Joseph has been at the forefront of trying to get this issue resolved. Most of the time he has been diplomatic, fair and rational about it (on *rare* instances he will lose his temper). But he has a troubling trait of admitting he's wrong on some points.
"em>that much of KDE is not legally distributable until they contact some people that are damned scarce these days and make the necessary arrangements."
The real problem is not that orginal KDE code uses Qt, but that a few bits of the code were borrowed from elsewhere and made to link to Qt. I believe that this specific issue can be resolved, but the argument keeps revolving around KDE as a whole. KDE has many packages, but only a few of them have this problem. Specifically, the heart of KDE, the libs, are under the LGPL, and thus there is no problem with kde-libs from *any* interpretations of the licenses involved. The problem lies with kfloppy, kghostview, and maybe some more packages (those two are the only ones I am aware of).
"In short, the GPL says that the whole program must be under its terms before you are allowed to distribute it. It makes a specific exception for things like proprietary libcs and the like, but the exception to that is that you can't distribute them both together, so we'd be stuck even if we considered Qt a system library."
Section 2 of the GPL, which talks about licensing the Program under the GPL, has very different language than what Joseph paraphrased it as. "These requirements apply to the modified work as a whole", "If identifiable sections of that work are not derived from the Program", "distribute them as separate works". Qt is a separate work from KDE. Qt is not distributed as part of KDE "as a whole". Is is merely distributed in aggregate with KDE.
When Joseph talks about exceptions for "proprietary" libraries, the GPL is saying something very different from what he implies. The exception allows Qt to be excluded from the distribution of KDE, because Qt is normally distributed with the operating system (at least it does on my Slackware). But even this is beside the point! Only the source to modules that KDE contains need be distributed under the Program as a whole. Even if the programmatic and copyrightable sense of the word "module" were confused, it is clear that even if Qt could be considered a "module of" KDE, is it not a "module contained" in KDE. kde-libs may be an exception, but again, that part of KDE is under the LGPL.
When Joseph says "we'd be stuck even if we considered Qt a system library", I wonder why debian is not similarly stuck with other GPL-incompatible system libraries such as Motif. I get the strong (but hopefully wrong) impression that there is a double standard at work here. Why does the exception count for Motif but not for Qt? In a lot of cases I have seen, the GPLd Motif applications disobey even the exception, since the Motif is statically linked, and thus is distributed inside the Program.
"We could distribute the source, but what would be the point in that?"
Because some people don't have as fast of connections as Joseph does. I have a DSL connection, and downloading the recent KDE-1.91 sources took fifteen minutes (and I had to wait several hours to get through). A lot of people have only 56K modems or less. Downloading KDE could take hours for some of them. Including the source to KDE may not be politically necessary for Debian, but it sure as hell is convenient for the user!
"We've been quite open to helping KDE fix the problems with their licenses."
A lot of KDE developers respect you for that Joseph. The problem is that Debian also contains a small for very vocal minority of members who are demanding instead of helping. Like any hacker in any community, the KDE hacker stubbornly resists demands. The more the push, they more the push back.
That's not freedom, that's nihilism and license. Freedom is not a complete lack of restrictions. Instead, it is the lack of restrictions within your personal domain. Your freedom to swing your fist ends at my nose because you are exiting your domain and entering mine. Any restrictions I place upon your hitting my nose is not a restriction on your freedom.
The BSD license does not allow you to remove the BSD license. In fact, it's only requirements are that you keep the permission and warranty statements intact!
What you *can* do is to include a BSDd source file as part of a GPLd package. The package as a whole will be under the GPL, but that individual file will still be under the BSD. Take a look at the linux kernel sources and you'll see instances of this. In fact, if it were possible to extract the BSDd code from the MS Kerberos implementation, you would be able to freely redistribute them.
Another thing that can be done is to distribute a binary version of a BSD program. The permissions and warrantees still have to be included in the docs though. Does Microsoft include the BSD license somewhere in it's documentation for Kerberos? I don't know, but if they don't they are violating copyright (something they have frequently done in the past, just ask Sun). This doesn't mean that I can freely redistribute MSKerberos, but it does mean that portions of it can be.
But what you cannot do is to change the license. I don't know who started this myth, but it's wrong.
