"When you leave your windows open without screens in, expect bugs come in the house. The internet is no different. If your server is designed to blindly respond to every request, expect to get requests you don't actually want. Just like physical screens for physical windows, virtual filters for virtual ports cost money."
You have changed your position in response to my criticism but pretend that you have not. There's a huge difference between your original position:
"If you walk in bad neighborhoods with $100 bills visibly sticking out of your pockets, people are justified in taking them from you."
and your new position:
"If you walk in bad neighborhoods with $100 bills visibly sticking out of your pockets, people are going to take them from you."
Yes, if you set up a service on the Internet that's capable of being abused, it's going to be abused. But that says nothing about what constitutes abuse or whether a particular type of access is justified or unjustified.
We accept that spam is abuse because it uses other people's services in a way they did not intend and forces them to choose between accepting both your spam (which they do not want) and other mail (which they do). When you filter ads, you do the same thing and force others into the same trade-offs.
"The money they spend on actually detecting fraudulent clicks is lost, but it can't be $1 billion... and if it is I'd like to meet the handful of programmers that are pulling in $100mm a year. But really, whatever they're truly losing (maybe $1mm a year tops?) is just part of doing business in a world with a small percentage of dishonest people. It sucks, but all businesses have to deal with this."
It's not like money spent to combat employee theft or things like that. Nobody shops at a store because they do such a good job of stopping employee theft. People run their ads through Google in part because of Google's reputation in combating click fraud and because this gives them a higher conversion rate.
"Second the HTTP client's request no representation to Google about it's intentions for requesting the URL's content, nor it's intended use of the URL's content."
This proves too much. If we really believed that, we could justify anything. Why can't I spam you? It's not my fault your mail server accepts the mail and delivers it to a person. As far as I know, it wants it and might even throw it away.
No, sorry. If you want to connect to servers run on other people's machines, you have an obligation to use them as they clearly wish you to use them, within reason.
If you don't believe me, try going to your local supermarket every day and taking every single copy of every "free" newspaper.
You cannot knowingly impose costs on other people and then say that you are free to do whatever the hell you want regardless.
Let's put it this way. The $1 billion Google "lost" they did not earn and were not entitled to. It is insane to say that fraud cost you money that you did not earn or was not entitled to. If someone loses $100,000 to a Nigeria scam, they are actually out $100,000 that they earned and that they are entitled to.
The argument you are making is incorrect for very subtle reasons. I'll try an analogy because I think that explains it best:
Suppose I run a video game store. Someone leaves 100 copies of Halo 3 on my doorstep. I have no idea where they came from. I do some investigation and determine that they are illegal copies. I could sell them for $50 each. But since I don't wish to participate in a fraud, I do not.
So did these copies being dumped on my doorstep cost me $5,000? If I hadn't detected that they were fraudulent and opted not to sell them, I'd be $5,000 richer.
Of course, that's $5,000 I did not earn and to which I'm not entitled.
It is the same with click fraud. This is fraudulent worthless "items" being dumped on Google's doorstep. They could sell them as if they were the genuine article and rip off their customers, but they're a smart reputable business, and so they don't. But they are not out the amount they could have sold the clicks for because they are not entitled to that money because they did not earn it.
"Who says they're not doing an analysis of caught vs. un-caught fraudulent clicks? If they have a decently accurate percentage figure for fake clicks that manage to pass their system, then that IS money lost, since it WAS paid out."
Err, what?! If they fail to detect a bogus click, they get *PAID* by the advertiser for that click. They pay out less than the advertiser pays.
Google is only scammed by bogus clicks indirectly.
For example, if I run an ad with Google and some other company, undetected bogus clicks will reduce the apparent effectiveness of the Google ad (the ratio of clicks I pay for to sales I get will be lower). This makes Google's ads worth less money.
In truth, Google benefits from click fraud in a very subtle and non-obvious way. Certain types of click fraud will travel with you from company to company. For example, if a competitor is trying to drain your ad budget, they will click on your ads whether they're hosted by Google or someone else. Google has the most sophisticated click fraud detection in the industry. So click fraud will actually decrease another advertiser's apparent effectiveness by more than Google's.
Google does not pay out for fraudulent clicks as much as its competitors do because it detects them better.
"Google discards those clicks voluntarily. If they hadn't, they could be charging the advertisers for each such click, and would be making more money."
So, is Google stupid? No, of course not.
