"It isn't a matter of right or wrong. It is a matter of HOW right and HOW wrong.
Newtonian mechanics is right enough for most everyday living"
Well, ptolemaic astronomy is right enough for most everyday living too, just as much as newtonian mechanics.
In fact, now that you told about forgetting GPS, you probably know that ptolemaic astronomy not only is good enough to navigate your way all around the world, by land, boat or plane, but the way navigation is tought to pilots and ship captains *is* ptolemaic, not copernican, right?
Why we don't stop teaching about Copernicus, then?
Yes, it is. Other ancient theories are crazily maddining wrong and certainly Newton's Principia is a shinning cathedral honoring the human intelligence but it still is utterly wrong.
Good you mention Asimov, since he was quite on the ontological path (against the pure mathematical path ala Dirac).
Now, forget about the numbers: it's about quality, not quantity. Newton thinks that there exists an absolute coordinate system and that things like mass, speed, length or time are therefore also absolute. Einstein demonstrates that he can't be any more wrong.
Ptolemy thought that the Earth is in the center of the universe and that objects in the sky circle around it by means of a dance of composed circles (epycicles and deferents) and it offers a magnificent math that "it's mostly right with many decimal places" and his model is also a magnificent show of human ingenuity. The problem with ptolemaic astronomy are not the numbers -ptolemaic astronomy can offer very precise numbers; it is the axioms: the Earth is not even near to the center of the solar system and there's no specific reason for orbits to be exclusively based on circles, so it is not a matter of how good its numbes are, just like there's no absolute coordinate system as Newton thought, no matter how good his numbers are.
As I see it, it's not that it would be mindblowing for the students to understand the basics of relativity or quantum mechanics but a matter of laziness from the teachers to find the proper way to teach them instead of "doing it as it has always been done".
"Well, it is one way to eliminate compiler bugs. By definition the compiler doesn't have any."
Yes, *a* compiler can be considered to be bug free per definition. But then, two different compilers will render two different outputs. What's the good one then?
IParent is not specifically talking about quantum mechanics, just non-classic physics. Special Relativity paper comes from 1905 and the general one, 1915.
Classical physics can be pointed back to Newton: 1687.
"is basically "correct" for 99.999%"
It is not. It is utterly wrong. It just happens to throw the right numbers most of the time.
"until then the classic approximation is pretty good for high school work."
It is not. It would be much better to explain non-classical even without the maths (but two dimensional and statistical approach can be offered since the maths are in the curriculum) and throw the classic maths as what they are: a (very) useful approximation.
"Not necessarily, anyone can contact the software authors"
Key word being "can" which is neither "should" nor "must". The user either contacts for renegotiation/clarification of the license, in which case there will be a paper track and it will be the paper what counts, or the user won't contact, in which case, it will be the paper what counts.
In neither case the licensor will have the chance to further explain his intent to the court.
"The reason what licensor actually said that counts is because the licensor is the entity that initiates legal proceedings to remedy perceived violations of the licence."
Taken to the extreme, the only thing that counts is what the court says about the meaning of the paper but, being not so extreme, the court will hear the meaning both the licensor and the licensee understand out of the paper and, as long as both interpretations are fair, it will be the interpretation of the licensee -NOT the licensor, the one that will be supported by the court.
"Any way the terms of the GPL are essentially a non-concern until you distribute."
It depends on "who" you are. If I get a GPL-released binary it will be me the one asking for the sources while still I didn't distribute anything.
"you can transfer that codebase to a thousand different aystems, windows, Linux, anything you want, you could even run it on a decemal based computer instead of digital. The program as it is written in an language would run the same every time."
Except, well, you know, it doesn't.
I don't know where your computer expertise comes from, but you can bet is not as exhaustive as you think it is.
"I saw a few companies adopt FreeBSD over Linux for that reason, back when the SCO UNIX case was going on, due to perceived legal risk."
