I'm thinking (with no particular evidence) that Trolltech is the company that may be involved in the rumored FSF GPL testcase.
Extremely unlikely. If it were true, RMS and his lawyers would walk out of the courtroom five minutes after they entered, then proceed to the men's room to wash the egg off of their faces.
The FSF can only sue on behalf of its own software. It does not hold copyright to any part of KDE or Qt. Furthermore, Trolltech doesn't even release anything under the GPL. What possible legal grounds could the FSF have against either of these entities?
I'm pretty sure a lot of the KDE developpers don't grasp the difference between Free software and Open Source.
So what is the difference?
I've read Richard Stallman's definition of Free Software, and I've read the OSS's definition of Open Source. As near as I can tell, anything that is Open Source must also be Free Software. Which one of RMS's four points does the OSS definition not cover?
There are several high quality toolkits out there without such restrictions...
Umm, you forget to list any restrictions. I can only find two restrictions that anyone can point to even vaguely onerous. Both of which, if you had read the article, are about to go away.
1) Patch clause. Gee, every OSS project I know of accepts only patches anyway. In any case, the only thing this restricts is forking.
2) No private distributions. If you distribute the software, you distribute the software. No if ands or buts.
I think it's valid to ask the question whether this is a good deal.
It's an excellent deal. Qt is Free Software if you're using it to write Free Software, but it's proprietary if you're writing proprietary. In a world where everyone keeps telling me "life isn't fair", this is a breath of fresh air.
And Qt is so expensive that I would have serious concerns about using it even as a purely commercial toolkit.
Have you priced out commercial toolkits? I don't mean those half-assed Microsoft offerings, I mean real-world high-quality tools.
That you have to resort to some of the most restrictive dictatorships to defend American society really bothers me in that context.
What really bothers me is that so many people assume that (insert your grievance here) only happens in the US. It's so nice to hear that there are no corporations in Europe and no patents in Asia. And of course, political coruption only occurs in in certain parts of North America.
No, the right thing would be to conform to the existing standard rather than extend it in some arbitrary way.
Please tell that to the GNU folks as well. When you use bash extensions, your scripts no longer work with the industry standard sh. Ditto for gmake, bison, etc.
The point is, there's nothing wrong with extending standards. The problem arises when you *change* the standard. bash still handles sh scripts and gmake handles standard Makefiles. And MS Kerberos can still use the industry standard Kerberos protocols.
Your virulent-anti-Microsoft frame of mind coupled with the lack of *equally* strong foundations to back it up make me wonder about your mental health.
It can't be the closed source nature of Microsoft, because you don't display the same hatred towards other closed source companies or even shareware authors. It can't be the size of Microsoft, because again you aren't displaying the same level of antagonism towards Oracle, IBM or Cisco. And it can't be the monopoly position because I haven't heard your utterances about TW/AOL brainwashing or "One (insert telco here) Way"
"I don't know how I'd react if any of my friends went and worked there"
When you go off the deep end in your dislike of Redmond inhabitants you do more harm to your own position than to Microsoft's. In fact, I think Bill and Steve would have loved to have you on the witness stand before Judge Penfield.
Then britain has very strange laws! Certainly the British bus passengers should expect legal rights and priviledges while they ride, but this is a very different thing than a contract. If this is the case in England, I suspect it derives not from the act of boarding the bus, from instead from the act of purchasing a fair (in the US, bills of sale can are construed as contracts, and is one of the basis' for the Commercial Code).
Of course, in the US, and especially with DMCA and UCITA, the act of tearing of shrink-wrap is considered more binding than a notarized signature. But so what. If Congress passed a law that said night was day and the blue was green, it still won't make it so.
I've read their license. Forwards and backward and even upside down. Those portions (and they are major portions) of KDE that are not under the GPL can be put in Debian with no problems other than the fact that no one is stepping up to the plate and doing it. Those portions of KDE that are under the GPL may also be distributed since all requirements of the GPL are indeed met. The problem comes when you examine GPL clauses out of context. You have to apply the GPL as a whole. This has been covered in minute detail elsewhere, including on the Debian lists and by attorneys, so I won't quote the GPL yet again.
