that is what Europe is, my friend, and I don't know which is worse.
Because you apparently don't know Europe. Germany did not take this position because some government official thought that would be a good idea. In fact, without our input, they'd probably support software patents (just like several other countries which are now against instead od pro).
You seem to think that Europe only has communist or fascist governments which decide everything on their own without any possibility to influence them.
suppose if Germany decides not to support the European Commission on changes in the law to software patents, then nobody can sway them otherwise because they are a sovereign state and don't have to comply with what the WIPO or the EC says.
First of all, as member of the EU, Germany has to comply with EU directives that are passed. Next, WIPO does not only not require software patents, it even forbids them (just like TRIPS).
The excuse used by software patent proponents regarding TRIPs, is article 27:
Patents shall be available for any inventions, whether products or processes,
in all fields of technology, provided that they are new, involve an inventive
step andare capable of industrial application.
This text however explicitly uses terms which are defined nowhere else in the treaty
(like "invention", "field of technology" and "inventive step"), so that signing
members can define these terms themselves in such a way that they fit best in
their existing laws.
According to article 52 of the the European Patent Convention, a computer
program can never constitute an invention. And in the Parliament proposal of the directive, "field of technology" is defined in such a way that computer programs, maths, business methods etc do cannot belong to one (even if they're executed on a computer).
And on top of that, there's articles 7 TRIPs which is interpreted by the WTO as that the measures as implemented must...
.. contribute to the promotion of technological innovation and to the transfer and dissemination of technology, to the mutual advantage of producers and users of technological knowledge and in a manner conducive to social and economic welfare
Most evidence points to the contrary as far as software patents are concerned.
So TRIPs does not require software patents, how does it forbid them?
Article 10 of the TRIPs treaty states:
Computer programs, whether in source or object code, shall be protected as
literary works under the Berne Convention (1971).
As opposed to what a first reading would suggest, namely that this simply means
that copyright protection must be available for computer programs, this article
goes further. The WTO states on its website regarding article 10.1:
The obligation to protect computer programs as literary works means
e.g. that only those limitations that are applicable to literary works may be
applied to computer programs.
Since patent protection is unavailable for literary works, it can't be
available for computer programs either according to TRIPs. Proponents of
software patents often counter this using their interpretation of "computer
program as such", which turns "computer programs with a further technical
effect" into "computer-implemented inventions", which in turn would supposedly not be
affected by this exclusion.
This interpretation is however invalid due to article 4 of the EU Software
Copyright directive from 1991. This article states that a computer program as
literary work includes the following (emphasis mine):
... the permanent or temporary reproduction of a computer program by any
means and in any form, in part or in whole. Insofar as loading, displaying,
running, transmission or storage...
The WIPO Copyright Treaty also contains applicable clauses (article 10):
(1) Contracting Parties may, in their national legislation, provide for
limitations of or exceptions to the rights granted to authors of literary and
artistic works under this Treaty in certain special cases that do not conflict
with a normal expl
Designing a novel, useful, original UI is as difficult as any other aspect of product development in the business world. If I can patent the design of a remote control [which wouldn't send the/. crew up in arms] why is it such a logical stretch to patent the interface for a software product that has the same type of functionality?
A patent is still a state-granted monopoly in exchange for something to society. What does society get from this? In case of a remote control, you get the schematics of how to build that thing. In case of a computer UI, society gets nothing in exchange for the monopoly: using the UI in a product is already the same as publishing it, so you don't need patents to encourage that.
And it's also not like Apple won't develop any new UI paradigms anymore if they can't get monopolies on them.
That's a big problem with UI (and business method) patents: the "inventors" get a monopoly in exchange for doing something they would do anyway (publish their innovation).
The original owners of the patent that's now asserted against JPEG also promised they would not sue anyone over it. Then they sold all their assets, Forgent got hold of the patent and did start to assert it. I'm not sure if a promise (or even a written license) from Cisco would still be valid if they were to sell that patent to someone else.
IBM allows anyone to license any and/or all of its patents. All you have to do is give them a call and they will be more than happy to work out an agreement with you.
The large patent portfolio is more of a defensive measure - if IBM is sued, it's more than likely that they can countersue with a few choice patent violations.
If that were the case, then why are they lobbying so hard to get software patents legalised in the EU, and not lobbying at all to get them banished from the US (and on the contrary, blocking studies that could show there is a problem with swpats)? You don't need defensive software patents if no-one can have any.
Assuming Parliment unanimously wants to create a law prohibiting software patents, and the Council unanimously wants to create a law to legalize software patents, which side has the final say to push through what they want? Or would it be a stalemate?
No. The last time the directive passes through Parliament (the third reading), the Parliament only has two options: yes and no. If they vote" yes", the directive as it came out the conciliation committee is accepted. If they vote "no", the whole directive is abandoned altogether (so then you simply don't get a directive, which would still be preferable to a directive that explicitly software patents).
