One of the problems of rapidly increasing complexity of applications, particularly Java/C#/C++ and other object oriented languages, is that performance is critically destroyed by design, and it's nearly impossible to recover it without re-writing the core of the application procedurally.
The problem here is that small dynamically allocated objects with heavy pointer threading flush even large caches quickly and yeild very low locality combined with very low efficiency of cache line memory faults... small amount of active useful data per cache line, and high overhead in data structures and useless data faulted in. This quickly applies at the page level as well, as locality remain poor between pages of bloated applications working set.
CPU performance scales, even L1/L2 performance scales reasonably well, but raw memory perforance is two orders of magnitude slower than processors and quickly becoming three orders of magnitude slower. It's nearly impossible to constrain page level working sets of OO programs to scale performance.
When more than 20% of your memory hits fault to raw memory, performance just crawls, and a faster machine, with lots more raw memory doesn't help,... in fact, just makes it worse.
Direct mapped caches, and small set associate caches do not work well with today's OO programming... needed are very large fully associated caches, and those are not cheap to build.
It's time to realize that for the last 5-10 years memory performace hasn't hardly (in comparison) improved at all... and it's past time to start looking at programming tools that optimize performance by design, not destroy it.
What good is a 4GHz processor, that runs at 100mHz memory speeds when it's all said and done because of poor cache line usage for every word or two referenced.
Actually... just the several inch leads on light fixtures and other electrical devices can become bi-directional antennas for the determined. You pretty much have to make the screen room such that power is completely issolated, and other other connections to the world severed.
There is something of a chicken and egg problem here. Good software can be produced in volume, but good software can not compete after cheap trash software consumes all the dollars (or available labor) in the market by first to ship. Both in commercial and open source markets.
As an oldie that started programming on tab equipment, IBM 1401's, 360's, Bendix G15's, Nova's, and a number of other long obsolete architectures, the problem is the unconstrained growth of nearly every application. Not just by the size of the application, but the minimum environment to just run "Hello World" in a GUI window.
Even if a software developer can get an application thru a GOOD Alpha and Beta test, the OS platform will change under it before or just after it goes to market. 30 years ago, when you targeted a vendors OS platform, the critical design concern was forward compatability and portability HAD to be perserved. Today the interactions are so complex, that isn't even an achievable design goal, even though some people try.
An interesting quote from the 1970's was a well known authors pride in the most difficult problem in managing the feature set of his compiler, was keeping all the good, but unnecessary cruft out. Today, software seldom has controlled complexity management, but rather not only is the kitchen sink tossed in to expand the side by side feature comparison set, but every kitchen sink on the planet. With this kind of pathing and feature competition, complexity spirals out of control, and the ability to regression test effectively is eliminated by the N to the Nth interaction matrix between features.
Add to that, the problem of library complexity, where everything is a.dll or.so, and you have the perfect example of being able to ship tested code, that since it is NOT statically linked, is very likely not to match field systems after the first "critical" OS update chasing bugs, security flaws, and creaping featurism. You can code around bugs, and sometimes accept them with statically linked code... that becomes impossible with a dynamic library environment.
Many difficult to debug code paths are the side effects of resource allocation issues/failures, which do not even show up until the system environment grows on certain platforms to trigger those issues, even though the applicaiton may have been stable on the platform for 5 years.
RedHat is trying to manage the problem, by forking Linux into a stable release path that doesn't have then entire OS and tool chain changing radically every 3-6 months. And avoid forcing applications developers from rolling new releases every 3-6 months as well. Many custom appliations, and the cost to maintain inhouse software, is driven by this instability of release migration. Even if you manage to stay on a particular MSWin or Linux release for several years to manage this cost, it comes back to bite you in that the migration inbetween gets so large that new off the self software will not support the 2 year old platform, and the coding changes to bring custom software forward and excessive.
There is a cost, a huge cost in rapid evolution... and that is unmanagable complexity and the resulting quality issues.
