Stallman Critical of OSDL Patent Project

PatPending writes to mention a News.com article about Richard Stallman's objections to the OSDL patent project. He argues that the project may actually be 'worse than nothing', as it will undermine certain legal tactics. From the article: "'Thus, our main chance of invalidating a patent in court is to find prior art that the Patent Office has not studied,' Stallman wrote. Second, patent applicants could use the prior art uncovered by the OSDL to write patent claims that simply avoid the technologies used in the tagged software. 'The Patent Office is eager to help patent applicants do this,' Stallman wrote. Finally, he wrote, a 'laborious half measure' such as the Open Source as Prior Art project could divert attention from the real problem: that software is patentable in the first place."

Comments

Aboslutly correct.

[geekoid]

"Finally, he wrote, a 'laborious half measure' such as the Open Source as Prior Art project could divert attention from the real problem: that software is patentable in the first place.""

Aboslutly correct.

Re:Aboslutly correct.

[Anonymous Coward]

> "Finally, he wrote, a 'laborious half measure' such as the Open Source as Prior Art project could divert attention from the real problem: that software is patentable in the first place.""

...and besides, it should have been called the Free Software as Prior Art project!

Re:Aboslutly correct.

[westlake]

Aboslutly correct.

We are two months from an important mid-term election, two years from a presidential election. Patent reform ranks somewhere below The Bridge to Nowhere on the national political agenda.

Stallman... half right

[xtaski]

Have to disagree. For those without $1M of FSF funds to run around fighting granted, bad patents, it's much more difficult to fight in a come-from-behind position when XYZ Patent Troll Inc. already has a patent granted. Litigating patents is expensive. It's cheaper to lop off the 10% that are obvious "spam patents" before they're ever granted than to let all 10% (that would be like 4,000 these days) go through and clean up later...

Re:Stallman... half right

[cswiger2005]

It's better to lop off as many software patents as we can, agreed. Litigating patents is usually expensive, but then, having to pay a patent troll royalties for a bogus claim usually becomes expensive, too. Having prior art available to refute a patent troll's claim makes litigating bogus patents a lot easier and a lot cheaper.

I don't believe anything which could be described as an algorithm should be patentable. I also don't believe that you should patent public API's, as such "programming interfaces" are by definition intended for use by other programs; public APIs normally are widely distributed in documentation, which at least prevents others from patenting the APIs which you might release (as your release will obviously constitute prior art). I suppose that if you implement computer software which is sufficiently original, not completely obvious after 5 minutes of thought, and is not representable as a mathematical algorithm, that might deserve the protection of a patent, but for the most part, simple copyright ought to provide enough protection....

If your kitchen counter was on fire...

[xtaski]

Would you wait for the rest of the house to burn down and rebuild the entire house from scratch... or put out the fire on the counter and replace the counter...?

Re:If your kitchen counter was on fire...

[Rogerborg]

Relevant how? This is like setting fire to the curtains and saying "How do you like that, kitchen counter?"

Moral correctness is not enough

[EmbeddedJanitor]

It's like saying nobody should steal, so I won't lock my car/house/whatever.

Sure, in the long term, and a perfect world, you might want to get rid of software patents. Right now however they are real and are here and measure that combat them face to face have some merit.

Re:Moral correctness is not enough

[Millenniumman]

Why does software not being a physical object make it less suitable to be patented? Patents are there so that if you invent something, someone else can't copy it and mass produce it cheaper than you can, without having paid anything for the development. If I invent a new software algorithm, and it is not patented, then someone can copy it (not a copy of the code, but of the process) into their own software, which they have far greater resources to distribute and undercut on price. Why should I spend time inventing new algorithms, then?

Re:Moral correctness is not enough

[vertinox]

If I invent a new software algorithm, and it is not patented, then someone can copy it (not a copy of the code, but of the process) into their own software, which they have far greater resources to distribute and undercut on price. Why should I spend time inventing new algorithms, then?

Because chances are the algorithm you created was most likely neither novel nor unique to the large amounts of algorithms created before you. As in... Your algorithm had to be based on some math that everyone knows or at the most an obscure math professor came up with years ago.

You aren't writing your own language but taking from knowledge of mankind and applying it to your own methods.