They are smart enough to know that if they charged for clicks they know are bogus, they would make *less* money. They'd have to charge less per ad, they'd lose customers, and so on.
These bogus clicks ARE NOT opportunities to make money. They're simply bogus.
There is no scenario where Google could convert these into revenue.
If you lose $1 billion due to something, that would have to mean that if not for that something, you'd make $1 billion more. If not for these fraudulent clicks, Google would make the same amount.
Yes, *IF* those fraudulent clicks were legitimate, Google would make $1 billion more. But they're simply aren't that many legitimate clicks, and that's not because of the fraudulent ones.
For example suppose they can afford to audit a random 1% of clicks they detect are bogus and 1% of clicks they detect are legitimate. Assume 1% is statistically significant. They can compute with known accuracy how many clicks fall into each of these categories: 1) Legitimate clicks detected as legitimate. 2) Bogus clicks detected as bogus. 3) Legitimate clicks detected as bogus. 4) Bogus clicks detected as legitimate.
Type 1 clicks, obviously, are great. Type 2 clicks don't harm Google significantly. Type 3 clicks are the big problem for Google as these result in lost revenue. Type 4 clicks result in income for Google but harm their reputation. Ultimately, this might be reflected in lower ad rates or fewer customers.
But your basic argument is flawed. They may be able to detect bogus clicks very reliable with a method that's impractical for use on a high percentage of clicks.
The logic behind this story is bogus. The $1 billion in money that these fraudulent clicks cost Google doesn't exist. If not for the bogus clicks, these clicks wouldn't exist.
It's like a software company claiming that false orders cost them $10 billion dollars last year because they received an bogus order for 100,000,000 copies of a $100 product. Had they not received the bogus order, they would not be $10 billion richer.
"Why? Why should an author get special treatment? If I make a toaster, I have special treatment only over that toaster. If I sell it to someone else, I revoke all rights and privlidges to control what they do with their new toaster. There is no reason books, music, etc should be any different."
They're really not any different. If I wanted to, before I sold you a toaster, I could first make you sign a contract stipulating that you won't open to toaster, or won't replicate the toaster's design, or pretty much any other limitation I wanted to impose.
The problem is, you might resell the toaster to someone else without telling them about this restrictive agreement. They might think they own the toaster free and clear when you don't actually have the right to sell them the toaster free and clear.
Copyright does not prohibit free and clear sales. If I want to sell you a book and include the right for you to write your own similar book or make copies of it, I can. And if you don't want to buy a book knowing that you have to agree not to make copies of it, you don't have to.
All copyright law does is protect innocent third parties. It prevents you from selling the book to someone, claiming they can do whatever they want with it, when you actually agreed not to copy it.
In other words, it simply sets the default agreement when I sell you a book differently from the default agreement when I sell you a toaster. But it doesn't make other arrangements impossible.
Arguing that there shouldn't be a copyright is simply arguing that there should be a different default agreement. There's never any answer to how you protect innocent third parties without copyright.
Today's desktops were yesterdays servers. The technology you find on the desktop today are the technologies you only found on servers a few years ago.
To give you a simple example of why this is a retarded idea, consider SMP (support for multiple cores/CPU). Five years ago, a rational desktop OS developer might have thought that SMP support should be dumped. It has significant overhead, and when will there be multiple CPU desktops?
Well, now.
The same goes for all kinds of things that were once unheard of on the desktop and are now on every desktop.
The beauty of Linux is that you don't have to fork it because you can compile in just what *your* desktop, server, or whatever needs.
"So, in short you recommend using a piece of software, that installs another piece of software, that stays on the system after uninstalling the first piece of software (How else could it work, if you have multiple pieces of software that uses it?), and, as you say service, I assume it runs while the original piece of software is not."
There's this really sophisticated technology. It's so sophisticated, all I can tell you is it's code name, "reference counting". As I understand it, and I don't have codeword clearance so this may not be quite right:
1) Each program checks to see if the module is already installed. If not, it installs it and sets the reference count to 1. (The reference count is usually stored in the registry on a Windows machine.)
2) If the module is already installed, it checks if this installer has an upgraded version, if so, it upgrades it. Either way, it bumps the reference count by one.
3) When a program is uninstalled, it decrements the reference count.
4) If the reference count hits zero, the module is uninstalled as well.
Now, this is all secret stuff and I have no way to verify it. But my aunt's cousin was once uninstalling some software, and it said something about whether she wanted to remove files that were no longer in use or leave them there in case something else might use them later.