Which is quite funny, since the first strong push for Linux instead of FreeBSD was because of the percived legal risk of BSD, back when the USL vs BSDi was going on.
"It doesn't matter what the aurthor of the GPL meant, it's what the licensor actually said that counts"
No, it isn't. What the licensor actually "said" is the text of the GPL itself, since that's how the work is licensed.
But then, since it is a license, a kind of contract that you don't get the chance to negotiate, it is not the interpretation of the licensor the one that counts, but the interpretation of the licensee, within reasonable bounds.
"The creators will tell you what it was intended to do and what it actually means within that narrow GPL context."
Too late. The contract is already signed and the license agreed upon.
"That is what is important and that's why we have the so called fine print in all agreements I have come across. No?"
No. It is the perfect opposite: the "fine print" is there exactly because you are *not* going to ask the text author about what was the intended meaning. The meaning is what the text says and no more.
Since it is a license, court will have to be liberal on the interpretation of the party subjected to the license, *not* the party producing it.
Luckily, Mr Moglen did a good job with the GPLv2 redaction and the intention will prevail.
"Why? They tried to squeeze money out of Google and failed; charging a nominal fee serves no purpose."
That they tried to squeeze money from Google is the stated true in press they even called the law "The Google Levvy". What they don't tell is that this isn't their only purpose. The real purpose of the law is to silent all voices not within AEDE. In Spain is not the press that it is the fourth power, AEDE is, and they want to stay that way.
"Most European cultures including Spain do not have such a culture. Instead, they would likely prefer to negotiate with google on the issue."
You just don't understand the issue here.
This legislation is so utterly stupid (I'm Spanish), so utterly against everybody but AEDE members (you see, they don't even represent the press but only the six biggest press companies), that everybody except AEDE and their government mates where *prying* for Google to stand its current position as the only hope for the law to be dismissed.
"Publishers do not have legislative powers. Democratically elected leaders do."
Have you thought a bit about what the "intellectual propiety business" involves, moreso when talking about levvies?
It basically gets down to the collecting agency sitting down waiting for the buckets full of money to arrive.
This means two things: 1) They have quite a lot of money. 2) They don't need to work a hard day for the money to arrive so they have time to do other things instead... like lobbying the government making use of the lot of money they gather (see one).
And then, let's talk about the subset of IP business we are talking here: press. They not only have the money and the inclination to lobby the government, they also control the tool politics are most afraid of: press. Now put together these three things and see that, yes, publishers have indeed factual legislative powers.
"All we have to do is figure out who is trying to extort whom."
AEDE, no doubt.
Their point is that they forcibly want a levvy for the "privilege" of indexing their headlines because "they are providing profit to the indexers". They obviously "forget" about the more or less 30% of traffic that comes to them straigth from google news.
Google can't just stop indexing AEDE members in order to avoid the tax because the levvy is for everyone in that activity no matter what, so they decided that since they don't want to pay, they'll close their doors in Spain.
And then, all of a sudden, AEDE members "remember" about that 30% of traffic that Google news is providing them and claim kindof "monopoly" on the part of Google so the government somehow should force Google to stay open and pay the levvy nevertheless.
AEDE (news), CEDRO (books), EGEDA (films) and SGAE (music) are just the Spanish versions of your RIAA, just so you know what kind of people are we talking about.
"The wording of the GPL is quite clear - it only requires the Makefiles to be included, and even adds an exception for the compiler when included with the OS as a runtime dependency. It doesn't say anything about the requirement to include the compiler."
In fact, it does. It adds an *exception* to include the compiler when it is already included and easily avaliable along the OS. This implies that when this is not the case, the compiler should be included since otherwise the source can't stand alone.
Anyway, I see your point. That's why I said "it is either not allowed by GPLv2 or a gap into the license that should be closed"
The vendor says: "The firmware of our modules is written in Pic Basic Pro [...] We provide all source code free of charge (GPLv2 licence)). The Pic Basic Pro compiler however is not for free"
And GPLv2 says:
"The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable. However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable."
