I personally don't really see the problem. You get charged based on how much electricity you use.
Yes, but they do not charge you a monthly fee for the privilege of getting further charged based on usage. Charge for the bandwidth used or charge an access fee. Not both.
The phone company used to do that. Now they're pretty much dumb pipes with flat long-distance fees in reaction to everybody leaving their service for flat-fee Internet-based services. This will eventually happen with ISPs, too, as soon as there is a sufficient disruptive technology. Some new tech will start to compete with those ISPs, and they'll be forced back to flat-rate services.
Heck, I might even go one step further and ask if the original poster or one of the poster's neighbors has a water timer that floods his or her yard at a particular time of day and adds craptons of capacitance on the line or something (because of bad insulation, mice, whatever).
By that standard, an ISP can provide gigabit to your home, but with everyone in town sharing a single T1 and call it gigabit service. That would be a textbook case of fraud. That's not what those numbers mean. They mean that the service from your ISP from your home to an average site should typically be that rate, barring congestion between the backbone and that site.
The speed test sites themselves are almost always very close to fast backbones. Thus, any significant contention on the routes between your ISP and those sites is almost invariably between your house and the backbones, not between the backbone routes and the speed test sites. Such poor performance almost invariably indicates either a local connection problem or an ISP that is massively overselling their upstream capacity. Sure, one speed test site giving a low speed could be a coincidence once in a while, but that's why there are dozens of those sites; if they're all showing 100 kilobit service, then I don't care how fast your pipe to the ISP is, you aren't getting 3 megabit service by any reasonable measure.
Actually, you can telnet or SSH into most routers these days and talk to the head end directly, including performing loopback tests (kind of like a ping, but only as far as your DSLAM). That can tell you a lot, including noise margins (both from the perspective of your device and what's being reported by the head end), attenuation, which tones the modem is actually using, etc.
I don't think it was the 2.7 Mbps that was the concern, but rather the "down to 0.1 or 0.2 Mbps in the evening". That's either the line retraining to the lowest possible fallback speed (bleed-through from somebody else's line, perhaps) or a massively over-committed upstream pipe from the DSLAM.
Had a Zoom once. After about two years, it started rebooting every few minutes. I don't have a Zoom now.
Step 1: Grab a different DSL router/modem.
Step 2: If that didn't fix it, call your service provider every single time it goes south until they're so sick of hearing from you that they send out a tech at the right time of night.
Step 3: If that didn't fix it, talk to your neighbors about sharing a trunk line.
The idea of messing with clunky DVDs, watching all the ads on them, sitting through a movie which may or may not be worth it, then trying to store/dispose of the DVD afterwards is a lot more than I'm willing to put up with for $5. That's worth $1 tops, less if the adverts are annoying. Now, a downloadable file of a movie I know I will like, without any ads or hassle of physical media, and the ability to play on any of my devices, that would be worth $5 or even $10.
See, from my perspective, the opposite is true. Short of a house fire, odds are 99% of those DVDs are still going to be playable in twenty years. For digital downloads, odds are that 0% will still be playable because the DRM servers they depend on will no longer exist.
And even if you have a DRM-free digital download, that still means that (assuming comparable quality and bit rate) just to download the movie portions of the Lord of the Rings extended edition Blu-Ray discs, without the extras, I would go through an eighth of a terabyte of bandwidth—half a month's data cap on many companies' services. And with my 3 Mbps DSL (the fastest I can get without selling my soul to AT&T's capped service), it would take almost four days of continuous downloading to get it.
After that, I would have to find a place to store and back up an eighth of a terabyte worth of data just for that one movie alone. With a digital download, it becomes my responsibility to keep backups of all those terabytes of data, whereas with physical media, its value is covered by by homeowner's insurance against loss. With hard drive reliability being as bad as it is, this seems like an absolutely insane proposition to me.
Sure, I could use less bandwidth if I settled for a lower quality copy, but that isn't really a fair comparison. You're comparing a high quality piece of physical media with a low-quality download. If the download were proportionally much cheaper to make up for the loss of extras, the fact that I'm paying for the bandwidth and storage directly, etc., then downloads might make sense. Until I can buy an average digital download for under a buck, it doesn't appeal to me. At all.
