Unless of course you think we should be learning morality from big business?
And that's exactly what's happened in the west and what you see happening in Bhutan at a speed that makes people finally take notice.
The propaganda is crafted to be as strong as possible. It's effective enough to make share holders allow corporations to spend millions on this. You cannot fight this with merely 'explaining' that there's more to life than consuming. The message is simply too strong, too ubiquitious.
I personally think the simplest way to solve this in a unix-compatible way is to get rid of the distinction between subdirectories and files.
You have streams, and you have named attributes. Of course, it must be possible to open such a node in two modes, one to read the main stream, and one to read it and the whole tree beneath it as a tar-like stream, but that's nothing a few O_* flags couldn't solve;)
It needs some work in the semantics area, but it probably needs the most work in the VFS layer.
Hold your horses, there. The microwaves modulated and there is a pulse sequence in the transmitted frames. In short, there is a strong LF component in cellphone radiation.
If you don't believe me, try putting a GSM close to a stereo and ring it.
Of course, the next step would be to interlink robots in such a way that if one robot is observed by another, you see an image of the owner instead of the robot.
At that point, it's only a tiny step away to skip the whole robot and camera business and to interact virtually only.
Of course, I sincerely hope to be dead before this becomes the only practical way of being in contact with people.
However, stating that "nothing can be known in an absolute sense" does not mean that we should not attempt to discover the truth.
True, that's why I have no objection to the scientific method, where one attempts to build a model of cause and effect relationships by creating such hypotheses and testing them. Even when the models generated by this or any other method are not "true", they are at least useful.
However, history is a special branch of science, because you cannot test any hypotheses, like "if the US wouldn't have been so pro-Israel, 9/11 wouldn't have happened", or "if Versailles wouln't have been so humiliating, Hitler wouldn't have happened", and similar.
Therefore, what I'm saying is that because it's not only impossible to find the "truth" about what's behind 9/11 from statements about how likely it is that a certain sequence of events happened, but that it's also impossible to build a scientific theory, because that requires testing hypotheses.
So it's not just solipsism. I'm saying that the scientific method cannot be applied to any conspiracy theory or history in general, and that this makes a productive discussion between skeptic and paranoid extremely difficult.
Of course, I no way am I saynig that that's an excuse to do nothing. But it does mean that the choice which step to take next is always a political one.
Should the US protect itself from the world in response to part of it holding a grudge against the US, or try to increase everybody's participation in the world? It's a political question, and nothing else. Eventually, and that's the point, hypotheses about 9/11 hardly matter. Theories could have mattered, even if they are no truths, but untestable hypotheses are completely useless for the purpose of making political choices.
The problem you seem to ignore is the fundamental one of induction: you can never be sure that two coinciding events have anything to do with each other.
The patterns, or the big picture you speak of, can not be deducted from the events themselves. You can only see them if, as you say, you step back and observe the whole. Are the patterns you'll see indeed the ones that actually caused the events? Possibly, but it's just as likely that they are not.
The human mind always tries to find patterns in the event soup, and by observing the whole, you'll give your mind a chance to do this. Is the resulting model in your mind "true"? It's impossible to know; the only thing you can say is that it may be useful to base further actions upon.
However, the problem with incidents such as September 11 is that it's never possible to refine the hypotheses about the patterns connecting their events, because the incidents do not repeat, and in history, it's impossible to experiment; you cannot change certain variables and keep others the same.
Everyone with a more or less fitting pattern may be "right". That's why you always see a clash between the sceptical and the paranoid that can never be resolved, because the random (says the sceptical) or ordered (says the paranoid) events occurred only once; the hypotheses about the causes behind them cannot be tested.
In short, I'm saying that you cannot deduct anything from what you see when you "step back". The only thing you can create is untestable hypotheses, that you may use personally, and that may have esthetic value for others if they fit well, but are otherwise useless.
What's bad is that they sadly got some backing by ESR in this, who says in his article at http://www.opensource.org/sco-vs-ibm.html: "A judgment in favor of SCO could do serious damage to the open-source community. SCO's implication of wider claims could turn Linux into an intellectual-property minefield, with potential users and allies perpetually wary of being mugged by previously unasserted IP claims [...]".