What pisses me off the most is that Debian won't include an AL licensed Cervisia (as an example) and it's LGPL licensed kdelibs. That they won't do even this indicates to me their true motive.
If the RIAA doesn't want people to link to certain units of their intellectual property, the solution is absurdly simple. Just GPL those files you want protected. Since you can't link from a non-GPL application to a GPL library, then neither should you be able to link from a non-GPL website to a GPL MP3. Instead of raising a huge legal ruckus of copyright infringement, just mention a possible GPL violation here on slashdot and the perps will be quickly intimidated.
Then when you install cvsup, make sure you use tag=RELENG_4
That's what I did this weekend. After a two hour build, it wouldn't install! I found little errors everywhere! After doing a bit of reading, I suddenly realized that STABLE is not necessarily stable. It is still a development release that changes from day to day. It safer than current, but you still don't stake your box on it. Stick with RELEASE unless you have a pressing need for the latest and greatest. The release cycle is quick enough that you can live with it.
Now I'm totally confused! A screw it all to hell, I'll just do what KDE Core says I can do and go ahead of distribute it. If I get arrested for legally distributing it, then I'll just resign from the English language.
In order for a contract to be a valid contract, both parties must accept the specific contract. Usually this acceptance is made by each party signing on the dotted line, or shaking hands, or saying "Yes, I Agree".
But in the case of the GPL we have a case where the user does not need to indicate any acceptance, and indeed, is not asked to. Merely saying "if you distribute then you indicate agreement" is just not good enough.
Furthermore, there is no specific contract. There is only a general contract. Even if I agree to the GPL "contract", the original author sure as hell hasn't, because he has no idea I even have his software! Contracts are always two-sided. That's why there must be consideration. In the case of the GPL-as-contract, we have only one side unilaterally setting terms to parties unknown, unnamed and unnumbered. If you go to a business of some kind and ask "does John Smith have a contract with you?", they will be able to definitively answer yes or no. But if you ask that of Richard Stallman, he will have no clue who John Smith is, let alone if he has agreed to the terms of the GPL.
"If the GPL is ruled invalid (which I regard as highly improbable) then the situation reverts to normal: that is, no redistribution allowed."
Wrong, sorry. I still have permission from the author to redistribute the software.
This case will not be about the right of the author to grant permission to users to redistribute his work. That one's a no brainer. I rather suspect that it will be about some vague or controversial clause. Like linking a GPL library to a non-GPL program, or whether the GPL applies to third parties, or whether the GPL is a contract or a permission statement, etc.
Then just write a turtle graphics class/library for C++, and Voila!
UCSD Pascal, that old standard of beginning programming before Turbo Pascal hit the scenes, had turtle graphics as a library, and it did wonders for teaching programming (I was a proctor at UCSD). Instant visual feedback is a powerful teacher.
So go write a turtle graphics class for C++ and get teaching!
And you advocate Java as an introductory language instead?
Java is so close to C++ that they might as well be different dialects of the same language. Java is a bit more "pure" OO while C++ is a bit more "down in the guts", but overall, they would be identical for new programmers.
C++ would make a great first language. First, they need learn only one language, instead of one for beginners, another for intermediate, and finally C++ for advanced. Two, the potentially confusing stuff in C++ can be left for later. They are not required to program. Classes can be introduced after mastering structured programming, and templates, STL and generic programming can be introduced after learning OOP. Third, teaching students to use proper coding styles makes any language easier.
I think this arguing against C++ as a beginners language arises out something other than it's difficulty. No proper programming language can be taught in 24 hours. There may only be 24 hours of lecture time, but this does not count the weeks spent on homework and labwork.
"Practical C++ Programming" is a great book for learning C++, even if one is not a programmer. Coupled with a comprehensive C++ reference, it would make a nice text for a beginning programming class.
What I was referring to was the so-called "system library" exception, which only counts if the "system library" is not distributed with the GPL'd software itself. So if both Qt and KDE are on the same CD, then the system library exception doesn't kick in.