Hmmm, the FSF lists its justifications for Free Software and the GPL in particular in an article entitle "Why Software Should Not Be Owned".
If you want the code to remain *yours*, why did you use a license based on the concept that it shouldn't belong to anyone?
As for making the *billions* of dollars, get real! No company has ever made that much money off of any single piece of free software, BSD or otherwise. But if you want to talk about *millions*, why don't you talk to Redhat, et al, who have made millions by not paying its volunteer developers.
There are licenses out there that will allow your code to be free beer for open source developers but require that you get a cut if someone makes a billion or two off of a closed source derivative. Go use one of them instead.
The QPL is perfectly sensible for making modifications. It only makes forking difficult.
I mean, come on! You would actually submit a Qt bug fix as a 2Meg tarball of the complete sources instead of a 100Byte patch? I bet the gtk people just love your submissions:-)
Those people who have had their GPLd code used in KDE have already been notified. None of them have yet protested. But even this is a non-issue. The GPL operates under copyright law, and not contract law. It cannot take away legal rights that the user already possesses. And since copyright law is completely silent on references, dynamic linkage to and from the GPL is permitted.
Maybe GPLv3 will remove that clause about copyright law and it will become a contract like the EULA, but for now it's legal to reference non-GPLd libraries and incorporate their interfaces in the binary.
"...the ONLY real solution to this mess is for the Trolls to release Qt under the GPL
Yes, it would solve the problem, but it is not the Trolls problem. You don't demand that a third party correct the actions of a first party. They weren't the ones writing or using GPLd code.
GNOME and KDE are intended for Windows users just making the switch.
Wrongo! The purpose of GNOME and KDE is to provide a desktop for X. You might not need or want a desktop, but that's no call to denigrate everyone who does. GNOME and KDE are attempting something never before achieved in X or Unix, and that is consistancy. Just because Windows and Mac also attempts this does not mean that consistancy is wrong.
KDE 1.0 apps work with Qt-1.4x and KDE 2.0 apps work with Qt-2.1x. During linkage, each app *knows* which library it wants, so you only need two libraries. One for the old apps and one for the new.
I don't know who told you that you needed twenty different versions, but they're wrong.
I am in the envious position of being a QA engineer in a company and culture that *respects* QA. But occasionally we get a new hire developer that smarts off to QA, so I understand that attitude you're describing. It's even worse with Open Source projects, since the developers have higher ego levels.
I've long since stopped submitting bugs to Open SOurce projects. They invariably get closed as "not a bug", "not our problem", or "submit it to that other project".
It's doubly hard since there are no specifications whatsoever for any of these projects. This is just insane. If there's no specs, then it's only a bug if it crashes the application (if it crashes the system, it was someone else's fault). Those bugs you quoted may or may not be real bugs. Without specs there's no way to know. He wrote "What here isn't working as designed?" and I'll answer "nothing because there is no design."
I hope that the KDE developers don't listen to this turkey, and trust the bug submitter that there might actually be a correctable defect.
People just don't know how to read anymore! I said *dominant* marketshare. Last time I looked Linux, XFree and gcc were not the market leaders. Give them a few years and they might be. But they aren't today.
First you start rattling on about dumb users, then finish up saying Corel should have gone into the server business. Are you saying client software is only for idiots? Because that's what it sounds like to me.
Why do you wish that they "fail big time", when you've already stated that you don't care. It's obvious that you *do* care, and care passionately about what Corel does. It seems to be a personal insult to you that some company doesn't include you in their target audience. BFD. Get used to it cuz the world don't revolve around your butt.
I absolutely agree with you about limiting the terms of copyright. I good starting number would be 50 years or 25 years after the death of the author, whichever comes first.
Copyright law (at least pre-DMCA) already allows you to reverse engineer. Most of these anti-RE licenses you find are not based on copyright law, but on contract law.
As for your utilitarian view, I can't agree with it because although it sounds very objective, it isn't. It still boils down to a very subjective opinion of what the public good is.
1) The Yawn (Moz=0%, Opr=0%) Mozilla will be released before Opera does. Why would anyone want to use Opera when a Free browser is available?