Secondly, copyrights protect someone from copying your code, but not your methods because if someone can simply recreate your algorithm simply by looking at the results and not even see any of your source code then again... your algorithm was neither novel nor unique and therby not deserve a patent.

However, your effort and code should be copyrighted and protected by such methods.

Re:Moral correctness is not enough

[F452]

Maybe you shouldn't, if you're concerned about not getting monopoly protection. No loss to the rest of us. Someone else will do it in your place. Patents are supposed to promote innovation for the benefit of society. Software patents just stifle people who are already working on making better algorithms with no interest in idea monopolies.

----
Freedom is on the March!
http://www.movingtofreedom.org

Re:Moral correctness is not enough

[LordLucless]

Building on what you said: the whole basis of a patent is that of a trade. The public gives the inventory a time-limited monopoly, the inventor tells the public how his invention works. The reason this is a good deal for the public is the non-obvious clause. Patents are designed to protect things that nobody else can figure out how to do. If someone can look at your invention and think "Eh, I know how to do that", then your invention is not worthy of patent - the public would be getting a bad deal, because they're trading a monopoly for something they already know. Patents are designed so that when someone invents something really cool, he doesn't keep it as a trade secret. If he does, and he dies, then the world loses knowledge. If he patents it, then everyone knows how to do it (even if they can't use it for another ten years) and knowledge is retained. That is how patents can (and should) act towards the good. The problem now is that the USPTO is betraying their purpose by making crappy deals on behalf of the public. They're giving away monopolies like candy, and reaping the kickbacks (errr, I mean processing fees).

Re:Moral correctness is not enough

[Chris Burke]

Why should I spend time inventing new algorithms, then?

The same reason you spend time writing software at all (that others with more resources may duplicate and undercut): The new algorithm solves a particular need of yours. It is useful.

This is why Sir Charles Hoare created the quicksort in 1960. It probably didn't even occur to him that this was something he should prevent others from using, and he still found it useful to invent. Thank God he did not -- could not -- patent it, or it would have been over a decade before people could have taken free advantage of the fastest-average-time general sorting algorithm known today. Imagine everyone else had been doing the same thing -- locking up merge sort, bin sort, r/b binary trees, avl binary trees, b-trees, etc etc. With all these foundations of computer science locked up in patents for 14-20 years, how much progress do you think the software world would have made compared to what it did with free access to all these ideas? Remember, we're talking about a fourth of the entire existence of computers.

Why does software not being a physical object make it less suitable to be patented?

Because software is math.

That's all it is. A program is just a series of mathematical operations performed by a computer. Now the computer is an invention. But the software is just a calculation. Patenting a software algorithm is like patenting a sequence of button pushes on your calculator, and by "like" I mean "is very literally the same".

Imagine if Sir Issac Newton had decided to lock away his Calculus? He might have had some legal troubles from Leibniz, but once they came to an amicable arangement, everyone else would have been out of luck. Is Calculus not a great invention? Of course! But like all math, it is an invention whose benefit comes from what could be built upon it. Thank goodness that Newton published his book and did not restrict others from using what it described, or you have to wonder where we'd be today.

Why shouldn't software be patentable? Because it's a patent on math, the fundamental language of the universe.

Re:Moral correctness is not enough

[Soldrinero]

Not that I'm a fan of the current patent system, but patents can and should cover things other than just "things that no one else can figure out." Non-obvious doesn't mean that you can't figure it out, but that you wouldn't think of doing it. For example, I've used a set of Allen wrenches that had specially beveled ends. The way they were shaped allowed you to turn the screw even with the wrench at a significant angle, like 20 degrees or so. Now, it's not at all difficult to figure out how they work, but the idea is brilliant and extremely useful, especially on an optics table.

Patents can work for the public good by encouraging innovation through financial incentives, i.e. limited-time monopoly protection. The trick is that if patents are being used as cudgels against competitors, especially if this is because the patents are too broad, then they're no longer serving the public good. Sadly, that's the state that we're in today.

Re:Moral correctness is not enough

[Shaper_pmp]

Patents are temporary monopolies, which inherently restrict people's freedom. Even the people who originally set up the patent law in the USA understood and agreed with this.

Given patents are restrictions, patents are only allowed to exist where necessary, and solely for the "greater good" of society.

