Slashdot Mirror
European Software Patents Not Dead Yet
Ensign Nemo writes "Software patents in Europe still being pushed.
They're at it again and they're not waisting any time. Even though opposition is there the backers of software patents are getting sneakier and sneakier." Poland, if you help us out again, I pledge to never, ever forget you.
One only has to look at the rampant achievements and success of Free Software and Open Source to see how much the rest of the industry is being held back by software patents and other "intellectual property" restrictions.
Try not. Do or do not, there is no try.
-- Dr. Spock, stardate 2822-3.
Um, I think you're a little confused.
Open source software, and free software *depend* on copyright. Yes, depend on it. Without copyright, then anyone could take the code, including large corporations, and modify it for their own interest, and sell it without releasing the source code. Basically, it'd defeat the point of the difference between "free", and "Free".
Software shouldn't be *patented* because you're patenting an algorithm. And computer code is a completely logical process. It'd be very similar to patenting a mathematical formula. They're both *discovered*, not really *created*. It also creates a lot of problems in enforceability, and in large corporations being able to sue anyone they please. It's not only logically wrong, but also effectively wrong. It's similar, in a way, to the DMCA in that it gives bigger corporations the power to control everyone.
Very very few people argue against copyright when it comes to software. Free software people/open source people argue against patents alone.
oh, and trademark has nothing to do with the issue. Don't lump the three under the whole "intellectual property" umbrella. You'll almost always be wrong when you do. They have very little to do with each other.
Very good post.
Copyright and patents were intended to encourage people to make stuff that otherwise wouldn't be done. When people and organizations are willing to create an entire operating system and a collection of thousands of programs (GNU/Linux and the thousands of associated programs), the basic premise of copyright and patents is nullified.
If people will create without incentive, there is no reason to impose needless costs on both consumers and creators by strangling the public domain with laws like copyright. If Microsoft refuses to develop Windows because it no longer has any copyrights, then Linux is there as a replacement, and it will become 10x what Windows ever was or Linux is today once it becomes unshackled from copyright and patent issues and has the customer base of MS Windows today.
And that is assuming that no grants are given to fund open source development. The Chinese government, among others, has shown that they are willing to fund open source work. A small amount of federal funds would replace a massive amount spent in retail software licenses.
If you don't want someone to know how you did something, close the source. If someone comes around 2 weeks later and duplicates your work.. well it wasn't really novel was it?
I've come up with many algorithms in the past that I've thought were novel... then found out someone thought of it and gave it a fancy name 50 years ago. I could have got a patent on it if I'd wanted - but I wouldn't have deserved it. I'm just not *that* good.
of course they're not dead. software patents benefit the corporations that control the governments, so they will eventually be installed in all countries. you'd think people would get the message from the backdoor process that was used to almost install them on eu countries this time. don't count on stopping it again.
"....the Ministers of Agriculture in their meeting of the EU Council of Agriculture and Fisheries on Monday are to give the nod to the controversial position of the EU Council of Ministers on the Directive on the Patentability of "computer-implemented inventions"".
Isn't that a bit like asking the Minister of Defence if they think the tax system should be revised? Or asking the Minister for Education what his thoughts are on creating a new highway?
I'm beginning to see why so many Europeans don't take the EU seriously.
I'm a creator, but I'm concerned that operating in a legal minefield of vague, overlapping and bogus patents will destroy my ability to make money.
Copyrights are a reasonable match for software, and that's all the protection it needs. Software is too maleable to support well-defined patent claims. Software doesn't need copyrights and patents. Other kinds of products don't get double IP coverage; why should software?
I'd rather take my chances that some someone duplicates the functionality of my products and competes in the marketplace than risk having some jackass submarine patent-holder pop out of the woodwork demanding cash out of my bank account.
patents create property. property is wealth. patents like all property create wealth. the more varieties of property in society the greater the wealth that can be created. the better off society as a whole becomes. you only need to look at the US versus poorer countries. it is the enforcement of property rights that has created wide spread wealth.
Weapons create property. property is wealth. Weapons like all property create wealth. the more varieties of weapons in society the greater the wealth that can be created. the better off society as a whole becomes. you only need to look at the US versus poorer countries. it is the enforcement of others property rights that has created wide spread wealth.
Well to begin with, patents are an imposition on freedom. You give someone a temporary and artificial monopoly in order to gain benefit to society. You don't have to justify why not give them patents, you have to show why you should. The benefits (at least in theory) are thus:
1) Companies have a greater incentive to innovate and create new products.
2) Little guys don't get their inventions stolen.
Except that neither of these apply to software patents. Now, clearly, there's no shortage of innovation and new product creation in the software market. I mean, really, there are fifty zillion software companies out there and myriad individuals just writing software in their spare time. And software patents are more like to hurt the little guy than help them. The little guy can't afford the cost of doing a patent search for every line of code he writes to make sure that no one else has used the same idea in the last 20 years.
