Slashdot Mirror
The Linux Kernel and Software Patents
batsman writes "The Linux VM system programmers are discussing the software patents that could block further development of important features. Alan Cox brings up several SGI patents covering the techniques they were considering, and Daniel Phillips has found some patents that affect features already present in Linux. Linus Torvalds thinks they should ignore these patents and pretend they don't exist until they cause troubles. How long before kernel developers are sued for patent infringement?"
If Linus has no problem violating patents, I would have no problem violating the GPL. Intellectual property is intellectual property, whether it's covered by a copyright or a patent or a trademark or whatever. If Linus intends to willfully violate someone else's IP, he should have no problem with me violating his IP.
"IP is IP"? Not quite. Software patents in particular are very legally questionable. It's only recently that they've been recognized inside the United States, and they aren't recognized at all in many other places; ditto for business algorithms.
Patent law was made to protect inventions -- physical pieces of hardware. I see no need to respect any extensions thereof, particularly when they have such unreasonable results.
Comment removed based on user account deletion
Any patent can be legally used for home or hobby use. The problem will lie with commercial exploitation of the patent. The developers are off the hook. Not so easy for Redhat et al.
They stab it with their steely knives,
But they just can't kill the beast.
1. Ignore patents.
2. If you get sued:
a. Replace the infringing code with code that does not infringe on that particular patent.
b. Sit back and watch patches back to the infringing code appear.
3. Laugh as the corps realize that they are completely ineffective.
We now return you to your regularly scheduled revolution.
-Peter
The patented parts of the kernel were deveopled in Malta, Egypt and Madagascar where no one holds said patents.
Also, kernel.org releases the SOURCE code. And it's more of a "reference implementation" or "proof of concept" than anything else. If someone actually COMPILES the kernel without first removing patented techniques, well, that's their own problem.
-... ---
How long before kernel developers are sued for patent infringement?
This brings up an interesting question. Who gets sued in this kind of situation? The one who writes the code, the one who compiles it, the one who distributes it or the user? Technically, there shouldn't be anything wrong with the source code itself, since it is not a product or a device. An example is that the ISO source code is freely distributable, even though there are many patent problems. Now it's it's not the developers, who is it? Unisys seems to have tried going after GIF users (web sites), while some others seem to try differt approaches. This is one really bad thing about software patents.
Opus: the Swiss army knife of audio codec
There is the old problem of referring to IP. Be more specific. A patent != license.
As usual, the average IQ of a slashdot poster is 50 right after an article gets posted. I think the people at Slashdot who reward based on who posts first to an article are idiots who greatly lower the quality of discussions of postings here.
Linus is saying that people should ignore patents, in the sense that people should come up with their own ideas independent of what other people may come up with at the same time. This way, if the patent goes to court, they can say "I invented the same thing with no knowledge of their patent"; obvious things can not hold up in court.
As the recent BT case shows, the patent courts are more reasonable than the average Slashdot loonet thinks they are.
I am not logging in; the Slashdot editors like bitchslapping people who don't hold their party line.
Just because a company holds a patent doesn't mean they have to enforce it, or plan to. If they wish they can grant Linux or the world an unlimited license to use the patented technology (which would be the friendliest approach) or they can simply ignore patent violations, which is at least neutral. (Ignoring, rather than granting license to use, is worse because it means they may change their minds at some point when the technology's already running enterprise servers throughout the world, at which point they pull a Rambus. In some ways it's worse than enforcing right away; at least if they enforce right away there's less damage to existing codebases.)
It's rare that you're presented with a knob whose only two positions are Make History and Flee Your Glorious Destiny.
No, the developer may be sued by the patent holder.
His freely available illegal implementation decreased the market value of the legal version, and can be held responsible from the patent publication date, not from the date he is notified of infringement.
This also assumes that the patented algorithms were released publicly before the patent was filed. Also you can't ignore a known patent infringement for many years. Like trademark infringement, you must act when you gain knowledge of infringement.
IANAL, this is my understanding of patent law in the US.
This is whythe term "intellectual property": it causes people to be confused into thinking that copyrights are the same thing as patents, when they are very different.
Did you know, for example, that many patents are invalid? That is, most patents are known by their owners to be so flawed that they carefully ask for just enough royalty so it's cheaper to pay than to go to court, but even so, about half the time a patent makes it to court get tossed out?