Of course, he's actually saying that SCO's/implications/ could create a perceived minefield, without there actually being one, but I'm sure MS' FUD department would be able to use this piece to their advantage ("See? Even a leading Open Source advocate is aware of the dangers. You better switch to an IP-safe product.").
Re:SCO has Dirty Hands. Will not be able to collec
on
SCO To Show Copied Code
·
· Score: 2, Insightful
You can split a sandwich among many people; you cannot digest it in a collective stomach.
That's why/intellectual/ property is different. Smart people understand this.
You will have a hard time of course because these policies have already been applied to software for decades.
They have been applied, but nobody has shown any results from applying these policies to software. You go a long way to show that some of these policies have been applied succesfully in other fields of technology, but software is a different thing. Even if you have stolen the blueprints of a plutonium fab, you can't produce any plutonium yet: you need lots of resources, experience and controlled materials to actually build it. Not so for software; the blueprint/is/ the product, and it's trivial to steal a copy without anybody even being aware of it.
Now the problem with open source is that there is no way to control it, so there is no way to implement the kind of policy outlined above, except to kill it (or discourage it), and have everyone use closed source, which can be controlled to a significant degree.
I think that's a harmful illusion. To someone not caring about the law, leave alone EULAs, there is no difference between Open Source and proprietary software in terms of restrictions on use or distribution.
The only thing Open Source makes easier for the non-law abiding citizen than proprietary software is improving the software. Not the use of it.
From section 5. W3C Royalty-Free (RF) Licensing Requirements:
"With respect to a Recommendation developed under this policy, a W3C Royalty-Free license shall mean a non-assignable, non-sublicensable license to make, have made, use, sell, have sold, offer to sell, import, and distribute and dispose of implementations of the Recommendation that: [...]
3. may be limited to implementations of the Recommendation, and to what is required by the Recommendation"
This means that a patent may only be licensed for free if applied in code that implements the W3C recommendation.
So code that implements a patended recommendation may still be non-free in the sense that you cannot go and do with it whatever you want as long as you comply with the author's copyright license; you may violate the royalty-free patent license if you use the code for something other than implementing the W3C standard.
This doesn't mean that the code cannot be licensed under the GPL, it just means that you may need to take a look at the patent status as well before creating a GPL'ed derived work from code that implements a W3C standard.
It's a shame, but still better than RAND or nothing -- IMHO. Having said that, I think the biggest problem remains, which is patented content creation algorithms. You can eg. patent compression and charge shitloads for a license, but make decompression royalty-free. That way, media corporations can still raise a barrier to entry to the "creative commons", which makes patenting harmful and have exact opposite effect from they were designed to achieve.
I suspect that revenue models are a bigger problem, combined with distribution. To earn enough from a game paid for in very small chunks (say a free demo, then paying for new levels), you'd need to be damn sure people would keep buying them. Also, you'd need to be sure that people were honest enough not to just slap then into their P2P apps...
Well, instead of trying hard and investing lots of energy to collect pay/after/ you released your product, why not have an auction site for game development projects to allow gamers to fund these/before/ the result is released to the world?
A team with a good reputation could outline its plans and say, we need two million euros to cover development costs, our expenses and to make a good living while we're developing this game. Please send your money here. If we don't receive our budget within three months, we cancel the project and pay everyone back 95% of what they contributed (5 % to cover auction costs and living expenses).
This would work for any creative product: literature, software, games, books, movies; it allows you to make money from supplying information without having to supply distribution- or other services, but prevents nasty things like copyrights and licenses.
The only thing is that people who have paid $100 and have received a great game, must learn not to whine anymore when others are playing that game for free.
If they think about it, they have little reason to either, because people who are enthusiastic about this game are likely to help make their favourite authoring team's next, bigger, better production possible.
Another nice property is that gamers are the ones actually investing here, instead of the banks funding initial costs. The latter tend to favour minimal risk over maximal fun and innovation; gamers may choose to strike a slightly different balance.
You just need some searching infrastructure to allow people to find the authors in whose products they want to invest, and a financial institution such as a bank that could be the trusted party to guarantee that people get paid back if not enough money was collected before the deadline.