Gee ... two I/Os to swap, One I/O to reclaim cache
on
Tuning Linux VM swapping
·
· Score: 2, Interesting
Swapping out any semi-active process, especially an interactive one, is utter foolishness which should be subject to termination grounds (unless of course you work for a big iron computer company that needs to boost revenue by purposefully slowing systems down). Any time you decide to clear memory and page a task out... you are creating two I/O's... one out, one back in... and doing it with shit poor locality since the swap area is probably at the other end of the disk from any other active I/O.... just discarding entries in the file cache only requires a single I/O to recover it..... so tell me... who is the idiot that believes two I/O's to kill the performance of an interactive process is better?
Because the outright public forum blackmail tactic being used by a sub-group inside KDE to demand artistic rights here smacks clearly in the face of gpl freedoms. If the discusion had stayed inside the KDE developer channels that's fine... they can work on policy issues there. What appears to be happening here is both the main stream KDE developers and FSF/GNU standing idly by while the very vocal group uses some very unethical means to blackmail RedHat into changing it's position.
If KDE isn't willing to take a firm stand on THEIR image, and FSF isn't either, then these noisy little power plays are going to be the poison pill that drives large corporate developers from adopting open source products in their platforms.
The open source community as a whole needs to provide the peer presure that FREE is FREE, for everyone. Those that do not want to play by the GPL rules, clearly need some external presure to modify this childish behavior. In this case, the general distrust of corporate america seems to encourage a hear no evil, see no evil, do no evil response.
Over the last two weeks a landmark crisis emerged between KDE and RedHat over GPL. The KDE team is probably the largest open source product, and has decided to ignore the FSF/GNU principles that true freedom explicitly means the freedom of NOT having to ask permission to change an open source GPL product and distribute it. In short, the KDE team is demanding "artistic control" and the right to censor the changes made by RedHat.
The FSF failed horribly by allowing this public debate to continue, rather than stepping in promptly and strongly affirming it's principles that free software, especially that lincensed under GPL, does not include the right of censorship that the KDE team is demanding.
While a large percentage of the open source community have open contempt for corporate America, and the growing contempt for and bashing of RedHat has grown beyound a friendly sport, this one issue/event may well be the turning point that forces corporate America to abandon the open source movement.
The bashing of RedHat by KDE over this issue is pure hypocrisy - freedom under GPL was supposed to be freedom for everyone, including corporate America. The FSF/GNU camp failed the movement over the last couple weeks by refusing to openly and agressively defend it's core principles. The KDE team failed the open source movement, by abandoning the core principle of the GPL, and using it's size and stature in the open source community to start what might well be the last debate of a failed ideal.
Power corrupts, and what we are watching unfold, is that the absolute power being demanding by the KDE team, is almost certainly going to corrupt the open source movement and GPL community absolutely as other ego's seek to grab their share of this new power.
Monday, at dawn I sent a written email request to gnu@gnu.org specifically asking three questions. There has been no answer, which is chilling.
1) From this sites web page, the free software definition includes the specific liberty of NOT having to ask permission to make changes and distribute them. Is the KDE team rightfully acting in the spirit of free software and the GPL licenses by demanding this contract term be ignored and RedHat abandon distribution of modified KDE software?
2) Does the FSF/GNU consider the forking of KDE a legal or moral violation of GPL?
3) Are the KDE teams actions regarding this issue, of publicly attacking RedHat for violating their artistic rights to control KDE's evolution consistant with the position of the FSF and GPL licensing?
I strongly urge everyone that placed their hopes on FSF/GPL to protect one core right to tinker to let gnu@gnu.org know that they MUST defend the principles they stand for, or watch THEIR movement fail as corporate America (nay that of the world) drop open source and GPL lest they be the next target of corporate bashing.
stealing one cd from the store is a misdemeanor and worth a slap on the hand and maybe a small fine. Do it two or three times and you are likely to go to jail a few days, steal 200 CD's from the store and you just commited a felony theft of about $2,000 that will send you to jail for a long time if you hit the three strikes provisions.
Many of us seldom handle paper money any more, in fact the vast majority of all "money" transactions do not actually use real government minted money anymore. The bits in the bank are the only money.
