I don't think you completely read the parent's post. He is saying that there cannot be such a thing as GPL-covered parts and BSD-covered parts in one program, as according to his interpretation, every single piece of a GPL-covered work must have originated as GPL-covered code, and that you cannot incorporate BSD code into a GPL-covered work without violating the GPL.
Why is it unethical to use the GPL but not to use a totalitarian closed license?
Because the BSD zealots like to support proprietary software makers, but don't want to support people developing software under the GPL, as they utterly hate that license.
The reason stated by Theo is that they can look at GPL software and see the code improvements made, but cannot touch (i.e. use) them. They think that it's better to not even be able to look at it, because then they won't know that somebody actually used the code.
So you are trying to say that all these years, using BSD code in GPL applications have meant violating the GPL? Could you point out the specific section of the GPL that says this? I have read the GPL several times, but I cannot recall ever reading that all parts of a GPL-covered work must have originated as GPL-covered code.
No, the complaint is that the GPL developers are stripping off the BSD copyright notices. Period. End of story. Full stop.
No, this is not the end of the story. Theo also complained that it was "unethical" to improve on BSD software without also licensing the improvements as BSD. All while claiming that he has no problems with corporations taking his code, improving upon it, and keeping the improved source proprietary. The "unethical to not give back under the BSDL" seemingly only applies to GPL developers.
The restrictions under the BSD license are much less restrictive than those of the GPL
They may be, but the BSD crowd are surely trying to make them sound just as restrictive as the GPL, while still arguing that the BSD license is "more free" than the GPL.
Let me break it down for you: msft can take the code and use it. It cannot take the code and rip the Copyright license off.
No, they can't. But no one other than Microsoft employees are ever going to see those copyright notices, so in practice, they could have ripped them out, without the BSD developers noticing.
An ethical FOSS developer would take a BSD driver, improve it, and release his/her changes under the original license.
So in other words, it is unethical to improve upon BSD code and release the improvements under the GPL? Well, I guess I'm unethical then, because I think that is fine, especially since the BSD crowd claims that this is an advantage of the BSD license compared with the GPL, and that this makes the BSD license "more free" than the GPL.
The people favoring the GPL want to protect their code from being taken proprietary, something the BSD license does not do. Which one is better, use the existing BSD software that explicitly allows this (but draws flak from the BSD crowd), or write your own GPL-only software? 'Cause I'm not going to release any software written on my spare time under a BSD license.
Software developed at my current place of work is different, since I work for a tax-funded entity (a university). The tax payers (including corporations) have already payed for it, so they should have a right to use it in any way they wish. For such cases, the BSD license is appropriate.
I can go and extract the BSD code and do what I please with it as long as it complied with the BSD license.
You can? That seems to fly in the face of the GPL derivative works clauses. Since GPL code was incorporated into BSD code, the derivative can only be licensed under the GPL. If you remove the new parts, it is a derivative of the derivative. And derivatives of GPL code can only be GPL.
But just from an ethical point of view, it just feels wrong for a free software advocate to take the software out of the hands of the BSD crowd.
Free software? I thought you were "open source" advocates, that can't stop mentioning how ok you are with people that take your code and makes derivatives unavailable to you. Except if they use the GPL. Then it is not ok, and derivatives "should" be made available as BSD code.
A user could probably (IANAL) remove all GPL-only parts and redistribute the dual-licensed code under both licenses again.
Wouldn't this be a derivative of a GPL-covered work? Such works can only be licensed under the GPL, regardless of whether the remaining code was originally BSD. That's why the BSD-ified BCM driver would still be an illegal derivative work when the GPL parts had been removed (as intended by the BSD developers).
But you don't get to establish new terms as you have no standing to do so. Saying the people who receive it from you can only use the GPL is imposing new terms. That's not a right you have.
So I cannot add my own code to a BSD-licensed source file and license my changes under the GPL? This would make the entire derivative work available under the GPL only, but if I read you right, this means imposing new terms on the BSD code, which I'm not allowed to do.
How is that "more free" again? Sounds exactly like the GPL. This is really confusing.
The linuxers just can't take it that they have to respect a different license.
You mean like Theo that can't take that he has to respect a different license. Because he obviously had a problem respecting the GPL in the BCM case. It is quite hypocritical to disrespect another license just to turn around and demand respect for his license.
Actually, he intends to liberate all those thousands of microsoft based applications and take them to linux.