If they had been eating that all along, perhaps. If they started today, either:
1. They would not have enough calories to satisfy the fat cells, and their metabolic rates would plummet, causing them to be lethargic and unhealthy, but they probably wouldn't lose much weight.
2. They would consume twice as much as you do, and wouldn't lose any weight.
"In many [western] countries, peoples' diet changed substantially in the second half of the twentieth century, generally with increases in consumption of meat, dairy products, vegetable oils, fruit juice, and alcoholic beverages, and decreases in consumption of starchy staple foods such as bread, potatoes, rice, and maize flour.
If they're talking about the U.S., that last part is exactly backwards. Our consumption of starchy staple foods has gone way, way up. Almost nobody eats a burger without fried potatoes, and the fast food is loaded with bread. The chicken is breaded, and half the time, you have a bun on top of that. And pizza is mostly bread. And so on. All of these are things that are rising in popularity. I have a hard time believing that bread consumption could possibly have gone down unless most people ate two meals a day of nothing but bread.:-)
No, based on what I've seen, what has changed most is:
A reduction of portion control. This means: 1. A gradual decline of buffet-style and cafeteria-style eating and a rise in fast food restaurants that serve the same, fixed-size portion to everyone. 2. Buffet pricing that makes people feel like they aren't getting their money's worth unless they consume more food than they would ordinarily eat.
An increase in fried and breaded foods. These tend to have more calories than foods that are baked or grilled.
An increase in the consumption of empty starches, such as rice (served at almost every Asian restaurant) and potatoes (mostly in the form of French fries).
An increase in overall stress levels and a decrease in sleep, both of which increase cortisol production, which is well understood to increase obesity, particularly around the waistline.
A huge decrease in the variety of vegetables available. Not counting salads or french fries, most fast food chains don't offer any vegetables at all, and if they do, it is usually limited to beans.
All of these things contribute to the problem. It isn't just one thing. It is an epidemic of really, really lousy options provided by our food industry. And in an era where many people simply do not have time to cook or sit down for an extended dinner at a nicer restaurant, there are painfully few healthy alternatives.
Any consumer who cares so little about it that he is willing to wait five years to get something for free is just as likely to download it illegally before that.
You're joking, right? I routinely buy movies that are way more than five years old out of the $5 movie bin. When a new movie comes out, I never buy it at $20 for the DVD or $30 for the Blu-Ray. I wait a few years for it to be below $10, then buy it. So people like me would basically no longer be contributing to the profits of those content creators because there would be no reason for the companies selling me those DVDs (e.g. Fry's) to pay them (the studios).
Similarly, a fair percentage of the books I have bought in my life have been older than five years by the time I bought them, not because I'm being cheap, but because a major publisher agreeing to publish more than a couple of books is a sign of a series that is probably good and will probably continue to produce new works. It isn't a perfect model, but it's a good first approximation.
For example, I recently bought Jim Butcher's Codex Alera series and about a year earlier, his Dresden Files series. Under a five-year copyright term, more than half of each series would have been out of copyright by the time I learned that the series existed, or about eight books out of 16. (Admittedly, this is in large part because a paperback copy of Storm Front sat on my shelves for years before I had time to read it, but that's neither here nor there.) It averages out to being about a quarter of all the books I bought in the past few months, but more than half of the books I've actually had time to read. Assuming that my experience is typical, which I suspect it is, that means that a five-year term would put a major dent in authors' ability to make a living, and it's hard enough for authors to make a living by writing even under the current system.
Now that book 1 is free, it provides great free publicity for book 5.
Except that most new authors' first books, assuming they write them in their spare time while doing another job, takes at least a couple of years to write, and I've been working on my trilogy (which had to be written in its entirety before the first book was published because of the complex relationship between the books) since 2001. I'm not done yet.
What this means is that the average new author will spend years of his or her life writing with no possibility of making any money. I don't care how much free publicity you get from it. You still won't have an income stream, and without that income stream, you'll never be able to free yourself from your actual job in order to spend the amount of time it takes to write at a reasonable speed, unless you just get exceptionally lucky and a professional publishing house decides to take a chance on a proposal before the book is written. (They won't, though, because there will be no incentive to pay authors when there is so much recently written public domain content that they can publish, knowing that the odds of another publisher randomly picking the same work is minimal.)