Likewise the sale of music has historically been physical, just like the handling of physical money. As we move into the electronic age, with computer bits for tangible assets, then a sale no longer is the replication of a physical CD, but the replication of the bits that represent the product. In this environment, unauthorized replication of the bits, is absolutely no differnet than unauthorized replication of a physical CD or record or tape. Your right to ownership of one copy hinges on your purchasing the right to make one copy (plus whatever fair use copies) - and that right is documents as a record of ownership, just as your right of ownership to the virtual bank money.
If someone transfers your bank account, they have stolen one or more units of payment for work you have done. If someone illegally distributes an artists work, they are depriving that artist (or other owner of the work) of one or more dollars of revenue for their work.
If you go back a few levels you will note that the poster I responded stated that bits had no value... in an electronic paperless world, bits do have value... both as money, and as real product.
All analogies have some flaw because they do not represent exactly the identical situation. The critial point here is that a music artist that has one copy of his work stolen is out some portion of the payment for his work. Anyone who has part of their bank account stolen is out of some portion of the payment they recieved for their work. The analogy was formed because of a flawed statement that electronic bits have no value, I respond with the arguement that virtual money and virtual music bits both represent a real asset of value to the rightful owners - legally and morally. Just because they are represented in bits, rather than physical paper doesn't mater a bit.
You say "Your bank-robber analogy is utterly wrong. The bank robber has denied you something: your assets. Once he takes them, those assets are not there and you cannot use them. Copying or record company) the ABILITY to sell the work. They still can and still do."
The bank robber stole a few days of your work, in the form of money - that does not deny you YOUR ability to sell your work either. So if a robber takes your money, or one or more of your sales, it's the same thing - each party had THEIR property stolen. Legally, and morally.
So the guy that has thought about the problem wisks off an answer in 30 seconds and appears a wizard. The guy who has not encountered that specific problem really searches the depths of his experiences and doesn't flash an answer until he is pretty sure it is the "best" answer - and after 15 minutes of very careful design places his answer on the board - and in comparison appears very slow.
Now what does that tell you about the performance of both applicants for a problem they haven't seen before?
You Say "Those bits in the bank's computer are a digital representation of a physical thing: the money in the bank's vault that belongs to me."
The music industry says EXACTLY the same thing, the MP3 bits are a digital representation of a physical thing - the music CD they own - and those bits belong to them till you purchase a copy.
I agree the bits in the banks computer are your money - was obvious from my assertion you would cry foul when somebody played with them.
But in exactly the same way, the bits that hold the pirated MP3 music are EVERY bit as tangible to the authors and copyright holders as is YOUR money. Both are totally intangable assets of value to the real owners.
To claim the music owners bits are different from yours, is well... self serving and short sighted.
"Fair Use" is not stealing, for all the reasons you state. However, the definition of "stealing" as defined in the english language as used in the US is much broader that you might try to ignore. Please checkout www.webster.com and what you will find is:
"1 : to take the property of another wrongfully and especially as an habitual or regular practice"
"1 a : to take or appropriate without right or leave and with intent to keep or make use of wrongfully"
"1 d : to appropriate to oneself or beyond one's proper share"
so, the courts have already ruled that P2P systems violate copyright by unauthorized distribution and the those downloading are willfully in violation since they have absolutely no "fair use" claim.
In the definition of stealing, the community involved in illegal distribution of copyrighted works, is in clear conformance with all three of the definitions above, so "stealing" is absolutely the correct term for this act.
First your comment "When you steal property, you deprive someone of it.", which is far to narrow a definition of "stealing". If you go to www.websters.com and look up "stealing" you will also find:
"1 : to take the property of another wrongfully and especially as an habitual or regular practice"
"1 a : to take or appropriate without right or leave and with intent to keep or make use of wrongfully"
"1 d : to appropriate to oneself or beyond one's proper share"
which by US law includes unauthorized duplication and distribution of non-tangable intelectual property of obvious value,especially where copyrighted.
