Slashdot Mirror
Judicial Order in MySQL AB vs. Nusphere Suit
bkuhn writes: "Judge Saris has ruled on the preliminary injunction motion. The Court recognizes in today's order that MySQL AB "seems to have the better argument" on the GNU GPL matter. The Court fully recognized the need for expert testimony at trial about the GNU GPL and the technical facts at hand, particularly as to why static linking of software components into a single, unified, compiled binary forms a derivative work of the original components."
jeez, I would say so. They wrote the program, they hold the trademark. It's theirs. They have every right to say how their name is used.
If you don't like the way someone runs their GPL project you have one choice: fork it, and call it something else. But if NuSphere wants to sell "NuSphere MySQL" I would think it would be in its best interest to respect the trademark and hard work of its owners and inventors.
No, Thursday's out. How about never - is never good for you?
I know this will be an unpopular viewpoint here, but I'm actually hoping for this to go against MySQL. The definitions of linking and derivative in the GPL are vague and confusing. Forcing the FSF to rewrite them or be given specific meaning by a judge would be tremendously helpful. This isn't as dangerous as it sounds because if the license is invalidated, users are granted no additional rights over a traditionally copyrighted work. So, code wouldn't be in danger of "escaping" in the meantime.
why static linking of software components into a single, unified, compiled binary forms a derivative work of the original components
Yes, a lot of us are curious about that too. It is clear that something is derivative if it "contains" the functionality of something else, but is it Just that the licensing agreement of a minor, or even insignificant, part of the derivative work should dominate the remaining portions? For example, if you're writing a program which decodes and formats web pages and you want to include a mechanism for playback of midi files, using a library distributed under the GPL is an impossibility, because even tho it might make up less than 1% of the total code of the product, linking to the library will force the terms of the GPL onto each and every other component. This is what the GPL says - if you dont like it, dont use the software. But is this using copyright to restrict rights beyond the intention of copyright law? And like the claim that Napster made about the RIAA, does the GPL try to "use their copyrights to extend their control to [new markets]?" Guess we have to hold our breath and wait.
How we know is more important than what we know.
The press release does not have much info on what the suit is actually about. Here are some links to explain the dispute:
I couldn't find any propaganda on the Nusphere site. I guess they're downplaying the story.
Many people complain that the GPL is not truly free.
That is mainly because of a misconception about what 'free' is. Modern theories of democracy are based on the concept of inalienable rights (see Locke or US Declaration of Independence). Inalienable rights are inherent rights that cannot be abandoned or taken away. For men to be free, they must ensure that these rights exist and are protected (hence, the need for government).
The GPL essentially defines the inalienable rights of software. As men have the rights of life, liberty, and the pursuit of happiness (or in their possesions if you are more Lockean), software has the right to be modified, redistributed, and derived from.
To simply grant these rights with no mechanism to preserve them would go against the fundamental principles of democracy. The GPL protects software not only from giving up it's fundamental rights.
But, the original author also has the choice under which license to distribute the work. He may choice to abandon certain rights. The important thing though, is that it is his choice and his choice alone.
I'm sorry, one simply cannot make the argument that the GPL is philosophically less 'free' than the BSD license. It's just not true.
BTW: If you consult the Unix History Tree, you will see that BSD is arguably older than Linux (the NetBSD base surely is).
BSD is not growing exponentially and Linux is surely not fragmenting. BSD's growth is also probably more related to the high quality of their operating system and less because of their license.
The BSD license is scary. If I had any intention of releasing my code so that it could be reused commerically without my permission, I would simply put in under the public domain. Of course, I'm not a communist, so there is little chance that I would ever release code in such a way.
int func(int a);
func((b += 3, b));
I'm curious, how do you argue that something is or isn't a derivative work? Especially software?
Do you say: amount of code? Amount of useful functionality created by the code (ie, the GPL code is used to implement a menu item that is not used in day-to-day usage of the program.) Do you make some kind of "user confusion" argument?
If you have two expert witnesses, wouldn't it just be an argument of "yes it is" vs. "no it isn't"?
I really never thought of this and assumed that there was something in the software copyright laws or some precedent that spelled this out clearly.
As much as I like the GPL I really hope that the writer of the License isn't the one who gets to define "derivative work" (ie, maybe microsoft someday would like to make their license cover any computer yours networks with, or something silly).
Agreed.
Therefore you are "forced" to distribute every other component under the GPL if you want to use the library.
Well, then you must agree that in order to use Windows, I am "forced" to pay Microsoft money. Or that in order to drive a car, I am "forced" to buy gas and therefore support Middle Eastern governments.
Force as defined by Websters dictionary is:
Force \Force\, v. t.
I hate to be a prick about it, but obviously, forced is not the right word for this situation. Choose is a better word to use. Your logic is better written as:
if you choose to use the library then you have to also choose to distribute every other component of your product under the GPL.
It's a mutual decision. It's like if I choose to have sex, I also have to choose to take the responsibility of also getting a woman pregant.
There is a significant difference between making a choice and being forced into something.
int func(int a);
func((b += 3, b));
I'm not quite sure what PD has to do with communism, unless you want to equate IP with physical property (which is honestly a kind of dicey proposition, IMO).
Releasing under something like the X/MIT license is probably better than PD, anyway, as PD leaves you legally liable for all sorts of fun things. You'll notice that the bulk of the X/MIT license is a disclaimer of liability.
DNA just wants to be free...
Back in 1996, Judge Saris made a common sense rulingin the case of State Street Bank vs. Signature Financial Group. In sumamry, SFG claimed that they had a patent on "multi-tiered" mutual funds and the software to manage it. Judge Saris ruled that the patent acquired by SGF was so broad, that no mutual fund company could do business without paying a royalty to them.