Also, no author would ever get paid for his work when big movie studios decide they want to make a movie of a well-known author's book. The book would be out of copyright, so they would have no obligation to pay for the right to create a derivative work. This means that authors would have to make all of their money from book sales (difficult even with long copyrights), donations/charity, or from doing something else to make an actual living. In short, a five-year-IP model is tantamount to indentured servitude for the actual content creators because anyone with an actual job loses any real opportunity to get out from under the thumb of abusive corporate contracts.
What this model fails to consider is that there is a huge difference between the ability of an individual and a corporation to monetize a work, and an even bigger difference between the speed at which different types of media can usefully be monetized. Works that take years to create tend to take years to pay back their creators. Five years is simply absurd unless your goal is to encourage everyone to stop creating works that are more than about a page or two long. If you want to live in a world where no movies exist that are longer than YouTube videos, and no books exist because they have been replaced by blog posts, feel free to create your own pseudo-utopian copyright-free zone, but count me out.
The original 28 years from the date of first publication (as opposed to five years from its creation) was a good length for copyright. It ensured that after about a generation, material fell out of copyright so that new generations of people could create interesting derivative works. It ensured that things fell out of copyright before people forgot about them. However, it also ensured that most books that might become movies would do so long before the copyright expired (with the obvious exception of Lord of the Rings, which took twice that long, but which was made long enough after the author's death to make the issue largely moot, IMO). It ensured that people who spent years working on something could realistically break even. And so on. But five years? Way, way, way too short.
Of course you're trolling, but what does Rupert Murdoch's gutter-level right-wing editorial service called Fox News have to do with a legitimate news operation?
Uh... any hardware codec chip (which all digital video cameras use in one form or another) has a maximum data rate even in VBR mode, and the video output of the camera, by definition, will never exceed that rate. As long as your storage exceeds that limit, you won't run out of storage. Similarly, as long as your network bandwidth exceeds that limit with a reasonable amount of room to spare, you shouldn't have problems there, either....
That makes the plane safer, but because, by definition, you are not on the plane, you are still no safer. Well, I suppose for the handful of people who happen to be in the ground track of the debris, you are slightly safer, but this is more than offset by the added risk to the people standing in line with the lunatic carrying a bomb.
And no respectable optometrist EVER recommends holding the book at that distance. Hold it at standard reading distance instead (say, 24 inches or higher) and run that calculation.
I measured 24 inches from my eyes, and in order to do so, I had to old the measuring tape up with my fingertips, with my arms fully extended. No optometrist in his right mind would recommend such a large reading distance. Maybe for TV viewing, but not reading.
Typical reading distance is 14-16 inches. So although 12 inches might be slightly close, your suggested reading distance is so far beyond normal that only an orangutan can achieve it.
Fair enough. So instead of dying, you'd probably just get really, really sick and die of cancer a few years later. The point was that critical mass is only meaningful when no pressure is applied, and that much smaller amounts can become dangerous under the right circumstances.
To put it in context, 3 pounds of Uranium is still more than the portion of the Uranium that actually contributed to the explosive yield of Little Boy (Glasstone and Dolan, Effects, pp. 12–13, as cited by Wikipedia). There's a lot of energy there that could theoretically be released under the right circumstances.
This will always be the case where a project has significant corporate backers.
True, but it is a far bigger concern when you have exactly one major corporate backer. When you have five or six companies contributing lots of code, the disadvantages caused by keeping private repositories in sync with the public ones usually outweigh any advantages any single company might get from keeping significant bits proprietary.
GPL just closes off the one possible line of attack, in that the company can't have their cake and eat it too- they can either abandon the open source project (and drop their favourite software at the same time), or they can keep their favourite software (and inevitably keep the open source project too).