The obvious intent of the GPL is for you to get a code in a way that allows you to work with it and get results. What this guys do is against this intent. It is either not allowed by GPLv2 or a gap into the license that should be closed but, in any case, this is not kosher.
"There is no allusion to any other model having such a limitation. This is down to, and I paraquote from the movie, the inevitability of the models developing a conscience."
A conscience like... hummm... Deckard?
Yes, Deckard's boss familiariy implies he has known him several years. And the script implies Deckard has been into his retirement some several years more. Several years + several years > 4, which is a no go for a conscient replicant.
And then, there's the meta-analisys of the script. It shows machines (replicants) becoming humans at the same time that humans become like machines. This is Deckard's paradigm, but he also represents the whole dehumanized society around him. It makes no sense if it becomes a story about machines hunting machines instead of a human that behaves like a machine hunting machines that in some aspects behave more humanly than himself.
"If he's a replicant, that means he's basically required to hunt and kill his own kind because non-replicant humans refuse to perform such a morally repugnant act."
ha, ha ha! humans refuse to perform such a morally repugnant... HA, HA HA!!!
"Deckard being a replicant is not nonsense just because you like the competing interpretation better, but neither is Deckard being human nonsense. Both interpretations are worthy of examination, and there's no reason why it has to be one or the other."
"I wish static linking was standard. kernels/hypervisors detect duplicate memory anyway, so dynamic linking has not a single advantage anymore."
Except, of course, not having half a dozen versions of foo runing around with a dozen different security bugs to patch -or not to patch, since the different dozen applications are defined to use exactly x.y.z version of foo and they'll break appart if you try to use x.y.z+1.
"the advantage of Docker and similar containers isn't static linking. The advantage is that you get most of the important benefits of whole operating system virtualization with much lower overhead in terms of disk space and resource use [...] for deploying identical instances of a configured application to a handful or even hundreds of machines with the smallest feasible overhead"
Please take a time to think about what you wrote. Once you deploy your app to "hundreds of machines" you are back to square one: you have again the overhead of hundreds of kernels and basic OS. And if you think "but I avoid moving around a 5GB image for the complete VM!" you are doing it wrong, since you don't move the VM image, but the VM's contents' definition: the land of puppet, chef, et al.
No. The point _is_ -even yourself accept it, that this way "...it doesn't matter if the host OS has a different version of libevent", so you are using your docker as a way to "statically link" your application, which is the wrong approach, as it was already discovered quite a lot of time ago.
Now, what's in fact the nut of the problem? It is undisciplined developers. Undisciplined developers that publish libfoo, without any guarantee or description of what version of libfoo is it going to be long term supported and when told about a bug just tend to answer "please use our last version, since we don't want to spend our time correcting past ones but, of course we do have time to spent on introducing new bugs" and undisciplined users of libfoo which when start developing just link against the last version of libfoo, not minding that it is just an interim one.
On this environment you end up needing half a dozen different versions of libfoo to support half a dozen of otherwise functionally similar applications and, instead of trying to solve the problem at its root, which is using one, you cover this failure with a bigger one: "I know, let's statically link the application to the versions they need and it's done".
We see this syndrome all around: from docker to pip requirements, to gemspec files, to turbogears virtualenvs, to all the other versioning systems over there to install/compile exactly version x.y.z of foo or an equivalent abstraction.
Slashdot Mirror
"It isn't a matter of right or wrong. It is a matter of HOW right and HOW wrong.
Newtonian mechanics is right enough for most everyday living"
Well, ptolemaic astronomy is right enough for most everyday living too, just as much as newtonian mechanics.
In fact, now that you told about forgetting GPS, you probably know that ptolemaic astronomy not only is good enough to navigate your way all around the world, by land, boat or plane, but the way navigation is tought to pilots and ship captains *is* ptolemaic, not copernican, right?
Why we don't stop teaching about Copernicus, then?
"So no, it's not "utterly wrong"."
