In the meantime, the only GPL-like license that actually closes the web services loophole (the Affero GPL), which is mentioned as a template for the GPLv3, ISNT GPL compatible!
As far as we know, a license that "closes the web services loophole" can neither be GPL compatible, nor a free software license. This "loophole" is necessary for a work to be free software; any license clause that attempts to "close" it will cause somebody to lose the ability to modify the work (exactly who depends on the clause). For example, the Affero GPL prohibits you from modifying and using the work if you do not have an HTTP server available (and many, many other cases).
Most creative types don't say to themselves, "I'm about to invest a couple-plus years of my life writing Cryptonomicon... but I'd better keep sucking up to my boss at the IHOP because my cultural history tells me I shouldn't expect a paycheck from book royalties, ever."
Obviously you are not a writer. Most of them do have to keep their day jobs. You shouldn't expect a paycheck from book royalties, because you probably aren't going to get one any time soon, and if you ever do, it's not going to be very much. Only the top end of the scale get paid enough to support themselves.
Yes, and it's also the same EDS who routinely wins UK government contracts by putting in the best bid, and then routinely drives the project into the ground, goes over budget, over deadline, and in the end still fails to deliver. They have been directly responsible for a disproportionately large number of the technology cock-ups in the UK public sector.
You throw a c/java/python/etc. program at a programmer from another language and they'll get the basic idea.
Oh really?
Care to explain this? I spent an hour or two deciphering it last week:
while ((low ^ (low+range)) < TOP || range < BOT &&
((range= -low & (BOT-1)),1)) {
code=(code << 8) | getc();
range <<= 8; low <<= 8;
}
There's above five thousand more lines of similar code in that one.
I've seen similar crap in java and python as well, and at least as often as I've seen it in perl. Bad programming is language-agnostic.
You throw a perl script written by somebody who is not crap at programming at somebody and they'll be able to see how it works. You give them anything written by a chucklehead like the one who wrote the above, and they're screwed regardless of language.
Funny that at a time when the rest of the world is actively privatizing various parts of their formerly-public economy in recognition of the fall of "communism" (actually socialism in practice, but communism in ideal) -- such as in Britain
Not really anything to do with communism or politics. Stuff in the UK got privatised and sold off simply to make the budget fit, with a government that was spending more money than it had.
It was a disaster all round. Everybody knows it was a disaster, and it was widely predicted to be a disaster when it happened. The resulting monopoly-based private industries are far worse than they ever were as public services. It'll probably get the current government unelected at the next election.
It's definitely a step in the right direction that Lessig has codified the Creative Commons license...
Creative Commons is a giant step backwards, because it's taken all the people who might have been interested in creating free non-software works, and persuaded them to release their stuff under a non-free, GPL-incompatible license, such that all of us working on free software can't really make any use of it. All of the Creative Commons version 2 licenses are broken and should be avoided, just like the GNU FDL, the OSL, etcetera. You can see a fairly complete summary of all the problems found so far at http://people.debian.org/~evan/ccsummary
This should not come as a great surprise. Have a look at the people responsible for creative commons. Note how they're mostly lawyers and corporate types. This is not a grassroots effort, despite the way they allude to being related.
Creative Commons has shown no interest in fixing their licenses. Consider why that might be.
and figure that the efforts to reduce global warming, even if they do nothing, are unlikely to have a significant negative impact
I'm pretty sure that abandoning all cars and airplanes will have a significant negative impact. You might want to be more careful about saying things like "efforts to reduce global warming". People have proposed all kinds of stuff, and none of it has turned into actual effort in the real world yet.
On the other hand, I'm pretty sure that if the guesses supporting carbon dioxide as responsible happen to be correct, then reducing global carbon dioxide production by 2% will not have a significant impact at all, and the Kyoto protocol is a load of hot air.
The "right" to break the law without having someone look over your shoulder?
