Freedom of Speech in Software 250
akpoff writes " I've been struggling with the question 'what's wrong with software patents' but haven't been able to find the right words. I was over at John Gilmore's website and found a link to John Salin's 'Freedom of Speech in Software' letter to the USPTO back in 1991! This is one of the best explanations I've seen. He reminds us that computer programs are essentially like literature or music -- they are expressions of ideas. Just because they run on a computer doesn't make them uniquely different from other creative mediums. We should think player piano (patentable) vs the music (copyrightable but not patentable) it plays. Europeans -- put this letter into the hands of your MEPs!"
Re:Why the focus on software patents? (Score:1, Interesting)
We can't allow the scope of the political debate to become too broad now, since that brings with it too many risks.
At some later stage, society should indeed revolt
against extreme copyright laws (both in duration and scope) and patent monopolies, since these really don't benefit society as a whole.
Tim Sefton
Winner takes it all (Score:3, Interesting)
One good thing about software patents. (Score:3, Interesting)
Although I would really prefer to not have software patents, I don't think that the case against them is so clear cut. There are many terrible software patents--vague, obvious, trivial, overly broad, and so on. But there are also software patents that are specific, novel, useful, innovative, implementable. And it is possible that software patents benefit us in a couple of ways.
First, companies are encouraged to publish details of their inventions that otherwise would have been held as trade secrets. In the database management world, most of the innovations have been made in industry, and before software was patentable most details were kept secret. For example, David Lomet tells me that Tandem held as secret the "repeating history" recovery scheme that was later re-invented by Mohan and published as part of IBM's ARIES system (parts of which were patented.). See ARIES [ibm.com] for details of that system and links to good patents.
If it weren't for software patents, it's doubtful that IBM would have published such details.
Lomet himself has a couple of dozen patents. Of the ones I've looked at, they are all high quality patents. On the question of patents encouraging innovation, he says:
I'm not sure that this effect is as significant, and the ill effects of all the low-quality software patents may outweigh the benefits, but I think it's important to admit that there are some good effects.
Re:Why the focus on software patents? (Score:3, Interesting)
Just because a system is abused, that doesn't make it a bad idea. The original concept of patents was a good idea, and it just needs a major overhaul.
And just for the record, I think software patents are one of those abuses of the system.
MEP's?? (Score:2, Interesting)
Re:What about the compiled programs? (Score:2, Interesting)
Even compiled it is still writing. I can't read German and make any sense of it. Does that mean it is not writing? Comprehension is not required for something to be considered writing.
"So isn't this whole "software patents" thing actually good for OSS, in that it will only make it harder to distribute proprietary software in its compiled form?"
Perhaps if that was all that software patents were being applied to, then one could make an argument that it is positive for OSS. I wouldn't agree with said argument, but that really doesn't matter. The problem lies in software patents being extended to cover the ideas behind software. Patenting things like "a process to do X" is a bad idea because it squelches competition, and in some cases, prevents a product that actually implements the idea from ever being produced thanks to firms that collect patents with no intention of implementing them.
Bring in peer review (Score:2, Interesting)
- The length of software patents needs to be severely reduced, perhaps to 7 years or so. The IT industry moves too quickly.
- Patents should not be granted for overly broad or trivial ideas. I would say pretty well all software patents are trivial.
- The patent office has no incentive to reject patents. They should be held liable, and penalised, for incorrectly granted patents.
- Patents should be open to the public for peer review, since clearly the patent office is not capable of distinguishing an ingenious idea from a trivial one.
Re:Why the focus on software patents? (Score:3, Interesting)
I know a programmer. He's good. I believe that he can program anything you throw at him. I believe that there are many more programmers who can do that. So, where is the invention if all is required is the skill of a good programmer? They all could easily run the risk of infringing just by applying their usual trade.
Another thing to consider: Patents last 20 years. That is a long time for software. Special arrangments have been made to allow drugs a somewhat longer protection. If there are software patents, they should be shorter, 10 years max. at best.
I'm a (Dutch) patent attorney. I will never write a software patent (and never have). I believe that software patents should be possible, for example if you come up with an algorithm that allows video to be broadcasted requiring only half the bandwidth. However, I have sincere fear that the system will not be able to come up with rules that do justice to real inventors and to society, which should not be hampered by patents that are granted too easily and subsequently overstretched in court. Even on more straightforward matters (such as mechanics) the USPTO is certainly not what it should be, and unfortunately the EPO is slowly deteriorating towards the USPTO level. In addition, it is extremely hard to determine what the state of the art is (how are you going to search?), so patents will be granted while they shouldn't have been. In patent law, the onus is on the person accused of infringment. Not a good thing if it is nearly impossible for you too to determine what was already there.