2) The Snub (Moz=0%, Opr=0%) I've tried Opera and I tell you that M37 beats it to hell and back. Just wait till Mozilla is released in a month or so.
3) The Announcement (Moz=0%, Opr=2%) Okay, Mozilla is released. Everyone can stop using Opera now. Finally we have a Free browser. Did you hear me? You can stop now.
4) The Promise (Moz=1%, Opr=5%) Just wait until 2.0 comes out and Mozilla is only a 5Meg download. In the meantime, we support 57 standards to their meager 54.
5) The Smuggery (Moz=2%, Opr=7%) Our distro doesn't include Opera! Unlike those evil corporate and commercial distributions, we only include Mozilla (and Gnuzilla beta). See how good and pure we are?
6) The Panic (Moz=2%, Opr=9%) Don't use Opera! You are only being dominated and enslaved by the corporate masters. Use Mozilla and be free! Opera is evil, Mozilla is good. We will free you. Don't be fooled by their 100% standards compliance, it's a trick!
7) The Plea (Moz=3%, Opr=12%) Why do you want to trade your freedom and morality for a lightweight, fast and crashproof browser? Sure, our browser is bloated, slow and takes down your entire system on a daily basis, but it will give you freedom, joy and eternal justice. Just try us. Please, before it's too late.
Epilogue (Moz=2%, Opr=16%) Stupid people, they got what they wanted. I hope they're satisfied.
Doing a bit of fortune-telling, are we? Reading the tea leaves and consulting the pigeon entrails?
Because that's what it is: fortune-telling. There is no stable open source, standards compliant and free browser today. In fact, the open source, standards compliant and free *anythings* that have won dominant marketshare are pretty rare. Apache, Perl, BIND, and that's all that I can think of.
You may have faith that open source is superior to closed source, but the typical user doesn't. All they care about is that something works and works well. But don't get your knickers in a twist, Opera is targeting a whole different audience than Mozilla, Netscape or IE. They are targeting those who are willing to pay $40 for a small fast browser. Looking at the free beer crowd (NS, IE and MZ), they'll have a lot of work to do to compete in the small and fast department.
It all depends on whether a song is the property of its creator or not. If the songwriter does have natural rights to his works, then copyright laws are not much different than laws against theft and trespass. But if the songwriter does not have natural rights to his works, then you have to follow this to its logical conclusion - that programmers have no rights to copyright their works. That means no GPL and no defense against encrypted binary-only derivations.
I don't know whether copyrights are legitimate or not. I *do* know that double standards never are. It's either all copyrights, or no copyrights at all. Anything in between is hypocrisy.
Re:What does it bring us?
on
Linux Mergers?
·
· Score: 2
"Just keep an open eye, so that they don't walk away with all our free software."
Uh, how can they?
"...use Debian..."
Why did I have a suspicion that this would be the conclusion of the post?
Everywhere I go, from the FSF, to OSI to even Debian's own social contract with its developers, I keep hearing "it's not free beer and it's okay to sell free software". So why does the Debian crowd always piss on the commercial distros?
I can't comment on Michael Jackson, so I'll stick with Liberace. Why was he making 5 million a year? Because his fans paid him that much! They wanted to!
One of his best quotes (paraphrased), to a member of the audience "Do you like my outfit? I'm glad you do because you're paying for it."
And what about the songwriter? Alvin can write a damn good song, but he can't sing or dance or operate the sound board. Bert takes Alvin's song for free and makes a million on his worldwide tour. Alvin gets nothing.
But, you say, Bert would certainly hire Alvin to keep writing his songs. True, except that Chuck already hired Alvin for 5$ an hour. Bert "stole" the song off of Alvin' and Chuck's demo release. Bert knows that he doesn't have to hire Alvin, he only needs to copy his stuff.
So Alvin and Chuck end up working at McDonalds to feed their family all because of some stupid philosophy that selling songs is evil.
Re:Closed Source != No Source
on
Java 2 For BSD
·
· Score: 2
Closed Source licenses do not necessarily mean that the source code is hidden. The source code to Java *IS* available. The only problem is getting Sun's signoff on a port. It may be entirely possible that an official port to BSD/OS can be simply recompiled for the other BSD's with no modifications.