Not true. If the GPL project requires copyright assignment, then the new owners of that copyright can do whatever they please, up to and including creating a closed-source fork. Likewise, if a single company owns the copyright on 90% of the code, they can rip out the 10% contributed by outsiders, hire people to rewrite those bits, and create a closed-source fork. The GPL closes off that line of attack only if there are multiple major contributors contributing large portions of the code, and in those circumstances, BSD-licensed software is also largely immune to that line of attack by virtue of the synchronization hassle factor, as I mentioned previously.
Linux vendors are unusual in that they've already thoroughly cast their lot with the GPL, for better or worse. They don't have any alternative short of not being a Linux vendor. Same goes for other companies that make their living by providing support for GPLed software.
So if you imply that I should license my software under a permissive license only to give in to the companies' unjustified fears, in the hope that their involvement will make my software better, then I disagree.
No, I'm just saying that if you want broad adoption, GPLv3 is a dead end. The fears, unjustified or not, are very prevalent, and are unlikely to go away. So you have to choose which is more important: broad adoption or preventing a theoretical abuse that statistically won't ever happen to the vast majority of projects regardless of license.
The situations where companies have created closed-source forks of BSD-licensed projects have almost invariably been cases where there was no set of benevolent core maintainers, e.g. the Apache Foundation. If the open source content is not changing significantly, there is minimal benefit of keeping up with public changes, and thus no real penalty for creating a closed branch. That's why I specifically stipulated that the ideal situation was a BSD license in which the core development was done by a benevolent foundation (and, by extension, in which the software is still being actively maintained and enhanced).
Further, it's worth noting that when companies create closed-source forks, they often do so by hiring the main/sole developer or buying the company that owns the copyrights. When that happens, the GPL provides no additional protection. The GPL provides protection only if a sizable percentage of the code was written by someone who won't accept money in exchange for the right to create a closed-source branch. In practice, this means that the GPL is usually largely ineffectual at preventing the sorts of situations that you describe; the fact that GPLed software falls victim to this less frequently has more to do with corporate aversion to touching the stuff, plus the existence of groups like the FSF that help back some of it, rather than because the license inherently prevents the creation of such a branch.
Finally, FWIW, most of the BSD-licensed technology that Apple uses is still released as open source. There's a giant pile of it on opensource.apple.com. I count 65 published projects for the current Lion release that are in whole or in part licensed under a BSD license.
Creating fusion is not hard. Philo T. Farnsworth (the TV guy) did it eighty years ago. Creating self-sustained fusion that produces more power than it consumes is hard.
It's still probably enough to easily go supercritical and kill you if you compress it sufficiently. It certainly doesn't require a factor of 30; the demon core was only 14 pounds, and killed two people in separate incidents.
Even the final 802.11n in the best case is not that much faster than 100BASE-T, and that requires having no more than a couple of devices on the network, all close to the base station. By its nature, wireless communication is a shared medium, not switched, so as soon as you have two devices, they're competing for bandwidth; MIMO will eventually help with this to some degree, but even that only goes so far. It doesn't take very many devices on a wireless network before the average per-device bandwidth is less than switched 10BASE-T would provide, much less switched 100BASE-T or gigabit.
Slashdot Mirror
Yes, but they do not charge you a monthly fee for the privilege of getting further charged based on usage. Charge for the bandwidth used or charge an access fee. Not both.
The phone company used to do that. Now they're pretty much dumb pipes with flat long-distance fees in reaction to everybody leaving their service for flat-fee Internet-based services. This will eventually happen with ISPs, too, as soon as there is a sufficient disruptive technology. Some new tech will start to compete with those ISPs, and they'll be forced back to flat-rate services.
Heck, I might even go one step further and ask if the original poster or one of the poster's neighbors has a water timer that floods his or her yard at a particular time of day and adds craptons of capacitance on the line or something (because of bad insulation, mice, whatever).
By that standard, an ISP can provide gigabit to your home, but with everyone in town sharing a single T1 and call it gigabit service. That would be a textbook case of fraud. That's not what those numbers mean. They mean that the service from your ISP from your home to an average site should typically be that rate, barring congestion between the backbone and that site.
