OK, normally I don't get involved in this argument, but your arguments seem to be from the gut, and not from the head, and this level of insensitivity pisses me off.
> The 'liberty' here involves ending the life of another person.
Define 'person', and when a mass of cells becomes a 'person'.
For instance: Is a sperm a person? What about the egg? What about a newly fertilized egg which hasn't yet been attached to a uterous? How about when it is 32 cells?... etc.
I find your statement to be ill-founded. Come up with a better foundation. I'll be absolutely shocked if you can come up with a defintion of 'person' that most people agree with here.
> There is a reason the founding fathers put life ahead of liberty. When the act of excersising a liberty causes the death of another person, the government must protect the life of the person. > Thus, my point still stands.
Youre being somewhat incoherent. What I believe you want to say is something to the effect of: Your right to liberty ends when it would infringe on someone else's right to life. Since gov't exists to protect these rights, it must protect people's lives so that they may exercise their liberty.
> Also, saying that your forcing someone to have an unwanted child is bs anyway.
No, actually it isn't. Please make a sensible argument about how the gov't wouldn't be forcing someone to have an unwanted child assuming that people have sex, and don't always want the child.
Saying "Well, they shouldn't have gotten themselves into that situation in the first place" doesn't change the fact that they do, and are then prevented from dealing with the pregnancy before an unwanted child is born.
> There seems to be this direct link between having sex and getting pregnant that people don't seem to get. The best way to keep from having a child is to not have sex.
You're right. But arguing that people should do without is simply unrealistic. Sex is a perfectly natural thing. It is ingrained and instinctual, and we're built to enjoy it, ostensibly so that we will procreate.
Since, as we're using too much of the planet's resources, that procreation may be a bad thing now (though catholics may disagree), there is a strong impetus (often economic) that people shouldn't have children.
> There is no god given right to sex in the constitution and no where does the constitution say that you have the right to the pursuit of happiness without concequences.
I love this line of reasoning.. You shouldn't be able to have sex (assuming that you cannot have an abortion) unless you're willing and able to deal with the consequences. This is fun! Eventually, when you take this line of reasoning to the extreme, what you're saying is that only rich people can have sex because only they are equipped to deal with the consequences. Fun!
> And before you bring up rape and incest, in those cases the woman will ussually get the morning after pill.
Incorrect. This can be labeled 'abortion', and some pharmicists DON'T DISPENSE THESE PRESCRIPTIONS. Some pharmicists won't even SELL THESE DRUGS if they're OTC. Even more fun, you get assholes who will not fill the prescription, and who will hold onto it so that no other pharmacist may fill it... EVEN MORE FUN, if there is an unwanted pregnancy in a underage child, they'll have to go get parental concent for the drugs in most places (if not all places now) in the US for a large class of the morning-after pills. This is especially fun when they've been raped by their dad, or uncle, etc.
> And in cases where the woman hides it till its too late to get a legal abortion (assuming that the state she lives in allows abortions up until a certain point) then they should be able to go to a judge and get due process of law to authorize the abortion.
You're making no sense. You're saying that a woman will go to a judge to get a dispensation to have an illegal abortion?
"AJA" makes a card with linux drivers for SD-SDI and HD-SDI video output.
"NVidia", also makes a card-- it (now) is a daughter card, which allows you to output SD-SDI or HD-SDI from the screen, or from an offscreen surface (e.g. glx-pbuffer). (Obviously, this is not a consumer card...).. lets see, another company that makes this kind of stuff is "DVS".
(Ya, I made linux boxes make pretty broadcast video for a living for a while...)
A tab == X spaces (at the beginning of a line, at least). This may seem obvious, but it is also powerful-- it means that I may indicate, symbolically (i.e. by usng tabs) the indentation level of a piece of code, etc.
Thus, I mix tabs and spaces according to the following: The number of tabs at the beginning of a line is equal to the number of open scopes. Everything from then onward is spaces. This means that whatever someone uses as their tabsize, the code will display correctly, regardless of whether that someone likes two-spaces per indent, 4, or 8, and it is trivial for them to change the format to display it at their preferred indentation level.