Can't agree more. The funny thing is that biological evolution also seems to prefer redundancy and an abundance of simple things over highly elegant, perfectly fitting solutions.
Throwing a load of generic units at the problem may be the easiest way to deal with complexity.
I refuse to dignify your straw dog with a response.
It was a bit harsh perhaps, but mostly because of you calling the guy's response childish, which is uncalled for, and because it left the impression that you think having an aversion to current corporate practice is childish, which is even more uncalled for IMHO.
I donate much of my spare time to the Debian project. What do you do to make the world a better place?
I try to tell people that humans can achieve more than animals, if we'd let go of the law of the jungle concept. We're free to choose, after all. I discuss politics with everyone who wants to. I donate money to the EFF and the FSF. I write Free Software. I preach about free software to my business relations, whenever possible. So it may not be much, but I try, too.
they didn't have the courtesy to contact him before resorting to lawyers.
You make it sound like "resorting to lawyers" is some sort of evil, violent deed. The fact of the matter is that real, professional businesspeople don't have a violent (and childish) aversion to lawyers. Lawyers are a routine part of the everyday business world, and have been since time immemorial. Freaking out about this merely makes the fellow look immature and foolish.
It may be not immediately evil, but it is agressive, for sure. Perhaps "profesisonal business people" don't have an aversion to lawyers because they bathe daily in the eat-or-be-eaten, right of the jungle world, and have come to like their style (who knows, perhaps even helped establish that style).
So I gather that in your view of the world, growing up means accepting the current way of doing business, and that means pure darwinism, i.e. everybody just fights as agressively as he can to get what he wants, and is obliged by the investors not to care how that is achieved?
That it is that way is one thing, but calling it immature not to agree with it is quite another.
Hell, that's one of the politest and friendliest C&D letters I've ever seen.
Which says just as much about C&D letters in general as about this particular one.
Of course, as long as they don't run any servers in the Netherlands, why should they get nervous about what the Dutch authorities consider illegal content?
They are not broadcasting the information or even publishing it in any way, not in the Netherlands. I specifically reach out with my request gor information all the way to the hosting center in the US, which kindly delivers it to me.
Now, if the Dutch authorities want to control what I see, they need to install content-aware firewalls at the national border. But how they can bother google with any request or demand whatsoever is beyond me, *if* google has no servers in the Netherlands that is.
In short, I don't really agree with your reasoning, and I really think it's strange (and worrying) if Google is really afraid of legislation from countries from which it doesn't server any information.
Because that would set a precedent that if I put up certain information on/my/ computer, that/I/ would become responsible for making sure that information is not sent to , not even upon explicit request from the client.
Which is not only a can of worms, technically and legally, but completely bollocks, IMHO.
Hold your horses, there. It's not context switches in themselves that make IPC inefficient, it's the address space switches, which empty your CPU's TLB.
Context switches among threads can potentially be very cheap, especially if the switch is not initiated by an interrupt (which would always switch to the kernel initially, before it can schedule a different thread).
It would be great if we'd have a non-application specific way (i.e. not dlopen) to load another image in a certain process' address space and relocate it, DOS-style, and creates a new thread for it. As long as neither application does anything stupid, you won't notice a thing.
That way, for a high speed webserver you could choose to put Apache, your userspace TCP stack, your userspace NIC driver, your userspace SCSI driver and your filesystem driver all in the same process space.
If the open, read, write, close functions of those drivers whould then be implemented using a good IPC channel, that uses local memory when talking to a thread in the same address space (think skbufs inside the kernel) and shared memory or ordinary pipes when talking to other processes, you would be completely free to define your system's compartmentalization.
For differently configured systems, you would only put your TCP/IP driver and your NIC driver in the same address space, and you keep your SCSI driver and filesystem in another.
So, applications using those drivers/outside/ their address space will incur the address space switch penalty for every call to them. But if you choose to run them in the same space, then you don't; context switches are theoretically just as nonexpensive as a setjmp()/longjmp(). It's just "slightly" less safe. If the application crashes, its drivers crash too.
