The GPL functions as much as a social contract as a legal document, but if a court gave people, and especially corporations, free reign to break it and steal GPL'd code, I don't doubt that some would.
If a court ruled the GPL invalid in some way, that would not give anyone the right to use code without giving back. For example, if the GPL's 'viral' provisions are found illegal, it would most likely mean you're not at all allowed to use GPLed code. Remember you don't have a license to use GPLed code unless you put your code under the GPL. If that requirement is illegal, then you don't have a license to use GPLed code at all!
The GPL quite clearly requires you to 'pay' for the code you use, and if the type of 'payment' it requires cannot be required for legal reasons, it does not mean that it suddenly requires no 'payment' at all. If someone offers a license to you that is legally invalid, you can't just use their copyrighted material, you'll have to wait for a valid offer (i.e. wait for the FSF to write a new version of the GPL).
Welcoem to the idea that regulation works to benefit the regulated (George Stigler expalined this long ago, and is teh "economist" in my subject line). Regulation, no matter how much it is touted as for the "public good," ultimately benefits the regulated.
True, but often (not always) it also benefits the public. 'Market forces' cannot apply before there is a track record to look at, but if you're in prison just because your lawyer didn't have any education (and was, due to a lack of regulation, not required to tell you), it won't help you that he'll eventually be out of business.
That's not to say that regulation is good in all cases, or that regulation as it's done now is perfect. A better regulation might be, for example, to require lawyers to make public what kind of education they had. Someone who only has a simple question might actually want advise from a 15-year-old who's not a 'lawyer'. Just make sure people know he's not.
If you research the Handelsregister (trade register) of Berlin, you'll find
this entry of Fully Licensed GmbH (information is incomplete, you need to pay for full information). One of their managers, Thomas Lopatic, was born in Unterschleißheim. What's wrong about that? Nothing, except it's where Microsoft Deutschland GmbH is located.
Also note that the trade register entry is from June 8, 2001. I cannot see if that was when they were founded or when the entry was last changed. Anyway, cracking a Microsoft product seems a bit strange as a company's first project to me.
Coincidence? Maybe, but given that they essentially said there would be no need for privacy concerns, I'd be careful with this source of information.
Note that they did not demand the package (package as in project) to be destroyed, i.e. to delete all the source code or something. That's a translation error.
What they did demand is that all packaging material (those boxes you buy commercial software in) be destroyed. This is pretty obvious, as the infringing name is usually printed on those boxes. However it shows that those lawyers have no idea about what open source software is, because this software is not packaged. You just download it.
but K can easily pull the "its a word... we could have called it Kstreet-light for gods' sake..".
...and Adobe can easily say "So why didn't you?".
Come on, they've used this name because it is similar to Adobe's well-known product. I agree that a trademark on the word 'Illustrator' is not a particularly good trademark, but the fact that the KIllustrator developers have a somewhat resonable excuse doesn't justify the choice of the name.
Open source projects are often asking companies not to enforce patents, trademarks etc. against open source, with the justification that such things should be the property of the public. Why can't the comapanies also ask open source to be respectful towards what they believe should be their property?
This makes me wonder what will happen to the SMT design. Alpha's overnext chip design (EV8) was desined with support for SMT (Simultaneous Multithreading (-tasking?)). I don't know a lot about that technology, but it's something like the processor running several tasks at the same time, in different pipelines.
The IA-64 runs only one process at a time, and relies on the compiler for optimization (which is a problem for JIT compilers). It also heavily relies on speculative execution, so many calculations might be thrown away later because the speculation was wrong. All of that draws lots of power, resulting in Intel processors being the most power-consuming CPUs around. German computer magazine c't recently tested an Itanium prototype system that had a heat production of 1.8 kilowatt!
I hope Intel didn't purchase Alpha just to get rid of competition, but somehow I doubt that we'll see a merger of IA-64 and EV8....
Well, that's just how a large group of people behaves. You look at what are the others doing, then you go the same way. Of course, the others also look at what you are doing. That's basically why a so-called 'mainstream' happens to exist, and why it's so hard for new ideas to get into the mainstream.
