Slashdot Mirror
LWN on the Patent Encumbrence of SELinux
Anonymous Coward writes "LWN has a story about patents in SELinux. The article says: "Much of the actual work in the implementation of SELinux was done by Secure Computing Corporation (SCC). SCC, in its implementation of SELinux, used a technology that it calls type enforcement. As it turns out, SCC has a patent on this technology." Sigh.
They had better consider quite carefully, and soon.
If I had one line of code of my opensource in thier distro, I'd be on them like mad.
If used well the patent could be a good thing, it could prevent any closed source implementations of the technology. Since it is released under the GPL they can't really prevent anyone from using or distributing it openly.
As x approaches total apathy I couldn't care less.
"SELinux includes Type Enforcement technology developed and patented by the Secure Computing Corporation, who still holds rights to all commercial use of the technology. Before a colo company, or anyone else uses the technology commercially, it will be necessary to negotiate a license with Secure Computing. If anyone wants to do so, I can help get the ball rolling with our Legal and BD folks.
--Tom
Dr. Tom Haigh, CTO
Secure Computing Corp."
- Further contact info ommited -
The guy was responding to a question (from Colo) that looked like they were ditributing GPL code without actually understang the GPL itself...
yep
its all stupid
patents should be on a process and as such software should not be patentable
its like saying I have copyrighted the ability to write trashy novels
there is no way that you could get away with this
simple because its too wide rangeing people argue that their software patents cover very specific things but unfortunatly you cant have them if you dont allow the wide rangeing
frankly its all a mess
just say no to software patents in europe
write to your MEP if you live in europe
Who is my MEP
regards
John Jones
From the story:
There will be no restrictions on the use of TE by the Linux open source community
+
Before a colo company, or anyone else uses the technology commercially, it will be necessary to negotiate a license with Secure Computing
=
How does this affect open source users? Granted these are both generalized statements about the license, but it would appear that the "license" is free for non-commercial uses... If you don't intend to sell the product, then you don't have to pay for the license...
Then, if they decide that it is incompatible with the GPL, we can panic. If we let our fears and emotions run away with this we could just drive them farther away from the Open Source Community. As one poem says:
If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other pertinent obligations
And since it is there own patent the sure can distibute it, or they should find a way to sue themself for distributing the software royalty-free. (That whould make an interesting case...)
They published the code under GPL they gave away the right to use and modify this code, they cannot withdraw that once given right under the terms of the GPL. Basically, I think they gave as they right to use, at least this implementation, of their patent at the moment the GPL'ed it. (But IANAL.)
I've heard the same phrase applied to capabilitiy-based architectures, but these are systems built around hardware enforcement, and I get the impression that SCC's scheme is software-based.
The GPL is a royalty-free license. The patent basically says that if you want to use their method, you must do it by citing credit to them and using a derivitive of their work. Fairly wierd, but still... It sums to say you just can't generate your own code from the ground up - but you could make a derivitive work that has nothing the same except the result. Either way, it must both remain under the GPL and the patent owner must recieve credit in the code.
SIG: HUP
Just because the GPL license file is included with the software does not mean that it was distributed under the license. The copyright owner does not need a license and is not bound by the license he includes for you!
... which is in clear contradiction to the GPL making it useless as far as open source is concerned. This is legal for them, but not nice ... the only way to stop it would be for GNU to get a trademark for GPL and have a certification process.
Of course if part of his project is other people's GPL'd code (for instance Linux) then this does not apply and you are correct.
My example isnt entirely academic though, for instance Lizardtech released their djvu code supposedly under the GPL but freely admits its patents are only valid when used to implement djvu
*cough* nVidia's binary closed source drivers drivers *cough*
(Do you know how hard it is to get that all out in one cough?)
Yeah, it's a patent-law versus GPL thing, but (at least to me) it appears to be cut-and-dry. Linux existed, with its use of the (L)GPL as its license. SCC held their patent and chose to modify Linux to make use of their technology. They are a 'consumer' of Linux, and Linux is NOT a consumer of their technology. They are, before any consideration of patent law is considered, required to comply with the GPL in regards to adding anything to Linux.
So they (we assume consciously) chose to work their code into Linux and redistribute it. This action implies intent to comply with the GPL. Considering the specific requirements of the GPL in regards to patents (must be royalty-free for ALL users who can receive the distribution), the only logical conclusion to draw is that SCC aggreed at the moment of distribution to permit use of their patent within Linux as royalty-free to the public (which includes individuals, governments, deities, and businesses alike).
it's a fairly a->b->c sort of thing. The only thing assumed is that SCC had the conscious intent to comply with the GPL... If they never intended to comply with the GPL, then this whole point is moot and they are in violation of the license! This certainly isn't a rocket-science concept to understand.
So what are our outcomes? (1) SCC provides royalty-free use of their patent in SELinux, (2) SCC withdraws their code, (3) SCC tries to play hardball by requiring licensing while violating the GPL and tries to fight the GPL in court. The likelihood of 3 doesn't seem too likely.
.... um, i lost you after "0110100001101001".