It's perhaps a bit radical, but it would give the best of both (microkernel and traditional) worlds, with the ability to choose your own memory protection barriers. QNX has been doing something similar for a while as well; although it still require the application to explicitly load another image (which must always be a shared library, i.e. position independent code, no relocation is done), instead of allowing applications to remain unaware of their actual configuration (shared space or not). It does run the Unix API through a message passing channel though, and all drivers and filesystems are purely user space.
Seriously though. The only things missing for good userspace drivers in Linux is a way to open/dev/irq/1 so that you can select() on it, and a modern version of the Unix pipe (one that supports atomic messages and transports them either using zero-copy local memory, shared memory or copying), so that the open/read/write/ioctl/mmap/select/close API can be implemented through it.
As someone pointed out earlier, though, where Reuters might find they've stepped in it is that they published the information they found. IMO, that was dumb, and that's where they overstepped their bounds.
Hmm, if done verbatim, you're probably right. But let's say there is some messaging system that automatically gives some financial numbers when you dial a certain phone number, which, although unlisted, is directly available through the public phone system. In such cases I don't think there's anything wrong legally with publishing information thus obtained.
Publishing the content of a human-to-human conversation is different. Because the other person was talking to a particular person, he may have revealed things in the conversation that he wouldn't want to have broadcast. There is a reasonable expectation of discretion there, and a journalist normally informs the other party in advance if he'd be on the record.
So your analogy doesn't hold, because a. the computer didn't establish the other's identity (no username/password protection), b. it was set up to answer anybody who asked. I.e. the journalist didn't misuse the trust placed in him by his conversation partner in order to give information to the public that would otherwise be unavailable.
As to your other question, I have no idea whether's its illegal to publish an unlisted phone number if that doesn't happen as part of publishing a big list of numbers, sorted alphabetically. IANAL either.
data that the plaintive could reasonably have expected to remain hidden?
He could not. If you put something on a/public/, passwordless directory of a webserver, then he has no grounds whatsoever to believe that it would remain hidden.
It has nothing to do with peeping either. There's no 'smaller hole' you have to go through technically in order to obtain the requested document from the server. http://www.company.com/secretreports.html is just as available as http://www.company.com/index.html. Site portals are just yellow pages that help you find those URLs. Am I forbidden to dial a phone number that I didn't find in the phone book?
If you want to protect a secret and assume that something will remain hidden, you need to take/reasonable/ measures./Any/ person with/any/ knowledge of computers and networking will say you/at least/ need username/password protection.
That is the very foundation of the Web...without it we have interactive television.
Sadly, nowadays people tend to say: "And the problem with that is, what?" because their ability to communicate stops little beyond sending output by their remote control and receving input from their TV.
If you see the balance of power between RIAA and consumer as a healthy one, then the government should not have signed the DMCA, which makes it illegal for people to circumvent the attempts of the RIAA to take away your fair use rights.
Even less should it even start to/think/ about mandating any sort of DRM.
In this climate, where the government knows that a sizeable percentage of votes can be bought by manipulative propaganda and other campaign techniques, each of which costs a fair amount of money, studies such as Gartner's, which may get politicians to understand that they can serve their constituent members of the RIAA even better by/not/ mandating or encouraging copy protection crap, are very, very welcome.
Guns *do* have substantial non-harmful uses that cannot be performed by any other device. It's actually less clear that FreeNet meets this standard.
Well, surprise me. I cannot imagine how humanity can benefit more from guns than from communication.
Not that I'd say weapons aren't needed to protect a country's sovereignty, but I truly cannot see how a democratic government's monopoly on warfare is more dangerous to society than unrestricted human communication.
I know that some Americans feel that when the constitution proves too weak to protect democracy (and it's sure starting to look like that, considering the amount of political power held by private corporations), they can use their weapons to re-establish it.
I think that's quite optimistic; since the industrial revolution, people have become so addicted to the abundance of cheap products that they're too easily bribed and too easily manipulated by commercial propaganda to start an armed revolt against any government that keeps them supplied with cheap gas, coca-cola and cable TV.
Personally, I highly doubt that there's any substitute for a government that is bound by its constitution and its constituents to protect democracy at all cost, fighting relentlessly against/all/ concentrations of power/within/ its state borders. (No, USA, power contentrations outside your borders may only be fought when they attack first. That's civilized rule of conduct since 1648).