While secretly inviting patent infringement lawsuits against Linux users by Microsoft, so that Microsoft can continue to talk about the "undisclosed balance sheet liabilities" of F/OSS.
This kind of apocalytic stories about millions of poor users being death scared about their poor software being GPLv3 is a projection of the *concerns* of a specific group of developers (in this case, BSD) upon the wider world
I'd rather say that it is a projection of the wishes of a certain group of developers. Many times I've seen BSD guys hoping that the Linux community will crash and burn, so that they can step in and take over the show.
I guess this is the BSD inferiority complex that shows its ugly rear.
I certainly would love to see GCC fork, and fork cleanly into (at least) two well supported versions. How about a fork that doesn't try to be a compiler for every frickin' language under the sun, under a BSD license - wouldn't that be spiffy?
It would also be impossible. Last time I checked, the copyrights of GCC are held by the FSF. They'll hardly allow a fork under a BSD license.
If you want a BSD licensed compiler, you'll have to write one yourself.
Yeah, the feel that nothing works. At least, that was what I went into when trying to do some minor tasks on a NetBSD box we have at work. When I get time, it'll definitely be switched to Linux (more specifically CentOS, as that is what we standardize on anyway).
Exactly. In their own presentation, they even state that they offer a path that minimizes the risk of changes to input specifications if this is of business value to the submitting party, and that a "good" standard today is better than a "perfect" one tomorrow ("good" and "perfect" are sales talk for "bad" and "good", respectively).
So in essence, what they offer is a quick rubber stamp on an input specification and a fast track to ISO. It is pretty obvious why Microsoft chose to use them.
But surely someone as clever as Miguel can extrapolate Microsoft's previous behaviour to the near future and see that it has a high probability to only be a bad thing?
No, he can't, since he is wearing those rosy Microsoft fanboy glasses.
So, the guy who started Ximian, the company that developed Evolution and main supporter of Gnome, not to mention Mono, one of the best projects for a cross-platform development tools, is trying to DESTROY open source in general . . . wow this is the more interesting conspirancy theory since the X Files show ended!
Well, he is trying to lure people into using more all-Microsoft technologies, just waiting for the patent trap to close.
I know Miguel in person and I don't think he is a Microsoft fan boy.
Well, maybe that is because you know him? You don't tend to think bad things about your friends.
To me, he is clearly a Microsoft fanboy. This OOXML-preaching is the best evidence there is. Standards bodies around the world have submitted thousands of comments on how the specification need to be fixed, with Microsoft having to resort to ballot stuffing to even have a shadow of a chance of passing the vote. And during all of this, Miguel sings the "Ooh Aah, OOXML is superb, everything else is just FUD" song in support of his masters in Redmond.
Bioshock, Call of Duty Airborne, Spore are three new games which if I wanted to play under linux I would have to scour through forums, usenet posting, etc just to get the game to launch
Slashdot Mirror
I don't think you completely read the parent's post. He is saying that there cannot be such a thing as GPL-covered parts and BSD-covered parts in one program, as according to his interpretation, every single piece of a GPL-covered work must have originated as GPL-covered code, and that you cannot incorporate BSD code into a GPL-covered work without violating the GPL.
No, I don't follow that interpretation.
Which is impossible due to the requirement to offer source code with GPL software.
Because the BSD zealots like to support proprietary software makers, but don't want to support people developing software under the GPL, as they utterly hate that license.
The reason stated by Theo is that they can look at GPL software and see the code improvements made, but cannot touch (i.e. use) them. They think that it's better to not even be able to look at it, because then they won't know that somebody actually used the code.
So you are trying to say that all these years, using BSD code in GPL applications have meant violating the GPL? Could you point out the specific section of the GPL that says this? I have read the GPL several times, but I cannot recall ever reading that all parts of a GPL-covered work must have originated as GPL-covered code.
No, this is not the end of the story. Theo also complained that it was "unethical" to improve on BSD software without also licensing the improvements as BSD. All while claiming that he has no problems with corporations taking his code, improving upon it, and keeping the improved source proprietary. The "unethical to not give back under the BSDL" seemingly only applies to GPL developers.
They may be, but the BSD crowd are surely trying to make them sound just as restrictive as the GPL, while still arguing that the BSD license is "more free" than the GPL.