The problem with software patents is that the more you look for, the more you will find; lots of basic techniques have been patented (often two or three times). If you look for and find these things, you either have to work around (very difficult or impossible in some cases), sue to invalidate the patent (expensive), or be subject to penalties for willful infringement. If you don't look, the patentholders have to slog through YOUR code looking for reasons to sue (and no willful infringement penalties). Why do their work for them?
(OK, maybe Linus wasn't right about the hit man thing. A hit man might be cheaper than an IP lawyer, but murder really IS unethical, and besides, you'd have to wipe out the whole corporation, and that gets almost as expensive as a lawsuit. But I assume Linus was speaking tongue-in-cheek there)
On Sun, 11 Aug 2002, Larry McVoy wrote:
>
> This issue is more complicated than you might think.
No, it's not. You miss the point.
> Big companies with
> big pockets are very nervous about being too closely associated with
> Linux because of this problem.
The point being that that is _their_ problem, and at a level that has
nothing to do with technology.
I'm saying that technical people shouldn't care. I certainly don't. The
people who _should_ care are patent attourneys etc, since they actually
get paid for it, and can better judge the matter anyway.
Everybody in the whole software industry knows that any non-trivial
program (and probably most trivial programs too, for that matter) will
infringe on _some_ patent. Ask anybody. It's apparently an accepted fact,
or at least a saying that I've heard too many times.
I just don't care. Clearly, if all significant programs infringe on
something, the issue is no longer "do we infringe", but "is it an issue"?
And that's _exactly_ why technical people shouldn't care. The "is it an
issue" is not something a technical guy can answer, since the answer
depends on totally non-technical things.
Ask your legal counsel, and I strongly suspect that if he is any good, he
will tell you the same thing. Namely that it's _his_ problem, and that
your engineers should not waste their time trying to find existing
patents.
Linus
god n. : the Supreme Being, indistinguishable from a good random number generator.
I think that much of the problem stems from the "non-obvious" bar being set too low.
When a company invests a lot of time and money to come up with an idea that the world would otherwise not have had, I think that the company should have a right to protect that investment through patents. Without such protection, they won't make the investment in the first place, so the idea won't be conceived, and society will be all the poorer for that.
The problem occurs when an inevitable idea becomes the property of whoever gets to the patent office first. When that happens, the law is taking an idea that would otherwise have belonged to society and general, and barring everybody except the owner from using it.
The big hope here, clearly, is for Linux developers to go ahead and do the obvious, and if some poor fool tries to sue over his VM idea being used in Linux, then Tux's Uncle IBM will drop mention of the umpty-three bajillion silly IBM patents they're violating, and the whole thing will go away.
Bonus question: if I attempt to enforce a patent used in GPLed code, what happens? Do I lose my rights to use/distribute that GPL code? Is the copyright and GPL on that code unenforcable?
Linus isn't interested in standing up for our right to code. He'd doesn't like politics, and would prefer to ignore the problem. But the problem won't go away. Rather than say immature stuff like that, it's time for him to use his notoriety to speak publicly about the problem and why it should be fixed.
Bruce
Bruce Perens.
I think what he is meaning by that, is the fact that patents are de facto validated in court and not by the USPTO, as the latter grants nearly every stupid and insignificant patent that comes by.
:)
In fact nowadays you can't write a single line of code without a chance of having a stupid patent somewhere which forbid it ! this is just slightly exaggerated.
So when and (if) someone dare to sue a Linux hacker about some stupid patent, considering all the interests now in stack, you will probably have someone with deep pockets (or a big defensive patent portfolio, can you say IBM) who will jump to defend it.
I think this is the best strategy, as Linux is probably already infringing hundreds of patents, and nobody can review all the thousands patents that may apply every time he writes a single line of code, and this why patents are bad and impractical
Anyway experience has shown that 95% of the time, Linus always does the right thing , (well this is just bit exaggerated to, but not far from the truth)
see here.
Patents are patents. You're confused, or you're thinking of copyright.
It's annoying, but it's a pretty solid and very common practice for anyone involved in software development. Dont Look At The Patents.
Patents have become such a complete minefield that if you spent time researching software patents while programming you might just as well quit the buisness. You cant develop anything beyond 'hello world' without risking running afoul of software patents. They're overly broad, they have decades of prior art, they're trivial and should never have been awarded in the first place, but they're there. And it will cost to get them overturned.
What it comes down to is that basically you have three choices. Either you dont know about them, or you know about them and violate them or you quit developing software.
And out of those choices not knowing about them is the least painful and/or dangerous one.
Patents has NEVER brought anything good to society. All patents do is create a monopoly.
