I think everyone should read this comment at least twice. My father was diagnosed schizophrenic many times. He had a partial lobotomy, was treated with lithium and several of the more recent treatments. (Be aware he died twenty years ago, so "more recent" should be taken with a grain of salt.) The last few years of his life he gave up on meds and he simply became a perpetual drunk. I think his life was no worse (and no more a problem for society) than when he was a "good" patient on meds.
I wish he had had a decent psychiatrist or someone like Jynx to help him work out his problems. When he talked with me, I really could not help -- I did not understand what was going on (I really still don't), but I think I saw him using some of the same techniques -- maybe he would have been more sucessful with encouragement.
I would also add a comment about trust. My father believed far too much of what people told him (doctors, his children, anyone he met), and I think this made him an easy victim. Seeing how he was manipulated and used, I've become a far more cynical person than I was before. And far less appreciative of the "good" people do for others....
Simple chemistry -- so a gallon of gasoline is as good as 5 pounds of beef at keeping you alive? Both have about the same number of calories and approximately the same chemical breakdown.
The important thing about polution is not about the elemental mix of the outputs (which I agree is the elemental mix of the inputs) -- it is the actual chemical compounds and even the physical form of the compounds that are left in the air.
There is hardly any nitrogen in most diesel and gasoline fuels, but nitrogen compounds are a major polution problem many places (made from the nitrogen in the air, which is not normally considered a polutant).
This is a very good point -- I get obviously better accelleration with my 2001 Prius than any comparable size compact car, and I get 10 mpg better gas mileage and comfort that is a lot closer to a Camry than a compact. It did cost more, but not by much if I compare it to another car similarly fitted out.
I think that is another point rarely mentioned in these comparisons. A subcompact that gets 35 mpg is a lot less comfortable and cannot get anything like the accelleration that I get from my Prius. And to get within 10 mpg if the Prius gas mileage I have to really sacrifice on the comfort front (further, my son is a mpg-nut and consistently gets 52 mpg from the Prius, but I'm more one of the "just drive it crowd" and get a piddling 45 mpg in my real world -- still not too bad?).
And I think I get closer to the EPA's 45 highway/55 city mileage (or whatever they claimed) out of it than the average Highlander owner gets to its 30 highway/22 city mileage....
I have to agree with you but I still think he is barking up the wrong tree as far as security goes, though.
A robust data repository that can be rebuilt on an uncorrupted machine from a known good starting point should be his goal, not an unbreakable system with perfect security. Even then, the change log records could have been modified. So there is no perfect solution.
My main irritation is that the messages I have read so far seem to imply that Stallman has assigned a scapegoat, and is really not interested in addressing the problem.
On the other hand, if the problem is that Savannah does not support recovery mechanisms (Microsoft's Sourcesafe is a good example of a code archive that fails miserably in this area), then that should be the reason given for the changeover, not that the system is "insecure" somehow.
Starting with 120 or so channels, you throw out the ones that overlap with nearby markets, that leaves you with 30-40 most places. Then you exclude the ones that are in adjacent channels to in-market stations and you wind up with 20 or so stations. ClearChannel broadcasts on, say 5-7 of these, they "own" another 3-6.
That leaves at least 7. Still not impossible to enter the market, but this is certainly an oligopoly (especially if all those 7, or most of them, are owned by buddies of the ClearChannel folks).
It does shed a lot of light. Stallman seems to have changed sides of this debate, No?
At the time of the quote, I suspect he would have told you to keep change logs and backups so a hacker doesn't have the opportunity to wipe out your data. [Much like Linus Torvalds backup strategy -- release the source to 1000 mirror sites and you don't really need to keep a backup copy....]
Now he seems to agree with the sysadmins -- security mechanisms, secure software, unbreakable passwords, whatever, of one form or another are how to best keep data secure?
Frankly 4 months is way too long for the site to be "not completely functional" and it can't help but make you doubt the quality of the administration of the site if there weren't sufficient provisions in place for this eventuality. Any website is a target so any webadmin should have a plan in place.
But from the discussions on savannah-hackers (?) and here, I would conclude that the administrators were the ones responsible for the the delay in bringing the site back up. They are also responsible for the switch to Gforge.
Do we have any reason to expect that the new system is necessarily any more reliable or secure than the original one?