Without patents there would be no protection for inventors' inventions, so the theory says they wouldn't invent anything, and society would suffer.

Large businesses already have large cash reserves and an infrastructure to quickly begin production of any new product or invention, so the people who patents are mostly designed to protect are independant inventors and small companies.

If inventions occurred without patent protection then there would be no need for patents.

The computer industry (specifically, the software industry) has, for the last thirty or forty years, self-evidently been the single most successful and fast-developing industry we've ever seen. Innovation and inventions have come on a regular (and ever-increasing) basis, for long before patents were issued on software. In fact, there are reams and reams of documented cases where innovation has been halted by the existence of patents, and not one where a patent has been proven to have helped innovation.

Unlike most physical devices, software is not generally a discrete invention. If even basic software algorithms are allowed to be patented it will be impossible to write software, since software is basically a collection of basic algorithms re-arranged into a novel order.

Because software is hard for patent examiners to understand they habitually issue overly-broad patents to basic algorithms. Because the only way to challenge patents is in court (with all the associated fees), software patents primarily benefit large companies who can afford to cross-licence or defend their patents.

So, patents are a necessary evil, and by design and intention may be used only where necessary.

Patents are required to stimulate innovation. Except in the software industry.
Patents reward novel and non-obvious innovation. Except in the software industry.
Patents protect the small inventor. Except in the software industry.

Have I made my point yet?

You can blather on all day about the philosophy of what should and shouldn't be patentable[1], but it's irrelevent - software shouldn't be patentable because there's no need.

Patents shouldn't be issued unless there's a good reason to do so, not merely because you can't see a reason not to.

[1] FWIW I agree that algorithms are ideas/math, and that ideas/math shouldn't be patentable. If you disagree, and believe ideas or math should be patentable you might as well ask why someone can't patent a colour, or why Isaac Newton couldn't patent gravity.

Software is a few very, very basic ideas (algorithms). Software merely consists of re-arranging them in a novel order, not "inventing" anything new.

Source code is an expression of ideas, and is (rightly) copyrightable. "Software" is just ideas, and ideas simply aren't patentable.

Re:Moral correctness is not enough

[Chris Burke]

Everything in the universe can be reduced to simpler elements that by themselves seem to defy property labels.

Except you don't have to reduce software to simpler elements to call it math.

Software is math, in the same way that "3 * y^2 + (7 / 6)x = z" is math.

Now you may argue that "3 * y^2 + (7 / 6)x" is actually a series of characters that represent math, but again, that's exactly the same as software. Software is a machine-readable representation of a series of mathematical statements, while the equation I wrote is a human-readable representation of mathematical statements. Actually, some programs would be able to parse the equation, as well.

Representations of mathematical concepts, aka software, can be covered by copyright, just like a book about math can be copyrighted. The concept represented, the math that the software is intended to convey, would not be patentable if it appeared in a book, but when it appears in source code, suddenly it is.

If you don't believe software is math, you just don't understand software.

Bad Analogies aplenty in this thread

[Mateo_LeFou]

If a bunch of people or organizations were doing reckless things, and semi-advertently causing a bunch of fires, would you pour research money into new fireproofing techniques or shut the bastards down?

Publically reject 'patent pledges' too.

[Anonymous Coward]

I don't want to be granted use of some companies software patents, I don't want software patents to exist at all!

Software patents aren't going anywhere

[RLiegh]

at least, not any time soon. Stallman is a great thinker, but he seems to have a reality distortion field that Steve Jobs would envy. I agree that the fact patents exist at all is a problem -but it's one which is not going to change, period. So the OSDL Patent project really is the only practical way of coping with the situation.

Re:Software patents aren't going anywhere

[Anonymous Coward]

What is it with this defeatist attitude?

Software patents are going somewhere, they are going to the bit-bucket in the sky. Do you know who's going to send them there?

We are!

I'm not saying it's going to be easy but you make the outcome sound inevitable, it isn't.

Re:Playing the odds

[jbrader]

1988 was 18 years ago.

Re:Thank God

[cloricus]

Oh god. I think I just read a Stallman point of view and agreed with every point. :| I didn't think that would ever happen...

I guess I'm going to be one of the people chanting "down with this project" now?

