outlawed "normal" sound cards and forced hardware manufacturers to stop making them.
If that happens, we've got bigger problems to worry about...
Slashdot Mirror
The most widely-used building block of modern computer circuits is the transistor. They come in 2 basic flavors, n-type and p-type. The difference, in layman's terms, is that n-types switch current on, and p-types switch current off. This was covered pretty well.
The computer world works in terms of binary logic functions, so we use transistors to build logic circuits. The simplest logic circuit that can be built is an inverter (NOT gate, etc). This takes 2 transistors (one of each type). Doing this on a molecule level is what IBM has just accomplished.
By itself, the NOT function is hardly useful, since it has only one input. To perform complex calculations, we use functions like OR, AND, NOR, NAND, XOR, etc, which have 2 inputs. As it turns out, the ones that human beings comprehend easily (AND and OR) are not the easiest to build, transistor-wise. NAND and NOR each require 4 transistors to build (two of each type), and have similar layouts. This is convenient because both functions (as mentioned) can be used exclusively to generate any possible binary function.
In practice, NAND gates are used most often in computer logic. Why? Although they are equally easy to build, the NOR gate requires you to stack two p-type transistors on top of each other. Due to their electrical characteristics, p-type transistors are significantly slower than n-type ones. Stacking them together only exacerbates the performance problem. Hence, we use NANDs whenever possible.
However, we still see a significant number of NOR gates in most computer architectures. The reason? NOR gates are the fundamental building block of flip-flops (a circuit that can "remember" a value), which are combined to make registers. Since RISC architectures tend to have a lot of registers (both visible and hidden in the pipeline system), we see plenty of NOR gates as well.
-- Brett
But it is policing! What they're doing is no different from a cop sitting by the side of the road catching speeders. The only difference is that they're also bypassing the rest of the legal process by deciding your guilt and then punishing you by shutting down your account.
Just because some dippy spokesperson "wouldn't call it policing" doesn't mean it's not!
-- Brett
Unfortunately, no one seems to be able to come up with many concrete applications of this idea (other than a few mentioned, like music sharing). The reason is that most of the functions we expect from a network are based on client-server technology. There seems to be this hazy notion of "conversations" between machines, in the same way that two friends could have a chat, and this is not the kind of fluffy computing model that I'm looking for when I visit my bank online to pay my bills.
I definitely agree that the HTTP protocol is ill-equipped to deal with the demands of today's network applications, but the solution is to standardize a better protocol, not throw out the entire model.
-- Brett
The GeForce3 is listed for $290 on pricewatch, which is certainly better than the $600 it was originally.
If I buy a car and have a miserable experience with the dealership or company, I can always get my next car somewhere else.
Games, unlike cars, are one of a kind. When a great new game comes out, you don't have the option of choosing which publisher to get it from; you take it or leave it.
Suppose Tribes 3 was published by Sierra in a few years. How many people here would honestly not get it because of what Sierra did to Dynamix?
1. Every mobile machine must be associated with a "home agent" machine, which cannot move (and therefore has a fixed IP adddress). All outgoing packets are tagged with the home agent's IP address, and return packets are sent there first using traditional routing methods. The home agent then forwards the packets to the mobile machine. To be able to do this, the mobile machine must continaully update the home agent on its current location. As the mobile machine changes location, it also changes IP addresses, and sends notifications to the home agent containing the new mobile IP address.
The advantage of this approach is that it is transparent to the upper application layers AND to the other end of the connection (it appears that the home agent is the final destination).
2. The other approach is more advanced, and requires the participation of both ends of the connection. The mobile machine, when it changes locations, sends a notification directly to the remote machine with its new IP address. This way, further packets can be sent directly to the new location, without the need of a home agent machine.
In short, mobile IP will always require somewhat of a "hack", and you already do change IP addresses when you change locations. Having those addresses correspond directly to the geographic location would simply eliminate some of the complications of routing.
Right now, our notion of the "destination" of a packet is based on IP addresses, which are somewhat arbitrarily chosen and have little relationship with the physical location of the target machine. To make this work, we've needed to employ complex routing tables and algortihms with relatively large upkeep and administration requirements.
As the 'net becomes more strongly connected, there will be even more paths for packets to take, and it seems logical to try and simplify routing. If the "address" of a machine were derived from its physical geographic location, then packets could be routed simply by "sending them in that general direction". Instead of complex routing tables, routers would only need to know their relative geographic location in order to send packets toward the target. Conventional routing methods could be used on a local scale to calculate the final hop or two for the packet.
Needless to say, this method would trivialize the problems posed in the article as well.
What this means is that most of the target audience for Tribes 2 has already bought the game. Management knows this, and so they will feel completely justified in cutting off T2 support, since most of the profits from the game have already been made.
And from upper management's point of view, things like "customer satisfaction" mean nothing in this case, because it doesn't have a dollar sign in front of it.
No matter what the regulators do, there will always be a software hack (non-standard audio player, etc) to ignore the watermark. The only way to ensure that the watermark can be detected is with a special sound card. Even if the regulators come up with such a thing, a discerning customer (aka geek) will always opt for the non-restrictive hardware. Therefore, the only way this could become truly ugly is if the government outlawed "normal" sound cards and forced hardware manufacturers to stop making them.
If that happens, we've got bigger problems to worry about...
-- Brett