When one uses spaces, this is often not possible. You can assume that N spaces means one indent level, but often people will get it wrong, and in order to change how you view the indentation, you actually have to change the file. If you're using some kind of versioning system, this may not be good-- If you forget to change it back, then you have modified the fie without adding any new content (i.e. created a spurious change). I'd hope that people would agree that being able to at least -view- a file with one's preferred indentation level will lead to a quicker understanding of the code.
There is no functional difference between offering a higher speed tier and a lower speed tier and then forcing all competetors to that lower speed tier, and degrading their service.... And even more interesting, you may allocate such a paucity of resources to the lower speed tier that it is effectively unusable. How (if you allow for such an arbitrary assignment of resources) do you prevent a company from degrading a competitors service simply because it makes financial sense to do so (e.g. voip competing with landlines, etc.).
Realize that, if you advocate non application-specific QoS (i.e. provider targeted QoS instead of application targeted QoS), you advocate a 'net on which service providers are free to extract money from providers and consumers without any recourse.
I advocate 'net neutrality because I see the 'net as a major source of communication and, for better and worse, culture. I find it the idea of requiring more money for some bits over other bits to be alien-- Am I paying for access to the 'net, or am I simply paying the telcos to get rich?
Service/application targeted QoS is a different matter, and it -does- make some sort of sense, but it too can be abused.
Keep in mind that the telcos are not as interested in making tiers for the consumers-- they're more focused on the providers. Essentially, they promise to deliver reduced service to any provider daring enough to -not- pay them. Ostensibly, they'd like a bidding war-- that way the telco (as the monopoly or member of oligopoly) can extract all the profit possible from the providers.
Analogy: There is a trucking company. They charge by the pound. One day, after looking at your SEC filing, they notice that you're actually making a profit. They decide that you will be charged more by the pound than your competetor. Unfortunately for you, the trucking company controls the rights of way over which all products must flow.
Replace 'trucking company' with 'telco', 'pound' by 'bit', and 'products' with 'information', and you'll realize that this analogy is a very strong one.
You're making the assumption that the voters are well-enough informed to make the decision to vote for C.
I'm not at all certain that is the case, and if it isn't the case, then C will not get elected. If C will not get elected, it is in your best interest to vote for the candidate who is still likely to win, and is least evil. Remember: in this example, C -cannot- win because the people may not even know about C.
The purpose of property tax is to keep raw resources in economic circulation-- Otherwise, you should be looking to tax income/profit. Treating software as an economic raw resource seems a bit silly to me. Software is something which takes very little resources to duplicate, and the usage of software by one party does not preclude another party from using it. Because of this, it should not be taxed under "property tax", im my humble opinion.
You make the point that software can aid you in making profits. Fine, tax the profits. Your analogy with physical entities doesn't hold because those entities are not easy or cheap to copy (in comparison to software).
Again, hardware has an high inherent cost (in tangible resources) as compared to software, and in any case, software itself is useless without the hardware. Tax the hardware if you will, but taxing ones-and-zeroes is taxation without a sound economic basis (when your "possession" of those bits does not preclude others from "possessing" them).
This argument is different when you're considering the IP rightsholders, who should, perhaps, be taxed a property tax, since the state artificially allows the rightsholders to create a scarcity of the resource.
Copyright was created to foster the creation of works, so as to create a larger body of works available to the commons..
Currently, 70 years plus life of author is effectively forever as most people would be dead before the copyright expires on anything created today. This is contrary to the original intent. I believe the original intent is laudable, and the current regime of copyright control is draconian, and actually prevents the creation of new and interesting works in many cases-- Remix works, where people look to other works as the instruments on which they play, are something that are a major part of todays creative culture, and yet, these are difficult to create legally.
Furthermore, most oceangoing ships are likely to cause more effect on the wind than a turbine-- ships are basically big, massive walls through which the wind does not pass...
The blades on a turbine generally have far less area.
If you extend the elevator out far away enough from earth, the entire line (including where it is attached to the Earth) will be under -tension-.
The centripital forces will keep the line (i.e. space elevator) under tension. You just want to make sure that the line is long enough that the centripitel acceleration outweighs that of gravity and any accelerations put upon the line as a result of lifting things out of the gravity well, no pun intended.
Firstly, it is a loss because many people will find it difficult to communicate effectively with people over there..