The speed test sites themselves are almost always very close to fast backbones. Thus, any significant contention on the routes between your ISP and those sites is almost invariably between your house and the backbones, not between the backbone routes and the speed test sites. Such poor performance almost invariably indicates either a local connection problem or an ISP that is massively overselling their upstream capacity. Sure, one speed test site giving a low speed could be a coincidence once in a while, but that's why there are dozens of those sites; if they're all showing 100 kilobit service, then I don't care how fast your pipe to the ISP is, you aren't getting 3 megabit service by any reasonable measure.
Actually, you can telnet or SSH into most routers these days and talk to the head end directly, including performing loopback tests (kind of like a ping, but only as far as your DSLAM). That can tell you a lot, including noise margins (both from the perspective of your device and what's being reported by the head end), attenuation, which tones the modem is actually using, etc.
I don't think it was the 2.7 Mbps that was the concern, but rather the "down to 0.1 or 0.2 Mbps in the evening". That's either the line retraining to the lowest possible fallback speed (bleed-through from somebody else's line, perhaps) or a massively over-committed upstream pipe from the DSLAM.
Had a Zoom once. After about two years, it started rebooting every few minutes. I don't have a Zoom now.
Step 1: Grab a different DSL router/modem.
Step 2: If that didn't fix it, call your service provider every single time it goes south until they're so sick of hearing from you that they send out a tech at the right time of night.
Step 3: If that didn't fix it, talk to your neighbors about sharing a trunk line.
See, from my perspective, the opposite is true. Short of a house fire, odds are 99% of those DVDs are still going to be playable in twenty years. For digital downloads, odds are that 0% will still be playable because the DRM servers they depend on will no longer exist.
And even if you have a DRM-free digital download, that still means that (assuming comparable quality and bit rate) just to download the movie portions of the Lord of the Rings extended edition Blu-Ray discs, without the extras, I would go through an eighth of a terabyte of bandwidth—half a month's data cap on many companies' services. And with my 3 Mbps DSL (the fastest I can get without selling my soul to AT&T's capped service), it would take almost four days of continuous downloading to get it.
After that, I would have to find a place to store and back up an eighth of a terabyte worth of data just for that one movie alone. With a digital download, it becomes my responsibility to keep backups of all those terabytes of data, whereas with physical media, its value is covered by by homeowner's insurance against loss. With hard drive reliability being as bad as it is, this seems like an absolutely insane proposition to me.
Sure, I could use less bandwidth if I settled for a lower quality copy, but that isn't really a fair comparison. You're comparing a high quality piece of physical media with a low-quality download. If the download were proportionally much cheaper to make up for the loss of extras, the fact that I'm paying for the bandwidth and storage directly, etc., then downloads might make sense. Until I can buy an average digital download for under a buck, it doesn't appeal to me. At all.
And even the fatty portion of meats is still digested more slowly than carbs.
If they had been eating that all along, perhaps. If they started today, either:
1. They would not have enough calories to satisfy the fat cells, and their metabolic rates would plummet, causing them to be lethargic and unhealthy, but they probably wouldn't lose much weight.
2. They would consume twice as much as you do, and wouldn't lose any weight.
If they're talking about the U.S., that last part is exactly backwards. Our consumption of starchy staple foods has gone way, way up. Almost nobody eats a burger without fried potatoes, and the fast food is loaded with bread. The chicken is breaded, and half the time, you have a bun on top of that. And pizza is mostly bread. And so on. All of these are things that are rising in popularity. I have a hard time believing that bread consumption could possibly have gone down unless most people ate two meals a day of nothing but bread. :-)
No, based on what I've seen, what has changed most is:
All of these things contribute to the problem. It isn't just one thing. It is an epidemic of really, really lousy options provided by our food industry. And in an era where many people simply do not have time to cook or sit down for an extended dinner at a nicer restaurant, there are painfully few healthy alternatives.
You're joking, right? I routinely buy movies that are way more than five years old out of the $5 movie bin. When a new movie comes out, I never buy it at $20 for the DVD or $30 for the Blu-Ray. I wait a few years for it to be below $10, then buy it. So people like me would basically no longer be contributing to the profits of those content creators because there would be no reason for the companies selling me those DVDs (e.g. Fry's) to pay them (the studios).