But finally, I'm most disappointed in that there seems to be nothing like the NTSB investigations to identify why the problem remained unknown for so long. For example, could checkins and backups be improved to minimize or even eliminate the problem in the future, since it is unlikely that even Gfourge is perfectly secure? Perhaps the analysis is just not public, however.
I have had a lot of trouble doing this myself, but I do agree it is the best of both worlds. As I get more experience doing it, I'm sure I will become more competent and it will become only marginally more difficult that simply grabbing a binary RPM and installing it.
Actually this political football is hardly a program at all. The quick summary ($500+ billion for an trip to Mars with all the preparation, rounded up to $800 billion for inflation, then adjust to $1 trillion so it's easier to say) is a pretty accurate rendition of the media story targeted by the article as I read it.
Of course the author of the article blew it too, when he said $1 trillion is 60% more than $800 billion.... Is that because of the silly 1 trillion = 1280x1280x1280 arithmetic thingie? Or because he was doing the same thing he is criticising (talk about inflation so we think he is considering it, then, without saying he doesn't believe in inflation, just discard that adjustment and point out that $1 trillion is 60% more than the original (low) estimate to put an unmanned probe on Mars before 2019). BTW, we did that, ahead of schedule, and under budget, I think.
Debunking is not a word I would have used for that article, though. Rant might be more accurate.
Um, how do the Democrats kill it? Take over Congress and the Presidency? That might work, but then if they actually decide NASA should spend money on the space program, they could make everything happen faster.
Don't bet any politician, especially Bush, has signed off on anything until the money actually goes where you think it went....
I think this is a very good example of someone pulling a story out of his a**. There are 6 insures listed for emergency room malpractice insurance in Nevada. That's one less than the number listed for Texas with a set of "written by the insurance industry itself" limits on lawsuits. And California, with one of the better insurance company oversight systems also has 6. Many states only have two or so providers. So what have the "reforms" got to do with anything. See Malpractice Insurance Providers by State. The company named (American Physicians Insurance) is listed there as only doing business in Texas. So Dr. Gondy is going to have to work in Texas, or so it seems.... And of course Texas has the highest homeowner and automobile insurance rates in the country, so she had better not plan on living in the state or buying a car there....
Another view is presented in Malpractice Rates Whitepaper. It appears that malpractice caps in California did not prevent them from having the highest insurance rates in the country (a 450% increase in the 13 years following their tort "reform"). That lead directly to Proposition 103, controlling the insurance companies directly and imposing a 20% rate reduction. Notice those insurance companies keep doing business in California because they MAKE MONEY doing so.
By the way, this directly contradicts your assertion that "the rate of growth in protected states is lower than that of unprotected states." But it does not keep the ACEP web site from asserting that MICRA has "lowered" rates 6% since 1988, forgetting to mention that from the passage of MICRA in 1975 to California's forced rollback of insurance rates in 1988, rates went up 450%. See the Q & A titled Was MICRA effective?.
And note that in Connecticut a "physician owned" insurance provider is jacking rates up 30% after a 20% increase in the same period that California rates dropped 6%: American Academy of Family Physicians "challenges".
Finally, I can't find any information anywhere about what rates really are other than urban legends about $300,000/year premiums. Doesn't that strike you as interesting, too?
19 tons of mostly silicon carbide does take up a lot of space, though.
The whole safety feature is based on having only a small amount of uranium in a lot of moderator (silicon carbide). I would not even be suprised if the pebbles float (haven't checked that though).
My main issue is who is going to pay for all this? Nuclear power has always boiled down to federal subsidies or it just doesn't make sense economically. If it were a lot better in some way or another, I'd go socialist, but for now I'm in favor of letting it compete economically (it can't, especially if the current subsidies are repealed -- free insurance, etc. -- I think, even though few reactors have been built recently, the subsidies are still in place in the US).
OK, pretty bad = 1 death in 500,000 (coal waste products),
really bad = 1 death in 2 (chernobl ground water a decade later)
I'd rather have the pretty bad stuff, because I don't trust anyone to hold that really bad stuff in a hole in the ground for 80-500 years. Even the best homeowners association can't be expected to protect your property values that long.
It's odd that everyone here is just spouting opinion (well, maybe not so odd, this is/.).
Some really dangerous byproducts of nuclear reactors are not particularly dangerous if you keep them in an asbestos cigarette box. Even skin can protect you from much of the damage. But human bodies are really good at filtering them out of water, air, food, etc, if they get released into the wild. And these insignificant levels of radiation are good at killing you over a 2-3 year period.