Secondly, and unfortunately more importantly for lots of people with money, it makes China more difficult to deal with as an investment. China is a huge potential market for many many companies, and some of these are internet companies. With China's stranglehold on its internet, breaking into the market may be problematic at best, and impossible at worst.
This is a problem for people on both sides of the Great Firewall...
The whole "no-reverse engineering" provision is pretty dang scary to me.... So I'd say it wasn't a good law by any stretch of the imagination, because it enforces vendor lock-in, which -may- have absolutely NOTHING to do with defending copyright... The prime example of this is the DVD region stuff. The main reason to do this, from a market perspective, is that you can sell to China at a lower price than in the U.S., and those people in the U.S. are prevented from playing those chinese discs on their american dvd players, regardless of whether or not those discs were legally purchased.
Oh yea? Well, that is cool. When I took the intro courses (it was quarter based back then), the intro courses were pseudocode, then pascal, respectively.
We were required to take a functional programming course (taught in lisp on the macs), however..
There are provisions in copyright law for multiple authors already. I don't think it would be any more complex than it already is. *shrugs*
Companies can already write whatever they want into the contract. The nice part about forcing the copyright holder to be a -human- is that there is only one standard for how long the copyright endures.
I do NOT like life of author duration for copyright. The purpose of copyright is to cause the creation of more content, expressly for the public good.
The law is supposed to balance the public good with the individual good. This is the 'deal' struck with copyright-- The public benefits when works go into the public domain (they may not -ever- be put up commercially for various economic reasons) after a reasonably short amount of time (i.e. while the works are still topical), and the author is compenstated by holding a temporary monopoly on the copying of those works.
The intent is to incent the author -just enough- to get the author to create, it is not intended to make authors rich enough that they are no longer incented to create.
This is why *your* works should be subject to "*my*" whims-- for the public good.
If, for instance, the card can render faster than the monitor can display, then you can do things like motion blur to account for some of the perceptual difference.
This implies that having the mouse sampled at a higher frequency than the output device is still useful.
I happen to agree that DRM is evil, but there is such a thing as a lesser of evils, and that is what DRM is without vendor-lock-in, as opposed to DRM with vendor-lock-in.
You're assuming that I'm defending Congress-- Incorrect assumption. I'm merely trying to discuss the issue. If I could have the DMCA repealed, I would. Furthermore, I'd like to see the term of various IP protections shortened by a fair amount. (70 years plus life of author is about two lifetimes too long for me.)
When given the choice between an evil and a lesser evil, I choose the lesser evil. If congress is seeing pressure to make loophole laws to circumvent the DMCA I'm going to be cheering it on, especially when I see someone using its terms to enforce vendor-lock-in, as is my perception of Apple's behaviour.
As for using programs to remove the DRM from the AAC-Fairplay files. This is now a felony (evil DMCA). Isn't that fun?
The fun thing about capitalism is that it destroys itself. Without laws regulating it, you don't have capitalism anymore. I place little stock in 'the rules of capitalism'. I do think, however, that the politicians should not be making laws for special-interests, but I'm not going to hold my breath and wait for that any time soon.
I don't disagree that Apple may have achieved its position in the market through innovation, which is to be commended and encouraged. I do see, however, that Apple is trying to use its iTunes and Fairplay technology to capture its market (via vendor-lock-in).
If you're a strong proponent of capitalism, then you should be encouraged that Congress is going to encourage competition-- without the DMCA it seems likely that either or both would have legally reversed-engineered the format.
"not because they have vendor-lock-in".. how can you tell this?
DRM becomes much much less evil when you can have interoperability. Currently, there is no guarantee that you can use any of the iTunes songs in the future should Apple decide to stop supporting that particular DRM. You may say that market forces would prevent them from doing this, but I disagree-- Sometimes the most profit lies in the path that is most destructive to the society as a whole.
If the point of the DRM is vendor lock-in, then that is a different issue than what Congress was trying to address with the DCMA, which is currently what makes it illegal to reverse-engineer the DRM on those songs, even for interoperability..
I've seen CPU speed increases actually -decrease- overall memory bandwidth (due to bus speed mismatch).
Overall the trend is that it increases, as you mention, but it is not always linear, and it is certainly not monotonicaly increasing.