Your second comment "When you steal something you otherwise would have paid for, you deprive someone of the revenue they would get from the sale. When you make a copy of some digital bits that you only are willing to copy precisely because it costs you next to nothing, and honestly (there's the rub) would rather do without if it cost you more, you do not deprive anyone of anything." is also false.
Take the example of someone hacking your bank account and transfering is entire value out of your account. If that person would never choose to work a normal wage paying job, then their transfer of your assets (not reall cash) out of your bank account by your definition is not really stealing since they would never work to earn it in the first place. If such a thief, empty's your bank account I strongly suspect you are going to be crying foul to the bank, the police, and the community for the theft. I call it "stealing" by the above definition. You can rationalize it however you might to ignore the obvious theft as already ruled by US Courts.
You might want to go to www.webster.com and check the definition which is not nearly as limited as you have mistakenly asserted:
"No, they are NOT. Stealing means depriving the original owner of the use of that material."
From Websters:
"1 : to take the property of another wrongfully and especially as an habitual or regular practice"
"1 a : to take or appropriate without right or leave and with intent to keep or make use of wrongfully"
"1 d : to appropriate to oneself or beyond one's proper share"
which by US law includes unauthorized duplication and distribution of non-tangable intelectual property of obvious value, especially where copyrighted.
So, in the United States, speaking in english, the description of illegal (already ruled so by the courts) P2P sharing of unauthorized copyrighted works as "stealing" is absolutely correct.
Copyright laws don't talk about selling, they do talk about distribution and unauthrized replication that is beyound "fair use", which the P2P community is in violation of.
There are lots of people that don't think money is worth the normal "work" effort it takes to "purchase" it... we call them robbers, thieves, scum. What's the difference with the property is something other than money?
Slashdot Mirror
One of the problems of rapidly increasing complexity of applications, particularly Java/C#/C++ and other object oriented languages, is that performance is critically destroyed by design, and it's nearly impossible to recover it without re-writing the core of the application procedurally.
... small amount of active useful data per cache line, and high overhead in data structures and useless data faulted in. This quickly applies at the page level as well, as locality remain poor between pages of bloated applications working set.
... in fact, just makes it worse.
... needed are very large fully associated caches, and those are not cheap to build.
... and it's past time to start looking at programming tools that optimize performance by design, not destroy it.
The problem here is that small dynamically allocated objects with heavy pointer threading flush even large caches quickly and yeild very low locality combined with very low efficiency of cache line memory faults
CPU performance scales, even L1/L2 performance scales reasonably well, but raw memory perforance is two orders of magnitude slower than processors and quickly becoming three orders of magnitude slower. It's nearly impossible to constrain page level working sets of OO programs to scale performance.
When more than 20% of your memory hits fault to raw memory, performance just crawls, and a faster machine, with lots more raw memory doesn't help,
Direct mapped caches, and small set associate caches do not work well with today's OO programming
It's time to realize that for the last 5-10 years memory performace hasn't hardly (in comparison) improved at all
What good is a 4GHz processor, that runs at 100mHz memory speeds when it's all said and done because of poor cache line usage for every word or two referenced.
Actually ... just the several inch leads on light fixtures and other electrical devices can become bi-directional antennas for the determined. You pretty much have to make the screen room such that power is completely issolated, and other other connections to the world severed.
so why not charge back if you can :)
Then why GPL with copyright rather than be truely open and Public Domain as has been done for a few centuries??
There is something of a chicken and egg problem here. Good software can be produced in volume, but good software can not compete after cheap trash software consumes all the dollars (or available labor) in the market by first to ship. Both in commercial and open source markets.
As an oldie that started programming on tab equipment, IBM 1401's, 360's, Bendix G15's, Nova's, and a number of other long obsolete architectures, the problem is the unconstrained growth of nearly every application. Not just by the size of the application, but the minimum environment to just run "Hello World" in a GUI window.
.dll or .so, and you have the perfect example of being able to ship tested code, that since it is NOT statically linked, is very likely not to match field systems after the first "critical" OS update chasing bugs, security flaws, and creaping featurism. You can code around bugs, and sometimes accept them with statically linked code ... that becomes impossible with a dynamic library environment.