I've heard theories that many new users are using Linux not for the ideals of {free|open source} software or even because they like Unix, but just are tired of Windows
And what's wrong with that? If it is the "freedom" that matters, we should encourage them to come over for their own good. But if you start imposing morality tests to ensure that new users think the same way you do, you have made a complete mockery of freedom.
Hmmm, if they were able to choose to abandon Windows in favor of Linux, maybe they already had freedom to begin with?
Is it now possible to sell a closed-source product built on top of GNU GPLd readline, without special permission from Richard Stallman?
I don't understand you point at all. If you are arguing that the LGPL is a better license than the QPL because it allows linkage to proprietary applications, I would have to agree with you. But if you are arguing that because you can't write a closed source app with Qt then Qt can't be free, I must strongly disagree and direct your attention to the Free Software definition.
It is not every Debian developer that wants to keep KDE out. But enough of them do that it can't come up to a vote. Some mirror site maintainers have even stated that they will not mirror KDE even if Debian voted for it and included it in main.
Having spent some time willingly and unwillingly (debian-legal posters frequently cross post to other lists) in several debian mailing lists, there is a small but very vocal minority that seem to have an irrational dislike for KDE. Right now, today, this very instant, there is no reason that the LGPL KDE *libraries* cannot be included in Debian. And once there, there is no reason not to include those KDE applications that fall under non-GPL licenses (like the most excellent Cervisia). But they have not done so even though this move has been brought up time and time again.
One particular developer said a while ago that the only thing stopping him for including KDE was the presence of KFloppy, which linked pre-KDE code to Qt. This is a valid point, and one of the few *real* legal problems with KDE/Qt licensing. However, removing KFloppy from KDE is trivial. So just do it!
If the plurality of Debian really wanted to include KDE, they could do so in some form. They could include just the base libraries and the non-GPL applications. They can include just the KDE source code (which sidesteps the entire linkage objection). But instead of doing this, they have spent the last two years endlessly arguing of minutia. The only reason I can fathom for them not included some form of KDE is because they just don't want to.
Slashdot Mirror
"until copyright dies ... this is how the world is.
And when Copyright dies, everyone will see once and for all how stupid this whole GPL-QPL argument truly was.
Joseph has been at the forefront of trying to get this issue resolved. Most of the time he has been diplomatic, fair and rational about it (on *rare* instances he will lose his temper). But he has a troubling trait of admitting he's wrong on some points.
"em>that much of KDE is not legally distributable until they contact some people that are damned scarce these days and make the necessary arrangements."
The real problem is not that orginal KDE code uses Qt, but that a few bits of the code were borrowed from elsewhere and made to link to Qt. I believe that this specific issue can be resolved, but the argument keeps revolving around KDE as a whole. KDE has many packages, but only a few of them have this problem. Specifically, the heart of KDE, the libs, are under the LGPL, and thus there is no problem with kde-libs from *any* interpretations of the licenses involved. The problem lies with kfloppy, kghostview, and maybe some more packages (those two are the only ones I am aware of).
"In short, the GPL says that the whole program must be under its terms before you are allowed to distribute it. It makes a specific exception for things like proprietary libcs and the like, but the exception to that is that you can't distribute them both together, so we'd be stuck even if we considered Qt a system library."
Section 2 of the GPL, which talks about licensing the Program under the GPL, has very different language than what Joseph paraphrased it as. "These requirements apply to the modified work as a whole", "If identifiable sections of that work are not derived from the Program", "distribute them as separate works". Qt is a separate work from KDE. Qt is not distributed as part of KDE "as a whole". Is is merely distributed in aggregate with KDE.
When Joseph talks about exceptions for "proprietary" libraries, the GPL is saying something very different from what he implies. The exception allows Qt to be excluded from the distribution of KDE, because Qt is normally distributed with the operating system (at least it does on my Slackware). But even this is beside the point! Only the source to modules that KDE contains need be distributed under the Program as a whole. Even if the programmatic and copyrightable sense of the word "module" were confused, it is clear that even if Qt could be considered a "module of" KDE, is it not a "module contained" in KDE. kde-libs may be an exception, but again, that part of KDE is under the LGPL.