1) There are no truly unique inventions. Everything is an evolution of everything else.
2) The car was patented. While the patent of the car was active cars were not built and they were low quality, etc. Ford got around the patent and made cars that everybody could buy and drive. Windsurfers were patented. The company windsurfer held onto the patent and in the last couple years of the patent sued the hell out of companies. Result companies went under and Windsurfer took the money and ran. Sure the original patent holders invented the car and windsurfer. But those same inventors did nothing to further the invention.
3) Time has shown again and again that ideas or concepts are worth nothing. Execution is worth everything. There are hardly any companies that survive only on patents. If you look at most big companies they survive because they know how to run a business.
4) Now and the future the only ones penalized by patents will be the "western" world. The rest of the world does not care about patents and they clone, etc. Why? Because patents introduce a penalty that only the "western" world can afford.
"You can't make a race horse of a pig"
"No," said Samuel, "but you can make very fast pig"
I've never met Linus, but from reading the Linux kernel mailing list over the years, I think he's doing just fine. Everyone is so quick to find some cause to rant about, but he's saying, I'm here to develop the kernel. He's not a lawyer, nor has he ever been known for his activism, but quite honestly, I don't want the head kernel architect to be a political spokesman, that just slows down projects and gets the developers thinking about the wrong things. Does this issue have to be addressed, definetly, but I think he knows he's not the one to do it. It's better suited for Redhat or maybe HP, not developers.
"Karma can only be portioned out by the cosmos." -Homer Simpson
What you miss is that a software developer's entire job is coming up with ideas and methods of implementing them and then describing a methodoly to implement those ideas in such a formal language that they can be executed by machine.
Which is to say... the only really important thing that a really good programmer ever does is come up with ideas; certainly there's a tremendous amount of drudgework involved in debugging, and making different systems talk to each other, and so forth... but the really core thing that extremely good programmers can do and only average programmers can't is come up with new algorithms.
If coming up with new algorithms is so core that it's part of the job, then why the hell should anyone else be prevented from implementing a newly developed algorithm?
If Linus (et al) are publishing source code, isn't this code protected by the first amendment no matter what patent law says? I mean, by sending out the source code, aren't they simply giving a description of *how* a patented thing works, not an implementation of that thing actually working? And since the patent requires that the patented thing be fully described, isn't source code simply a different way of saying something that is already public knowledge?
Couldn't a developer who is being sued for patent infringement simply say, "I'm just exercising first amendments rights.. and besides I'm not saying anything more than you've already said in your patent filing. I'm just saying it in a different language than you."
?
Key to financial independence: Spend less than you earn. Save and invest the difference. Do it for a long time.
is one of the defining characteristics of a valid patent.
if someone, or several someones, without knowledge of said patent, come up with the same solution to the same problem, then it clearly fails the non-obvious test, and as such is not a valid patent.
seems like these patents would get thrown out on that basis.
The difference between Theory and Practice is greater in Practice than in Theory.
> I guess he was just pissed.
Either that, or he's been playing Grand Theft Auto 3 recently.
Funny, the Constitution says (Art. I, Sect. 8)
There is nothing there that says the "discovery" needs to be a machine and not an algorithm.Thomas Jefferson thought patents should be just for machines, but he was not the king of the U.S., and others thought differently. The Patent Act of 1793 states that the inventor of
is entitled to a patent. Note that "arts," not just machines, are entitled to patents. The 1952 Patent Act revised this to read,Again, not just machines, but processes were elegible according to the letter of the law to be patented. Algorithms and business plans seem to me to be processes and hence, are not automatically excluded from the wording of the historical patent laws.
By not knowing whether any given algorithm or method has been patented (and chances are most any real project has some 'infringing' methods) you not only protect yourself against accusations of willful violations, and against negligence for failing to find *all* violations, but you also weaken the patent itself, since by *policy* you didn't refer to the original claim and thus definitely came up with your solution by yourself.
While I respect your opinion, I have to flatly disagree and ask what it is you're smoking.
Patents on hyperlinks? Patents the parent poster mentioned? Patents on chat bots. Come on -- all it would take is one guy with a BS in comp. sci. and 5 minutes with Google to evaluate any of those.
This is the government we're talking about here. If any 24 year old comp. sci. major can figure that out, shouldn't we expect the same of a government body which regulates patents that result in high dollar lawsuits?
~Dalcius
Rome wasn't burnt in a day.
I didn't like it either. How do you "whack" Microsoft? Fly a jet plane into their Redmond campus? I'd rather not be identified with the sort of person who does that.