... and that is unmanagable complexity and the resulting quality issues.
Even if a software developer can get an application thru a GOOD Alpha and Beta test, the OS platform will change under it before or just after it goes to market. 30 years ago, when you targeted a vendors OS platform, the critical design concern was forward compatability and portability HAD to be perserved. Today the interactions are so complex, that isn't even an achievable design goal, even though some people try.
An interesting quote from the 1970's was a well known authors pride in the most difficult problem in managing the feature set of his compiler, was keeping all the good, but unnecessary cruft out. Today, software seldom has controlled complexity management, but rather not only is the kitchen sink tossed in to expand the side by side feature comparison set, but every kitchen sink on the planet. With this kind of pathing and feature competition, complexity spirals out of control, and the ability to regression test effectively is eliminated by the N to the Nth interaction matrix between features.
Add to that, the problem of library complexity, where everything is a
Many difficult to debug code paths are the side effects of resource allocation issues/failures, which do not even show up until the system environment grows on certain platforms to trigger those issues, even though the applicaiton may have been stable on the platform for 5 years.
RedHat is trying to manage the problem, by forking Linux into a stable release path that doesn't have then entire OS and tool chain changing radically every 3-6 months. And avoid forcing applications developers from rolling new releases every 3-6 months as well. Many custom appliations, and the cost to maintain inhouse software, is driven by this instability of release migration. Even if you manage to stay on a particular MSWin or Linux release for several years to manage this cost, it comes back to bite you in that the migration inbetween gets so large that new off the self software will not support the 2 year old platform, and the coding changes to bring custom software forward and excessive.
There is a cost, a huge cost in rapid evolution
Swapping out any semi-active process, especially an interactive one, is utter foolishness which should be subject to termination grounds (unless of course you work for a big iron computer company that needs to boost revenue by purposefully slowing systems down). Any time you decide to clear memory and page a task out ... you are creating two I/O's ... one out, one back in ... and doing it with shit poor locality since the swap area is probably at the other end of the disk from any other active I/O .... just discarding entries in the file cache only requires a single I/O to recover it. .... so tell me ... who is the idiot that believes two I/O's to kill the performance of an interactive process is better?
what kind of lidar is non-optical ?????
Because the outright public forum blackmail tactic being used by a sub-group inside KDE to demand artistic rights here smacks clearly in the face of gpl freedoms. If the discusion had stayed inside the KDE developer channels that's fine ... they can work on policy issues there. What appears to be happening here is both the main stream KDE developers and FSF/GNU standing idly by while the very vocal group uses some very unethical means to blackmail RedHat into changing it's position.
If KDE isn't willing to take a firm stand on THEIR image, and FSF isn't either, then these noisy little power plays are going to be the poison pill that drives large corporate developers from adopting open source products in their platforms.
The open source community as a whole needs to provide the peer presure that FREE is FREE, for everyone. Those that do not want to play by the GPL rules, clearly need some external presure to modify this childish behavior. In this case, the general distrust of corporate america seems to encourage a hear no evil, see no evil, do no evil response.
Over the last two weeks a landmark crisis emerged between KDE and RedHat over GPL. The KDE team is probably the largest open source product, and has decided to ignore the FSF/GNU principles that true freedom explicitly means the freedom of NOT having to ask permission to change an open source GPL product and distribute it. In short, the KDE team is demanding "artistic control" and the right to censor the changes made by RedHat.
& file=article&sid=202 and http://www.linuxandmain.com/modules.php?name=News& file=article&sid=203.
The FSF failed horribly by allowing this public debate to continue, rather than stepping in promptly and strongly affirming it's principles that free software, especially that lincensed under GPL, does not include the right of censorship that the KDE team is demanding.
While a large percentage of the open source community have open contempt for corporate America, and the growing contempt for and bashing of RedHat has grown beyound a friendly sport, this one issue/event may well be the turning point that forces corporate America to abandon the open source movement.