I think that this is going to come down as a matter of opinion. There's many ways to interpret the GPL, and I can certainly see how you do. But do you at least understand my interpretation? Qt is distributed on the same CD as KDE, and if it is indeed a part of the Program (KDE) that needs to be GPLd, then it needs to be distributed according to sections 1 and 2. But Qt is not "derived from the Program, and can be reasonably considered independent and separate works in" itself", so it is excluded from the requirements of the GPL.
The "proof" you cite says nothing about KDE infringing anyone's copyright, only that distributing KDE is illegal (and it's still your opinion).
When you keep your cool, your arguments are rational and thought provocating. But I find it impossible to attribute those qualities to you when you lose your temper.
p.s. KDE is indeed a legal entity, though whether Mr. Bastian is a member of that entity is another matter.
KDE is not licensed under the QPL. There is not requirement in the KDE licenses to distribute modifications as patches. Only the Qt library is under that restriction.
"Whether Qt is a "system library" or not is pretty questionable, and in any case it would be distributed on the same CD as KDE"
Wait! This is either a double standard or an artful dodge...
Stop mixing and matching clauses to fit your theory. Way before the "exception" clause, there is section 2, which specifically disassociated the GPL from software distributed aggregate to the Program. The exception clause is excepting from the exception stuff that is included "inside" the Program, and not stuff that is distributed elsewhere on the CD.
"The recent $3000 offer to KDE to fix their license problems went unclaimed."
I wanted to claim it:-) I really did. And it was offered to me. But I had to turn it down because the offer was in error. Someone grabbed my name from the kde cvs user list, but didn't check to see whether I was actually a KDE developer (I did some work on some web pages).
Receiving this message was doubly ironic since I am the proud author of one of only five (last I checked) "legal" Qt programs included with Debian.
Slashdot Mirror
I'm thinking (with no particular evidence) that Trolltech is the company that may be involved in the rumored FSF GPL testcase.
Extremely unlikely. If it were true, RMS and his lawyers would walk out of the courtroom five minutes after they entered, then proceed to the men's room to wash the egg off of their faces.
The FSF can only sue on behalf of its own software. It does not hold copyright to any part of KDE or Qt. Furthermore, Trolltech doesn't even release anything under the GPL. What possible legal grounds could the FSF have against either of these entities?
the only substantive reason a distribution would include Qt is so it can include KDE
Oh wow! This must be why Debian includes the Qt-2.x library! Thanks for clearing that up for me.
I'm pretty sure a lot of the KDE developpers don't grasp the difference between Free software and Open Source.
So what is the difference?
I've read Richard Stallman's definition of Free Software, and I've read the OSS's definition of Open Source. As near as I can tell, anything that is Open Source must also be Free Software. Which one of RMS's four points does the OSS definition not cover?
There are several high quality toolkits out there without such restrictions...
Umm, you forget to list any restrictions. I can only find two restrictions that anyone can point to even vaguely onerous. Both of which, if you had read the article, are about to go away.
1) Patch clause. Gee, every OSS project I know of accepts only patches anyway. In any case, the only thing this restricts is forking.
2) No private distributions. If you distribute the software, you distribute the software. No if ands or buts.
I think it's valid to ask the question whether this is a good deal.
It's an excellent deal. Qt is Free Software if you're using it to write Free Software, but it's proprietary if you're writing proprietary. In a world where everyone keeps telling me "life isn't fair", this is a breath of fresh air.
And Qt is so expensive that I would have serious concerns about using it even as a purely commercial toolkit.
Have you priced out commercial toolkits? I don't mean those half-assed Microsoft offerings, I mean real-world high-quality tools.
That you have to resort to some of the most restrictive dictatorships to defend American society really bothers me in that context.
What really bothers me is that so many people assume that (insert your grievance here) only happens in the US. It's so nice to hear that there are no corporations in Europe and no patents in Asia. And of course, political coruption only occurs in in certain parts of North America.
No, the right thing would be to conform to the existing standard rather than extend it in some arbitrary way.
Please tell that to the GNU folks as well. When you use bash extensions, your scripts no longer work with the industry standard sh. Ditto for gmake, bison, etc.
The point is, there's nothing wrong with extending standards. The problem arises when you *change* the standard. bash still handles sh scripts and gmake handles standard Makefiles. And MS Kerberos can still use the industry standard Kerberos protocols.