Horns Of A Dilemna

[rumblin'rabbit]

On Stallman's (probably long) list of things he doesn't like, the following rank at or near the top:

• Software patents.
• Proprietary (that is, closed) software.

Here's the thing: Probably the best defense against having to deal with software patents is to keep the software closed. Don't make the code public and don't tell how it works. If people don't know you've violated their patent, they are not likely to sue you, and their software patent won't be worth very much.

Such a strategy is not dishonest - even when behaving with the highest integrity, inadvertent patent violation is not only possible, but likely. You should not knowingly violate patents, but you aren't required to help the patent holders identify offenders either.

By hating both simultaneously, RMS has given himself a very tough row to hoe. Open software is highly vulnerable to patent litigation.

Re:patent GPL?

[Mateo_LeFou]

One kinda important difference is that copyright automatically attaches to any software that is written. Patents do not. So getting a patent on software would be an active (evil) step to take, but getting copyright on software is totally inadvertent. And rms correctly realized that putting software into the public domain was -- well, not evil -- but not as good as it could be.

Re:patent GPL?

[illuminatedwax]

Because if someone patents something, you can't make a free version of it yourself. A software patent closes off all versions and iterations of that software completely.

Stallman's issue isn't with copyright - his issue is with people not voluntarily giving up their code to the community. He is all for copyright and ownership of code. His problem is that software is not something you should be able to patent, and that the OSDL initiative distracts from this point.

Absolutism is impractical

[RareButSeriousSideEf]

Yes, okay, sure... the Real Problem is that software is patentable in the first place. But does Stallman think that we're going to change that fact quickly just by being absolutist about it?

If not, how many ideas do we want to see slip into proprietary hands while we maintain our moral purism about software patents? This is a political issue, and political agendas live, eat, sleep & breathe half-measures. According to Stallman, "If we are not careful, this can sap the pressure for a real solution." Erm, what pressure? Where is the well-funded, politically connected lobby that's creating more pressure for a Real Solution than beneficiaries of software patents can create in the opposite direction? `Cause short of that, we all know that we're not going to see a Real Solution anytime soon.

IMO, the idea of "Open Source as Prior Art" is basically a good one that needs some tweaking. If Stallman is correct (and I have no way of knowing whether he is) that "when prior art is considered by the Patent Office during the patent-granting process, it usually loses any weight it might have had in a court case," then that might be a problem. However, I can't understand how patent holders of a variation on an OSDL-tagged thingamabob could (a) claim their idea is patentable because of some variation, and still (b) go after the FOSS-derived works. Wouldn't the very granting of the patent in spite of the OSDL art be the basis for establishing non-infringement?

Stallman wants the very idea of ideas to be irrevocably tied to freedom. That's a beautiful vision, and God bless him if he can ever pull it off; I'm (sadly) not optimistic. Meanwhile, I'll settle for having as many of them as possible stay clean from proprietary claims. Failing both, anonymous/pseudonymous coding & releasing might be our only refuge.

You have never, and will never invent an algorithm

[Anonymous Coward]

Nobody has ever invented any algorithm, algorithms are truths, they are fundamental facts. Nobody invents physics, math, chemistry, they just discover the truth. Most software doesn't even contain any discoveries, and none of it contains inventions.

a grey solution is no solution

[fatcop]

Firstly I am completely convinced software patents are a big mistake and all available effort should directly to completely abolishing them, no less.

This debate has been rehashed so much, but I really get tired of hearing some things like:

"Software patents are here to stay get used to it."

Bloody hell, how defeatest is that. They aren't here to stay, but they will be if that is going to be people's attitude. Also aiming at stop gap measures is a complete waste of resources and will only give software patents more time to get a strangle hold. Allowing some "strong-er" software patents through simply opens legal loopholes and abuse.

"Richard Stallman is a raggy looking, hippy, commi, socialist, fanatic protagonist ..."

So the hell what. So people feel more comfortable hearing carefully crafted corporate messages from some clean shaven multi-millionaire in an expensive suite. It that really what makes people feel more comfortable about issues. Though I don't have to agree with all his beliefs, *each* issue deserves individual attention.