In my opinion, the tree laws of democracy would seem to be:
1. protect the equal distribution of the state's power among its citizens, 2. do not go to war, unless that conflicts with the first law (eg. when any group, local or foreign, tries to grab the state's power), 3. work in the public interest, but only if that does not conflict with the first or the second law.
Slashdot Mirror
On the mark, except that ssh doesn't use ssl/tls, but is different protocol based on a number of the same cyphers and hashes.
Unless of course you think we should be learning morality from big business?
And that's exactly what's happened in the west and what you see happening in Bhutan at a speed that makes people finally take notice.
The propaganda is crafted to be as strong as possible. It's effective enough to make share holders allow corporations to spend millions on this. You cannot fight this with merely 'explaining' that there's more to life than consuming. The message is simply too strong, too ubiquitious.
Money should not be permitted free speech.
I suspect that once the world is dominated by one country, one culture, one affinity, that we will actually know the truth.
Ah, so the aliens brought weapons and Hollywood?
I personally think the simplest way to solve this in a unix-compatible way is to get rid of the distinction between subdirectories and files.
;)
You have streams, and you have named attributes. Of course, it must be possible to open such a node in two modes, one to read the main stream, and one to read it and the whole tree beneath it as a tar-like stream, but that's nothing a few O_* flags couldn't solve
It needs some work in the semantics area, but it probably needs the most work in the VFS layer.
Is something like that feasible?
Hold your horses, there. The microwaves modulated and there is a pulse sequence in the transmitted frames. In short, there is a strong LF component in cellphone radiation.
If you don't believe me, try putting a GSM close to a stereo and ring it.
I'll be honest, all the "scary GUI changes" Microsoft has made over the years has never phased me once.
Not even that positively frightening Teletubby landscape? Not even the happy, rounded "My First Windows" widgets?
Gah. You really must have no taste.
Of course, the next step would be to interlink robots in such a way that if one robot is observed by another, you see an image of the owner instead of the robot.
At that point, it's only a tiny step away to skip the whole robot and camera business and to interact virtually only.
Of course, I sincerely hope to be dead before this becomes the only practical way of being in contact with people.
However, stating that "nothing can be known in an absolute sense" does not mean that we should not attempt to discover the truth.
True, that's why I have no objection to the scientific method, where one attempts to build a model of cause and effect relationships by creating such hypotheses and testing them. Even when the models generated by this or any other method are not "true", they are at least useful.
However, history is a special branch of science, because you cannot test any hypotheses, like "if the US wouldn't have been so pro-Israel, 9/11 wouldn't have happened", or "if Versailles wouln't have been so humiliating, Hitler wouldn't have happened", and similar.
Therefore, what I'm saying is that because it's not only impossible to find the "truth" about what's behind 9/11 from statements about how likely it is that a certain sequence of events happened, but that it's also impossible to build a scientific theory, because that requires testing hypotheses.
So it's not just solipsism. I'm saying that the scientific method cannot be applied to any conspiracy theory or history in general, and that this makes a productive discussion between skeptic and paranoid extremely difficult.
Of course, I no way am I saynig that that's an excuse to do nothing. But it does mean that the choice which step to take next is always a political one.
Should the US protect itself from the world in response to part of it holding a grudge against the US, or try to increase everybody's participation in the world? It's a political question, and nothing else. Eventually, and that's the point, hypotheses about 9/11 hardly matter. Theories could have mattered, even if they are no truths, but untestable hypotheses are completely useless for the purpose of making political choices.
The problem you seem to ignore is the fundamental one of induction: you can never be sure that two coinciding events have anything to do with each other.
The patterns, or the big picture you speak of, can not be deducted from the events themselves. You can only see them if, as you say, you step back and observe the whole. Are the patterns you'll see indeed the ones that actually caused the events? Possibly, but it's just as likely that they are not.
The human mind always tries to find patterns in the event soup, and by observing the whole, you'll give your mind a chance to do this. Is the resulting model in your mind "true"? It's impossible to know; the only thing you can say is that it may be useful to base further actions upon.
However, the problem with incidents such as September 11 is that it's never possible to refine the hypotheses about the patterns connecting their events, because the incidents do not repeat, and in history, it's impossible to experiment; you cannot change certain variables and keep others the same.