Yes, it is. Other ancient theories are crazily maddining wrong and certainly Newton's Principia is a shinning cathedral honoring the human intelligence but it still is utterly wrong.
Good you mention Asimov, since he was quite on the ontological path (against the pure mathematical path ala Dirac).
Now, forget about the numbers: it's about quality, not quantity. Newton thinks that there exists an absolute coordinate system and that things like mass, speed, length or time are therefore also absolute. Einstein demonstrates that he can't be any more wrong.
Ptolemy thought that the Earth is in the center of the universe and that objects in the sky circle around it by means of a dance of composed circles (epycicles and deferents) and it offers a magnificent math that "it's mostly right with many decimal places" and his model is also a magnificent show of human ingenuity. The problem with ptolemaic astronomy are not the numbers -ptolemaic astronomy can offer very precise numbers; it is the axioms: the Earth is not even near to the center of the solar system and there's no specific reason for orbits to be exclusively based on circles, so it is not a matter of how good its numbes are, just like there's no absolute coordinate system as Newton thought, no matter how good his numbers are.
As I see it, it's not that it would be mindblowing for the students to understand the basics of relativity or quantum mechanics but a matter of laziness from the teachers to find the proper way to teach them instead of "doing it as it has always been done".
"Well, it is one way to eliminate compiler bugs. By definition the compiler doesn't have any."
Yes, *a* compiler can be considered to be bug free per definition. But then, two different compilers will render two different outputs. What's the good one then?
"Over 100 years old is a bit of a stretch"
IParent is not specifically talking about quantum mechanics, just non-classic physics. Special Relativity paper comes from 1905 and the general one, 1915.
Classical physics can be pointed back to Newton: 1687.
"is basically "correct" for 99.999%"
It is not. It is utterly wrong. It just happens to throw the right numbers most of the time.
"until then the classic approximation is pretty good for high school work."
It is not. It would be much better to explain non-classical even without the maths (but two dimensional and statistical approach can be offered since the maths are in the curriculum) and throw the classic maths as what they are: a (very) useful approximation.
"Not necessarily, anyone can contact the software authors"
Key word being "can" which is neither "should" nor "must". The user either contacts for renegotiation/clarification of the license, in which case there will be a paper track and it will be the paper what counts, or the user won't contact, in which case, it will be the paper what counts.
In neither case the licensor will have the chance to further explain his intent to the court.
"The reason what licensor actually said that counts is because the licensor is the entity that initiates legal proceedings to remedy perceived violations of the licence."
Taken to the extreme, the only thing that counts is what the court says about the meaning of the paper but, being not so extreme, the court will hear the meaning both the licensor and the licensee understand out of the paper and, as long as both interpretations are fair, it will be the interpretation of the licensee -NOT the licensor, the one that will be supported by the court.
"Any way the terms of the GPL are essentially a non-concern until you distribute."
It depends on "who" you are. If I get a GPL-released binary it will be me the one asking for the sources while still I didn't distribute anything.
"He distributed a product which practiced the patent and thereby gave an implicit license to the patent for use of that product.
Where it gets a little hairyer is when the initial customer then passes it on to a third .. does the implied patent license transfer or not?"
Have you ever read the license? Is not that long and it is not that unclear.
What you call "a little hairyer" is right there for you to read.
"you can transfer that codebase to a thousand different aystems, windows, Linux, anything you want, you could even run it on a decemal based computer instead of digital. The program as it is written in an language would run the same every time."
Except, well, you know, it doesn't.
I don't know where your computer expertise comes from, but you can bet is not as exhaustive as you think it is.
"Once you have a working compiler, the code you write for it will be interpreted correctly and the same way everytime."
Rrrrright.
And then, what's the compiler we should use? GCC, Turbo C, Intel's?
You may tell "whatever is the first one that came to happen" but then, what would be the difference with common law using precedents?
"I wonder if it is possible to create a communication language that has the reproducibility and stability of a programming language."
