Slashdot Mirror
MySQL AB and Nusphere Go to Court Over GPL
A little fairy whispered in our ear: "MySQL AB is seeking a temporary injunction against NuSphere, even though they've finally released the source code for Gemini and MySQL Advantage. According to the GPL, NuSphere lost the right to redistribute when they violated #3 by not providing the source code originally. The FSF will testify tomorrow in court, according to this Newsforge article." Newsforge and Slashdot are both part of OSDN. We've done a couple of previous stories about the MySQL AB vs. Nusphere conflict: the original story, a follow-up, and a note about a countersuit.
Update: 02/26 21:15 GMT by T :
bkuhn (Bradley Kuhn of the Free Software Foundation) writes: "The FSF has a press release on the matter and affidavit that we filed is also available."
What will be the ramifications if the GPL doesn't hold up in court? If the GPL hasn't been challenged in court thus far, this could have far reaching implications within the Linux community. If the GPL doesn't hold up, does that mean Microsoft is free to take large chunks of GPL'd software and make it proprietary?
Sounds like Bush should add NuSphere to his "Axis of Evil" list.
I, for one, use MySQL all of the time, but had no idea about the history of it, or the fact that NuSphere didn't create it.
Pretty pathetic, NuSphere. I, for one, hope you get nailed in court, and all of your source code gets handed over to MySQL AB. You deserve it!
NuSphere clearly didn't do "the right thing" and I hope they get their butts kicked.
Unfortunately, "credible" in this context may boil down to, "defended by the most expensive lawyers/lobbying on earth". The only real hope that the GPL has may be defense from folks like IBM, not the FSF, or any of the small, independent developers or software houses using it.
Common sense and existing law may say that making a single digital copy of a piece of music, software, etc. should fall under "fair use," but the RIAA and MPAA can still get trash like the DMCA made into law, and defended in court. Remember, this is America, where money == power, period.
So the article said that NuSphere registered MySQL.org, and AFAIK the real MySQL web page is at MySQL.com.. but --
[xaxxon@chopper log]$ nslookup www.mysql.com
Server: localhost
Address: 127.0.0.1
Non-authoritative answer:
Name: www.mysql.com
Address: 64.28.67.70
[xaxxon@chopper log]$ nslookup www.mysql.org
Server: localhost
Address: 127.0.0.1
Non-authoritative answer:
Name: www.mysql.org
Address: 64.28.67.70
Did I mis-read?
EULAs are a strange beast. They are a non-negotiated contract made through click-through or breaking a seal for something that you purchased already. The theory behind a EULA is that you contract to the EULA. You do not need a license to run software. If you need to clikc Agree to use the software, have you enterred into a contract? That's an INTERESTING legal question.
GPL is MUCH less interesting. By default, you have NO right to distribute software. The GPL is a distribution license.
This Slashdot mental masturbation is childish. The odds of the GPL being overturned and everyone's software under license being made public domain is pretty close to 0%. It is only a concern on Slashdot.
The GPL hasn't been to court because every violator has reached a settlement.
This case sounds like NuSphere is fucked. The portion in question suggests that if you violate the terms of the license the license is voided. This is pretty standard stuff.
Here is the question that the court will answer.
If I break the GPL, I can be sued for damages, etc., and must stop distribution. My license is revoked, etc., etc. Can I then go out, download a fresh copy and distribute under the terms of the GPL? Stallman says no, I'm not certain. That's where this case is questionable.
However, this is a good test case for the GPL. The question of derivative work is interesting. I'm not certain that the linking scenario creates a derivative work. However, since this company distributed a modified MySQL with their additions, they are CLEARLY distributing the work.
They need to establish that they have a separate license or did so under the GPL.
Regardless, the GPL being invalidated would not make things Public Domain. Without license you cannot distribute, so if the license falls, no distribution under GPL v2. FSF releases GPL v2.1 within a week and any provision that includes (or later version) is fine, everyone else needs to update.
Alex
It is also acceptable for the license to require that, if you have distributed a modified version and a previous developer asks for a copy of it, you must send one.
and also...
3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following:
a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
c) Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.)
I don't see, anywhere in the liscense, a timeline specified as to when any changes to source code must be supplied. Anyone know of a quote from the GPL that specifies that the source must be made available at the time of binary release? AFAIK, they don't have to release the code to anyone who doesn't specifically request it, and who has a copy of the binaries; and I don't think time-frame is brought into it.
put the what in the where?