I think this is evidence that this is not just a rumor or myth, but that they really are@#07a239-29u CARRIER LOST > OK
I think I was clear that I was talking about Windows XP and Windows Media Player, neither of which include a DVD decoder. Perhaps someone else was vague.
Please read the following web page: Plugins "If you are running Windows XP, you can enhance your DVD playback experience by purchasing a DVD decoder pack."
"Why are we waiting until it's a crisis to deal with it?"
Ironically, the longer you wait to deal with it, the cheaper it may be!
There are some obvious reasons why waiting longer makes it cost more, but there are quite a few subtle reasons why it's cheaper to wait. For example:
1) If your current hardware is not IPv6 capable and you buy new IPv6-capable hardware now, it may reach end-of-life before you need the IPv6 capability.
2) IPv6 routes take more memory than IPv4 routes. The longer you wait, the cheaper it will be to add this memory. (Note that we're not just talking cheap main memory, we're talking expensive CAM and custom chip memory.)
3) Research and development are constantly progressing. The longer you wait, the better researched the solution you ultimately deploy may be. (To a limit, of course. You also lose the chance to gain experience.)
On balance, I think we're progressing at a sensible pace, perhaps a bit slower than perfect. People are continuing to do test deployments to see how IPv6 will work and make sure they'll be able to implement it for real when the demand comes. But they're not wasting money replacing working hardware or increasing network instability on the real, live Internet we all depend on for our daily (hourly? half-hourly?)/. fix.
"On the other hand, electronic, cryptographic technology has many failure points which only a few people understand."
Translation: I can't actually find a flaw in the system, so I'll just assume one exists.
"Each electronic device is a black box where human actions input and a lot of cryptographic gibberish output on the other side."
Exactly. The beautiful part is, if you design the system correctly, it becomes mathematically infeasible for the output to differ from the human input.
"Tweak just one machine in the whole chain so that it reports a result to humans and a different one into the system, and the whole process gets compromised;"
Perhaps you have no idea about the properties of the systems you are talking about are, but the primary property they have is that it is not possible for the machine to report a different result to humans and into the system. Because the results reported to humans are cryptographically signed, it becomes easy to prove whether or not the results passed onto the system are identical to the ones passed to humans.
"and just a few persons in the whole country would be able to tell if a particular machine has something wrong in it."
Not true. The systems discussed have the property that they can be audited on-site by multiple organizations including political parties and civics groups.
I think you are confusing properties of particular cryptographic systems with theoretical limits of cryptographic systems. I think you also don't realize that I'm on your side. My position would be that because it's possible to design a system that can cryptographically assure that its results are valid, we shouldn't settle for a system that can't. That is, I oppose the use of pretty much every electronic voting scheme used to date, precisely because they don't provide these assurances, and there's no good reason they don't.
We're not talking about adding license text that is misleading, we're talking about *removing* license text. Sure, you'd be right if they added a notice saying the work was in the public domain, since it's *not*.
Nothing in copyright law requires all the rights you might have to a work to be documented in that work itself. You most certainly could remove license notices if the various licenses didn't require you to keep them intact.
Please show me what in copyright law says that when you have the right to modify a work and to distribute modified versions of that work you must retain all possible ways another person might obtain the right to modify or distribute that work.
I think your confusion is that you think that the BSD and GPL licenses are *USE* licenses or EULAs. They are not. Neither of them has any effect on ordinary use.
This is certainly true, but what Theo doesn't seem to understand is that NOBODY CHANGED THE TERMS. They simply changed the *notice*.
If you place a work under a dual license, somebody can obtain the right to modify that work and distribute those modification from *either* license. Neither license requires them to retain the other license's *NOTICE*.
That is, when you offer a work under the GPL, whether among other licenses or not, you grant people the right to remove anything from that file that they want to, other than the GPL itself. Why? Because this is what the GPL *says*.
Note that this has no effect on anyone's actual rights to use the file. It just means you *can* remove the notice. Why? Because the GPL says so.
This complaint is no better. It claims they "detected an individual" who is "distributing". But they don't actually detect any distribution. Their "download and/or distribute" language makes no sense, since they never detect anybody downloading anything.
This has nothing to do with multi-core CPUs. This exact same attack would work on the run-of-the-mill SMP systems that have been around for decades. Multi-core CPUs just make SMP systems more common.