Everyone with a more or less fitting pattern may be "right". That's why you always see a clash between the sceptical and the paranoid that can never be resolved, because the random (says the sceptical) or ordered (says the paranoid) events occurred only once; the hypotheses about the causes behind them cannot be tested.
In short, I'm saying that you cannot deduct anything from what you see when you "step back". The only thing you can create is untestable hypotheses, that you may use personally, and that may have esthetic value for others if they fit well, but are otherwise useless.
That's very likely, yes.
/implications/ could create a perceived minefield, without there actually being one, but I'm sure MS' FUD department would be able to use this piece to their advantage ("See? Even a leading Open Source advocate is aware of the dangers. You better switch to an IP-safe product.").
What's bad is that they sadly got some backing by ESR in this, who says in his article at http://www.opensource.org/sco-vs-ibm.html: "A judgment in favor of SCO could do serious damage to the open-source community. SCO's implication of wider claims could turn Linux into an intellectual-property minefield, with potential users and allies perpetually wary of being mugged by previously unasserted IP claims [...]".
Of course, he's actually saying that SCO's
You can split a sandwich among many people; you cannot digest it in a collective stomach.
/intellectual/ property is different. Smart people understand this.
That's why
You will have a hard time of course because these policies have already been applied to software for decades.
/is/ the product, and it's trivial to steal a copy without anybody even being aware of it.
They have been applied, but nobody has shown any results from applying these policies to software. You go a long way to show that some of these policies have been applied succesfully in other fields of technology, but software is a different thing. Even if you have stolen the blueprints of a plutonium fab, you can't produce any plutonium yet: you need lots of resources, experience and controlled materials to actually build it. Not so for software; the blueprint
Now the problem with open source is that there is no way to control it, so there is no way to implement the kind of policy outlined above, except to kill it (or discourage it), and have everyone use closed source, which can be controlled to a significant degree.
I think that's a harmful illusion. To someone not caring about the law, leave alone EULAs, there is no difference between Open Source and proprietary software in terms of restrictions on use or distribution.
The only thing Open Source makes easier for the non-law abiding citizen than proprietary software is improving the software. Not the use of it.
From section 5. W3C Royalty-Free (RF) Licensing Requirements:
This means that a patent may only be licensed for free if applied in code that implements the W3C recommendation.
So code that implements a patended recommendation may still be non-free in the sense that you cannot go and do with it whatever you want as long as you comply with the author's copyright license; you may violate the royalty-free patent license if you use the code for something other than implementing the W3C standard.
This doesn't mean that the code cannot be licensed under the GPL, it just means that you may need to take a look at the patent status as well before creating a GPL'ed derived work from code that implements a W3C standard.
It's a shame, but still better than RAND or nothing -- IMHO. Having said that, I think the biggest problem remains, which is patented content creation algorithms. You can eg. patent compression and charge shitloads for a license, but make decompression royalty-free. That way, media corporations can still raise a barrier to entry to the "creative commons", which makes patenting harmful and have exact opposite effect from they were designed to achieve.
I suspect that revenue models are a bigger problem, combined with distribution. To earn enough from a game paid for in very small chunks (say a free demo, then paying for new levels), you'd need to be damn sure people would keep buying them. Also, you'd need to be sure that people were honest enough not to just slap then into their P2P apps...
/after/ you released your product, why not have an auction site for game development projects to allow gamers to fund these /before/ the result is released to the world?
Well, instead of trying hard and investing lots of energy to collect pay
A team with a good reputation could outline its plans and say, we need two million euros to cover development costs, our expenses and to make a good living while we're developing this game. Please send your money here. If we don't receive our budget within three months, we cancel the project and pay everyone back 95% of what they contributed (5 % to cover auction costs and living expenses).
This would work for any creative product: literature, software, games, books, movies; it allows you to make money from supplying information without having to supply distribution- or other services, but prevents nasty things like copyrights and licenses.
The only thing is that people who have paid $100 and have received a great game, must learn not to whine anymore when others are playing that game for free.
If they think about it, they have little reason to either, because people who are enthusiastic about this game are likely to help make their favourite authoring team's next, bigger, better production possible.