No, the right to not break the law without having someone look over your shoulder. If you are willing to break the law then you don't have to have someone look over your shoulder - you can just ditch the tracker. These things can only track honest people. Note that people are only allowed out on probation if it is reasonably believed that they will not break the law again. We're not talking career criminals here, we're talking about people who made one stupid mistake and will now be persecuted for the next decade or so. People exactly like you, the reader, with the one difference that they got caught, and you probably haven't been caught yet (there's no real chance that you've never committed a crime, with the proliferation of laws these days).
This is the same tired old fundamental problem. All these sorts of ideas fail in the same way: they forget that criminals do not obey the law, by definition. Anything which relies on criminals obeying the law is idiotic. (The objectionable parts of the DMCA are a good example of laws based on this kind of idiocy)
There is now an overwhelming body of real science on this issue to be read if you just go out and look for it.
However, it mostly says "Maybe. We're not sure. Oh, and here's a few other theories, and we're not sure about those either. One of them could be true, or none of them, ask us again in a couple hundred years".
Nobody has any answers to this puzzle, just a whole bunch of questions.
The pro-firearms people always say that "guns don't kill people, people do" as their main defense. But same applies to cars, and still driving a car is not a constitutional right, but a special privilege granted only for those, who have obtained a driver's license. And yet killing living things -- including, but not limited to humans -- is the primary function guns are actually designed for, but this is definitely not the case with cars.
On the other hand, it's worth noting that gun control doesn't actually help much. The UK effectively banned all guns a few years ago. It more or less eliminated gun crime - that's not a surprise. It had little effect on violent crime, as people just shifted to using knives and stuff instead. That's not really a surprise either. The majority of violent crime is domestic.
The problem with global warming instead just seems to be that when people are not absolutely sure that global warming happens and that CO2 emissions caused by humans are actually contributing to it, they are willing to do nothing, as they feel that the preventative measures are too expensive to take without certainty of their necessity and effectiveness.
There's a difference between doing things because global warming might be occurring, and you don't want to take the risk, and doing things because you know that driving cars makes the world flood. In the latter case the solution is simple and obvious: fix the damn cars. But since there isn't any particularly compelling evidence to support the theory that car emissions are causing global warming (just a whole bunch of speculation), mindlessly pursuing that answer could easily be as bad.
The thing that global warming pundits often fail to consider is: what if they're wrong? What if the world is getting hotter, but they have the wrong reasons? What if the world is actually going to get a lot colder, and their purported claims would make things worse? Both of these options have plenty of arguments in their favour that are about as convincing. Or it could be something else entirely that we haven't thought of yet - and given humanity's track record at this sort of thing, this is the most likely one.
Maybe reducing emissions will make things better. Maybe it will make them much worse. Most likely it will do nothing; it's sheer hubris to think that humans can affect anything as big as a planet. But the risk in both directions is there, and advocating one of them simply because you're ignoring the other is not really a good idea.
Redhat has an OK system, but Microsoft has a nice tool (software update services) that allows me to download the patches in one place and push them out to all the machines on my network. This will only get better when MS releases the next update to this tool (windows update services).
I haven't seen a similar thing from any of the linux vendors.
You tell your sysadmin, and he sets it up. This sort of thing is utterly trivial for a unix sysadmin. That's what you're paying him for. The vendors don't bother to include useless fluff like that.
This is one of the reasons I will never buy a console. Console games are geared to be throw-away games. i.e. You spend $50 on a FPS, and you are stuck with whatever maps the publisher sees fit to let you have.
Well, yes, that's true for multiplayer FPS games on consoles, which are such a blatantly stupid idea that only Microsoft are really pushing them. The reason for consoles is to play Final Fantasy; all else is just cost amortisation. People always seem to forget that nowadays.
Again, more simply: consoles are for plot-driven games.
Right, as opposed to somebody who has spent their time in a senior management position at a company whose business model is to buy successful game companies and run them into the ground, then scrap them. Remember folks, EA does not do 'original' or 'creative'. EA does 'asset liberation'.
I think projects of that scope should stage such large developments, start with a general specification for the system and the desired end result and interoporability, then develop and roll-out modules progressively.
Oh, they do that often enough. Then you find that you've developed and debugged all the core modules and you have run out of time and money without actually producing any new functionality.