You might argue that this behavior is not reasonable, and while this is true to a certain extent (for example, the mainstream sometimes doesn't choose the technically best solution), it also has some advantages, like building up de facto standards, which help in other areas like add-on services, education etc.
Of course, there are always some who don't go with the mainstream, but, as the word mainstream suggests, those are not the majority.
Using this method you'd solve most of the problems that come with shared libraries, but you'd also lose most of the advantages. If a user installs a bugfixed version of a library (that is otherwise completely compatible with the previous version) no application will ever get to use it.
To find out what version a library is, versions numbers are the best solution.
Of course, it's the library developer's responsibility to make sure that different versions changed only in the minor version number are actually compatible, as the numbering suggests.
unless of course you have signed a contract with the software vendor that prohibits you to do so
... I should add that generic OEM licenses that the customers agree on when installing the software don't count here, as those aren't individual contracts, and certain provisions are illegal under german law unless the contract has been individually negotiated.
One thing I'd like to comment on, however, is the "Microsoft Licensing" question. The justification for licenses which prohibit resale given here is that this prevents persons or organizations who are eligible to receive a less expensive license from reselling it. That's true, but preventing resale is not the only way of doing this.
Some of you might know that the highest german court has recently ruled that reselling OEM software is legal (unless of course you have signed a contract with the software vendor that prohibits you to do so).
Still, the court has agreed that there might be different types of usage for a software, and a product may be sold for one specific type of usage only. For example, it is still illegal to sell or use a single user software as a network software, or an educational version as a standard version.
(In fact, the court has simply denied that OEM versions are a seperate type of usage. OEM versions are considered a special low price offer that anyone is of course free to make, however that doesn't give the vendor any additional rights.)
This allows software vendors to have lower pricing for schools (or higher pricing for network versions), while still allowing customers to resell the software -- i.e., one school may sell its software to another school, but not to a business.
I'm not saying the one or the other system is better (though the no-resale licensing seems less consumer friendly to me), I just wanted to point out that there are alternative solutions as well.
Slashdot Mirror
If a court ruled the GPL invalid in some way, that would not give anyone the right to use code without giving back. For example, if the GPL's 'viral' provisions are found illegal, it would most likely mean you're not at all allowed to use GPLed code. Remember you don't have a license to use GPLed code unless you put your code under the GPL. If that requirement is illegal, then you don't have a license to use GPLed code at all!
The GPL quite clearly requires you to 'pay' for the code you use, and if the type of 'payment' it requires cannot be required for legal reasons, it does not mean that it suddenly requires no 'payment' at all. If someone offers a license to you that is legally invalid, you can't just use their copyrighted material, you'll have to wait for a valid offer (i.e. wait for the FSF to write a new version of the GPL).
Welcoem to the idea that regulation works to benefit the regulated (George Stigler expalined this long ago, and is teh "economist" in my subject line). Regulation, no matter how much it is touted as for the "public good," ultimately benefits the regulated.
True, but often (not always) it also benefits the public. 'Market forces' cannot apply before there is a track record to look at, but if you're in prison just because your lawyer didn't have any education (and was, due to a lack of regulation, not required to tell you), it won't help you that he'll eventually be out of business.
That's not to say that regulation is good in all cases, or that regulation as it's done now is perfect. A better regulation might be, for example, to require lawyers to make public what kind of education they had. Someone who only has a simple question might actually want advise from a 15-year-old who's not a 'lawyer'. Just make sure people know he's not.
If you research the Handelsregister (trade register) of Berlin, you'll find this entry of Fully Licensed GmbH (information is incomplete, you need to pay for full information). One of their managers, Thomas Lopatic, was born in Unterschleißheim. What's wrong about that? Nothing, except it's where Microsoft Deutschland GmbH is located.
Also note that the trade register entry is from June 8, 2001. I cannot see if that was when they were founded or when the entry was last changed. Anyway, cracking a Microsoft product seems a bit strange as a company's first project to me.
Coincidence? Maybe, but given that they essentially said there would be no need for privacy concerns, I'd be careful with this source of information.