You have never faced a bug or a divergence between compilers, do you?
"I saw a few companies adopt FreeBSD over Linux for that reason, back when the SCO UNIX case was going on, due to perceived legal risk."
Which is quite funny, since the first strong push for Linux instead of FreeBSD was because of the percived legal risk of BSD, back when the USL vs BSDi was going on.
"It doesn't matter what the aurthor of the GPL meant, it's what the licensor actually said that counts"
No, it isn't. What the licensor actually "said" is the text of the GPL itself, since that's how the work is licensed.
But then, since it is a license, a kind of contract that you don't get the chance to negotiate, it is not the interpretation of the licensor the one that counts, but the interpretation of the licensee, within reasonable bounds.
"The creators will tell you what it was intended to do and what it actually means within that narrow GPL context."
Too late. The contract is already signed and the license agreed upon.
"That is what is important and that's why we have the so called fine print in all agreements I have come across. No?"
No. It is the perfect opposite: the "fine print" is there exactly because you are *not* going to ask the text author about what was the intended meaning. The meaning is what the text says and no more.
Since it is a license, court will have to be liberal on the interpretation of the party subjected to the license, *not* the party producing it.
Luckily, Mr Moglen did a good job with the GPLv2 redaction and the intention will prevail.
"Why? They tried to squeeze money out of Google and failed; charging a nominal fee serves no purpose."
That they tried to squeeze money from Google is the stated true in press they even called the law "The Google Levvy". What they don't tell is that this isn't their only purpose. The real purpose of the law is to silent all voices not within AEDE. In Spain is not the press that it is the fourth power, AEDE is, and they want to stay that way.
"Most European cultures including Spain do not have such a culture. Instead, they would likely prefer to negotiate with google on the issue."
You just don't understand the issue here.
This legislation is so utterly stupid (I'm Spanish), so utterly against everybody but AEDE members (you see, they don't even represent the press but only the six biggest press companies), that everybody except AEDE and their government mates where *prying* for Google to stand its current position as the only hope for the law to be dismissed.
"2. Corruption in Spain is rampant, in every aspect of life."
No, not in every aspect of life but, unluckily it's rampant on politicians.
"Publishers do not have legislative powers. Democratically elected leaders do."
Have you thought a bit about what the "intellectual propiety business" involves, moreso when talking about levvies?
It basically gets down to the collecting agency sitting down waiting for the buckets full of money to arrive.
This means two things:
1) They have quite a lot of money.
2) They don't need to work a hard day for the money to arrive so they have time to do other things instead... like lobbying the government making use of the lot of money they gather (see one).
And then, let's talk about the subset of IP business we are talking here: press. They not only have the money and the inclination to lobby the government, they also control the tool politics are most afraid of: press. Now put together these three things and see that, yes, publishers have indeed factual legislative powers.
"All we have to do is figure out who is trying to extort whom."
AEDE, no doubt.
Their point is that they forcibly want a levvy for the "privilege" of indexing their headlines because "they are providing profit to the indexers". They obviously "forget" about the more or less 30% of traffic that comes to them straigth from google news.
Google can't just stop indexing AEDE members in order to avoid the tax because the levvy is for everyone in that activity no matter what, so they decided that since they don't want to pay, they'll close their doors in Spain.
And then, all of a sudden, AEDE members "remember" about that 30% of traffic that Google news is providing them and claim kindof "monopoly" on the part of Google so the government somehow should force Google to stay open and pay the levvy nevertheless.
AEDE (news), CEDRO (books), EGEDA (films) and SGAE (music) are just the Spanish versions of your RIAA, just so you know what kind of people are we talking about.
"The wording of the GPL is quite clear - it only requires the Makefiles to be included, and even adds an exception for the compiler when included with the OS as a runtime dependency. It doesn't say anything about the requirement to include the compiler."
In fact, it does. It adds an *exception* to include the compiler when it is already included and easily avaliable along the OS. This implies that when this is not the case, the compiler should be included since otherwise the source can't stand alone.