This project is classified as a failure.
Modularity doesn't help. Monolithic construction doesn't help. Only budgeting time and money correctly, and then producing the project within budget, will help. The problem is corruption and incompetence; different design approaches, regardless of the approach, are not going to solve it. It's not like the people doing the work have any particular incentive to succeed anyway; they'll get paid regardless, and, contrary to what capitalist propaganda would have you believe, they'll get the next contract regardless. Market forces are unreliable at the best of times, and they do not function at all for government contracts, which are awarded for political reasons and nothing else.
The problems are not new or poorly understood. Nor are the solutions difficult to see. But there is absolutely no interest in implementing them. This is politics, not engineering. Success is not important to the politicians who make the decisions. Rather than expending resources to succeed, they can lie to you and tell you that they want the project to succeed, for the same effect on their reelection prospects. There's usually more going on than this, but in the trivial case it comes down to a choice between company A, who will succeed for a given amount of money, and company B, who will fail for much less money. The politician will choose company B every time, because failure doesn't matter to them; it won't happen for ages, and it's not their fault.
Since major companies like IBM have chosen to produce compilers that perform best with FORTRAN.
It's not a choice. Fortran is easier to compile efficiently in a number of respects, that's all. Notably, it doesn't have anywhere near the amount of trouble that C compilers do with pointer aliasing, which can make a significant difference for scientific code.
That's not to say that fortran is a good language. However, we still don't have any more modern language that does any better in this respect, which is kinda pathetic really.
We're talking about the same EU which has not managed to balance its accounts in ten years. That doesn't mean they lost money in the sense of not making a profit - it means they lost money in the sense of "money has gone and they do not know where it went". That fine is going to be embezzeled, of course.
> the world works in a hierarchical fashion not because it can but because in fact this works well
...
> But it costs too much to identify these folks
What you have actually demonstrated here is that capitalism implies hierarchial structure. This is neither new nor particularly insightful; it is an elementary tenet of anarchism that capitalism is the cause of this. That doesn't mean it works well, it just means that capitalist societies will inevitably do it. It's a demonstration of why capitalism sucks.
Proprietary software invariably *sucks*. Proprietary firmware invariably *sucks*. I have lost count of the number of proprietary things I've seen which badly needed obvious, simple improvements in the way they worked. Cameras are no exception. Especially the interface. There's no excuse for them being that clumsy. I should *not* be waiting a perceptable length of time for a response to a button push, even if it starts some long-running operation.
Note that the absence of these flaws in *your* camera is not an excuse for this, because most of the consumer ones *do* have these flaws.
So, what is wrong with people doing a good job and creating value, getting a little bit of that value back so they can pay rent ?
That would be the part where they make it impossible for anybody else to develop the thing any further, so that it suits their needs when the original developer has no interest in them. There's a reason why proprietary software sucks.
This is a classic one. Take a group of people with widely divided opinions on an issue. Take the point you want to prove, stated in a sufficiently broad manner. Take the inverse of that point.
Now, divide the group into people who support your point and people who support the inverse, without asking them directly, but rather based on reading related things which they have written.
The fallacy here is that most of them actually haven't stated an opinion on your point at all - you have excluded the middle possibility that the response might be "no comment", and redistributed this group between the extreme cases. It's not actually in "How to lie with statistics" but it could have been.
Most of these papers say things like "Here is an interesting correlation that I found". These are valuable scientific papers, but it must be understood that correlation does not imply a connection. It merely implies that there is something here worth investigating and explaining. Most of the people who wrote those papers are fully aware of this and were not stating an opinion on what was going on - they were just offering something interesting that they discovered. History teaches us that correlations in the physical sciences are occasionally connections but more commonly they are things which we don't understand at all.
You can't prove anything outside a laboratory. (Or more accurately, it's not proof unless it's possible to replicate in a laboratory by anybody with the right tools).
You can argue all day about what you want people to do (personally, I don't care), but you're going to have to do it without proof. Because there isn't any, and we can't pack a solar system into a laboratory to investigate it.