This is true but misleading. The dual license still applies to, and only to, contributions by those individual authors who contributed under a dual license.
For example, suppose I am contributing to a file under a dual license. All I need is the right to modify the work or create a derivative work. Once I have that right, I can contribute. And the only rights you get to my contributions are those that I am required to grant you or choose to grant you. There are two ways I can do that:
1) The BSD license permits me to make modifications and distribute them. It does not require me to offer my changes under the BSD license, so I am not required to do so. It does not require me to leave the GPL license offer in the code, so I am not required to do so.
2) The GPL license permits me to make modifications and distribute them. It does not require me to keep the BSD license text intact, so I am not required to do so.
Note that none of this affects the license on the original code, but they do affect the license on my modifications and thus what you can do with the integrated work.
Unfortunately, it's really impractical for Linux to keep code in the main kernel source tree under a dual license. The problem is that you can't cut/paste code from one file into another because code that was contributed under the GPL cannot be re-offered under a dual license.
Consider the following nightmare scenario:
1) Linux gets a filesystem that's under a dual-license. They try to keep it that way.
2) Someone adds a new hook that all filesystems should support. The add support to the ext2 and ext3 filesystems, which are under the GPL.
3) Someone tries to add support for that hook into the filesystem mentioned in part 1, but they cannot cut/paste any of the implementation from ext2/ext3 into the file the implementation has to go into.
It is extremely difficult for a project to have to deal with code sections that are under a different license. The exception would be if those code sections aren't under active development inside the project. But every file in the Linux kernel distribution, almost, is under active development. Things like filesystems and drivers clearly so.
I don't think they have any legal authority to ignore a DMCA notice without losing their safe harbor rights. I think you could consider this a defect in the DMCA. Recipients of take down notices should be able to delay acting on the notice if they have a good faith belief that the notice may be bogus, they should be allowed to take diligent steps to verify the notice, and should be permitted to keep the material up while retaining safe harbor rights so long as they have such a good faith belief.
Slashdot Mirror
"When you leave your windows open without screens in, expect bugs come in the house. The internet is no different. If your server is designed to blindly respond to every request, expect to get requests you don't actually want. Just like physical screens for physical windows, virtual filters for virtual ports cost money."
You have changed your position in response to my criticism but pretend that you have not. There's a huge difference between your original position:
"If you walk in bad neighborhoods with $100 bills visibly sticking out of your pockets, people are justified in taking them from you."
and your new position:
"If you walk in bad neighborhoods with $100 bills visibly sticking out of your pockets, people are going to take them from you."
Yes, if you set up a service on the Internet that's capable of being abused, it's going to be abused. But that says nothing about what constitutes abuse or whether a particular type of access is justified or unjustified.
We accept that spam is abuse because it uses other people's services in a way they did not intend and forces them to choose between accepting both your spam (which they do not want) and other mail (which they do). When you filter ads, you do the same thing and force others into the same trade-offs.
"The money they spend on actually detecting fraudulent clicks is lost, but it can't be $1 billion... and if it is I'd like to meet the handful of programmers that are pulling in $100mm a year. But really, whatever they're truly losing (maybe $1mm a year tops?) is just part of doing business in a world with a small percentage of dishonest people. It sucks, but all businesses have to deal with this."
It's not like money spent to combat employee theft or things like that. Nobody shops at a store because they do such a good job of stopping employee theft. People run their ads through Google in part because of Google's reputation in combating click fraud and because this gives them a higher conversion rate.
Google makes money from click fraud.
"Second the HTTP client's request no representation to Google about it's intentions for requesting the URL's content, nor it's intended use of the URL's content."
This proves too much. If we really believed that, we could justify anything. Why can't I spam you? It's not my fault your mail server accepts the mail and delivers it to a person. As far as I know, it wants it and might even throw it away.
No, sorry. If you want to connect to servers run on other people's machines, you have an obligation to use them as they clearly wish you to use them, within reason.
If you don't believe me, try going to your local supermarket every day and taking every single copy of every "free" newspaper.
You cannot knowingly impose costs on other people and then say that you are free to do whatever the hell you want regardless.
Let's put it this way. The $1 billion Google "lost" they did not earn and were not entitled to. It is insane to say that fraud cost you money that you did not earn or was not entitled to. If someone loses $100,000 to a Nigeria scam, they are actually out $100,000 that they earned and that they are entitled to.