Anyway, I see your point. That's why I said "it is either not allowed by GPLv2 or a gap into the license that should be closed"
"The project is still fully open source."
That's quite arguably.
The vendor says: "The firmware of our modules is written in Pic Basic Pro [...] We provide all source code free of charge (GPLv2 licence)). The Pic Basic Pro compiler however is not for free"
And GPLv2 says:
"The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable. However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable."
The obvious intent of the GPL is for you to get a code in a way that allows you to work with it and get results. What this guys do is against this intent. It is either not allowed by GPLv2 or a gap into the license that should be closed but, in any case, this is not kosher.
"There is no allusion to any other model having such a limitation. This is down to, and I paraquote from the movie, the inevitability of the models developing a conscience."
A conscience like... hummm... Deckard?
Yes, Deckard's boss familiariy implies he has known him several years. And the script implies Deckard has been into his retirement some several years more. Several years + several years > 4, which is a no go for a conscient replicant.
And then, there's the meta-analisys of the script. It shows machines (replicants) becoming humans at the same time that humans become like machines. This is Deckard's paradigm, but he also represents the whole dehumanized society around him. It makes no sense if it becomes a story about machines hunting machines instead of a human that behaves like a machine hunting machines that in some aspects behave more humanly than himself.
"If he's a replicant, that means he's basically required to hunt and kill his own kind because non-replicant humans refuse to perform such a morally repugnant act."
ha, ha ha! humans refuse to perform such a morally repugnant... HA, HA HA!!!
"Deckard being a replicant is not nonsense just because you like the competing interpretation better, but neither is Deckard being human nonsense. Both interpretations are worthy of examination, and there's no reason why it has to be one or the other."
How long does a replicant survive?
How long did Harry Bryant know him?
I think this resolves the question.
"Deckard was a replicant."
In fact, Deckard neither was a replicant nor a human: he was a Schordinger factoid.
"I wish static linking was standard. kernels/hypervisors detect duplicate memory anyway, so dynamic linking has not a single advantage anymore."
Except, of course, not having half a dozen versions of foo runing around with a dozen different security bugs to patch -or not to patch, since the different dozen applications are defined to use exactly x.y.z version of foo and they'll break appart if you try to use x.y.z+1.
"the advantage of Docker and similar containers isn't static linking. The advantage is that you get most of the important benefits of whole operating system virtualization with much lower overhead in terms of disk space and resource use [...] for deploying identical instances of a configured application to a handful or even hundreds of machines with the smallest feasible overhead"
Please take a time to think about what you wrote. Once you deploy your app to "hundreds of machines" you are back to square one: you have again the overhead of hundreds of kernels and basic OS. And if you think "but I avoid moving around a 5GB image for the complete VM!" you are doing it wrong, since you don't move the VM image, but the VM's contents' definition: the land of puppet, chef, et al.
No. The point _is_ -even yourself accept it, that this way "...it doesn't matter if the host OS has a different version of libevent", so you are using your docker as a way to "statically link" your application, which is the wrong approach, as it was already discovered quite a lot of time ago.
Now, what's in fact the nut of the problem? It is undisciplined developers. Undisciplined developers that publish libfoo, without any guarantee or description of what version of libfoo is it going to be long term supported and when told about a bug just tend to answer "please use our last version, since we don't want to spend our time correcting past ones but, of course we do have time to spent on introducing new bugs" and undisciplined users of libfoo which when start developing just link against the last version of libfoo, not minding that it is just an interim one.
On this environment you end up needing half a dozen different versions of libfoo to support half a dozen of otherwise functionally similar applications and, instead of trying to solve the problem at its root, which is using one, you cover this failure with a bigger one: "I know, let's statically link the application to the versions they need and it's done".
We see this syndrome all around: from docker to pip requirements, to gemspec files, to turbogears virtualenvs, to all the other versioning systems over there to install/compile exactly version x.y.z of foo or an equivalent abstraction.