No one in their right mind would start a new project with this patent-apocalypse clause. With all the patent litigation lately it would reflect poorly on Sun to kick off such a large project using a poorly considered license that doesn't cover the legal issues developers face today.
Most new OSI-approved licenses seem to be ill-considered.
Of course, if you (or anyone else here for that matter) are complaining without actually knowing the implications of this license, you should consider the number of people who would be quite happy to annihilate free software projects by colliding them with their own, at the loss of both projects. That's what these clauses really do. They don't provide any protection, they just give mutually assured maximum destruction.
OpenSSL. Many IRCds and clients these days support encryption.
Speaking both as an author of an ircd and somebody with a comprehensive understanding of what SSL does: worthless. It is no defence at all against an intelligent attacker. It has all the same problems as when it is deployed with http. Read CRYPTO-GRAM and some of Schneier's books.
If I were just a little more paranoid then I would suggest that the CIA started this SSL-frenzy in order to make people feel more secure while standardising the method for intercepting their communication.
You need to define "derivative". You used it a lot in your post, but it doesn't address the core problem: can I or can I not link to a GPL library with my application.
That is for the courts to decide. You just have to make a guess that you're willing to defend in court if you have to.
Copyright law does not address linking, but it does define "derivative work". The mere act of linking (or otherwise calling a function by its signature) is not sufficient to create derivation.
That is what I said. The mere method by which your program uses the library is irrelevant. Derivation is an intrinsic property of the code, or not, as the case may be.
Mindlessly linking a library that you do not need is not derivation. Building an application that includes features from some other code is derivation, regardless of whether or not that code is in a library. Those are the clear cases in both directions; anything between these two is necessarily going to be difficult to define, because that's how copyright works. The reason we have the courts is to resolve questions like this.
In all cases, the method by which copyright is infringed does not matter. The question is merely whether it is infringed, or not, and whether the license permits this form of infringement. The GPL is a set of simple rules about what you have to do for each type of infringement (distribution, derivation, etc); the definition of what constitutes each type is a matter of law.
Slashdot Mirror
In the meantime, the only GPL-like license that actually closes the web services loophole (the Affero GPL), which is mentioned as a template for the GPLv3, ISNT GPL compatible!
As far as we know, a license that "closes the web services loophole" can neither be GPL compatible, nor a free software license. This "loophole" is necessary for a work to be free software; any license clause that attempts to "close" it will cause somebody to lose the ability to modify the work (exactly who depends on the clause). For example, the Affero GPL prohibits you from modifying and using the work if you do not have an HTTP server available (and many, many other cases).
Most creative types don't say to themselves, "I'm about to invest a couple-plus years of my life writing Cryptonomicon... but I'd better keep sucking up to my boss at the IHOP because my cultural history tells me I shouldn't expect a paycheck from book royalties, ever."
Obviously you are not a writer. Most of them do have to keep their day jobs. You shouldn't expect a paycheck from book royalties, because you probably aren't going to get one any time soon, and if you ever do, it's not going to be very much. Only the top end of the scale get paid enough to support themselves.
Yes, and it's also the same EDS who routinely wins UK government contracts by putting in the best bid, and then routinely drives the project into the ground, goes over budget, over deadline, and in the end still fails to deliver. They have been directly responsible for a disproportionately large number of the technology cock-ups in the UK public sector.
Oh really?
Care to explain this? I spent an hour or two deciphering it last week:There's above five thousand more lines of similar code in that one.
I've seen similar crap in java and python as well, and at least as often as I've seen it in perl. Bad programming is language-agnostic.
You throw a perl script written by somebody who is not crap at programming at somebody and they'll be able to see how it works. You give them anything written by a chucklehead like the one who wrote the above, and they're screwed regardless of language.
This is actually pretty clever stuff, the banks can be many things, but they're not stupid, you don't last long in financial circles if you are.
They didn't arrive at this by design. They arrived at it by exhaustion, endlessly trying different schemes and having them cracked.