The argument you are making is incorrect for very subtle reasons. I'll try an analogy because I think that explains it best:
Suppose I run a video game store. Someone leaves 100 copies of Halo 3 on my doorstep. I have no idea where they came from. I do some investigation and determine that they are illegal copies. I could sell them for $50 each. But since I don't wish to participate in a fraud, I do not.
So did these copies being dumped on my doorstep cost me $5,000? If I hadn't detected that they were fraudulent and opted not to sell them, I'd be $5,000 richer.
Of course, that's $5,000 I did not earn and to which I'm not entitled.
It is the same with click fraud. This is fraudulent worthless "items" being dumped on Google's doorstep. They could sell them as if they were the genuine article and rip off their customers, but they're a smart reputable business, and so they don't. But they are not out the amount they could have sold the clicks for because they are not entitled to that money because they did not earn it.
"Who says they're not doing an analysis of caught vs. un-caught fraudulent clicks? If they have a decently accurate percentage figure for fake clicks that manage to pass their system, then that IS money lost, since it WAS paid out."
Err, what?! If they fail to detect a bogus click, they get *PAID* by the advertiser for that click. They pay out less than the advertiser pays.
Google is only scammed by bogus clicks indirectly.
For example, if I run an ad with Google and some other company, undetected bogus clicks will reduce the apparent effectiveness of the Google ad (the ratio of clicks I pay for to sales I get will be lower). This makes Google's ads worth less money.
In truth, Google benefits from click fraud in a very subtle and non-obvious way. Certain types of click fraud will travel with you from company to company. For example, if a competitor is trying to drain your ad budget, they will click on your ads whether they're hosted by Google or someone else. Google has the most sophisticated click fraud detection in the industry. So click fraud will actually decrease another advertiser's apparent effectiveness by more than Google's.
Google does not pay out for fraudulent clicks as much as its competitors do because it detects them better.
"Google discards those clicks voluntarily. If they hadn't, they could be charging the advertisers for each such click, and would be making more money."
So, is Google stupid? No, of course not.
They are smart enough to know that if they charged for clicks they know are bogus, they would make *less* money. They'd have to charge less per ad, they'd lose customers, and so on.
These bogus clicks ARE NOT opportunities to make money. They're simply bogus.
There is no scenario where Google could convert these into revenue.
If you lose $1 billion due to something, that would have to mean that if not for that something, you'd make $1 billion more. If not for these fraudulent clicks, Google would make the same amount.
Yes, *IF* those fraudulent clicks were legitimate, Google would make $1 billion more. But they're simply aren't that many legitimate clicks, and that's not because of the fraudulent ones.
This is a nonsensical argument.
For example suppose they can afford to audit a random 1% of clicks they detect are bogus and 1% of clicks they detect are legitimate. Assume 1% is statistically significant. They can compute with known accuracy how many clicks fall into each of these categories:
1) Legitimate clicks detected as legitimate.
2) Bogus clicks detected as bogus.
3) Legitimate clicks detected as bogus.
4) Bogus clicks detected as legitimate.
Type 1 clicks, obviously, are great. Type 2 clicks don't harm Google significantly. Type 3 clicks are the big problem for Google as these result in lost revenue. Type 4 clicks result in income for Google but harm their reputation. Ultimately, this might be reflected in lower ad rates or fewer customers.
But your basic argument is flawed. They may be able to detect bogus clicks very reliable with a method that's impractical for use on a high percentage of clicks.
The logic behind this story is bogus. The $1 billion in money that these fraudulent clicks cost Google doesn't exist. If not for the bogus clicks, these clicks wouldn't exist.
It's like a software company claiming that false orders cost them $10 billion dollars last year because they received an bogus order for 100,000,000 copies of a $100 product. Had they not received the bogus order, they would not be $10 billion richer.
Duh.
"Why? Why should an author get special treatment? If I make a toaster, I have special treatment only over that toaster. If I sell it to someone else, I revoke all rights and privlidges to control what they do with their new toaster. There is no reason books, music, etc should be any different."
They're really not any different. If I wanted to, before I sold you a toaster, I could first make you sign a contract stipulating that you won't open to toaster, or won't replicate the toaster's design, or pretty much any other limitation I wanted to impose.
The problem is, you might resell the toaster to someone else without telling them about this restrictive agreement. They might think they own the toaster free and clear when you don't actually have the right to sell them the toaster free and clear.