Similarly, a fair percentage of the books I have bought in my life have been older than five years by the time I bought them, not because I'm being cheap, but because a major publisher agreeing to publish more than a couple of books is a sign of a series that is probably good and will probably continue to produce new works. It isn't a perfect model, but it's a good first approximation.
For example, I recently bought Jim Butcher's Codex Alera series and about a year earlier, his Dresden Files series. Under a five-year copyright term, more than half of each series would have been out of copyright by the time I learned that the series existed, or about eight books out of 16. (Admittedly, this is in large part because a paperback copy of Storm Front sat on my shelves for years before I had time to read it, but that's neither here nor there.) It averages out to being about a quarter of all the books I bought in the past few months, but more than half of the books I've actually had time to read. Assuming that my experience is typical, which I suspect it is, that means that a five-year term would put a major dent in authors' ability to make a living, and it's hard enough for authors to make a living by writing even under the current system.
Except that most new authors' first books, assuming they write them in their spare time while doing another job, takes at least a couple of years to write, and I've been working on my trilogy (which had to be written in its entirety before the first book was published because of the complex relationship between the books) since 2001. I'm not done yet.
What this means is that the average new author will spend years of his or her life writing with no possibility of making any money. I don't care how much free publicity you get from it. You still won't have an income stream, and without that income stream, you'll never be able to free yourself from your actual job in order to spend the amount of time it takes to write at a reasonable speed, unless you just get exceptionally lucky and a professional publishing house decides to take a chance on a proposal before the book is written. (They won't, though, because there will be no incentive to pay authors when there is so much recently written public domain content that they can publish, knowing that the odds of another publisher randomly picking the same work is minimal.)
Also, no author would ever get paid for his work when big movie studios decide they want to make a movie of a well-known author's book. The book would be out of copyright, so they would have no obligation to pay for the right to create a derivative work. This means that authors would have to make all of their money from book sales (difficult even with long copyrights), donations/charity, or from doing something else to make an actual living. In short, a five-year-IP model is tantamount to indentured servitude for the actual content creators because anyone with an actual job loses any real opportunity to get out from under the thumb of abusive corporate contracts.
What this model fails to consider is that there is a huge difference between the ability of an individual and a corporation to monetize a work, and an even bigger difference between the speed at which different types of media can usefully be monetized. Works that take years to create tend to take years to pay back their creators. Five years is simply absurd unless your goal is to encourage everyone to stop creating works that are more than about a page or two long. If you want to live in a world where no movies exist that are longer than YouTube videos, and no books exist because they have been replaced by blog posts, feel free to create your own pseudo-utopian copyright-free zone, but count me out.
The original 28 years from the date of first publication (as opposed to five years from its creation) was a good length for copyright. It ensured that after about a generation, material fell out of copyright so that new generations of people could create interesting derivative works. It ensured that things fell out of copyright before people forgot about them. However, it also ensured that most books that might become movies would do so long before the copyright expired (with the obvious exception of Lord of the Rings, which took twice that long, but which was made long enough after the author's death to make the issue largely moot, IMO). It ensured that people who spent years working on something could realistically break even. And so on. But five years? Way, way, way too short.
Wait... I thought that was the whole reason the Internet was created....
See also rule 34.
Of course you're trolling, but what does Rupert Murdoch's gutter-level right-wing editorial service called Fox News have to do with a legitimate news operation?
Uh... any hardware codec chip (which all digital video cameras use in one form or another) has a maximum data rate even in VBR mode, and the video output of the camera, by definition, will never exceed that rate. As long as your storage exceeds that limit, you won't run out of storage. Similarly, as long as your network bandwidth exceeds that limit with a reasonable amount of room to spare, you shouldn't have problems there, either....
That makes the plane safer, but because, by definition, you are not on the plane, you are still no safer. Well, I suppose for the handful of people who happen to be in the ground track of the debris, you are slightly safer, but this is more than offset by the added risk to the people standing in line with the lunatic carrying a bomb.
Err... had to hold the measuring tape....