Funny that at a time when the rest of the world is actively privatizing various parts of their formerly-public economy in recognition of the fall of "communism" (actually socialism in practice, but communism in ideal) -- such as in Britain
Not really anything to do with communism or politics. Stuff in the UK got privatised and sold off simply to make the budget fit, with a government that was spending more money than it had.
It was a disaster all round. Everybody knows it was a disaster, and it was widely predicted to be a disaster when it happened. The resulting monopoly-based private industries are far worse than they ever were as public services. It'll probably get the current government unelected at the next election.
It's definitely a step in the right direction that Lessig has codified the Creative Commons license...
Creative Commons is a giant step backwards, because it's taken all the people who might have been interested in creating free non-software works, and persuaded them to release their stuff under a non-free, GPL-incompatible license, such that all of us working on free software can't really make any use of it. All of the Creative Commons version 2 licenses are broken and should be avoided, just like the GNU FDL, the OSL, etcetera. You can see a fairly complete summary of all the problems found so far at http://people.debian.org/~evan/ccsummary
This should not come as a great surprise. Have a look at the people responsible for creative commons. Note how they're mostly lawyers and corporate types. This is not a grassroots effort, despite the way they allude to being related.
Creative Commons has shown no interest in fixing their licenses. Consider why that might be.
and figure that the efforts to reduce global warming, even if they do nothing, are unlikely to have a significant negative impact
I'm pretty sure that abandoning all cars and airplanes will have a significant negative impact. You might want to be more careful about saying things like "efforts to reduce global warming". People have proposed all kinds of stuff, and none of it has turned into actual effort in the real world yet.
On the other hand, I'm pretty sure that if the guesses supporting carbon dioxide as responsible happen to be correct, then reducing global carbon dioxide production by 2% will not have a significant impact at all, and the Kyoto protocol is a load of hot air.
Of course, who wants to be on the side of ignoring or supporting the widespread destruction of the planet by humans? Therein likes the rub...
The real tricky point is, which side is that? Just because one currently has the approval of the media doesn't make them right.
The "right" to break the law without having someone look over your shoulder?
No, the right to not break the law without having someone look over your shoulder. If you are willing to break the law then you don't have to have someone look over your shoulder - you can just ditch the tracker. These things can only track honest people. Note that people are only allowed out on probation if it is reasonably believed that they will not break the law again. We're not talking career criminals here, we're talking about people who made one stupid mistake and will now be persecuted for the next decade or so. People exactly like you, the reader, with the one difference that they got caught, and you probably haven't been caught yet (there's no real chance that you've never committed a crime, with the proliferation of laws these days).
This is the same tired old fundamental problem. All these sorts of ideas fail in the same way: they forget that criminals do not obey the law, by definition. Anything which relies on criminals obeying the law is idiotic. (The objectionable parts of the DMCA are a good example of laws based on this kind of idiocy)
There is now an overwhelming body of real science on this issue to be read if you just go out and look for it.
However, it mostly says "Maybe. We're not sure. Oh, and here's a few other theories, and we're not sure about those either. One of them could be true, or none of them, ask us again in a couple hundred years".
Nobody has any answers to this puzzle, just a whole bunch of questions.
The pro-firearms people always say that "guns don't kill people, people do" as their main defense. But same applies to cars, and still driving a car is not a constitutional right, but a special privilege granted only for those, who have obtained a driver's license. And yet killing living things -- including, but not limited to humans -- is the primary function guns are actually designed for, but this is definitely not the case with cars.
On the other hand, it's worth noting that gun control doesn't actually help much. The UK effectively banned all guns a few years ago. It more or less eliminated gun crime - that's not a surprise. It had little effect on violent crime, as people just shifted to using knives and stuff instead. That's not really a surprise either. The majority of violent crime is domestic.
The problem with global warming instead just seems to be that when people are not absolutely sure that global warming happens and that CO2 emissions caused by humans are actually contributing to it, they are willing to do nothing, as they feel that the preventative measures are too expensive to take without certainty of their necessity and effectiveness.