Copyright does not prohibit free and clear sales. If I want to sell you a book and include the right for you to write your own similar book or make copies of it, I can. And if you don't want to buy a book knowing that you have to agree not to make copies of it, you don't have to.
All copyright law does is protect innocent third parties. It prevents you from selling the book to someone, claiming they can do whatever they want with it, when you actually agreed not to copy it.
In other words, it simply sets the default agreement when I sell you a book differently from the default agreement when I sell you a toaster. But it doesn't make other arrangements impossible.
Arguing that there shouldn't be a copyright is simply arguing that there should be a different default agreement. There's never any answer to how you protect innocent third parties without copyright.
Today's desktops were yesterdays servers. The technology you find on the desktop today are the technologies you only found on servers a few years ago.
To give you a simple example of why this is a retarded idea, consider SMP (support for multiple cores/CPU). Five years ago, a rational desktop OS developer might have thought that SMP support should be dumped. It has significant overhead, and when will there be multiple CPU desktops?
Well, now.
The same goes for all kinds of things that were once unheard of on the desktop and are now on every desktop.
The beauty of Linux is that you don't have to fork it because you can compile in just what *your* desktop, server, or whatever needs.
"So, in short you recommend using a piece of software, that installs another piece of software, that stays on the system after uninstalling the first piece of software (How else could it work, if you have multiple pieces of software that uses it?), and, as you say service, I assume it runs while the original piece of software is not."
There's this really sophisticated technology. It's so sophisticated, all I can tell you is it's code name, "reference counting". As I understand it, and I don't have codeword clearance so this may not be quite right:
1) Each program checks to see if the module is already installed. If not, it installs it and sets the reference count to 1. (The reference count is usually stored in the registry on a Windows machine.)
2) If the module is already installed, it checks if this installer has an upgraded version, if so, it upgrades it. Either way, it bumps the reference count by one.
3) When a program is uninstalled, it decrements the reference count.
4) If the reference count hits zero, the module is uninstalled as well.
Now, this is all secret stuff and I have no way to verify it. But my aunt's cousin was once uninstalling some software, and it said something about whether she wanted to remove files that were no longer in use or leave them there in case something else might use them later.
I think this is evidence that this is not just a rumor or myth, but that they really are@#07a239-29u
CARRIER LOST
> OK
"...you may be in for a real treat when the time comes for your kids to pick your nursing home."
Wasn't this the one we saw on that episode of 60 Minutes?
So how would you feel about candy stores that refused to sell to overweight people? I mean, why should they promote obesity?
I think I was clear that I was talking about Windows XP and Windows Media Player, neither of which include a DVD decoder. Perhaps someone else was vague.
Please read the following web page:
Plugins
"If you are running Windows XP, you can enhance your DVD playback experience by purchasing a DVD decoder pack."
"Why are we waiting until it's a crisis to deal with it?"
/. fix.
Ironically, the longer you wait to deal with it, the cheaper it may be!
There are some obvious reasons why waiting longer makes it cost more, but there are quite a few subtle reasons why it's cheaper to wait. For example:
1) If your current hardware is not IPv6 capable and you buy new IPv6-capable hardware now, it may reach end-of-life before you need the IPv6 capability.
2) IPv6 routes take more memory than IPv4 routes. The longer you wait, the cheaper it will be to add this memory. (Note that we're not just talking cheap main memory, we're talking expensive CAM and custom chip memory.)
3) Research and development are constantly progressing. The longer you wait, the better researched the solution you ultimately deploy may be. (To a limit, of course. You also lose the chance to gain experience.)
On balance, I think we're progressing at a sensible pace, perhaps a bit slower than perfect. People are continuing to do test deployments to see how IPv6 will work and make sure they'll be able to implement it for real when the demand comes. But they're not wasting money replacing working hardware or increasing network instability on the real, live Internet we all depend on for our daily (hourly? half-hourly?)
Then neither does Windows XP. Windows media player does not include a DVD decoder, for legal reasons.
"On the other hand, electronic, cryptographic technology has many failure points which only a few people understand."
Translation: I can't actually find a flaw in the system, so I'll just assume one exists.
"Each electronic device is a black box where human actions input and a lot of cryptographic gibberish output on the other side."
Exactly. The beautiful part is, if you design the system correctly, it becomes mathematically infeasible for the output to differ from the human input.