But lets discount that. We don't need to worry if only our children and grandchildren are sterile -- it won't hurt their grandchildren, will it?
I suspect the reason is the same as his first one. Hauling all the tools necessary for a self sufficient nuclear power system would probably cost a lot more than anyone is willing to spend.
Of course if you get George to foot the bill and you still get to keep the profits, it might work out....
I think it scales as well as any other tool. In years past I have use ballots that were easily 10 pages long (probably 30 constitutional amendments and 50 or more races, mostly unopposed). And the only buffer overflow exploit I've seen with paper involved computer output, not input....
Of course if the phrase "fair and balanced" is descriptive, it is very difficult to trademark, and if it is accurate, then it is not protectable. So to challenge Fox most easily,/. would only have to have to prove that Fox news can accurately be described as fair and balanced. Or that/. could be....
Oh well, perhaps there are other defenses,....
Remember Ferroelectric memory?
on
MRAM in 2004?
·
· Score: 3, Insightful
Does this remind anyone but me of the ferroelectric memory cells of about a decade ago?
Smaller than DRAM cells, faster than SRAM and nonvolatile as well. They did actually make it out into the real world, several devices made today include a dozen or so F-RAM cells, but they certainly did not take over the world.
One thing that does shout "vaporware" to me is that the articles I can find are all really sparse on details.
Also, how compatible is this technology with common (or esoteric, for that matter) silicon technology? If it's not, can we use the same technology to build processors, etc.?
How soon do we actually get to see a 256 MBit MRAM device? How much will it cost in 2005? The answer to those questions will tell me a lot about whether this is enough to make people show interest in Motorola's stock again....
Honorable Mark may have a really good point there (I don't have a license or GIF on my web site, but I bet I'm in the minority there, one way or the other....).
If an officer of Microsoft Corporation was authorized by the board to distribute the souce code to Windows and was acting in his capacity as an officer of the company when he did in fact hack MSDN and put GPL licenses and the Windows souce code up there, I would say that in fact it would be free. I think any US court would agree, too. It would be odd for him to do it that way instead of just instructing the webmaster to add the page in his normal work schedule, but perhaps that's because Bill wanted to REALLY SUPRISE everyone! Or he just wanted to prove he could still hack MSDN.
And as far as the viral nature of GPL: it is a viral license to that extent and it was intended to be so. It does not allow any code you have explicitly released to be withdrawn. I don't believe I have signed any contracts that gave me rights like that in the last 40 years. And I don't think any corporate lawyer should let his clients sign a contract to use software that did not include such a prohibition. Do you think Walmart would install cash registers with the posibility that they would be liable for copyright violation any time SCO, or whoever, wanted to change the license?
I think the real issue is that the rendering errors never appear when running the benchmark in its normal mode.
I used to write graphics drivers (and BIOSes) for one of the very big names in the video card business back then, and I can see how this can happen without any intentional "cheating". In the development cycle we would test with benchmarks and demo programs since those programs give you repeatable frames and you can actually see what was happening and you could see that you fixed a bug.
On the other hand, we had a large number of testers that just played games. They would see issues that did not show up on our canned scripts. Those bugs were a lot harder to find and fix. And to go along with that we actually wrote programs that pushed the driver into corner cases that were intended to create failures. This work was not usually approved of by management because it made our products less attractive, not more attractive, to reviewers.
So you can say that eliminating those corner case tests and the random game tests (reduces cost, reduces time to market) is good management, not cheating, but the effect is really the same, a lower quality product.
On the other hand, the fact that real world tests do not show improvement and the 3Dmark does is a pretty clear indication that someone knew some kind of buggy behavior (cheating is just a more emotionally charged word for the same thing) was going on (I presonally would be very suspicious if I saw a 30% improvement on one test and none on a second -- it would be time to investigate what the effective difference between the tests was).
So I'd say nVidia's engineering team may not have known they were "cheating" but I'm sure they were suspicious and ought to have put more effort into identifying the reason for the differences. Or they were told to not ask and not tell (to steal a phrase from the military lexicon).
I have to agree. The result over time has been what several folks have called a move toward PC content and comarketing (movies, TV, toys, games, web sites, you name it!) to summarize how I understand the comments.