There's a difference between doing things because global warming might be occurring, and you don't want to take the risk, and doing things because you know that driving cars makes the world flood. In the latter case the solution is simple and obvious: fix the damn cars. But since there isn't any particularly compelling evidence to support the theory that car emissions are causing global warming (just a whole bunch of speculation), mindlessly pursuing that answer could easily be as bad.
The thing that global warming pundits often fail to consider is: what if they're wrong? What if the world is getting hotter, but they have the wrong reasons? What if the world is actually going to get a lot colder, and their purported claims would make things worse? Both of these options have plenty of arguments in their favour that are about as convincing. Or it could be something else entirely that we haven't thought of yet - and given humanity's track record at this sort of thing, this is the most likely one.
Maybe reducing emissions will make things better. Maybe it will make them much worse. Most likely it will do nothing; it's sheer hubris to think that humans can affect anything as big as a planet. But the risk in both directions is there, and advocating one of them simply because you're ignoring the other is not really a good idea.
Redhat has an OK system, but Microsoft has a nice tool (software update services) that allows me to download the patches in one place and push them out to all the machines on my network. This will only get better when MS releases the next update to this tool (windows update services).
I haven't seen a similar thing from any of the linux vendors.
You tell your sysadmin, and he sets it up. This sort of thing is utterly trivial for a unix sysadmin. That's what you're paying him for. The vendors don't bother to include useless fluff like that.
Well, yes, that's true for multiplayer FPS games on consoles, which are such a blatantly stupid idea that only Microsoft are really pushing them. The reason for consoles is to play Final Fantasy; all else is just cost amortisation. People always seem to forget that nowadays.
Again, more simply: consoles are for plot-driven games.
Right, as opposed to somebody who has spent their time in a senior management position at a company whose business model is to buy successful game companies and run them into the ground, then scrap them. Remember folks, EA does not do 'original' or 'creative'. EA does 'asset liberation'.
I think projects of that scope should stage such large developments, start with a general specification for the system and the desired end result and interoporability, then develop and roll-out modules progressively.
Oh, they do that often enough. Then you find that you've developed and debugged all the core modules and you have run out of time and money without actually producing any new functionality.
This project is classified as a failure.
Modularity doesn't help. Monolithic construction doesn't help. Only budgeting time and money correctly, and then producing the project within budget, will help. The problem is corruption and incompetence; different design approaches, regardless of the approach, are not going to solve it. It's not like the people doing the work have any particular incentive to succeed anyway; they'll get paid regardless, and, contrary to what capitalist propaganda would have you believe, they'll get the next contract regardless. Market forces are unreliable at the best of times, and they do not function at all for government contracts, which are awarded for political reasons and nothing else.
The problems are not new or poorly understood. Nor are the solutions difficult to see. But there is absolutely no interest in implementing them. This is politics, not engineering. Success is not important to the politicians who make the decisions. Rather than expending resources to succeed, they can lie to you and tell you that they want the project to succeed, for the same effect on their reelection prospects. There's usually more going on than this, but in the trivial case it comes down to a choice between company A, who will succeed for a given amount of money, and company B, who will fail for much less money. The politician will choose company B every time, because failure doesn't matter to them; it won't happen for ages, and it's not their fault.
Since major companies like IBM have chosen to produce compilers that perform best with FORTRAN.
It's not a choice. Fortran is easier to compile efficiently in a number of respects, that's all. Notably, it doesn't have anywhere near the amount of trouble that C compilers do with pointer aliasing, which can make a significant difference for scientific code.
That's not to say that fortran is a good language. However, we still don't have any more modern language that does any better in this respect, which is kinda pathetic really.
We're talking about the same EU which has not managed to balance its accounts in ten years. That doesn't mean they lost money in the sense of not making a profit - it means they lost money in the sense of "money has gone and they do not know where it went". That fine is going to be embezzeled, of course.
> the world works in a hierarchical fashion not because it can but because in fact this works well
...
> But it costs too much to identify these folks
What you have actually demonstrated here is that capitalism implies hierarchial structure. This is neither new nor particularly insightful; it is an elementary tenet of anarchism that capitalism is the cause of this. That doesn't mean it works well, it just means that capitalist societies will inevitably do it. It's a demonstration of why capitalism sucks.