Re:How to bring across your message to MEPs (Score:3, Interesting)
I have written to one of the MEPs of the party I voted for, the European Liberal Democrats (Note that Liberal means something very different in Europe than it does in the States; it's actually a right-wing party), about my concerns regarding software patents. She replied with an amendmend to the proposed directive, drafted by the commission she is in. From this directive I gather that these politicians do share our concerns to some extend. A few choice quotes from the amendment (taken from the argumentation of each revision to the Directive):
"It is clear however that the Directive, despite argmentation by the Commission, will open the way for a broader use of patents as a means to protect computer software".
"The goal of patent law is not to ensure that patent holders enjoy certain privileges; the privileges granted to patent holders are only an instrument to benefit the invention process, to benefit society as a whole".
"The requirement of an effort of invention, and a significant improvement to existing technology, are fundamental if one wants to prevent patents for trivial "inventions".
And last but not least: "3bis: Exclusion from patentability: An invention implemented in a computer is NOT to be seen as a technical contribution only because it implies the use of a computer or other device. Also excluded are inventions that use computer programs, business methods or mathematical algorithms, and that do not carry any technical implications other than processing and rendering of information in a computer system or network. This means you cannot patent some widely practiced activity X as "Activity X on a computer/the Internet".
(Translated badly from Dutch by me).
Reading these amendmends, I get the warm and fuzzy feeling that some politicians at least share our concern. What remains to be seen is how they weigh these concerns against other interests.
Computer programs are not generally writings (Score:4, Interesting)
Yes, there may some cases where this is so, such as might be entries in the "obfuscated C code contest". But this is very seldom what we are speaking of when we talk about "computer programs".
More so than it is similar to an essay, the computer program is similar to the collection of specific gears, sprockets, pins, and their particular arrangements which cause a machine to work in a useful and well-defined way. These parts, similarly, like components of computer programs, have yet more primitive components and specifications which provide for those at the higher level.
Perhaps a more useful example due to its comparative complexity is a clock.
That the analogy of gears, pins, sprokets and their arrangement, to software and the computer on which it runs, is successful, is illustrated by the case of the swapped watch internals.
Suppose that we have an existing mechanical clock. Now, suppose we take a duplicate watch casing, but inside we insert a "general purpose gear, ping, and sproket emulator". Like the computer, this device, in order to function in a useful manner, requires a set of instructions. Specifically, these instructions must instruct the emulator how to perform operations with equivalent results to those that were in our existing clock performed by real, rather than emulated gears, pins, and sprokets.
The success of the clockmaker is determined not by creativity in ordering or commentary on the natural world. Rather, success is measured by efficiency, and correctness. Optimality is measured (when it is computationally possible) by precise mathematical metrics. Correctness is determined not by asthetic appeal, creativity, or insight, but by nothing more than the solution being provably correct through the employment of established mathematical techniques. In fact, we might wonder what "correctness" should mean, if anything at all, in the context of an essay.
In the case of the computer program, creatvity or imagination are desirable not in the structure, choice, or ordering of the instructions, but rather in coming upon a correct solution or in visualizing the problem.
A source listing's utility is perhaps directly proportional to its lack of creativity.
Re:Exclaim!!! (Score:2, Interesting)
AEL has collected even more useful material
http://wiki.ael.be/index.php/FightingSW
and there is a nice Letter of Tord Jansson
We have so many statements, and arguments. The EU conducted so many studies...
It is not a lack of intellect or arguments but a lack of power.
http://www.bustpatents.org
The movement for world-wide software patents is coming to a new high we are now 10 days ahead of the time when the european parliament is about to follow the US into patentability of software, business methods and it's not even clear to them because the words used have been very much stretched into endless terms of patentability(e.g. "technical")
If the EU adopts software patents, they will become global enforceable. That's when the real problem will begin.
Please help us to spread the word to attract people to a street performance on August 27th in Brussels(Old Europe). We also hope that webmasters against software patents can replace their
What if your child asks you one day: Where have you been when they
introduced software patents?
Lame argument (Score:2, Interesting)
Engineers in other fields have had to deal with this issue for two centuries. Get over it. The same argument could be made for the developers of player pianos, which the author says *should* be patentable. If he wanted to argue against all patents, this would be a valid argument, but that's not what he's saying.
There are other problems as well with this "software is free speech" approach. Source code is text - so what? Circuits expressed as VHDL are text too, so are plans for any mechanical device. Since all of these are just text and pictures, why single out software?
Flawed Argument (Score:3, Interesting)
Music is not a functional, algorithmic implementation of a transformative process. There is no mapping of a range to a domain inherent in a piece of music.