:-)
Expect a common-sense ruling from her in this case as well.
Where was she when the "1-Click" patent was challenged?
More than anything else, I'm looking for a ruling somewhere, somehow that manages to draw nice lines around where a program starts and ends for copyright and patent purposes.
We have APIs, libraries, modules, remote function calls, socket connections, pipes, shared memory, and a whole host of different ways in which code A communicates with code B. What we really need to define in a legal sense is exactly what constitutes a "program" for the purposes of code use. I definitely want to be able to isolate code which can potentially restrict "my" code's licensing (whether through a patent, restrictive copyright, GPL, or whatever). Until we get a good definition of what is "external" to a program, this will remain a legal quagmire.
Honestly, it's a hard decision. But it needs to be made. And the sooner, the better, for all parties: Free Software, Open Source, and Proprietary.
-Erik
There are always four sides to every story: your side, their side, the truth, and what really happened.
"Modern theories of democracy are based on the concept of inalienable rights"
Ehhh... not entirely true. Modern applications of democracy take into account these "inalienable rights." They do this by making sure that there's a mechanism to protect the rights of the individual from the abuses of the majority (which will happen in a true democracy). I think it was Madison that noted that it was a democracy that decreed the hemlock one day and statues the next.
"Inalienable rights are inherent rights that cannot be abandoned or taken away. For men to be free, they must ensure that these rights exist and are protected."
Either they can be taken away or they can't. Pick one.
"To simply grant these rights with no mechanism to preserve them would go against the fundamental principles of democracy. The GPL protects software not only from giving up it's fundamental rights."
Democracy is a pretty name for "mob rule," where the majority always wins over the minority and the individual. The GPL is democratic in the sense that it requires coders to surrender some of their coding rights as an individual (restricting what they can do with the code, even if they've heavily modified it) as they become a part of the group that works on the program (ie. the majority). The GPL protects the rights of the group over that of the individual by more or less forcing the individual to join the group.
"But, the original author also has the choice under which license to distribute the work. He may choice to abandon certain rights. The important thing though, is that it is his choice and his choice alone."
Ah, but who is the author? Isn't somebody that modifies the code and thereby improving it just as much of an author? By enacting the GPL, the original coder declares himself first among equals in the new group of coders that work on the app. The original author then requires that you join this group as a subordinate before you are allowed to make and distribute any modifications.
This more or less denies the existance of derivative works. It's based on the principle that the "original" author wrote the code ex nihilo while the work of anybody else is nothing more than adding to the original idea. Anybody that came along after the code's genesis (literally) is incapable of actually having an original idea (legally if not philosophically). If fire were released under the GPL, Bob the Caveman would get all the credit of the Industrial Revolution.
I personally have difficulty understanding how people can complain about the abuses of US copyright law by the MPAA and RIAA one day and support the GPL the next.
"I'm sorry, one simply cannot make the argument that the GPL is philosophically less 'free' than the BSD license. It's just not true."
It's untrue only so long as you pretend that a group and the individuals that make up that group are one and the same. Psychology and history tell us otherwise.
"BSD is not growing exponentially and Linux is surely not fragmenting."
Linux is growing exponentially because it requires anybody that does anything with the code to become part of the Linux group. BSD has no restrictions, so coders aren't forced to join the BSD camp. Like it or not, the folks that call the GPL "viral licensing" have a point. In this sense the BSD license is more free as it is not forced to grow at such a rate.
"Of course, I'm not a communist, so there is little chance that I would ever release code in such a way."
No, the GPL is the more communistic choice. It is a "coder's paradise," free from the opression of the bourgeois corporations but subject to another aristocracy just the same. Coders are more or less required to unite into one single group and produce as a whole, and any credit from the accomplishments of any single coder is both given to the group as a whole and the original Lenin figure that started the app in the first place. There is no room for the individual here except for the person that applied the GPL to begin with.
I'm curious, how do you argue that something is or isn't a derivative work? Especially software?
I'm curious whether they might consider the linked object to be a "compilation" in the literary sense - i.e. a collection of separate works published together without significant modification to each, such as an anthology.
As much as I like the GPL I really hope that the writer of the License isn't the one who gets to define "derivative work" (ie, maybe microsoft someday would like to make their license cover any computer yours networks with, or something silly).
Rule of thumb: When one side writes the contract, the courts construe the words as much in favor of the other side as possible. It's up to the author of the contract terms to make them clear, understandable, and legal.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
if you want to use the library then you have to distribute every other component of your product under the GPL. Therefore you are "forced" to distribute every other component under the GPL if you want to use the library.
I think this is clearly the point of contention. Does the use of a library make a program a derivative work ?
Remember, functionality is not copyrightable - only the specific expression in the source code is. So, for starters, if multiple libraries provide the same functionality, linking to a library does not make something a derivative work. This is a concept of a fixed boundary in copyright. The API is defined, there are multiple ways to fulfill that API, so whatever does fulfill that API does not turn the linking program into a derivative work (more specifically - it doesn't require any specific expression to work - because multiple expressions exist that allow it to work).
So, it is 100% possible to link to a GPL'd program and not be open source and not violate the GPL. There are other issues though.
If a library is unique, then it becomes a murky area as to whether a calling program is a derivative.
If a library is staticly linked, another gray area.
These are some of the things the court will sort out. However, I would not be surprised if the court finds in general that any linking to GPLd libraries does not make something a derivative.
After all, the intended purpose of a library is to have something link to it using a public API. Therefore, it seems silly to make anything that does that a derivative. However, you could argue that any program can be turned into a library, so that using the intent of making something a library may not be good either.
In any case, it should get interesting.