They should consider doing research on things people are really want to have. Like nuclear rockets. Oh, wait.
Note that they did not demand the package (package as in project) to be destroyed, i.e. to delete all the source code or something. That's a translation error.
What they did demand is that all packaging material (those boxes you buy commercial software in) be destroyed. This is pretty obvious, as the infringing name is usually printed on those boxes. However it shows that those lawyers have no idea about what open source software is, because this software is not packaged. You just download it.
but K can easily pull the "its a word... we could have called it Kstreet-light for gods' sake..".
...and Adobe can easily say "So why didn't you?".
Come on, they've used this name because it is similar to Adobe's well-known product. I agree that a trademark on the word 'Illustrator' is not a particularly good trademark, but the fact that the KIllustrator developers have a somewhat resonable excuse doesn't justify the choice of the name.
Open source projects are often asking companies not to enforce patents, trademarks etc. against open source, with the justification that such things should be the property of the public. Why can't the comapanies also ask open source to be respectful towards what they believe should be their property?
You might want to try scrolling past the first paragraph. Looks almost like English to me, what do you think?
This makes me wonder what will happen to the SMT design. Alpha's overnext chip design (EV8) was desined with support for SMT (Simultaneous Multithreading (-tasking?)). I don't know a lot about that technology, but it's something like the processor running several tasks at the same time, in different pipelines.
The IA-64 runs only one process at a time, and relies on the compiler for optimization (which is a problem for JIT compilers). It also heavily relies on speculative execution, so many calculations might be thrown away later because the speculation was wrong. All of that draws lots of power, resulting in Intel processors being the most power-consuming CPUs around. German computer magazine c't recently tested an Itanium prototype system that had a heat production of 1.8 kilowatt!
I hope Intel didn't purchase Alpha just to get rid of competition, but somehow I doubt that we'll see a merger of IA-64 and EV8....
Well, that's just how a large group of people behaves. You look at what are the others doing, then you go the same way. Of course, the others also look at what you are doing. That's basically why a so-called 'mainstream' happens to exist, and why it's so hard for new ideas to get into the mainstream.
You might argue that this behavior is not reasonable, and while this is true to a certain extent (for example, the mainstream sometimes doesn't choose the technically best solution), it also has some advantages, like building up de facto standards, which help in other areas like add-on services, education etc.
Of course, there are always some who don't go with the mainstream, but, as the word mainstream suggests, those are not the majority.
Using this method you'd solve most of the problems that come with shared libraries, but you'd also lose most of the advantages. If a user installs a bugfixed version of a library (that is otherwise completely compatible with the previous version) no application will ever get to use it.
To find out what version a library is, versions numbers are the best solution.
Of course, it's the library developer's responsibility to make sure that different versions changed only in the minor version number are actually compatible, as the numbering suggests.
... part of the monitor?
... I should add that generic OEM licenses that the customers agree on when installing the software don't count here, as those aren't individual contracts, and certain provisions are illegal under german law unless the contract has been individually negotiated.
Wow! What a great article!
One thing I'd like to comment on, however, is the "Microsoft Licensing" question. The justification for licenses which prohibit resale given here is that this prevents persons or organizations who are eligible to receive a less expensive license from reselling it. That's true, but preventing resale is not the only way of doing this.
Some of you might know that the highest german court has recently ruled that reselling OEM software is legal (unless of course you have signed a contract with the software vendor that prohibits you to do so).
Still, the court has agreed that there might be different types of usage for a software, and a product may be sold for one specific type of usage only. For example, it is still illegal to sell or use a single user software as a network software, or an educational version as a standard version.(In fact, the court has simply denied that OEM versions are a seperate type of usage. OEM versions are considered a special low price offer that anyone is of course free to make, however that doesn't give the vendor any additional rights.)
This allows software vendors to have lower pricing for schools (or higher pricing for network versions), while still allowing customers to resell the software -- i.e., one school may sell its software to another school, but not to a business.
I'm not saying the one or the other system is better (though the no-resale licensing seems less consumer friendly to me), I just wanted to point out that there are alternative solutions as well.