When Joseph says "we'd be stuck even if we considered Qt a system library", I wonder why debian is not similarly stuck with other GPL-incompatible system libraries such as Motif. I get the strong (but hopefully wrong) impression that there is a double standard at work here. Why does the exception count for Motif but not for Qt? In a lot of cases I have seen, the GPLd Motif applications disobey even the exception, since the Motif is statically linked, and thus is distributed inside the Program.
"We could distribute the source, but what would be the point in that?"
Because some people don't have as fast of connections as Joseph does. I have a DSL connection, and downloading the recent KDE-1.91 sources took fifteen minutes (and I had to wait several hours to get through). A lot of people have only 56K modems or less. Downloading KDE could take hours for some of them. Including the source to KDE may not be politically necessary for Debian, but it sure as hell is convenient for the user!
"We've been quite open to helping KDE fix the problems with their licenses."
A lot of KDE developers respect you for that Joseph. The problem is that Debian also contains a small for very vocal minority of members who are demanding instead of helping. Like any hacker in any community, the KDE hacker stubbornly resists demands. The more the push, they more the push back.
Your freedom is my prison
My freedom is your prison
That's not freedom, that's nihilism and license. Freedom is not a complete lack of restrictions. Instead, it is the lack of restrictions within your personal domain. Your freedom to swing your fist ends at my nose because you are exiting your domain and entering mine. Any restrictions I place upon your hitting my nose is not a restriction on your freedom.
The BSD license does not allow you to remove the BSD license. In fact, it's only requirements are that you keep the permission and warranty statements intact!
What you *can* do is to include a BSDd source file as part of a GPLd package. The package as a whole will be under the GPL, but that individual file will still be under the BSD. Take a look at the linux kernel sources and you'll see instances of this. In fact, if it were possible to extract the BSDd code from the MS Kerberos implementation, you would be able to freely redistribute them.
Another thing that can be done is to distribute a binary version of a BSD program. The permissions and warrantees still have to be included in the docs though. Does Microsoft include the BSD license somewhere in it's documentation for Kerberos? I don't know, but if they don't they are violating copyright (something they have frequently done in the past, just ask Sun). This doesn't mean that I can freely redistribute MSKerberos, but it does mean that portions of it can be.
But what you cannot do is to change the license. I don't know who started this myth, but it's wrong.
Hmmm, the FSF lists its justifications for Free Software and the GPL in particular in an article entitle "Why Software Should Not Be Owned".
If you want the code to remain *yours*, why did you use a license based on the concept that it shouldn't belong to anyone?
As for making the *billions* of dollars, get real! No company has ever made that much money off of any single piece of free software, BSD or otherwise. But if you want to talk about *millions*, why don't you talk to Redhat, et al, who have made millions by not paying its volunteer developers.
There are licenses out there that will allow your code to be free beer for open source developers but require that you get a cut if someone makes a billion or two off of a closed source derivative. Go use one of them instead.
You hit the nail on the head. The important thing is not that it's an Open Source program, but that it is an open standard.
The QPL is perfectly sensible for making modifications. It only makes forking difficult.
:-)
I mean, come on! You would actually submit a Qt bug fix as a 2Meg tarball of the complete sources instead of a 100Byte patch? I bet the gtk people just love your submissions
Those people who have had their GPLd code used in KDE have already been notified. None of them have yet protested. But even this is a non-issue. The GPL operates under copyright law, and not contract law. It cannot take away legal rights that the user already possesses. And since copyright law is completely silent on references, dynamic linkage to and from the GPL is permitted.
Maybe GPLv3 will remove that clause about copyright law and it will become a contract like the EULA, but for now it's legal to reference non-GPLd libraries and incorporate their interfaces in the binary.
"...the ONLY real solution to this mess is for the Trolls to release Qt under the GPL
Yes, it would solve the problem, but it is not the Trolls problem. You don't demand that a third party correct the actions of a first party. They weren't the ones writing or using GPLd code.
GNOME and KDE are intended for Windows users just making the switch.
Wrongo! The purpose of GNOME and KDE is to provide a desktop for X. You might not need or want a desktop, but that's no call to denigrate everyone who does. GNOME and KDE are attempting something never before achieved in X or Unix, and that is consistancy. Just because Windows and Mac also attempts this does not mean that consistancy is wrong.