The bashing of RedHat by KDE over this issue is pure hypocrisy - freedom under GPL was supposed to be freedom for everyone, including corporate America. The FSF/GNU camp failed the movement over the last couple weeks by refusing to openly and agressively defend it's core principles. The KDE team failed the open source movement, by abandoning the core principle of the GPL, and using it's size and stature in the open source community to start what might well be the last debate of a failed ideal.
Power corrupts, and what we are watching unfold, is that the absolute power being demanding by the KDE team, is almost certainly going to corrupt the open source movement and GPL community absolutely as other ego's seek to grab their share of this new power.
Monday, at dawn I sent a written email request to gnu@gnu.org specifically asking three questions. There has been no answer, which is chilling.
1) From this sites web page, the free software definition includes the specific liberty of NOT having to ask permission to make changes and distribute them. Is the KDE team rightfully acting in the spirit of free software and the GPL licenses by demanding this contract term be ignored and RedHat abandon distribution of modified KDE software?
2) Does the FSF/GNU consider the forking of KDE a legal or moral violation of GPL?
3) Are the KDE teams actions regarding this issue, of publicly attacking RedHat for violating their artistic rights to control KDE's evolution consistant with the position of the FSF and GPL licensing?
For those that have missed the quiet war, please check out http://www.linuxandmain.com/modules.php?name=News
I strongly urge everyone that placed their hopes on FSF/GPL to protect one core right to tinker to let gnu@gnu.org know that they MUST defend the principles they stand for, or watch THEIR movement fail as corporate America (nay that of the world) drop open source and GPL lest they be the next target of corporate bashing.
using your same source, then lookup
[http://www.dictionary.com/search?q=steal]
which says absolutely nothing about taking physical possession of something in an exclusive fashion, and says everything about:
"To take (the property of another) without right or permission."
The issue here, legal and moral, is that the music belongs to the artist or copyright holder, and to take a copy without permission is stealing.
stealing one cd from the store is a misdemeanor and worth a slap on the hand and maybe a small fine. Do it two or three times and you are likely to go to jail a few days, steal 200 CD's from the store and you just commited a felony theft of about $2,000 that will send you to jail for a long time if you hit the three strikes provisions.
Many of us seldom handle paper money any more, in fact the vast majority of all "money" transactions do not actually use real government minted money anymore. The bits in the bank are the only money.
... in an electronic paperless world, bits do have value ... both as money, and as real product.
Likewise the sale of music has historically been physical, just like the handling of physical money. As we move into the electronic age, with computer bits for tangible assets, then a sale no longer is the replication of a physical CD, but the replication of the bits that represent the product. In this environment, unauthorized replication of the bits, is absolutely no differnet than unauthorized replication of a physical CD or record or tape. Your right to ownership of one copy hinges on your purchasing the right to make one copy (plus whatever fair use copies) - and that right is documents as a record of ownership, just as your right of ownership to the virtual bank money.
If someone transfers your bank account, they have stolen one or more units of payment for work you have done. If someone illegally distributes an artists work, they are depriving that artist (or other owner of the work) of one or more dollars of revenue for their work.
If you go back a few levels you will note that the poster I responded stated that bits had no value
All analogies have some flaw because they do not represent exactly the identical situation. The critial point here is that a music artist that has one copy of his work stolen is out some portion of the payment for his work. Anyone who has part of their bank account stolen is out of some portion of the payment they recieved for their work. The analogy was formed because of a flawed statement that electronic bits have no value, I respond with the arguement that virtual money and virtual music bits both represent a real asset of value to the rightful owners - legally and morally. Just because they are represented in bits, rather than physical paper doesn't mater a bit.
You say "Your bank-robber analogy is utterly wrong. The bank robber has denied you something: your assets. Once he takes them, those assets are not there and you cannot use them. Copying or record company) the ABILITY to sell the work. They still can and still do."
The bank robber stole a few days of your work, in the form of money - that does not deny you YOUR ability to sell your work either. So if a robber takes your money, or one or more of your sales, it's the same thing - each party had THEIR property stolen. Legally, and morally.