Another nice property is that gamers are the ones actually investing here, instead of the banks funding initial costs. The latter tend to favour minimal risk over maximal fun and innovation; gamers may choose to strike a slightly different balance.
You just need some searching infrastructure to allow people to find the authors in whose products they want to invest, and a financial institution such as a bank that could be the trusted party to guarantee that people get paid back if not enough money was collected before the deadline.
Why not? Any reasons why this wouldn't work?
Can't agree more. The funny thing is that biological evolution also seems to prefer redundancy and an abundance of simple things over highly elegant, perfectly fitting solutions.
Throwing a load of generic units at the problem may be the easiest way to deal with complexity.
I try to tell people that humans can achieve more than animals, if we'd let go of the law of the jungle concept. We're free to choose, after all. I discuss politics with everyone who wants to. I donate money to the EFF and the FSF. I write Free Software. I preach about free software to my business relations, whenever possible. So it may not be much, but I try, too.
So I gather that in your view of the world, growing up means accepting the current way of doing business, and that means pure darwinism, i.e. everybody just fights as agressively as he can to get what he wants, and is obliged by the investors not to care how that is achieved?
That it is that way is one thing, but calling it immature not to agree with it is quite another.
Which says just as much about C&D letters in general as about this particular one.
Of course, as long as they don't run any servers in the Netherlands, why should they get nervous about what the Dutch authorities consider illegal content?
/my/ computer, that /I/ would become responsible for making sure that information is not sent to , not even upon explicit request from the client.
They are not broadcasting the information or even publishing it in any way, not in the Netherlands. I specifically reach out with my request gor information all the way to the hosting center in the US, which kindly delivers it to me.
Now, if the Dutch authorities want to control what I see, they need to install content-aware firewalls at the national border. But how they can bother google with any request or demand whatsoever is beyond me, *if* google has no servers in the Netherlands that is.
In short, I don't really agree with your reasoning, and I really think it's strange (and worrying) if Google is really afraid of legislation from countries from which it doesn't server any information.
Because that would set a precedent that if I put up certain information on
Which is not only a can of worms, technically and legally, but completely bollocks, IMHO.
Hold your horses, there. It's not context switches in themselves that make IPC inefficient, it's the address space switches, which empty your CPU's TLB.
/outside/ their address space will incur the address space switch penalty for every call to them. But if you choose to run them in the same space, then you don't; context switches are theoretically just as nonexpensive as a setjmp()/longjmp(). It's just "slightly" less safe. If the application crashes, its drivers crash too.
/dev/irq/1 so that you can select() on it, and a modern version of the Unix pipe (one that supports atomic messages and transports them either using zero-copy local memory, shared memory or copying), so that the open/read/write/ioctl/mmap/select/close API can be implemented through it.
Context switches among threads can potentially be very cheap, especially if the switch is not initiated by an interrupt (which would always switch to the kernel initially, before it can schedule a different thread).
It would be great if we'd have a non-application specific way (i.e. not dlopen) to load another image in a certain process' address space and relocate it, DOS-style, and creates a new thread for it. As long as neither application does anything stupid, you won't notice a thing.
That way, for a high speed webserver you could choose to put Apache, your userspace TCP stack, your userspace NIC driver, your userspace SCSI driver and your filesystem driver all in the same process space.
If the open, read, write, close functions of those drivers whould then be implemented using a good IPC channel, that uses local memory when talking to a thread in the same address space (think skbufs inside the kernel) and shared memory or ordinary pipes when talking to other processes, you would be completely free to define your system's compartmentalization.
For differently configured systems, you would only put your TCP/IP driver and your NIC driver in the same address space, and you keep your SCSI driver and filesystem in another.
So, applications using those drivers
It's perhaps a bit radical, but it would give the best of both (microkernel and traditional) worlds, with the ability to choose your own memory protection barriers. QNX has been doing something similar for a while as well; although it still require the application to explicitly load another image (which must always be a shared library, i.e. position independent code, no relocation is done), instead of allowing applications to remain unaware of their actual configuration (shared space or not). It does run the Unix API through a message passing channel though, and all drivers and filesystems are purely user space.