Anybody who takes the 'whack' comment seriously is an idiot, and probably believes that penguins really can charge at you in excess of 100 miles per hour.
Linus isn't interested in standing up for our right to code. He'd doesn't like politics, and would prefer to ignore the problem. But the problem won't go away. Rather than say immature stuff like that, it's time for him to use his notoriety to speak publicly about the problem and why it should be fixed.
I interpreted Linus's outburst as an admission that he really does care about this issue a great deal but hates the idea of his core developers being distracted by it. Personally I don't want to see Linus at all distracted by this either. His energy is much better spent quarterbacking the development effort. Actually, it's up to people like you, Bruce, to take the lead.
Have you got your LWN subscription yet?
What's the obvious thing to do here? Ignore the problem. I'm absolutely serious.
Either you ignore the patents or you stop coding. There is no other solution. You can't be a patent lawyer and a coder at the same time. You don't have enough time to do both. And unless you're a patent lawyer, you will never reasonably know that you haven't infringed on something.
I strongly suspect, to the point of certain belief, that 99% of the patents in question are bogus and that Linus and Co. would win a court case. But again, there's not enough time in the world to both go to court and to code. Of the remaining 1%, you don't know which ones they are. The only way to avoid them is to remove all suspect code. You can't simply recode it in another algorithm, because that other algorithm may be patented as well. So you remove the code and are left with nothing more than a README file.
There are times when you must ignore the law of man and obey the law of God. And the law of God says that you only have 24 hours in a day. If you're going to be a coder, you have to ignore patents.
A Government Is a Body of People, Usually Notably Ungoverned
for you asseration that patent holders can't ignore infringement.
Patents are NOT trademark like.
In some legal senses trademarks are not owned. One merely retains the *temporary* right to associate one's business with a particular mark. That right is maintained only so long as in the opinion of the *public* that business is uniquely identified with the mark. That's very important. It is the *public* that determines the validity of a mark. I cases where a mark is called into question in the courts the court only rules whether or not a particular mark uniquely identifies a business in the *public* mind. He does NOT assign *ownership,* only the rights for USE. One *registers* one's use of a mark to show that it was in use by you at a particular time. One does not have *title* to it. Thus for a mark to remain current the courts have ruled that one must defend it's association with one's business vigorously and a mark abandoned becomes once again available or even in the public domain.
A patent is completely different. It is considered true property, like your house, and like your house you can allow people to use it as you will, even to the point of ignoring neighbor's children using it as if it were their own while 'capriciously and discriminatorially' prosecuting another neighbor for trespass. One is given *title* to a patent, just as one is given title to a house, and many of the same legal principles apply. Evidence of this is as near as the headlines, as nearly every day some company discovers they own title to some patent that they didn't even know they had and begins enforcing it, often times against only one or two specific 'people' while continuing to allow all others to freely ignore it.
Kind of like allowing one neighbor to use your lawnmower without asking. It doesn't cease to be your property and you retain the right to, at any time, deny him it's use, or to prosecute a burgler for stealing it.
All perfectly proper, legal, and within the general philosophical framework that governs all property law.
Developer: (click-click), "#!/usr/bin/python"
Paper Clip: "Hi! It appears as if you are violating a patent! Would you like to:
Of course that would be violating a patent too
You have to come to an agreement with the patent owner!
If he doesn't want to come to an agreement, you post their contact info on Slashdot so they can be driven crazy by anti-patent nazis.
If somebody sues you, you change the algorithm or you just hire a hit-man to whack the stupid git.
You go, Linus! Teach 'em who's boss! Send Guido and Boris to have a discussion with the patent holder's kneecaps!
1) There was no assertion that inventions created and patented were, in any sense, bad. The assertion was that they were an evolutionary developement that would have occurred anyway. This may not be provably correct, but it is a very different assertion.
2) There has, as far as I am aware, been only one Thomas Edison. Perhaps there are reasons? You might consider them.
3) Patents are advantageous to those who hold them AFTER they hold them IF they have enough money to defend them. These conditions are not commonly met among inventors. Patents have been described as "a license to sue and be sued". This is nearly correct, but it ignores the effects of patent pools, which only serve to increase the degree of monopoly in an industry.
4) Were I to accept the limited degree of benefit that you ascribe to patents, I would still need to consider the costs that they impose. And these costs are formidable.