But I think it is not so much "politically correct", otherwise we would see only what lines up with Murdock's rather reactionary politics. It is more a tendency to maximize profit from minimal cost -- perhaps we should call it "capitalistically correct"? One should never offend a potential customer unless he gets a larger return from a sure-fire customer by doing so.
However, I really believe our limitations are more based on the fact that there are fewer and fewer decision makers left. [Not really trying to set off a flame war, but] Anime is no better or worse than other cartoon genres, but it was developed in a relatively small non-US market. The creators were not necessarily more willing to take risks (though they might have been), but they interpreted the risks and rewards differently because they were different people. The style was a product of minds not integrated into our large media conglomerates. Now that the world is more cohesive and orderly, I doubt we will see another creative explosion like that for some time.
Music, like cartoons, is also filled with stylization, minimalism and cultural idioms. And it seems to show the very same same shift to mass production over the last 100 years that cartoons demonstrated over the last 30 years. A quote from the Collier's Encycolpedia describes the shift of "the music industry" from a business oriented toward [competent] performers to one oriented toward [mass] consumers. Saturday morning cartoons were never a community or performer driven medium, but they were, I think, a far less lucrative market, so they could accomodate a more creative class of products.
I suspect other forms of communication are or have done the same and whatever is created in the future will follow the same pattern. After Motzart, Bill Gates and Walt Disney happen, we seal up the stargate to make sure it never happens again -- to mix all the metaphors I can in one sentence.
Re:One major error in the article...
on
The Law and P2P
·
· Score: 1
And one more error. At least when I signed up last year, you could sign up month-to-month as well (it's just a bit more expensive, like with an ISP). It also carries a lot of independent, psuedo-non-label music (to be listed you have to have a label, but serveral bands I and my son like are self published and the label is pretty near the same as the musician....
I would like to point out that Universal already is doing a variant of this. emusic is a subsidiary of Vivendi/Uinversal and has a fair number of subscribers -- they just charge so much a month for the downloads and the files are pretty much stock MP3s. I liked the service when I signed up for a free trial, and if I listened to more music, I'd certainly subscribe: they have a lot of the older stuff I really do like and they don't charge an absurd amount for it.
BTW, you actually own the music you download -- the link above makes it clear that after you terminate your subscription your rights to the music don't end.
Slashdot Mirror
I think everyone should read this comment at least twice. My father was diagnosed schizophrenic many times. He had a partial lobotomy, was treated with lithium and several of the more recent treatments. (Be aware he died twenty years ago, so "more recent" should be taken with a grain of salt.) The last few years of his life he gave up on meds and he simply became a perpetual drunk. I think his life was no worse (and no more a problem for society) than when he was a "good" patient on meds.
I wish he had had a decent psychiatrist or someone like Jynx to help him work out his problems. When he talked with me, I really could not help -- I did not understand what was going on (I really still don't), but I think I saw him using some of the same techniques -- maybe he would have been more sucessful with encouragement.
I would also add a comment about trust. My father believed far too much of what people told him (doctors, his children, anyone he met), and I think this made him an easy victim. Seeing how he was manipulated and used, I've become a far more cynical person than I was before. And far less appreciative of the "good" people do for others....
Simple chemistry -- so a gallon of gasoline is as good as 5 pounds of beef at keeping you alive? Both have about the same number of calories and approximately the same chemical breakdown.
The important thing about polution is not about the elemental mix of the outputs (which I agree is the elemental mix of the inputs) -- it is the actual chemical compounds and even the physical form of the compounds that are left in the air.
There is hardly any nitrogen in most diesel and gasoline fuels, but nitrogen compounds are a major polution problem many places (made from the nitrogen in the air, which is not normally considered a polutant).
This is a very good point -- I get obviously better accelleration with my 2001 Prius than any comparable size compact car, and I get 10 mpg better gas mileage and comfort that is a lot closer to a Camry than a compact. It did cost more, but not by much if I compare it to another car similarly fitted out.
I think that is another point rarely mentioned in these comparisons. A subcompact that gets 35 mpg is a lot less comfortable and cannot get anything like the accelleration that I get from my Prius. And to get within 10 mpg if the Prius gas mileage I have to really sacrifice on the comfort front (further, my son is a mpg-nut and consistently gets 52 mpg from the Prius, but I'm more one of the "just drive it crowd" and get a piddling 45 mpg in my real world -- still not too bad?).