Basically, MySQL AB is arguing that because NuSphere violated the GPL *in the past*, they have forfeited their right to distribute a GPL app, even though the source code in question has since been released in compliance with the terms of the GPL. What are the ramifications if they succeed? Does this mean NuSphere will be unable to distribute any GPL applications? Will the NuSphere contributions be deemed invalid? Will MySQL AB then distribute MySQL with Gemini tables while NuSphere is not allowed to? (Gemini tables are a NuSphere contribution.)
Isn't it sadly ironic? The first time GPL is tested in court, and I hope they loose, for all our sakes. No matter which way the decision goes, the dispute hurts the entire free software community, and will make corporate contributions to code base that much harder to come by. And think of what Craig Mundie from Microsoft will have to say about this! All over what was originally a domain name dispute. (If your new to this conflict, NuSphere paid MySQL a huge chunk of change for the right to distribute MySQL, (that's right, paid for the right to distribute a GPL app.). In the process, they created a mysql.org web page that did not give any credit to MySQL AB, and indeed, only made SQL Source Code available to registered users. MySQL AB charged NuSphere with GPL violation. In response, NuSphere made the source code more available and released the source code for their own proprietary modules. It is still not clear to me, however, what exactly MySQL AB thinks the millions they were paid by NuSphere was supposed to be for. Somehow, the right to distribute a GPL app for a limited time just doesn't make sense.)
MySQL AB, I thank you very much for your hard work and development of MySQL. But please, stop pissing on the GPL to advance your own agenda. You can debate whether the money you were paid gave NuSphere the right to hijack MySQL trademark until the cows come home; but all GPL issues have been put to bed months ago!
The question here is whether if party A has distributed party B's source-code without party B's permission, party B can deny party A permission to distribute from then on, no matter what party A does. This is a subtle point. (Item 4 in the GPL applies.) For an example, this is the exact situation that KDE wound up on the wrong side of which RMS pointed out in the "forgiveness" speech. (A speech that ignorant morons all over flamed him about, but that is another story.) Having once not been in compliance with the GPL, you need explicit forgiveness from the copyright holder to regain your rights under it later.
If the GPL wins, it demonstrates that not only does the GPL hold, but a very strict interpretation of the GPL holds which gives copyright holders the power to bargain with the question of whether forgiveness will ever be granted. (This forgiveness is customarily granted by the FSF.)
If the GPL loses, it only casts doubt on whether the copyright holder needs to give forgiveness. There is no question of whether it is legal to distribute in violation of the GPL's terms. (The injunction seeks to stop a distribution that includes source-code.)
IANAL, but I suspect that the FSF will have an easy decision here. But the cynical bastard in me notes that other recipients of the forked code have rights to distribute the whole thing under the GPL. If one of them distributes back to the original creator, under item 6 of the GPL the creator now gets a fresh license. What if the creator then distributes using that brand spanking new license?
The one interesting thing here I observe is that once you have been found in violation of the GPL only the holder of the copyright can reinstate your ability to distribute derivative works-- and there is no obligation to do so.
This doesn't seem right. Why can't I, a random person with a fresh copy of MySQL plus source, modify MySQL very slightly then pass this derived work onto the defendant under the GPL, thus reinstating their right to redistribute.
The article refers to article 4 of the GPL, but there's nothing there that makes special reference to the copyright holder. In fact, clause 6 says that if I redistribute the program then the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions.
Surely that means that even if NuSphere were once in violation of the GPL, the copyright holders cannot take away their right to distribute derivative works so long as someone (e.g. Red Hat, Debian, the cat's mother...) is willing to distribute to them.
The early version of NuSphere was clearly in violation of the GPL and thus _de facto_ renounced their rights to redistribute that version of MySQL. However, it seems that they've cleaned up their act for the next version of NuSphere.
What I'm wondering is if that revocation is permanent, absent forgiveness by the licensor of the MySQL code? It will obviously be so for that particular version of MySQL. But what about later versions of MySQL? New code, new license. Does their initial unrepentent infringment of the GPL mean that they are forever barred from redistributing newer versions of MySQL whose license they haven't violated?