In software how novel things are is very unclear. There would be nothing wrong with patenting things which truly are completely novel, but if ten programmers are given a problem, four or five of them are going to come up with the same answer. If we allow the first one to come up with an idea to patent it, it means that the others aren't allowed to solve the problem in the best way even though they would have on their own.
And more to the point what will actually happen is that the second, third, fourth, etc. guy will come to the problem, have no idea how the first guy solved it, but come up with the same algorithm and implement it. They won't know what the other guy named his algorithm, so they won't be able to find the patent and the software will be produced. Then at some later point, the first guy's company will discover this and hold the other guys' software hostage with a patent lawsuit. Patent is meant to prevent really stealing people's ideas, but when you look at the patents which have been granted for software in places where such is allowed, the level of novelty required is so low that multiple independent creation of the idea patented is not only likely, but in many cases inevitable.
Further, showing prior art in physical devices is fairly easy, but showing it in software where other people's source code is hidden from you, becomes almost impossible. So maybe in the above scenario, it's not the first guy who patented it, but the third, but the first and second ones didn't see it as novel or didn't work for companies which worry about patenting things. But by the time the lawsuit roll around 10 years later, the first two have completely forgotten all about it and their code is each buried inside some obsolete product whose source code is locked away in a secret vault somewhere.
Now, you may note that this also makes it harder to find the lawsuits to press to begin with, and that's true. But there is one very obvious section of software which is more vulnerable to suits over software patents: open source software. I personally know of at least one major, well known open source project used by a significant portion of the user base of this web site which accidentally stepped on the patents of a large company. Now, in this case, the company has done a really terrific thing, they've not filed a lawsuit and aren't mentioning it in public (which is why I'm being so vague) so as to not hurt the reputation of the project or in any way cause FUD. But, in general, software patents are likely to hurt open source because they'll be enforced on open source products much more often than on closed source products.
Keith
... and write to your minister of agriculture, if you're a EU citizen!
Let them know that software patents are important enough to deserve a proper discussion in the parliament. Add why *you* think they are important (and wrong).
For the Belgians and the Dutch: http://www.softwarepatenten.be/landbouwraad.
For the rest of us: http://ffii.org/.
Please, take this 5 minutes, it's worth it.
Chris.
Because that is not what usually happens. Software patents are used as weapons to destroy competition, not as protection for true innovation. If patent holders would limit themselves to only truly innovative, non-trivial solutions I doubt anyone would care. As it is we (anyone making a living in software) are under constant threat because someone might take out a patent on some completely trivial technique tomorrow and shut you down for the next twenty years.
I asked this in the previous patent discussion as well, but let me ask again: does anyone know of any "good" software patents? I.e. non-trivial, innovative, and realistic? The only one that came out of that previous discussion was the RSA patent, which I agree meets these criteria and as such is worthy of patent protection.
But right now most software patents do not fall in this category. Instead they patent utterly trivial "inventions", often in ways that have been standard practice within the industry for years.
Let me add a prediction for what will happen within a year of software patents becoming a reality in Europe: with patent protection finally possible in all major markets, Microsoft will make its big move against open source. It will attack Apache, Samba, Mozilla, Open Office, and perhaps some others, stating they all violate Microsoft-held patents. I suspect Linux itself will be allowed to exist as a sort of token competitor to Windows, but there will not be any useful software left to run on it, which accomplishes Microsofts goal as well as destroying the OS itself.
Moreover, the people who worked so hard to make these incredible projects a reality will be painted as "thieves" in the media. After all, they "stole" "intellectual property" that so clearly belongs to our beloved "innovator", Microsoft.
One final question on my mind: who keeps adding software patents to all sorts of irrelevant agenda's in Europe? It is very annoying having to fight this battle every two weeks, especially since our adversary is some nameless civil servant who apparently got one bribe too many...
Supporting Open Source software by not introducing software patents is the best thing we can do for our trade balance. By using more and more Open Source instead of paying overseas countries for commodity software like Operating Systems and Office suites the more likely it is that Europes own software packages sell more than we import from Oracle, Microsoft, Adobe etc.
:).
That is a solid argument for the MEPs. Something they can understand.
Im all for copyright legislation because it protects Open Source aswell as commercial endeavours - that is very important, but patents on software algorithm and ideas should not be possible, next thing you know someone is patenting pi - or atleast the first five decimals
There are software ideas that are unique enough to support a patent. The problem is that every software idea ends up being patented, unique or not.
This effectively makes it very hard for small and independent software developers to exist as they don't have the money to get a patent. Don't have the money to protect against lawsuits for unjustified patents (it takes a lot of money to get a patent overturned). And finally, if they some way get a patent, they can't afford to protect it.