And I think I get closer to the EPA's 45 highway/55 city mileage (or whatever they claimed) out of it than the average Highlander owner gets to its 30 highway/22 city mileage....
I have to agree with you but I still think he is barking up the wrong tree as far as security goes, though.
A robust data repository that can be rebuilt on an uncorrupted machine from a known good starting point should be his goal, not an unbreakable system with perfect security. Even then, the change log records could have been modified. So there is no perfect solution.
My main irritation is that the messages I have read so far seem to imply that Stallman has assigned a scapegoat, and is really not interested in addressing the problem.
On the other hand, if the problem is that Savannah does not support recovery mechanisms (Microsoft's Sourcesafe is a good example of a code archive that fails miserably in this area), then that should be the reason given for the changeover, not that the system is "insecure" somehow.
I disagree!
Starting with 120 or so channels, you throw out the ones that overlap with nearby markets, that leaves you with 30-40 most places. Then you exclude the ones that are in adjacent channels to in-market stations and you wind up with 20 or so stations. ClearChannel broadcasts on, say 5-7 of these, they "own" another 3-6.
That leaves at least 7. Still not impossible to enter the market, but this is certainly an oligopoly (especially if all those 7, or most of them, are owned by buddies of the ClearChannel folks).
It does shed a lot of light. Stallman seems to have changed sides of this debate, No?
At the time of the quote, I suspect he would have told you to keep change logs and backups so a hacker doesn't have the opportunity to wipe out your data. [Much like Linus Torvalds backup strategy -- release the source to 1000 mirror sites and you don't really need to keep a backup copy....]
Now he seems to agree with the sysadmins -- security mechanisms, secure software, unbreakable passwords, whatever, of one form or another are how to best keep data secure?
But from the discussions on savannah-hackers (?) and here, I would conclude that the administrators were the ones responsible for the the delay in bringing the site back up. They are also responsible for the switch to Gforge.
Do we have any reason to expect that the new system is necessarily any more reliable or secure than the original one?
But finally, I'm most disappointed in that there seems to be nothing like the NTSB investigations to identify why the problem remained unknown for so long. For example, could checkins and backups be improved to minimize or even eliminate the problem in the future, since it is unlikely that even Gfourge is perfectly secure? Perhaps the analysis is just not public, however.
I have had a lot of trouble doing this myself, but I do agree it is the best of both worlds. As I get more experience doing it, I'm sure I will become more competent and it will become only marginally more difficult that simply grabbing a binary RPM and installing it.
Actually this political football is hardly a program at all. The quick summary ($500+ billion for an trip to Mars with all the preparation, rounded up to $800 billion for inflation, then adjust to $1 trillion so it's easier to say) is a pretty accurate rendition of the media story targeted by the article as I read it.
Of course the author of the article blew it too, when he said $1 trillion is 60% more than $800 billion.... Is that because of the silly 1 trillion = 1280x1280x1280 arithmetic thingie? Or because he was doing the same thing he is criticising (talk about inflation so we think he is considering it, then, without saying he doesn't believe in inflation, just discard that adjustment and point out that $1 trillion is 60% more than the original (low) estimate to put an unmanned probe on Mars before 2019). BTW, we did that, ahead of schedule, and under budget, I think.
Debunking is not a word I would have used for that article, though. Rant might be more accurate.
Um, how do the Democrats kill it? Take over Congress and the Presidency? That might work, but then if they actually decide NASA should spend money on the space program, they could make everything happen faster.
Don't bet any politician, especially Bush, has signed off on anything until the money actually goes where you think it went....
I think this is a very good example of someone pulling a story out of his a**. There are 6 insures listed for emergency room malpractice insurance in Nevada. That's one less than the number listed for Texas with a set of "written by the insurance industry itself" limits on lawsuits.
And California, with one of the better insurance company oversight systems also has 6. Many states only have two or so providers. So what have the "reforms" got to do with anything. See Malpractice Insurance Providers by State. The company named (American Physicians Insurance) is listed there as only doing business in Texas. So Dr. Gondy is going to have to work in Texas, or so it seems.... And of course Texas has the highest homeowner and automobile insurance rates in the country, so she had better not plan on living in the state or buying a car there....