Increasing the cache speed will increase the speed at which bits are fetched out of the L1 and L2, and maybe if you're lucky, even the L3 cache, but it doesn't generally speed up memory. Sometimes, in fact, you have to decrease the speed at which memory operates in order to increase the speed at which your CPU operates. In these cases, certainly, for memory-bound processes the cpu speed is not a win.
A real world example: Take your IP stack. You think that it computes the CRC after the packet is put together? Wrong! It takes a few cycles to look up data out of memory, so it goes ahead and computes the CRC in the shadow of the memory lookup... Basically the CRC computation happens 'for free' because of the memory latency-- I.e. the process is memory-bound, and most assuredly not compute-bound.
I actually happen to know about 3d card bottlenecks, as I worked in an industry for 5 years whereby I dealt directly with lots of video and graphics, and I can tell you that there are certain IO limitations of current hardware that are -just- beginning to be addressed with pci-express. Asymmetric bandwidth to and from the card, for instance, was a -real- problem with AGP that seems better under PCI-Express.
PCI bandwidth does get saturated when you're doing lots of IO (oh, lets say sending uncompressed 10 bit graphics out to some IO card).
I would say that Apple -does- want to deal with all this DRM stuff. It [the DRM stuff] represents a significant barrier-to-entry for new players since new players are unlikely to be able to get their music in the proper DRM format for the hardware, or are unlikely to get licenses to sell the DRM music from the same sources that Apple has.
That is assuming you're compute-bound, instead of memory-bandwidth, harddrive-bandwidth, or some other kind of IO bound.
This may not be the case for many applications out there in the wild these days, so the performance gain is likely to be less than linear for those applications.
Slashdot Mirror
OK, normally I don't get involved in this argument, but your arguments seem to be from the gut, and not from the head, and this level of insensitivity pisses me off.
... etc.
.. EVEN MORE FUN, if there is an unwanted pregnancy in a underage child, they'll have to go get parental concent for the drugs in most places (if not all places now) in the US for a large class of the morning-after pills. This is especially fun when they've been raped by their dad, or uncle, etc.
> The 'liberty' here involves ending the life of another person.
Define 'person', and when a mass of cells becomes a 'person'.
For instance: Is a sperm a person? What about the egg? What about a newly fertilized egg which hasn't yet been attached to a uterous?
How about when it is 32 cells?
I find your statement to be ill-founded. Come up with a better foundation. I'll be absolutely shocked if you can come up with a defintion of 'person' that most people agree with here.
> There is a reason the founding fathers put life ahead of liberty. When the act of excersising a liberty causes the death of another person, the government must protect the life of the person.
> Thus, my point still stands.
Youre being somewhat incoherent. What I believe you want to say is something to the effect of:
Your right to liberty ends when it would infringe on someone else's right to life. Since gov't exists to protect these rights, it must protect people's lives so that they may exercise their liberty.
> Also, saying that your forcing someone to have an unwanted child is bs anyway.
No, actually it isn't. Please make a sensible argument about how the gov't wouldn't be forcing someone to have an unwanted child assuming that people have sex, and don't always want the child.
Saying "Well, they shouldn't have gotten themselves into that situation in the first place" doesn't change the fact that they do, and are then prevented from dealing with the pregnancy before an unwanted child is born.
> There seems to be this direct link between having sex and getting pregnant that people don't seem to get. The best way to keep from having a child is to not have sex.
You're right. But arguing that people should do without is simply unrealistic.
Sex is a perfectly natural thing. It is ingrained and instinctual, and we're built to enjoy it, ostensibly so that we will procreate.
Since, as we're using too much of the planet's resources, that procreation may be a bad thing now (though catholics may disagree), there is a strong impetus (often economic) that people shouldn't have children.
> There is no god given right to sex in the constitution and no where does the constitution say that you have the right to the pursuit of happiness without concequences.
I love this line of reasoning.. You shouldn't be able to have sex (assuming that you cannot have an abortion) unless you're willing and able to deal with the consequences. This is fun! Eventually, when you take this line of reasoning to the extreme, what you're saying is that only rich people can have sex because only they are equipped to deal with the consequences.
Fun!
> And before you bring up rape and incest, in those cases the woman will ussually get the morning after pill.