Furthermore a problem with software is that you patent an idea and not a solution. Eg you can patent "buying items with one click" instead of the actual implementation.
My biggest gripe is that I don't see any benefits from software patents. If you can demonstrate that it has some benefits I'd be interested in hearing about it.
Patents which are too ambiguous and too general, those having no real use to an overall industry or those too similar to existing patents or existing work, should not be condoned. Companies should not even bother---it makes them seem disrespectful to the industry.
In perfect world this wouldn't happen and therefore software patents should be possible. But in the real world, where companies try to patent anything that they can, patent offices are not able to investigate wich patents are too general to be allowed. This is why you get patens for one-click, double-click, scrollbar...
Other industries have existed long before patenting and on those industries most of the common tools and technigues have been availeble for everybody to innovate. When it has gotten more complex, then came the patents. But with more complex structures you need more R&D or something really innovative to prevent anybody else of doing something basic stuff.
software industry is just too young and evolving to be limited with patents.
Not being allowed to exactly duplicate someone else's work will encourage people to try alternate solutions, or experiment with variables.
Or just forget their ideas, because they have not enough resources to find out what parts of their innovation has been already been patented. Why risking everything, when you can join the big company and work there as they have allways done.
Eventually someone's bound to slip up and say "YES".
GJC
Gregory Casamento
## Chief Maintainer for GNUstep
"For example, electronic companies want to keep the same type of patent coverage for the next generation of their products (with more embedded code) as they are used to having. And what's wrong with that?"
A hardware idea simply re-implemented in software is not a new idea and not patentable.
A pure software algo is a mathematical discovery and not patentable. You cannot patent mathematics.
I think you misunderstood what the EU parliament did, the wording rules out pure data processing patents. It does not rule out software as part of an invention.
So electronics manufacturers can still patent their widgets because the software is part of the new widget.
The other change was to beef up the 'technical' requirement, to prevent business process patents disguised as software patents. e.g. electronic shopping lists, one click ordering, assigning rating numbers to how powerful a computer is in order to decide if games will run or not...
Who could argue that that isn't a good thing!?
And that may exactly be the FTC report concluded that patents in the computer hardware industry and semi-conductors are generally considered to be not very efficient either.
Further, even though the intellectual achievement is quite similar (the routine and layout of the chips may require some extra time), in practice you often need either a chip fabrication line, or have to buy stock of those chips with someone else. In case of software development, almost the only investment that really counts is human capital. You don't have to setup a new assembly line for each program.
There are indeed general problems with the patent system (e.g. triviality). However, software patents turn out to be especially sensitive to those issues, and they are in fact different. Otherwise, why would all those studies make a special case out of software?Donate free food here
How desperate are these people?
Agriculture Ministers to pass Software patents?
Please... somebody get these people jobs !paid by SCO.... I feel pity for the pro-patent people, I honestly think that they are flogging a dead horse, trying to get this completely undeomocratic legislation passed.
Honestly, the patent people should just admit defeat, with some grace... it's sad at this stage.
As an example there is a patent on using pointer exchange to swap two large data structures instead of swapping each member of the two structures. The patent applies to any software that is implementing a CPU emulation. So a very old technique is patented because nobody has documented that it has been used in a CPU emulator, even though it is just the same as any other case.
This is the problem, software is easy. It is trivially easy such that a teenager can master the most advanced of techniques. The only difficult thing is not making mistakes and missing bits out when you translate the logic into source code. That just takes discipline to write down each and every step in full and knowing how the language you are using actually works. It is thus certainly not patentable. However, I beleive that things like the psycological model that an mp3 encoder uses to select information to throw away might be reasonable to patent.
But many devices are just the sampling of a signal, then transformation, followed by the application of the result through an output device. The input device should be patentable if it is inventive and difficult, and so should the output device. The transformation used to be difficult when it consisted of cogs and cables and things. With software, when you know how the transformation should go, it is trivially easy, and thus should not be patentable.
I think this "Computer implemented invention" thing is supposed to cover the cost of discovering the necessary transformation. That could be reasonable if the transformation is a truly new and difficult thing to grasp. But I think a patent is still wrong for that. The effort has gone into mathematical calculations of the relationship between the input and output and the acceptable margins, and that effort should be protected enough to allow people to produce without cheap duplicates running the original producer out of business.
I think that the discovery of scientific knowledge for business purposes should be protected. That is what patents are supposed to do, but that has nothing to do with software, and no software technique should be patentable. Only the application of a specific transformation to convert a particular defined input to a particular defined output to complete a given physical (*not* logical) task should be covered. How you make the software do that is always trivial and should be always unpatentable. Making levers and cogs do it is hard and could be intrinsically patentable but not software.