And this, folks, is why over-reliance on automatic spell checking is harmful. *sigh*
I measured 24 inches from my eyes, and in order to do so, I had to old the measuring tape up with my fingertips, with my arms fully extended. No optometrist in his right mind would recommend such a large reading distance. Maybe for TV viewing, but not reading.
Typical reading distance is 14-16 inches. So although 12 inches might be slightly close, your suggested reading distance is so far beyond normal that only an orangutan can achieve it.
Fair enough. So instead of dying, you'd probably just get really, really sick and die of cancer a few years later. The point was that critical mass is only meaningful when no pressure is applied, and that much smaller amounts can become dangerous under the right circumstances.
To put it in context, 3 pounds of Uranium is still more than the portion of the Uranium that actually contributed to the explosive yield of Little Boy (Glasstone and Dolan, Effects, pp. 12–13, as cited by Wikipedia ). There's a lot of energy there that could theoretically be released under the right circumstances.
True, but it is a far bigger concern when you have exactly one major corporate backer. When you have five or six companies contributing lots of code, the disadvantages caused by keeping private repositories in sync with the public ones usually outweigh any advantages any single company might get from keeping significant bits proprietary.
Not true. If the GPL project requires copyright assignment, then the new owners of that copyright can do whatever they please, up to and including creating a closed-source fork. Likewise, if a single company owns the copyright on 90% of the code, they can rip out the 10% contributed by outsiders, hire people to rewrite those bits, and create a closed-source fork. The GPL closes off that line of attack only if there are multiple major contributors contributing large portions of the code, and in those circumstances, BSD-licensed software is also largely immune to that line of attack by virtue of the synchronization hassle factor, as I mentioned previously.
Linux vendors are unusual in that they've already thoroughly cast their lot with the GPL, for better or worse. They don't have any alternative short of not being a Linux vendor. Same goes for other companies that make their living by providing support for GPLed software.
No, I'm just saying that if you want broad adoption, GPLv3 is a dead end. The fears, unjustified or not, are very prevalent, and are unlikely to go away. So you have to choose which is more important: broad adoption or preventing a theoretical abuse that statistically won't ever happen to the vast majority of projects regardless of license.
The situations where companies have created closed-source forks of BSD-licensed projects have almost invariably been cases where there was no set of benevolent core maintainers, e.g. the Apache Foundation. If the open source content is not changing significantly, there is minimal benefit of keeping up with public changes, and thus no real penalty for creating a closed branch. That's why I specifically stipulated that the ideal situation was a BSD license in which the core development was done by a benevolent foundation (and, by extension, in which the software is still being actively maintained and enhanced).
Further, it's worth noting that when companies create closed-source forks, they often do so by hiring the main/sole developer or buying the company that owns the copyrights. When that happens, the GPL provides no additional protection. The GPL provides protection only if a sizable percentage of the code was written by someone who won't accept money in exchange for the right to create a closed-source branch. In practice, this means that the GPL is usually largely ineffectual at preventing the sorts of situations that you describe; the fact that GPLed software falls victim to this less frequently has more to do with corporate aversion to touching the stuff, plus the existence of groups like the FSF that help back some of it, rather than because the license inherently prevents the creation of such a branch.
Finally, FWIW, most of the BSD-licensed technology that Apple uses is still released as open source. There's a giant pile of it on opensource.apple.com. I count 65 published projects for the current Lion release that are in whole or in part licensed under a BSD license.
Creating fusion is not hard. Philo T. Farnsworth (the TV guy) did it eighty years ago. Creating self-sustained fusion that produces more power than it consumes is hard.
It's still probably enough to easily go supercritical and kill you if you compress it sufficiently. It certainly doesn't require a factor of 30; the demon core was only 14 pounds, and killed two people in separate incidents.
Even the final 802.11n in the best case is not that much faster than 100BASE-T, and that requires having no more than a couple of devices on the network, all close to the base station. By its nature, wireless communication is a shared medium, not switched, so as soon as you have two devices, they're competing for bandwidth; MIMO will eventually help with this to some degree, but even that only goes so far. It doesn't take very many devices on a wireless network before the average per-device bandwidth is less than switched 10BASE-T would provide, much less switched 100BASE-T or gigabit.