So the guy that has thought about the problem wisks off an answer in 30 seconds and appears a wizard. The guy who has not encountered that specific problem really searches the depths of his experiences and doesn't flash an answer until he is pretty sure it is the "best" answer - and after 15 minutes of very careful design places his answer on the board - and in comparison appears very slow.
Now what does that tell you about the performance of both applicants for a problem they haven't seen before?
You Say "Those bits in the bank's computer are a digital representation of a physical thing: the money in the bank's vault that belongs to me."
The music industry says EXACTLY the same thing, the MP3 bits are a digital representation of a physical thing - the music CD they own - and those bits belong to them till you purchase a copy.
I agree the bits in the banks computer are your money - was obvious from my assertion you would cry foul when somebody played with them.
... self serving and short sighted.
But in exactly the same way, the bits that hold the pirated MP3 music are EVERY bit as tangible to the authors and copyright holders as is YOUR money. Both are totally intangable assets of value to the real owners.
To claim the music owners bits are different from yours, is well
and the "could be useful" by definition equates to "value" and economic losses.
"Fair Use" is not stealing, for all the reasons
you state. However, the definition of "stealing" as defined in the english language as used in the US is much broader that you might try to ignore. Please checkout www.webster.com and what you will find is:
"1 : to take the property of another wrongfully and especially as an habitual or regular practice"
"1 a : to take or appropriate without right or leave and with intent to keep or make use of wrongfully"
"1 d : to appropriate to oneself or beyond one's proper share"
so, the courts have already ruled that P2P systems violate copyright by unauthorized distribution and the those downloading are willfully in violation since they have absolutely no "fair use" claim.
In the definition of stealing, the community involved in illegal distribution of copyrighted works, is in clear conformance with all three of the definitions above, so "stealing" is absolutely the correct term for this act.
First your comment "When you steal property, you deprive someone of it.", which is far to narrow a definition of "stealing". If you go to www.websters.com and look up "stealing" you will also find:
"1 : to take the property of another wrongfully and especially as an habitual or regular practice"
"1 a : to take or appropriate without right or leave and with intent to keep or make use of wrongfully"
"1 d : to appropriate to oneself or beyond one's proper share"
which by US law includes unauthorized duplication and distribution of non-tangable intelectual property of obvious value,especially where copyrighted.
Your second comment "When you steal something you otherwise would have paid for, you deprive someone of the revenue they would get from the sale. When you make a copy of some digital bits that you only are willing to copy precisely because it costs you next to nothing, and honestly (there's the rub) would rather do without if it cost you more, you do not deprive anyone of anything." is also false.
Take the example of someone hacking your bank account and transfering is entire value out of your account. If that person would never choose to work a normal wage paying job, then their transfer of your assets (not reall cash) out of your bank account by your definition is not really stealing since they would never work to earn it in the first place. If such a thief, empty's your bank account I strongly suspect you are going to be crying foul to the bank, the police, and the community for the theft. I call it "stealing" by the above definition. You can rationalize it however you might to ignore the obvious theft as already ruled by US Courts.
You might want to go to www.webster.com and check the definition which is not nearly as limited as you have mistakenly asserted:
"No, they are NOT. Stealing means depriving the original owner of the use of that material."
From Websters:
"1 : to take the property of another wrongfully and especially as an habitual or regular practice"
"1 a : to take or appropriate without right or leave and with intent to keep or make use of wrongfully"
"1 d : to appropriate to oneself or beyond one's proper share"
which by US law includes unauthorized duplication and distribution of non-tangable intelectual property of obvious value, especially where copyrighted.
So, in the United States, speaking in english, the description of illegal (already ruled so by the courts) P2P sharing of unauthorized copyrighted works as "stealing" is absolutely correct.
I said "largely the same" which implies a large intersection, not equivalence. I'm glad you are proud to be outside that group.
Copyright laws don't talk about selling, they do talk about distribution and unauthrized replication that is beyound "fair use", which the P2P community is in violation of.
There are lots of people that don't think money is worth the normal "work" effort it takes to "purchase" it ... we call them robbers, thieves, scum. What's the difference with the property is something other than money?