Another view is presented in Malpractice Rates Whitepaper. It appears that malpractice caps in California did not prevent them from having the highest insurance rates in the country (a 450% increase in the 13 years following their tort "reform"). That lead directly to Proposition 103, controlling the insurance companies directly and imposing a 20% rate reduction. Notice those insurance companies keep doing business in California because they MAKE MONEY doing so.
By the way, this directly contradicts your assertion that "the rate of growth in protected states is lower than that of unprotected states." But it does not keep the ACEP web site from asserting that MICRA has "lowered" rates 6% since 1988, forgetting to mention that from the passage of MICRA in 1975 to California's forced rollback of insurance rates in 1988, rates went up 450%. See the Q & A titled Was MICRA effective?.
And note that in Connecticut a "physician owned" insurance provider is jacking rates up 30% after a 20% increase in the same period that California rates dropped 6%: American Academy of Family Physicians "challenges".
Finally, I can't find any information anywhere about what rates really are other than urban legends about $300,000/year premiums. Doesn't that strike you as interesting, too?
19 tons of mostly silicon carbide does take up a lot of space, though.
The whole safety feature is based on having only a small amount of uranium in a lot of moderator (silicon carbide). I would not even be suprised if the pebbles float (haven't checked that though).
My main issue is who is going to pay for all this? Nuclear power has always boiled down to federal subsidies or it just doesn't make sense economically. If it were a lot better in some way or another, I'd go socialist, but for now I'm in favor of letting it compete economically (it can't, especially if the current subsidies are repealed -- free insurance, etc. -- I think, even though few reactors have been built recently, the subsidies are still in place in the US).
OK, pretty bad = 1 death in 500,000 (coal waste products),
really bad = 1 death in 2 (chernobl ground water a decade later)
I'd rather have the pretty bad stuff, because I don't trust anyone to hold that really bad stuff in a hole in the ground for 80-500 years. Even the best homeowners association can't be expected to protect your property values that long.
Ohhh, I did it, didn't I.... Oh well, it's just karma....
It's odd that everyone here is just spouting opinion (well, maybe not so odd, this is /.).
Some really dangerous byproducts of nuclear reactors are not particularly dangerous if you keep them in an asbestos cigarette box. Even skin can protect you from much of the damage. But human bodies are really good at filtering them out of water, air, food, etc, if they get released into the wild. And these insignificant levels of radiation are good at killing you over a 2-3 year period.
But lets discount that. We don't need to worry if only our children and grandchildren are sterile -- it won't hurt their grandchildren, will it?
I suspect the reason is the same as his first one. Hauling all the tools necessary for a self sufficient nuclear power system would probably cost a lot more than anyone is willing to spend.
Of course if you get George to foot the bill and you still get to keep the profits, it might work out....
I think it scales as well as any other tool. In years past I have use ballots that were easily 10 pages long (probably 30 constitutional amendments and 50 or more races, mostly unopposed). And the only buffer overflow exploit I've seen with paper involved computer output, not input....
Of course if the phrase "fair and balanced" is descriptive, it is very difficult to trademark, and if it is accurate, then it is not protectable. So to challenge Fox most easily, /. would only have to have to prove that Fox news can accurately be described as fair and balanced. Or that /. could be. ...
Oh well, perhaps there are other defenses,....
Does this remind anyone but me of the ferroelectric memory cells of about a decade ago?
Smaller than DRAM cells, faster than SRAM and nonvolatile as well. They did actually make it out into the real world, several devices made today include a dozen or so F-RAM cells, but they certainly did not take over the world.
One thing that does shout "vaporware" to me is that the articles I can find are all really sparse on details.
Also, how compatible is this technology with common (or esoteric, for that matter) silicon technology? If it's not, can we use the same technology to build processors, etc.?
How soon do we actually get to see a 256 MBit MRAM device? How much will it cost in 2005? The answer to those questions will tell me a lot about whether this is enough to make people show interest in Motorola's stock again....
Ah so!
....).
Honorable Mark may have a really good point there (I don't have a license or GIF on my web site, but I bet I'm in the minority there, one way or the other
I don't think so.
If an officer of Microsoft Corporation was authorized by the board to distribute the souce code to Windows and was acting in his capacity as an officer of the company when he did in fact hack MSDN and put GPL licenses and the Windows souce code up there, I would say that in fact it would be free. I think any US court would agree, too. It would be odd for him to do it that way instead of just instructing the webmaster to add the page in his normal work schedule, but perhaps that's because Bill wanted to REALLY SUPRISE everyone! Or he just wanted to prove he could still hack MSDN.