KDE 1.0 apps work with Qt-1.4x and KDE 2.0 apps work with Qt-2.1x. During linkage, each app *knows* which library it wants, so you only need two libraries. One for the old apps and one for the new.
I don't know who told you that you needed twenty different versions, but they're wrong.
I am in the envious position of being a QA engineer in a company and culture that *respects* QA. But occasionally we get a new hire developer that smarts off to QA, so I understand that attitude you're describing. It's even worse with Open Source projects, since the developers have higher ego levels.
I've long since stopped submitting bugs to Open SOurce projects. They invariably get closed as "not a bug", "not our problem", or "submit it to that other project".
It's doubly hard since there are no specifications whatsoever for any of these projects. This is just insane. If there's no specs, then it's only a bug if it crashes the application (if it crashes the system, it was someone else's fault). Those bugs you quoted may or may not be real bugs. Without specs there's no way to know. He wrote "What here isn't working as designed?" and I'll answer "nothing because there is no design."
I hope that the KDE developers don't listen to this turkey, and trust the bug submitter that there might actually be a correctable defect.
The Consitution says that is the purpose, but it doesn't say that is the means with which to implement them.
People just don't know how to read anymore! I said *dominant* marketshare. Last time I looked Linux, XFree and gcc were not the market leaders. Give them a few years and they might be. But they aren't today.
First you start rattling on about dumb users, then finish up saying Corel should have gone into the server business. Are you saying client software is only for idiots? Because that's what it sounds like to me.
Why do you wish that they "fail big time", when you've already stated that you don't care. It's obvious that you *do* care, and care passionately about what Corel does. It seems to be a personal insult to you that some company doesn't include you in their target audience. BFD. Get used to it cuz the world don't revolve around your butt.
I absolutely agree with you about limiting the terms of copyright. I good starting number would be 50 years or 25 years after the death of the author, whichever comes first.
Copyright law (at least pre-DMCA) already allows you to reverse engineer. Most of these anti-RE licenses you find are not based on copyright law, but on contract law.
As for your utilitarian view, I can't agree with it because although it sounds very objective, it isn't. It still boils down to a very subjective opinion of what the public good is.
The Seven Stages of Denial
1) The Yawn (Moz=0%, Opr=0%)
Mozilla will be released before Opera does. Why would anyone want to use Opera when a Free browser is available?
2) The Snub (Moz=0%, Opr=0%)
I've tried Opera and I tell you that M37 beats it to hell and back. Just wait till Mozilla is released in a month or so.
3) The Announcement (Moz=0%, Opr=2%)
Okay, Mozilla is released. Everyone can stop using Opera now. Finally we have a Free browser. Did you hear me? You can stop now.
4) The Promise (Moz=1%, Opr=5%)
Just wait until 2.0 comes out and Mozilla is only a 5Meg download. In the meantime, we support 57 standards to their meager 54.
5) The Smuggery (Moz=2%, Opr=7%)
Our distro doesn't include Opera! Unlike those evil corporate and commercial distributions, we only include Mozilla (and Gnuzilla beta). See how good and pure we are?
6) The Panic (Moz=2%, Opr=9%)
Don't use Opera! You are only being dominated and enslaved by the corporate masters. Use Mozilla and be free! Opera is evil, Mozilla is good. We will free you. Don't be fooled by their 100% standards compliance, it's a trick!
7) The Plea (Moz=3%, Opr=12%)
Why do you want to trade your freedom and morality for a lightweight, fast and crashproof browser? Sure, our browser is bloated, slow and takes down your entire system on a daily basis, but it will give you freedom, joy and eternal justice. Just try us. Please, before it's too late.
Epilogue (Moz=2%, Opr=16%)
Stupid people, they got what they wanted. I hope they're satisfied.
Doing a bit of fortune-telling, are we? Reading the tea leaves and consulting the pigeon entrails?
Because that's what it is: fortune-telling. There is no stable open source, standards compliant and free browser today. In fact, the open source, standards compliant and free *anythings* that have won dominant marketshare are pretty rare. Apache, Perl, BIND, and that's all that I can think of.