Incorrect. This can be labeled 'abortion', and some pharmicists DON'T DISPENSE THESE PRESCRIPTIONS.
Some pharmicists won't even SELL THESE DRUGS if they're OTC.
Even more fun, you get assholes who will not fill the prescription, and who will hold onto it so that no other pharmacist may fill it.
> And in cases where the woman hides it till its too late to get a legal abortion (assuming that the state she lives in allows abortions up until a certain point) then they should be able to go to a judge and get due process of law to authorize the abortion.
You're making no sense. You're saying that a woman will go to a judge to get a dispensation to have an illegal abortion?
http://en.wikipedia.org/wiki/Internet_censorship_i n_mainland_China
Look for "Great Firewall" it is another name for the "Golden Shield Project", which has been policy in China since 1998.
"AJA" makes a card with linux drivers for SD-SDI and HD-SDI video output.
.. lets see, another company that makes this kind of stuff is "DVS".
"NVidia", also makes a card-- it (now) is a daughter card, which allows you to output SD-SDI or HD-SDI from the screen, or from an offscreen surface (e.g. glx-pbuffer).
(Obviously, this is not a consumer card...)
(Ya, I made linux boxes make pretty broadcast video for a living for a while...)
A tab == X spaces (at the beginning of a line, at least).
This may seem obvious, but it is also powerful-- it means that I may indicate, symbolically (i.e. by usng tabs) the indentation level of a piece of code, etc.
Thus, I mix tabs and spaces according to the following:
The number of tabs at the beginning of a line is equal to the number of open scopes. Everything from then onward is spaces.
This means that whatever someone uses as their tabsize, the code will display correctly, regardless of whether that someone likes two-spaces per indent, 4, or 8, and it is trivial for them to change the format to display it at their preferred indentation level.
When one uses spaces, this is often not possible. You can assume that N spaces means one indent level, but often people will get it wrong, and in order to change how you view the indentation, you actually have to change the file.
If you're using some kind of versioning system, this may not be good-- If you forget to change it back, then you have modified the fie without adding any new content (i.e. created a spurious change).
I'd hope that people would agree that being able to at least -view- a file with one's preferred indentation level will lead to a quicker understanding of the code.
$ vimtutor
This is an effective little program for learning basic vi/vim.
There is no functional difference between offering a higher speed tier and a lower speed tier and then forcing all competetors to that lower speed tier, and degrading their service. ... And even more interesting, you may allocate such a paucity of resources to the lower speed tier that it is effectively unusable.
How (if you allow for such an arbitrary assignment of resources) do you prevent a company from degrading a competitors service simply because it makes financial sense to do so (e.g. voip competing with landlines, etc.).
Realize that, if you advocate non application-specific QoS (i.e. provider targeted QoS instead of application targeted QoS), you advocate a 'net on which service providers are free to extract money from providers and consumers without any recourse.
I advocate 'net neutrality because I see the 'net as a major source of communication and, for better and worse, culture. I find it the idea of requiring more money for some bits over other bits to be alien-- Am I paying for access to the 'net, or am I simply paying the telcos to get rich?
Service/application targeted QoS is a different matter, and it -does- make some sort of sense, but it too can be abused.
Keep in mind that the telcos are not as interested in making tiers for the consumers-- they're more focused on the providers. Essentially, they promise to deliver reduced service to any provider daring enough to -not- pay them. Ostensibly, they'd like a bidding war-- that way the telco (as the monopoly or member of oligopoly) can extract all the profit possible from the providers.
Analogy: There is a trucking company. They charge by the pound. One day, after looking at your SEC filing, they notice that you're actually making a profit. They decide that you will be charged more by the pound than your competetor. Unfortunately for you, the trucking company controls the rights of way over which all products must flow.
Replace 'trucking company' with 'telco', 'pound' by 'bit', and 'products' with 'information', and you'll realize that this analogy is a very strong one.
You're making the assumption that the voters are well-enough informed to make the decision to vote for C.
I'm not at all certain that is the case, and if it isn't the case, then C will not get elected.
If C will not get elected, it is in your best interest to vote for the candidate who is still likely to win, and is least evil.
Remember: in this example, C -cannot- win because the people may not even know about C.