And as far as the viral nature of GPL: it is a viral license to that extent and it was intended to be so. It does not allow any code you have explicitly released to be withdrawn. I don't believe I have signed any contracts that gave me rights like that in the last 40 years. And I don't think any corporate lawyer should let his clients sign a contract to use software that did not include such a prohibition. Do you think Walmart would install cash registers with the posibility that they would be liable for copyright violation any time SCO, or whoever, wanted to change the license?
I think the real issue is that the rendering errors never appear when running the benchmark in its normal mode.
I used to write graphics drivers (and BIOSes) for one of the very big names in the video card business back then, and I can see how this can happen without any intentional "cheating". In the development cycle we would test with benchmarks and demo programs since those programs give you repeatable frames and you can actually see what was happening and you could see that you fixed a bug.
On the other hand, we had a large number of testers that just played games. They would see issues that did not show up on our canned scripts. Those bugs were a lot harder to find and fix. And to go along with that we actually wrote programs that pushed the driver into corner cases that were intended to create failures. This work was not usually approved of by management because it made our products less attractive, not more attractive, to reviewers.
So you can say that eliminating those corner case tests and the random game tests (reduces cost, reduces time to market) is good management, not cheating, but the effect is really the same, a lower quality product.
On the other hand, the fact that real world tests do not show improvement and the 3Dmark does is a pretty clear indication that someone knew some kind of buggy behavior (cheating is just a more emotionally charged word for the same thing) was going on (I presonally would be very suspicious if I saw a 30% improvement on one test and none on a second -- it would be time to investigate what the effective difference between the tests was).
So I'd say nVidia's engineering team may not have known they were "cheating" but I'm sure they were suspicious and ought to have put more effort into identifying the reason for the differences. Or they were told to not ask and not tell (to steal a phrase from the military lexicon).
I have to agree. The result over time has been what several folks have called a move toward PC content and comarketing (movies, TV, toys, games, web sites, you name it!) to summarize how I understand the comments.
But I think it is not so much "politically correct", otherwise we would see only what lines up with Murdock's rather reactionary politics. It is more a tendency to maximize profit from minimal cost -- perhaps we should call it "capitalistically correct"? One should never offend a potential customer unless he gets a larger return from a sure-fire customer by doing so.
However, I really believe our limitations are more based on the fact that there are fewer and fewer decision makers left. [Not really trying to set off a flame war, but] Anime is no better or worse than other cartoon genres, but it was developed in a relatively small non-US market. The creators were not necessarily more willing to take risks (though they might have been), but they interpreted the risks and rewards differently because they were different people. The style was a product of minds not integrated into our large media conglomerates. Now that the world is more cohesive and orderly, I doubt we will see another creative explosion like that for some time.
Music, like cartoons, is also filled with stylization, minimalism and cultural idioms. And it seems to show the very same same shift to mass production over the last 100 years that cartoons demonstrated over the last 30 years. A quote from the Collier's Encycolpedia describes the shift of "the music industry" from a business oriented toward [competent] performers to one oriented toward [mass] consumers. Saturday morning cartoons were never a community or performer driven medium, but they were, I think, a far less lucrative market, so they could accomodate a more creative class of products.
I suspect other forms of communication are or have done the same and whatever is created in the future will follow the same pattern. After Motzart, Bill Gates and Walt Disney happen, we seal up the stargate to make sure it never happens again -- to mix all the metaphors I can in one sentence.
And one more error. At least when I signed up last year, you could sign up month-to-month as well (it's just a bit more expensive, like with an ISP). It also carries a lot of independent, psuedo-non-label music (to be listed you have to have a label, but serveral bands I and my son like are self published and the label is pretty near the same as the musician....
I would like to point out that Universal already is doing a variant of this. emusic is a subsidiary of Vivendi/Uinversal and has a fair number of subscribers -- they just charge so much a month for the downloads and the files are pretty much stock MP3s. I liked the service when I signed up for a free trial, and if I listened to more music, I'd certainly subscribe: they have a lot of the older stuff I really do like and they don't charge an absurd amount for it.
BTW, you actually own the music you download -- the link above makes it clear that after you terminate your subscription your rights to the music don't end.