My general conclusions are that the US patent system is so broken that we would be better off without any at all. I also feel that a limited patent system might be desireable, in which patent pools should be explicitly prohibited and if a patent was ruled "silly", then the defense fees would need to be paid by the clerk that granted the patent. And that no penalties could be collected under a patent until a panel of experts in the field ruled that it was non-obvious. These experts to be selected by lot from a population that volunteered (for some minimal recompense) and declared at the time of volunteering what their field of expertise was. This is not sufficient to totally rule out silly patents being enforced, but is obviously better than the current system. Also, that this same panel could throw out any patent on the grounds that it was too broad. Not just decline to enforce some particular provision, throw out the entire thing.
But even with these limitations, I am dubious that a decent patent system could be constructed. The examiners are required to be specialists in too many disciplines, and this is clearly impossible.
I think we've pushed this "anyone can grow up to be president" thing too far.
Ignoring the problem is a viable solution since patents expire. In fact having all these frivolous patents in the public record prevents them from being filed in the future.
Linus is merely saying that coders should code and legal matters should be handled seperately. Remember the statement was made in a kernel development thread not a political or legal forum. It only hurts software development efforts for coders to proactively go out and seek out stupid patents that might possibly cover what they are doing.
Come on wasn't it Shakespeare who said "The first thing we do, let's kill all the lawyers" I think Linus was mostly just making his point when suggesting whacking stupid people.
Here's how it works: if you read a patent and decide it doesn't apply, and then you get sued and lose, your liability automatically triples because you violated it flagrantly. If you didn't read it, the violation was incidental. Many big companies have policies forbidding their engineering staff from reading patents, for just that reason.
(Those of you who notice a similarity with the Catholic notion of mortal and venal sins may feel smug.)
Patents do not foster creativity. They foster ways to beat the patent and not evolve the product. This is cloning with a twist.
Will all scientific information become a trade secret? Some yes and guess what there is plenty that is a trade secret anyways. Why? because even with patents once the cat is out of the bag it is out of the bag.
Patents could have protected a small inventor from true monopolizers, but in fact the small time inventor is hurt by patents today. These days to get a world wide patent costs several hundred thousand dollars. In the EU it costs about 30,000 Euros and in the US about 20,000 dollars. This is not money that a small inventor has. However, it is something large corps can afford. And then large corps use that money to fend off competitors. In other words a big corp can be inefficient.
Now I am going to throw you a concept. Imagine that there was no patent protection and there was rampant cloning or evolution? Could an industry survive? Absolutely... Examples include, books, music, sports, fashion, food (wine, beer, spirits), etc.
All of these industries are based on selling a product to make it interesting for the client. For example how many receipes are there for mashed potatoes? But yet people buy receipe books in droves. Or what about jeans? Remember when jeans were are all the rage in the eighties and jeans were cloned rampantly? Or how about sports? First people started with simple surfing, then skate boards then roller blades, etc. THAT is true innovation not hampered by patents!
About IP theft, and wanting to eat? Gee whiz, but I think the engineers still eat in those industries. The difference is that the engineers have to actually be part of a business. A monopoly grants the inventor a license of bad business and money gouging.
"You can't make a race horse of a pig"
"No," said Samuel, "but you can make very fast pig"
The Wright Brothers were precisely the amateurs you mention. They did patent their invention and spent the next ten years fighting off other competing companies for exclusive rights to this body of work. They prevailed in court and yet completely failed to prevent rivals such as Glenn Curtiss from building airplanes. Moreover, they wasted years of useful energy and a tremendous fortune in the process. Most importantly, Curtiss' wing design was superior to theirs, and the movable surfaces which replaced the Wrights wing-warping system is the basis for the ailerons of virtually all aircraft built since.
-Hope
I do believe that Linus' point just might be that a patent is supposedly for a "non-obvious" idea. If you or he can come up with an idea without benefit of digging it out of someone else's patent, the idea can not have been as obscure and "original" as the jackass that patented the idea thought, or the over-worked, under endowed patent office clerk assumed. The fact that some mediocre mind has a brain cramp and actually has an idea, doesn't mean that many others, more nimble minded had not already devised,would not,or could not devise the self-same idea independently at need, because to them it was and always had been and would always be obvious. Far too many of the mentally lame file patents these days shrilly proclaiming their invention of the obvious. Worse, they turn about and sell their patents to the unscrupulous, who then display all the worst behaviour of parasites.
In fact, I suspect that many creative minds don't bother patenting ideas, while those who do, do so because of the fear that they may never again experience the rare event of a creative thought.
------ The only greater hazard to your liberty than n politicians is n+1 politicians.