You forgot the most important one:
0. Turn it into a better camera
Proprietary software invariably *sucks*. Proprietary firmware invariably *sucks*. I have lost count of the number of proprietary things I've seen which badly needed obvious, simple improvements in the way they worked. Cameras are no exception. Especially the interface. There's no excuse for them being that clumsy. I should *not* be waiting a perceptable length of time for a response to a button push, even if it starts some long-running operation.
Note that the absence of these flaws in *your* camera is not an excuse for this, because most of the consumer ones *do* have these flaws.
That would be the part where they make it impossible for anybody else to develop the thing any further, so that it suits their needs when the original developer has no interest in them. There's a reason why proprietary software sucks.
This is a classic one. Take a group of people with widely divided opinions on an issue. Take the point you want to prove, stated in a sufficiently broad manner. Take the inverse of that point.
Now, divide the group into people who support your point and people who support the inverse, without asking them directly, but rather based on reading related things which they have written.
The fallacy here is that most of them actually haven't stated an opinion on your point at all - you have excluded the middle possibility that the response might be "no comment", and redistributed this group between the extreme cases. It's not actually in "How to lie with statistics" but it could have been.
Most of these papers say things like "Here is an interesting correlation that I found". These are valuable scientific papers, but it must be understood that correlation does not imply a connection. It merely implies that there is something here worth investigating and explaining. Most of the people who wrote those papers are fully aware of this and were not stating an opinion on what was going on - they were just offering something interesting that they discovered. History teaches us that correlations in the physical sciences are occasionally connections but more commonly they are things which we don't understand at all.
You can't prove anything outside a laboratory. (Or more accurately, it's not proof unless it's possible to replicate in a laboratory by anybody with the right tools).
You can argue all day about what you want people to do (personally, I don't care), but you're going to have to do it without proof. Because there isn't any, and we can't pack a solar system into a laboratory to investigate it.
No one in their right mind would start a new project with this patent-apocalypse clause. With all the patent litigation lately it would reflect poorly on Sun to kick off such a large project using a poorly considered license that doesn't cover the legal issues developers face today.
Most new OSI-approved licenses seem to be ill-considered.
Of course, if you (or anyone else here for that matter) are complaining without actually knowing the implications of this license, you should consider the number of people who would be quite happy to annihilate free software projects by colliding them with their own, at the loss of both projects. That's what these clauses really do. They don't provide any protection, they just give mutually assured maximum destruction.
OpenSSL. Many IRCds and clients these days support encryption.
Speaking both as an author of an ircd and somebody with a comprehensive understanding of what SSL does:
worthless. It is no defence at all against an intelligent attacker. It has all the same problems as when it is deployed with http. Read CRYPTO-GRAM and some of Schneier's books.
If I were just a little more paranoid then I would suggest that the CIA started this SSL-frenzy in order to make people feel more secure while standardising the method for intercepting their communication.
It is security theatre. Not security.
You need to define "derivative". You used it a lot in your post, but it doesn't address the core problem: can I or can I not link to a GPL library with my application.
That is for the courts to decide. You just have to make a guess that you're willing to defend in court if you have to.
Copyright law does not address linking, but it does define "derivative work". The mere act of linking (or otherwise calling a function by its signature) is not sufficient to create derivation.
That is what I said. The mere method by which your program uses the library is irrelevant. Derivation is an intrinsic property of the code, or not, as the case may be.
Mindlessly linking a library that you do not need is not derivation. Building an application that includes features from some other code is derivation, regardless of whether or not that code is in a library. Those are the clear cases in both directions; anything between these two is necessarily going to be difficult to define, because that's how copyright works. The reason we have the courts is to resolve questions like this.
In all cases, the method by which copyright is infringed does not matter. The question is merely whether it is infringed, or not, and whether the license permits this form of infringement. The GPL is a set of simple rules about what you have to do for each type of infringement (distribution, derivation, etc); the definition of what constitutes each type is a matter of law.