The purpose of property tax is to keep raw resources in economic circulation-- Otherwise, you should be looking to tax income/profit. Treating software as an economic raw resource seems a bit silly to me. Software is something which takes very little resources to duplicate, and the usage of software by one party does not preclude another party from using it. Because of this, it should not be taxed under "property tax", im my humble opinion.
You make the point that software can aid you in making profits. Fine, tax the profits. Your analogy with physical entities doesn't hold because those entities are not easy or cheap to copy (in comparison to software).
Again, hardware has an high inherent cost (in tangible resources) as compared to software, and in any case, software itself is useless without the hardware. Tax the hardware if you will, but taxing ones-and-zeroes is taxation without a sound economic basis (when your "possession" of those bits does not preclude others from "possessing" them).
This argument is different when you're considering the IP rightsholders, who should, perhaps, be taxed a property tax, since the state artificially allows the rightsholders to create a scarcity of the resource.
Copyright was created to foster the creation of works, so as to create a larger body of works available to the commons..
Currently, 70 years plus life of author is effectively forever as most people would be dead before the copyright expires on anything created today. This is contrary to the original intent. I believe the original intent is laudable, and the current regime of copyright control is draconian, and actually prevents the creation of new and interesting works in many cases-- Remix works, where people look to other works as the instruments on which they play, are something that are a major part of todays creative culture, and yet, these are difficult to create legally.
Furthermore, most oceangoing ships are likely to cause more effect on the wind than a turbine-- ships are basically big, massive walls through which the wind does not pass...
The blades on a turbine generally have far less area.
No no no..
If you extend the elevator out far away enough from earth, the entire line (including where it is attached to the Earth) will be under -tension-.
The centripital forces will keep the line (i.e. space elevator) under tension. You just want to make sure that the line is long enough that the centripitel acceleration outweighs that of gravity and any accelerations put upon the line as a result of lifting things out of the gravity well, no pun intended.
No, they just have to sue him until he runs out of funds.
Actually it is a loss in many ways.
Firstly, it is a loss because many people will find it difficult to communicate effectively with people over there..
Secondly, and unfortunately more importantly for lots of people with money, it makes China more difficult to deal with as an investment. China is a huge potential market for many many companies, and some of these are internet companies. With China's stranglehold on its internet, breaking into the market may be problematic at best, and impossible at worst.
This is a problem for people on both sides of the Great Firewall...
The whole "no-reverse engineering" provision is pretty dang scary to me.. .. So I'd say it wasn't a good law by any stretch of the imagination, because it enforces vendor lock-in, which -may- have absolutely NOTHING to do with defending copyright. .. The prime example of this is the DVD region stuff. The main reason to do this, from a market perspective, is that you can sell to China at a lower price than in the U.S., and those people in the U.S. are prevented from playing those chinese discs on their american dvd players, regardless of whether or not those discs were legally purchased.
There are other problems.
Oh yea? Well, that is cool.
When I took the intro courses (it was quarter based back then), the intro courses were pseudocode, then pascal, respectively.
We were required to take a functional programming course (taught in lisp on the macs), however..
There are provisions in copyright law for multiple authors already. I don't think it would be any more complex than it already is. *shrugs*
Companies can already write whatever they want into the contract. The nice part about forcing the copyright holder to be a -human- is that there is only one standard for how long the copyright endures.
I do NOT like life of author duration for copyright. The purpose of copyright is to cause the creation of more content, expressly for the public good.
The law is supposed to balance the public good with the individual good. This is the 'deal' struck with copyright-- The public benefits when works go into the public domain (they may not -ever- be put up commercially for various economic reasons) after a reasonably short amount of time (i.e. while the works are still topical), and the author is compenstated by holding a temporary monopoly on the copying of those works.
The intent is to incent the author -just enough- to get the author to create, it is not intended to make authors rich enough that they are no longer incented to create.
This is why *your* works should be subject to "*my*" whims-- for the public good.
License it. You retain copyright, they get to exercize it, and the duration is still the same.
I'm not arguing for either side here, just playing devil's advocate.
It can make sense to have them decoupled.
If, for instance, the card can render faster than the monitor can display, then you can do things like motion blur to account for some of the perceptual difference.