Your virulent-anti-Microsoft frame of mind coupled with the lack of *equally* strong foundations to back it up make me wonder about your mental health.
It can't be the closed source nature of Microsoft, because you don't display the same hatred towards other closed source companies or even shareware authors. It can't be the size of Microsoft, because again you aren't displaying the same level of antagonism towards Oracle, IBM or Cisco. And it can't be the monopoly position because I haven't heard your utterances about TW/AOL brainwashing or "One (insert telco here) Way"
"I don't know how I'd react if any of my friends went and worked there"
When you go off the deep end in your dislike of Redmond inhabitants you do more harm to your own position than to Microsoft's. In fact, I think Bill and Steve would have loved to have you on the witness stand before Judge Penfield.
Then britain has very strange laws! Certainly the British bus passengers should expect legal rights and priviledges while they ride, but this is a very different thing than a contract. If this is the case in England, I suspect it derives not from the act of boarding the bus, from instead from the act of purchasing a fair (in the US, bills of sale can are construed as contracts, and is one of the basis' for the Commercial Code).
Of course, in the US, and especially with DMCA and UCITA, the act of tearing of shrink-wrap is considered more binding than a notarized signature. But so what. If Congress passed a law that said night was day and the blue was green, it still won't make it so.
I've read their license. Forwards and backward and even upside down. Those portions (and they are major portions) of KDE that are not under the GPL can be put in Debian with no problems other than the fact that no one is stepping up to the plate and doing it. Those portions of KDE that are under the GPL may also be distributed since all requirements of the GPL are indeed met. The problem comes when you examine GPL clauses out of context. You have to apply the GPL as a whole. This has been covered in minute detail elsewhere, including on the Debian lists and by attorneys, so I won't quote the GPL yet again.
What pisses me off the most is that Debian won't include an AL licensed Cervisia (as an example) and it's LGPL licensed kdelibs. That they won't do even this indicates to me their true motive.
If the RIAA doesn't want people to link to certain units of their intellectual property, the solution is absurdly simple. Just GPL those files you want protected. Since you can't link from a non-GPL application to a GPL library, then neither should you be able to link from a non-GPL website to a GPL MP3. Instead of raising a huge legal ruckus of copyright infringement, just mention a possible GPL violation here on slashdot and the perps will be quickly intimidated.
Pardon me for asking, but what the hell do you need SMP on a gateway for?
Then when you install cvsup, make sure you use tag=RELENG_4
That's what I did this weekend. After a two hour build, it wouldn't install! I found little errors everywhere! After doing a bit of reading, I suddenly realized that STABLE is not necessarily stable. It is still a development release that changes from day to day. It safer than current, but you still don't stake your box on it. Stick with RELEASE unless you have a pressing need for the latest and greatest. The release cycle is quick enough that you can live with it.
illegal to legally distribute
Now I'm totally confused! A screw it all to hell, I'll just do what KDE Core says I can do and go ahead of distribute it. If I get arrested for legally distributing it, then I'll just resign from the English language.
SysV init scripts are much less of a pain when you need to add some service automatically...
How frequent are your service changes that you need to be able to do this automatically? I'm really confused as to your whole point here.
"Hang on, it's Three O'Clock, time to change runlevels again..."
acceptance
In order for a contract to be a valid contract, both parties must accept the specific contract. Usually this acceptance is made by each party signing on the dotted line, or shaking hands, or saying "Yes, I Agree".
But in the case of the GPL we have a case where the user does not need to indicate any acceptance, and indeed, is not asked to. Merely saying "if you distribute then you indicate agreement" is just not good enough.
Furthermore, there is no specific contract. There is only a general contract. Even if I agree to the GPL "contract", the original author sure as hell hasn't, because he has no idea I even have his software! Contracts are always two-sided. That's why there must be consideration. In the case of the GPL-as-contract, we have only one side unilaterally setting terms to parties unknown, unnamed and unnumbered. If you go to a business of some kind and ask "does John Smith have a contract with you?", they will be able to definitively answer yes or no. But if you ask that of Richard Stallman, he will have no clue who John Smith is, let alone if he has agreed to the terms of the GPL.
The key here is that nothing but the GPL grants me the right to distribute a derivative work.