Patents are awarded for Inventions (also sometimes termed Implementations), and copyrights for Expressions. If someone were to take an existing well known algorithm, say the bubble sort, and code it up in his language of choice, that work is an expression and is protectable by copyright.
If somebody were to examine and existing piece of mathematics and realize for the first time that it has applicability in solving a sorting problem, and then code it up, he would have both an invention (use of the math in the contect of the sorting problem on a computer) and an expression (the computer code). Clearly this qualifies in some protean sense as an invention.
Heck, the author would even be able to take the result as a working model to the patent office - on his laptop computer, satisfying the old 19th century requirement for a working model.
The concept of a working model of a piece of music just does make sense.
Nor does the concept that a piece of software is solely an expression.
Math invention (Score:1, Interesting)
Its easy to dtermine that a pure algorithm is unpatentable. The difficulty arises when physical utility is associated with that new algorithm.
What if I improve anti-skid brakes by an improved algorithm in the logic of the device? Is this patentable?
What about an invention that is 90% software ("math") and 10% mechanical.
80%/20%?
20%/80%?
How could anyone possibly make the determination that one invention is a little too much algorithm dependent and very similar invention is not?
The basic concept behind all IP laws (Laws that for some unknown reason have been determined to be necessary in all developed nations, yes even China, and have withstood the cries of patent abolitionists for hundreds of years) is that they serve as an incentive for investment into new technology.
Thus the reasoning should follow that these laws should apply to anything which provides utility and would not have been undertaken or would have been undertaken at a later date without the incentive that IP protection provides.
It follows that the argument should in reality be centered upon the amount of difficulty required (obviousness factors) in order to arrive at the new utility.
There are difficult problems that can be solved via software (again, 90% sw/ 10% mechanical, 70% sw
The real solution is to disallow the trivial SW patents that would have been arrived at no sooner in the absence of IP protection.
Re:But if code is like music... (Score:3, Interesting)
Mind if I ask why? How can you possibly make such a distinction? A series of operating instructions is a series of operating instructions, whether they are ideas expressed verbally to a carpenter building a house or written ideas given to a machine for processing information or mixing dough or baking bread. How are the operating instructions given to the processor in a GPS fundamentally different? If you want to patent a truly novel, non-obvious GPS device, fine, but patenting a specific set of instructions to the device is no different than patenting an idea.
Re:Math invention (Score:3, Interesting)
I said that doubling video compression and RSA encryption didn't even rate the title "software invention", those are actually math "inventions". Once the math is known it is blindingly obvious to apply it to software. Neither one would be a "real" invention.
What about an invention that is 90% software ("math") and 10% mechanical.
That's fine! It doesn't matter what the percents are, so long as it is not 100% software. Just test the patentability criteria. The criteria are (1) utility (2) novelty and (3) nonobviousness. And you go with the patent rules we've always had - the rules before idiots started changing them to make software patentable. The rules said math and software are not patentable subject matter, so you ignore and all any portion about what's going on in the software. Then you look to see if all the patentability criteria are met. (1) Does it have physical utility? (2) Is there novelty? It has to be something novel outside of the software. (3) Is there something non-obvious? It has to be something non-obvious outside of the software.
Your invention can be 99% software and 1% hardware, if it's useful and there is novelty outside of the software and non-obviousness outside of the software then you've invented something and you have a valid patent. You can use as much software in you invention as you like, but the software can't BE the invention.
On the other hand if there is zero novelty outside of the software and zero non-obviousness outside of the software then your "invention" actually lies entirely within software. It is therefore 100% software 0% hardware. You are trying to patent "software per se". You haven't invented anything, all you've done is write an ordinary program and programs are entitled to a copyright, not a patent.
You can have your patent so long as your "invention" isn't pure software.
What if I improve anti-skid brakes by an improved algorithm in the logic of the device? Is this patentable?
Sounds like you are trying to patent a pure software invention there. You mention nothing new aside from writing a new program. You are entitled to copyright protection on any new program.
Had you made some sort of novel and non-obvious change to the brake pads that allowed you to use a different algorithm that reduced heat build-up and increased traction then you get a patent.
-
My explanantion (Score:3, Interesting)
One point I think I made more clearly, that should have been made in that piece and would have fit perfectly, is that to my knowledge, only software is covered by both Copyright and Patents; it should be no surprise that two systems that were never designed to work together, basically don't! Copyright fits software much better then patents; that's a sign it should be treated under the copyright system and not the patent system. IMHO, I also did a somewhat more thorough job of exploring that point.
Still, for a 1991 essay, that was pretty darned good.