"Tweak just one machine in the whole chain so that it reports a result to humans and a different one into the system, and the whole process gets compromised;"
Perhaps you have no idea about the properties of the systems you are talking about are, but the primary property they have is that it is not possible for the machine to report a different result to humans and into the system. Because the results reported to humans are cryptographically signed, it becomes easy to prove whether or not the results passed onto the system are identical to the ones passed to humans.
"and just a few persons in the whole country would be able to tell if a particular machine has something wrong in it."
Not true. The systems discussed have the property that they can be audited on-site by multiple organizations including political parties and civics groups.
I think you are confusing properties of particular cryptographic systems with theoretical limits of cryptographic systems. I think you also don't realize that I'm on your side. My position would be that because it's possible to design a system that can cryptographically assure that its results are valid, we shouldn't settle for a system that can't. That is, I oppose the use of pretty much every electronic voting scheme used to date, precisely because they don't provide these assurances, and there's no good reason they don't.
We're not talking about adding license text that is misleading, we're talking about *removing* license text. Sure, you'd be right if they added a notice saying the work was in the public domain, since it's *not*.
Nothing in copyright law requires all the rights you might have to a work to be documented in that work itself. You most certainly could remove license notices if the various licenses didn't require you to keep them intact.
Please show me what in copyright law says that when you have the right to modify a work and to distribute modified versions of that work you must retain all possible ways another person might obtain the right to modify or distribute that work.
I think your confusion is that you think that the BSD and GPL licenses are *USE* licenses or EULAs. They are not. Neither of them has any effect on ordinary use.
This is certainly true, but what Theo doesn't seem to understand is that NOBODY CHANGED THE TERMS. They simply changed the *notice*.
If you place a work under a dual license, somebody can obtain the right to modify that work and distribute those modification from *either* license. Neither license requires them to retain the other license's *NOTICE*.
That is, when you offer a work under the GPL, whether among other licenses or not, you grant people the right to remove anything from that file that they want to, other than the GPL itself. Why? Because this is what the GPL *says*.
Note that this has no effect on anyone's actual rights to use the file. It just means you *can* remove the notice. Why? Because the GPL says so.
This complaint is no better. It claims they "detected an individual" who is "distributing". But they don't actually detect any distribution. Their "download and/or distribute" language makes no sense, since they never detect anybody downloading anything.
This should be rejected summarily as well.
This has nothing to do with multi-core CPUs. This exact same attack would work on the run-of-the-mill SMP systems that have been around for decades. Multi-core CPUs just make SMP systems more common.
This is true but misleading. The dual license still applies to, and only to, contributions by those individual authors who contributed under a dual license.
For example, suppose I am contributing to a file under a dual license. All I need is the right to modify the work or create a derivative work. Once I have that right, I can contribute. And the only rights you get to my contributions are those that I am required to grant you or choose to grant you. There are two ways I can do that:
1) The BSD license permits me to make modifications and distribute them. It does not require me to offer my changes under the BSD license, so I am not required to do so. It does not require me to leave the GPL license offer in the code, so I am not required to do so.
2) The GPL license permits me to make modifications and distribute them. It does not require me to keep the BSD license text intact, so I am not required to do so.
Note that none of this affects the license on the original code, but they do affect the license on my modifications and thus what you can do with the integrated work.
Unfortunately, it's really impractical for Linux to keep code in the main kernel source tree under a dual license. The problem is that you can't cut/paste code from one file into another because code that was contributed under the GPL cannot be re-offered under a dual license.
Consider the following nightmare scenario:
1) Linux gets a filesystem that's under a dual-license. They try to keep it that way.
2) Someone adds a new hook that all filesystems should support. The add support to the ext2 and ext3 filesystems, which are under the GPL.
3) Someone tries to add support for that hook into the filesystem mentioned in part 1, but they cannot cut/paste any of the implementation from ext2/ext3 into the file the implementation has to go into.
It is extremely difficult for a project to have to deal with code sections that are under a different license. The exception would be if those code sections aren't under active development inside the project. But every file in the Linux kernel distribution, almost, is under active development. Things like filesystems and drivers clearly so.
I don't think they have any legal authority to ignore a DMCA notice without losing their safe harbor rights. I think you could consider this a defect in the DMCA. Recipients of take down notices should be able to delay acting on the notice if they have a good faith belief that the notice may be bogus, they should be allowed to take diligent steps to verify the notice, and should be permitted to keep the material up while retaining safe harbor rights so long as they have such a good faith belief.