You may have faith that open source is superior to closed source, but the typical user doesn't. All they care about is that something works and works well. But don't get your knickers in a twist, Opera is targeting a whole different audience than Mozilla, Netscape or IE. They are targeting those who are willing to pay $40 for a small fast browser. Looking at the free beer crowd (NS, IE and MZ), they'll have a lot of work to do to compete in the small and fast department.
It all depends on whether a song is the property of its creator or not. If the songwriter does have natural rights to his works, then copyright laws are not much different than laws against theft and trespass. But if the songwriter does not have natural rights to his works, then you have to follow this to its logical conclusion - that programmers have no rights to copyright their works. That means no GPL and no defense against encrypted binary-only derivations.
I don't know whether copyrights are legitimate or not. I *do* know that double standards never are. It's either all copyrights, or no copyrights at all. Anything in between is hypocrisy.
"Just keep an open eye, so that they don't walk away with all our free software."
Uh, how can they?
"...use Debian..."
Why did I have a suspicion that this would be the conclusion of the post?
Everywhere I go, from the FSF, to OSI to even Debian's own social contract with its developers, I keep hearing "it's not free beer and it's okay to sell free software". So why does the Debian crowd always piss on the commercial distros?
I can't comment on Michael Jackson, so I'll stick with Liberace. Why was he making 5 million a year? Because his fans paid him that much! They wanted to!
One of his best quotes (paraphrased), to a member of the audience "Do you like my outfit? I'm glad you do because you're paying for it."
And what about the songwriter? Alvin can write a damn good song, but he can't sing or dance or operate the sound board. Bert takes Alvin's song for free and makes a million on his worldwide tour. Alvin gets nothing.
But, you say, Bert would certainly hire Alvin to keep writing his songs. True, except that Chuck already hired Alvin for 5$ an hour. Bert "stole" the song off of Alvin' and Chuck's demo release. Bert knows that he doesn't have to hire Alvin, he only needs to copy his stuff.
So Alvin and Chuck end up working at McDonalds to feed their family all because of some stupid philosophy that selling songs is evil.
Closed Source licenses do not necessarily mean that the source code is hidden. The source code to Java *IS* available. The only problem is getting Sun's signoff on a port. It may be entirely possible that an official port to BSD/OS can be simply recompiled for the other BSD's with no modifications.
I've heard theories that many new users are using Linux not for the ideals of {free|open source} software or even because they like Unix, but just are tired of Windows
And what's wrong with that? If it is the "freedom" that matters, we should encourage them to come over for their own good. But if you start imposing morality tests to ensure that new users think the same way you do, you have made a complete mockery of freedom.
Hmmm, if they were able to choose to abandon Windows in favor of Linux, maybe they already had freedom to begin with?
Is it now possible to sell a closed-source product built on top of GNU GPLd readline, without special permission from Richard Stallman?
I don't understand you point at all. If you are arguing that the LGPL is a better license than the QPL because it allows linkage to proprietary applications, I would have to agree with you. But if you are arguing that because you can't write a closed source app with Qt then Qt can't be free, I must strongly disagree and direct your attention to the Free Software definition.
It is not every Debian developer that wants to keep KDE out. But enough of them do that it can't come up to a vote. Some mirror site maintainers have even stated that they will not mirror KDE even if Debian voted for it and included it in main.
Having spent some time willingly and unwillingly (debian-legal posters frequently cross post to other lists) in several debian mailing lists, there is a small but very vocal minority that seem to have an irrational dislike for KDE. Right now, today, this very instant, there is no reason that the LGPL KDE *libraries* cannot be included in Debian. And once there, there is no reason not to include those KDE applications that fall under non-GPL licenses (like the most excellent Cervisia). But they have not done so even though this move has been brought up time and time again.
One particular developer said a while ago that the only thing stopping him for including KDE was the presence of KFloppy, which linked pre-KDE code to Qt. This is a valid point, and one of the few *real* legal problems with KDE/Qt licensing. However, removing KFloppy from KDE is trivial. So just do it!
If the plurality of Debian really wanted to include KDE, they could do so in some form. They could include just the base libraries and the non-GPL applications. They can include just the KDE source code (which sidesteps the entire linkage objection). But instead of doing this, they have spent the last two years endlessly arguing of minutia. The only reason I can fathom for them not included some form of KDE is because they just don't want to.