OK, but what I'm interested in is what happens to the source that's already out there, with a GPL license on it. If they now try and enforce their patent, they void their GPL compliance, and lose the right to continue distributing code based on or linked with GPL code. But it's already out there. So what happens to their (copyrighted) source that was distributed under the GPL and which is now in the hands of many individuals?
They can't retroactively remove the GPL granted rights from that source, but on the other hand, they can't apply the GPL to it now or continue to distribute it. So if I have a copy of it (and I do), can I continue to modify and distribute their source? I didn't violate the GPL, and I'm not applying patent restrictions, so why should I (and the potential recipients) suffer from their patent lockdown? But then it means that I can keep distributing their source with a GPL license on it, but they can't, which gives me more powers. But heck, they can still sue me for patent infringement, because patent law is separate from copyright law. The GPL gives me the right to copy, modify and distribute their source, but their patent stops me from using it!
This looks like a bit of a legal minefield. I'm usually fairly clear on where the GPL leaves me, but in this case I'm stumped.
If you were blocking sigs, you wouldn't have to read this.
OK, I shouldn't have dived straight in with the inflammatory language, sorry about that. I'm afraid that software patents have always upset me quite deeply, so I tend to lose my rag about them.
:-)
I don't see that patenting Open Source ideas will encourage the discovery of a single algorithm that would not otherwise have been thought of, so all that would occur is that some of the limited monetary resources of the Free Software community would be diverted to propping up the structure of the patent office.
You say that $350 is a small price to pay. How many patentable ideas do you think are contained in the body of Free Software code Debian GNU/Linux (which only integrates a fraction of the software available) currently runs to about 5GB of compressed source code --- there are a lot of ideas in there, and that's only the tip of the iceberg.
Assuming that someone decided to go through that and patent all the as yet unpatented ideas, what purpose would that serve?
How about ideas that were thought up by people in teritories that do not recognise sotware patents?
As a non-lawyer, I'd assume that that fact that all that source has been published would mean that it would act as prior art, but I have a feeling that the USPTO allows after the event registration (feel free to enlighten me).
If it is prior art, haw can one ever register a Free Software idea, given that the publication is inherent in the development process.
If it is not prior art, what is to stop someone else from claiming to have originated these ideas, and patenting them themselves?
Assuming that it turns out that the an was patentable, and what you suggest was done, in what sense would it be more useful that simply publishing the code under the GPL (apart perhaps from the fact of preempting someone else's patenting of the same idea)?
I can see that it would allow one to prevent others from using that idea, but that is pretty much entirely against the principles of Free Software (if someone else wants to use an idea from one of my programs, and they go to the effort of reimplementing it, they're welcome to it).
I can also see that it is likely to be much more difficult to win a case relying on patent law, rather than copyright law, which means that the likelihood is that the party with the deeper pockets has an unfair advantage, which is not likely to be the Free Software developer.
Was that better? (I've calmed down, now that they've given me one of my dried frog tablets
As to your point about the patent system being there to stay, it seems that there are likely to be parts of the world where that will not be the case for some time, and in places like South Africa and India they've been moving in the opposite direction (at least as far as pharmaceuticals are concerned). Reform of the USPTO has even been rumoured, and parts of Europe seem less than keen to get involved in this form of silliness.
Debian: GNU/Linux done the Linux way
We need a statement from SELinux on this. Having a patent in itself isn't necessarily evil; enforcing it is. Having the patent itself can be good, it prevents some patent whore from seeing what was done and then patenting ti and claiming he owns the idea [Not that such a thing would ever be done ;-) ]. What needs to happen now is for SELinux to make it clear that this patent will not be enforced against the Linux community. Or, if they want a fight, ......
I'm an American. I love this country and the freedoms that we used to have.
In so doing, they will blow away their own ability to distribute Linux (possibly permanently). However, they will also be setting up a situation where nobody else is free to distribute (or use?) their patented stuff either. Those people will then be unable to distribute Linux either- to the extent that it depends on the patented stuff. They'd be able to distribute OTHER Linux dists that did not contain the patents.
So it's a doomsday switch: having anything patented in Linux provides a chance for the patent holder to blow themselves away and also take out everybody else, to the extent that the patented stuff is indispensable. If there was a patented thing that was absolutely indispensable to Linux, it would be a tactic that could shut down the whole movement (causing it to be GNU/ with nothing after it ;) ). It would require that the patent holder blow away their own work and, as someone else said, be a 'patch looking for a kernel', which is simply a measuring of relative value: is the financial hit of ruining the value of this IP less than, say, 40 billion dollars from MS for killing off Linux and making everyone start over with the Hurd?
For these reasons I'd say, totally reject patents in the context of free software. Any patents, even 'defensive' ones, can potentially cause this situation, and I would have thought the technolibertarian 'let's make our own patent pool and fight it out rather than ask that the rules be changed' types would be the FIRST to accept that, if offered enough money, anybody'll crack. In a perfect idealistic world, maybe 'open source patent pools' would be safe, but we don't live in one, and in practice it's more like stockpiling dynamite and using it as barricades. Stupid!