Seriously though. The only things missing for good userspace drivers in Linux is a way to open
As someone pointed out earlier, though, where Reuters might find they've stepped in it is that they published the information they found. IMO, that was dumb, and that's where they overstepped their bounds.
Hmm, if done verbatim, you're probably right. But let's say there is some messaging system that automatically gives some financial numbers when you dial a certain phone number, which, although unlisted, is directly available through the public phone system. In such cases I don't think there's anything wrong legally with publishing information thus obtained.
Publishing the content of a human-to-human conversation is different. Because the other person was talking to a particular person, he may have revealed things in the conversation that he wouldn't want to have broadcast. There is a reasonable expectation of discretion there, and a journalist normally informs the other party in advance if he'd be on the record.
So your analogy doesn't hold, because a. the computer didn't establish the other's identity (no username/password protection), b. it was set up to answer anybody who asked. I.e. the journalist didn't misuse the trust placed in him by his conversation partner in order to give information to the public that would otherwise be unavailable.
As to your other question, I have no idea whether's its illegal to publish an unlisted phone number if that doesn't happen as part of publishing a big list of numbers, sorted alphabetically. IANAL either.
data that the plaintive could reasonably have expected to remain hidden?
/public/, passwordless directory of a webserver, then he has no grounds whatsoever to believe that it would remain hidden.
/reasonable/ measures. /Any/ person with /any/ knowledge of computers and networking will say you /at least/ need username/password protection.
He could not. If you put something on a
It has nothing to do with peeping either. There's no 'smaller hole' you have to go through technically in order to obtain the requested document from the server. http://www.company.com/secretreports.html is just as available as http://www.company.com/index.html. Site portals are just yellow pages that help you find those URLs. Am I forbidden to dial a phone number that I didn't find in the phone book?
If you want to protect a secret and assume that something will remain hidden, you need to take
That is the very foundation of the Web...without it we have interactive television.
/was/ clickable TV!"
Sadly, nowadays people tend to say: "And the problem with that is, what?" because their ability to communicate stops little beyond sending output by their remote control and receving input from their TV.
"Uhuh, I thought Internet
If you see the balance of power between RIAA and consumer as a healthy one, then the government should not have signed the DMCA, which makes it illegal for people to circumvent the attempts of the RIAA to take away
/think/ about mandating any sort of DRM.
/not/ mandating or encouraging copy protection crap, are very, very welcome.
your fair use rights.
Even less should it even start to
In this climate, where the government knows that a sizeable percentage of votes can be bought by manipulative propaganda and other campaign techniques, each of which costs a fair amount of money, studies such as Gartner's, which may get politicians to understand that they can serve their constituent members of the RIAA even better by
As much as I admire your relentless accuracy, I'm worried that the boredom from which you're suffering is approaching dangerous levels.
I'd suggest doing something USEFUL fast, before it really gets the better of you. Just some free advice from another well meaning person.
Guns *do* have substantial non-harmful uses that cannot be performed by any other device. It's actually less clear that FreeNet meets this standard.
/all/ concentrations of power /within/ its state borders. (No, USA, power contentrations outside your borders may only be fought when they attack first. That's civilized rule of conduct since 1648).
Well, surprise me. I cannot imagine how humanity can benefit more from guns than from communication.
Not that I'd say weapons aren't needed to protect a country's sovereignty, but I truly cannot see how a democratic government's monopoly on warfare is more dangerous to society than unrestricted human communication.
I know that some Americans feel that when the constitution proves too weak to protect democracy (and it's sure starting to look like that, considering the amount of political power held by private corporations), they can use their weapons to re-establish it.
I think that's quite optimistic; since the industrial revolution, people have become so addicted to the abundance of cheap products that they're too easily bribed and too easily manipulated by commercial propaganda to start an armed revolt against any government that keeps them supplied with cheap gas, coca-cola and cable TV.
Personally, I highly doubt that there's any substitute for a government that is bound by its constitution and its constituents to protect democracy at all cost, fighting relentlessly against
In my opinion, the tree laws of democracy would seem to be:
1. protect the equal distribution of the state's power among its citizens,
2. do not go to war, unless that conflicts with the first law (eg. when any group, local or foreign, tries to grab the state's power),
3. work in the public interest, but only if that does not conflict with the first or the second law.