No, they can't. But no one other than Microsoft employees are ever going to see those copyright notices, so in practice, they could have ripped them out, without the BSD developers noticing.
Which is pretty useless, as they are rotting away in a Microsoft safe.
So in other words, it is unethical to improve upon BSD code and release the improvements under the GPL? Well, I guess I'm unethical then, because I think that is fine, especially since the BSD crowd claims that this is an advantage of the BSD license compared with the GPL, and that this makes the BSD license "more free" than the GPL.
The people favoring the GPL want to protect their code from being taken proprietary, something the BSD license does not do. Which one is better, use the existing BSD software that explicitly allows this (but draws flak from the BSD crowd), or write your own GPL-only software? 'Cause I'm not going to release any software written on my spare time under a BSD license.
Software developed at my current place of work is different, since I work for a tax-funded entity (a university). The tax payers (including corporations) have already payed for it, so they should have a right to use it in any way they wish. For such cases, the BSD license is appropriate.
You can? That seems to fly in the face of the GPL derivative works clauses. Since GPL code was incorporated into BSD code, the derivative can only be licensed under the GPL. If you remove the new parts, it is a derivative of the derivative. And derivatives of GPL code can only be GPL.
Free software? I thought you were "open source" advocates, that can't stop mentioning how ok you are with people that take your code and makes derivatives unavailable to you. Except if they use the GPL. Then it is not ok, and derivatives "should" be made available as BSD code.
Wouldn't this be a derivative of a GPL-covered work? Such works can only be licensed under the GPL, regardless of whether the remaining code was originally BSD. That's why the BSD-ified BCM driver would still be an illegal derivative work when the GPL parts had been removed (as intended by the BSD developers).
So I cannot add my own code to a BSD-licensed source file and license my changes under the GPL? This would make the entire derivative work available under the GPL only, but if I read you right, this means imposing new terms on the BSD code, which I'm not allowed to do.
How is that "more free" again? Sounds exactly like the GPL. This is really confusing.
But not before Theo started a flame war against the IMO quite diplomatic BCM developers.
You mean like Theo that can't take that he has to respect a different license. Because he obviously had a problem respecting the GPL in the BCM case. It is quite hypocritical to disrespect another license just to turn around and demand respect for his license.
While secretly inviting patent infringement lawsuits against Linux users by Microsoft, so that Microsoft can continue to talk about the "undisclosed balance sheet liabilities" of F/OSS.
I'd rather say that it is a projection of the wishes of a certain group of developers. Many times I've seen BSD guys hoping that the Linux community will crash and burn, so that they can step in and take over the show.
I guess this is the BSD inferiority complex that shows its ugly rear.
It would also be impossible. Last time I checked, the copyrights of GCC are held by the FSF. They'll hardly allow a fork under a BSD license.
If you want a BSD licensed compiler, you'll have to write one yourself.
Yeah, the feel that nothing works. At least, that was what I went into when trying to do some minor tasks on a NetBSD box we have at work. When I get time, it'll definitely be switched to Linux (more specifically CentOS, as that is what we standardize on anyway).
In Sweden, the version I've heard substitutes communist with red, which also includes the social democrats.
Exactly. In their own presentation, they even state that they offer a path that minimizes the risk of changes to input specifications if this is of business value to the submitting party, and that a "good" standard today is better than a "perfect" one tomorrow ("good" and "perfect" are sales talk for "bad" and "good", respectively).
So in essence, what they offer is a quick rubber stamp on an input specification and a fast track to ISO. It is pretty obvious why Microsoft chose to use them.
No, he can't, since he is wearing those rosy Microsoft fanboy glasses.
Well, he is trying to lure people into using more all-Microsoft technologies, just waiting for the patent trap to close.
I know Miguel in person and I don't think he is a Microsoft fan boy.Well, maybe that is because you know him? You don't tend to think bad things about your friends.
To me, he is clearly a Microsoft fanboy. This OOXML-preaching is the best evidence there is. Standards bodies around the world have submitted thousands of comments on how the specification need to be fixed, with Microsoft having to resort to ballot stuffing to even have a shadow of a chance of passing the vote. And during all of this, Miguel sings the "Ooh Aah, OOXML is superb, everything else is just FUD" song in support of his masters in Redmond.
Fanboy it is, no doubt about it.
You are already playing Spore?
You are? Does that mean that Russia is violating the treaty by resuming their constant nuclear bomber patrols?