This implies that having the mouse sampled at a higher frequency than the output device is still useful.
pyroelectric-- Converts heat energy into electrical energy
piezoelectric-- Converts kinetic energy into electrical energy
In this experiment, they heat up a (Lithium tantalate) crystal which reacts by creating a very high charge.. etc.
In other words, the crystal is a pyroelectric crystal, and not necessarily piezoelectric.
I happen to agree that DRM is evil, but there is such a thing as a lesser of evils, and that is what DRM is without vendor-lock-in, as opposed to DRM with vendor-lock-in.
You're assuming that I'm defending Congress-- Incorrect assumption. I'm merely trying to discuss the issue. If I could have the DMCA repealed, I would. Furthermore, I'd like to see the term of various IP protections shortened by a fair amount. (70 years plus life of author is about two lifetimes too long for me.)
When given the choice between an evil and a lesser evil, I choose the lesser evil. If congress is seeing pressure to make loophole laws to circumvent the DMCA I'm going to be cheering it on, especially when I see someone using its terms to enforce vendor-lock-in, as is my perception of Apple's behaviour.
As for using programs to remove the DRM from the AAC-Fairplay files. This is now a felony (evil DMCA). Isn't that fun?
The fun thing about capitalism is that it destroys itself. Without laws regulating it, you don't have capitalism anymore. I place little stock in 'the rules of capitalism'. I do think, however, that the politicians should not be making laws for special-interests, but I'm not going to hold my breath and wait for that any time soon.
I don't disagree that Apple may have achieved its position in the market through innovation, which is to be commended and encouraged. I do see, however, that Apple is trying to use its iTunes and Fairplay technology to capture its market (via vendor-lock-in).
If you're a strong proponent of capitalism, then you should be encouraged that Congress is going to encourage competition-- without the DMCA it seems likely that either or both would have legally reversed-engineered the format.
"not because they have vendor-lock-in"
DRM becomes much much less evil when you can have interoperability. Currently, there is no guarantee that you can use any of the iTunes songs in the future should Apple decide to stop supporting that particular DRM. You may say that market forces would prevent them from doing this, but I disagree-- Sometimes the most profit lies in the path that is most destructive to the society as a whole.
If the point of the DRM is vendor lock-in, then that is a different issue than what Congress was trying to address with the DCMA, which is currently what makes it illegal to reverse-engineer the DRM on those songs, even for interoperability..
I've seen CPU speed increases actually -decrease- overall memory bandwidth (due to bus speed mismatch).
Overall the trend is that it increases, as you mention, but it is not always linear, and it is certainly not monotonicaly increasing.
Increasing the cache speed will increase the speed at which bits are fetched out of the L1 and L2, and maybe if you're lucky, even the L3 cache, but it doesn't generally speed up memory. Sometimes, in fact, you have to decrease the speed at which memory operates in order to increase the speed at which your CPU operates. In these cases, certainly, for memory-bound processes the cpu speed is not a win.
A real world example: Take your IP stack. You think that it computes the CRC after the packet is put together? Wrong! It takes a few cycles to look up data out of memory, so it goes ahead and computes the CRC in the shadow of the memory lookup... Basically the CRC computation happens 'for free' because of the memory latency-- I.e. the process is memory-bound, and most assuredly not compute-bound.
I actually happen to know about 3d card bottlenecks, as I worked in an industry for 5 years whereby I dealt directly with lots of video and graphics, and I can tell you that there are certain IO limitations of current hardware that are -just- beginning to be addressed with pci-express. Asymmetric bandwidth to and from the card, for instance, was a -real- problem with AGP that seems better under PCI-Express.
PCI bandwidth does get saturated when you're doing lots of IO (oh, lets say sending uncompressed 10 bit graphics out to some IO card).
I would say that Apple -does- want to deal with all this DRM stuff. It [the DRM stuff] represents a significant barrier-to-entry for new players since new players are unlikely to be able to get their music in the proper DRM format for the hardware, or are unlikely to get licenses to sell the DRM music from the same sources that Apple has.
That is assuming you're compute-bound, instead of memory-bandwidth, harddrive-bandwidth, or some other kind of IO bound.
This may not be the case for many applications out there in the wild these days, so the performance gain is likely to be less than linear for those applications.