Three buts:
1) If the GPL is not a contract, then can a permission be tied to a condition? Or even more esoteric, does one have to agree to a permission?
2) Linking to a library is not derivation under copyright law. Yet the FSF claims that it is.
3) Copyright law gives me a couple of exceptions that allow me to redistribute parts of copyrighted works under special circumstances.
"If the GPL is ruled invalid (which I regard as highly improbable) then the situation reverts to normal: that is, no redistribution allowed."
Wrong, sorry. I still have permission from the author to redistribute the software.
This case will not be about the right of the author to grant permission to users to redistribute his work. That one's a no brainer. I rather suspect that it will be about some vague or controversial clause. Like linking a GPL library to a non-GPL program, or whether the GPL applies to third parties, or whether the GPL is a contract or a permission statement, etc.
Then just write a turtle graphics class/library for C++, and Voila!
UCSD Pascal, that old standard of beginning programming before Turbo Pascal hit the scenes, had turtle graphics as a library, and it did wonders for teaching programming (I was a proctor at UCSD). Instant visual feedback is a powerful teacher.
So go write a turtle graphics class for C++ and get teaching!
And you advocate Java as an introductory language instead?
Java is so close to C++ that they might as well be different dialects of the same language. Java is a bit more "pure" OO while C++ is a bit more "down in the guts", but overall, they would be identical for new programmers.
C++ would make a great first language. First, they need learn only one language, instead of one for beginners, another for intermediate, and finally C++ for advanced. Two, the potentially confusing stuff in C++ can be left for later. They are not required to program. Classes can be introduced after mastering structured programming, and templates, STL and generic programming can be introduced after learning OOP. Third, teaching students to use proper coding styles makes any language easier.
I think this arguing against C++ as a beginners language arises out something other than it's difficulty. No proper programming language can be taught in 24 hours. There may only be 24 hours of lecture time, but this does not count the weeks spent on homework and labwork.
"Practical C++ Programming" is a great book for learning C++, even if one is not a programmer. Coupled with a comprehensive C++ reference, it would make a nice text for a beginning programming class.
What I was referring to was the so-called "system library" exception, which only counts if the "system library" is not distributed with the GPL'd software itself. So if both Qt and KDE are on the same CD, then the system library exception doesn't kick in.
I think that this is going to come down as a matter of opinion. There's many ways to interpret the GPL, and I can certainly see how you do. But do you at least understand my interpretation? Qt is distributed on the same CD as KDE, and if it is indeed a part of the Program (KDE) that needs to be GPLd, then it needs to be distributed according to sections 1 and 2. But Qt is not "derived from the Program, and can be reasonably considered independent and separate works in" itself", so it is excluded from the requirements of the GPL.
The "proof" you cite says nothing about KDE infringing anyone's copyright, only that distributing KDE is illegal (and it's still your opinion).
When you keep your cool, your arguments are rational and thought provocating. But I find it impossible to attribute those qualities to you when you lose your temper.
p.s. KDE is indeed a legal entity, though whether Mr. Bastian is a member of that entity is another matter.
KDE is not licensed under the QPL. There is not requirement in the KDE licenses to distribute modifications as patches. Only the Qt library is under that restriction.
Sorry, TrollTech isn't that large. They have about maybe half a dozen KDE guys, leaving several hundred employed elsewhere.
"Whether Qt is a "system library" or not is pretty questionable, and in any case it would be distributed on the same CD as KDE"
Wait! This is either a double standard or an artful dodge...
Stop mixing and matching clauses to fit your theory. Way before the "exception" clause, there is section 2, which specifically disassociated the GPL from software distributed aggregate to the Program. The exception clause is excepting from the exception stuff that is included "inside" the Program, and not stuff that is distributed elsewhere on the CD.
"The recent $3000 offer to KDE to fix their license problems went unclaimed."
:-) I really did. And it was offered to me. But I had to turn it down because the offer was in error. Someone grabbed my name from the kde cvs user list, but didn't check to see whether I was actually a KDE developer (I did some work on some web pages).
I wanted to claim it
Receiving this message was doubly ironic since I am the proud author of one of only five (last I checked) "legal" Qt programs included with Debian.