"Software patents aren't any different normal patents"

I am not going to reiterate what countless people have said about the difference. Look at the history of how patents came into being. But what about killing, is that ok? We kill animals and eat them. That is acceptable (by most), but we certainly don't condone killing humans. So don't go generalising and saying all "patents" are ok, just as we clearly don't consider all killing to be equal. Their are philosophical and logical reasons why software patents are different. If you're interesting in protecting a fairplay capitalist market in the future, don't just give into the political and corporate pressure. We lose.

(flame on)

And another thing...

[Anonymous Coward]

Why should I spend time inventing new algorithms, then?

The other posts to the grandparent post are well and good, though I would like to add my own two cents.

As it turns out, lots of programmers find it very rewarding to develop new algorithms without any profit motive at all. Sure, not EVERYONE likes this, and not all of those who do ever come up with anything novel or useful. However, one need only glance at the linux community to see an example of the amounts of work that people are willing to do without pay.

Many programmers, you see, find the knowledge that what they produced was useful, and is in use by others, to be very gratifying.

So, the argument (not really stated by the GP but implied) that "without patents, nobody will innovate" does not hold water. There will always be innovators, and there will always be ways to turn an innovation into a revenue stream, without patents.

I would also like to point out that, generally speaking, patents do not do any good for individuals. They are a bit expensive, and even when an individual does obtain his patent, the process of enforcing it requires a lot of cash. Often, actually enforcing a patent is simply beyond the financial means of an individual, and the big wealthy companies win, whether they have a patent or not.

So, patents do not, generally speaking, protect the underdog. They do not act as a blanket of protection for the people, but usually, as a blanket of oppression. They deny the little guy his right to code, and that is neither morally or economically acceptable in my book.

Lastly, the definition of a "software invention" is quite vague and fuzzy around the edges. Did Microsoft invent IsNot? They sure do have a patent on it! Part of the reason the patent office gives out so many bad patents is because it is not always obvious which claims are overbroad, which part of an algorithm really constitutes a novel invention, and which parts are just necessary plumbing to make it all hang together. Often the plumbing gets patented right along with the invention, and then NOBODY can go to the bathroom. It is a serious problem, it manifests in a unique way for each "invention" (thus preventing any blanket policy changes from being able to address it), and the only real solution is to drop the idea of software patents altogether.

Say it aint so!

[novus ordo]

Critics have pointed out that many of the large companies behind patent reform efforts in the United States, such as Microsoft and IBM--both also OSDL members and supporters of Open Source as Prior Art--are the same companies aggressively pushing for software patents to be legitimized in Europe.

Umm...there is no contradiction there...let us explain...

But the truth is they WANT software patents. I don't see them lobbying to change the status quo. All they are interested in the "quality":

"OSDL supports the USPTO's drive to improve the quality of software patents."

Stallman:

"Thus, our main chance of invalidating a patent in court is to find prior art that the patent office has not studied. Furthermore, patent applicants can use this information to write patent claims that cover important activities while avoiding the known prior art that could invalidate the claims. The patent office is eager to help patent applicants do this."

Ding? If not:
IBMicrosoft: "Thank you little critters for making our patent applications the strongest on the planet. Now we can make sure that nobody will ever overturn our patent on quicksort, mergesort, bubblesort, mydickinyourassort...and don't forget the youinventedandIpatentedsort"

Backfire is an understatement...

Re:I work at OSDL...

[sgtrock]

Do not use anything from a company that patents software.

Not a workable solution. Virtually the entire software industry is taking out patents for defensive reasons if nothing else. Even companies that otherwise are FOSS friendly are. Who on earth, for example, would you recommend that we buy highly specialized trust account software from?

Re:Thank God

[orasio]

The problem is that the guy in the past was talking about bad things that would happen if you didn't follow his lead. He sounded a little nutty, talking about an oppressive future, for example whn he discussed Palladium, or "Treacherous Computing".
Right now, the problems he was talking about are more real, DRM, and trusted computing are real problems that affect you right now.
He hasn't changed, but the reality did change into what he said it would.
Software patents right now have a harmful effect, but they could be much worse if you let them advence further.
It's not as hard right now to imagine a future time where every free software project is effectively killed by software patents, and the only practical software available is a restricted TV-like personal computer, completely controlled by content providers.
Of course, there are countries without software patents, but for example my country will get them through an FTA with the US, and in most other places, this could happen too.