But taking another toke from it myself before coughing and hurling, as I'm opposed to broad/silly/etc. (software) patents myself :
You can't have patents without broad/silly/... ones. And software patents by their very nature are bound to have a much higher percentage of broad/trivial ones. See e.g. this post of mine and this page by FFII.
The first problem is that although there's a level playing field in the EU, there isn't in the US which means there isn't a level playing field for large multinationals situating in both areas.
It's indeed annoying for multinationals that the law isn't the same everywhere. That doesn't mean you should import a bad law to somewhere else however, but kill the bad law.
The second issue of EU companies patenting stuff in the US is valid and something grandparent probably didn't keep in mind. Paranoia may strike and cause concern that US companies may be favored by the USPTO, however.
I don't think so. There have already been Eolas-like cases involving European companies before, e.g. AllVoice. It's nothing new.
On the third issue, the simple thinking is this:
EU > EU ; no patenting
US > EU ; no patenting
EU > US ; patenting, but there's a good chance the patent's already held
US > US ; patenting up the wazoo
I.e. there's a disadvantage to the 'EU' company, and it's this disadvantage that *some* legislators want to iron out. Unfortunately they're using a steamroller instead of a velvet iron:)
I still fail to see where the disadvantage is. Whether an EU or US company wants to patent something in the US, the chances that this thing is already patented there are equal, and the fact that others are "patenting up the wazoo" in the US is equal as well.
The only thing that's different, is that US companies already have this problem immediately in their home country, while a European company's home base is currently a safe haven. I honestly still fail to see how introducing software patents in Europe would help European companies.
Indeed, they're simply ignored. In practice, it amounts to more or less the same thing.
It's part of the law making process.
Unfortunately, it's true that legally, the Council is not bound at all by what the Parliament fought and voted for. It's merely "advice" which they "have to take into account". That doesn't mean that the fact that they take full advantage of this hole in European law and that tabling a counter proposal written by, of all people, patent office administrators is not a subversion of the democratic process.
The proposal bounces between comission an parliament several times until one side accepts the proposal of the other. The comisions proposal will hit the parliament after the election, so elect wisely.....
It bounces between Council and Parliament actually. The problem is that in the second reading in Parliament, the Parliament can only reinstate what it voted in first reading with an absolute majority (nr_of_MEPS / 2). So it becomes much harder. If it still doesn't pan out, there's a reconciliation committee of MEPs, people from the Commission and people of the Council.
Of course, most of the people in the Parliament directly responsible for this directive were pro-software patents (as the whole purpose was to legalise software patents, and not "clarification" and "harmonisation" like the Commission claims). Let's hope indeed the next Parliament will be ready to show its teeth if we can't get the Council to reconsider.
PS: Here's the whole codecision procedure in pseudo-java. I wonder whether this means that the "underlying principles and processes" of it should be patentable as well...
During the past years, more and more subjects that the EU makes laws about have fallen into the process of codecision, which means that the Parliament has to agree to every law the Commission or the Council proposes
The software patents are also handled by codecision. The problem is that you first have a proposal from the Commission, which the Parliament then can only change by majority. Next you have the Council (as it is now in the swpat case), which does not have to take into account anything voted by the Parliament, because the amendements we fought for in the Parliament are merely "advice" to the Council. They can basically draw up anything they like to submit to the Parliament for second reading.
If the Council destroys the Parliament's amendements, the Parliament can only restore the amendements they approved in first reading (i.e., they can't add anything new) in second reading, but only with absolute majority this time... So it becomes much harder.
Secondly, many innovations can be implemented in either software or hardware (is ASIC chip hardware or software?)
Chips are not patentable either, they're covered by a special "sui generis" right (in Europe at least), because chip design is not considered to be creative enough to fall under copyright law and not inventive enough to warrant granting patents.
Anyway, you're right on one count: the way an invention is implemented is irrelevant. Let me give two examples to illustrate this.
You have a fully automated weaving machine, software/computer controlled. You then invent a way to turn the knives while cutting so that they can cut the threads much more easily. The only thing you have to do to implement this, is adjust some parameters in your software.
This is perfectly patentable under the directive that the parliament voted, because the invention is not the fact or how you changed those parameters in your program, but how and when you have to rotate those knives. In this case, you also get a patent on the turning of those knives, not the way you implemented it in software. As such, even if someone would do that using pure mechanical means, they'd still violate your patent.
You have this same weaving machine, and decide it would be useful to see all sorts of parameters of the inner workings of this machine at all times. You stick it full of off-the-shelf sensors, connect them to the cpu, add a new display, write software and presto: great graphics. This would not be patentable, even though your implementation requires a lot of hardware (physical stuff).
The reason is that an invention must be new and non-obvious (and industrially applicable) to be patentable. Those sensors, the display and the connections are not new (nor non-obvious, probably). The only new thing is your software and the presentation of information it does. As such, there is no (physical/material/technical/...) invention present, and you cannot get a patent.
Regarding "no secrets", in addition to the etymological derivation of "patent" (Latin for open) and its secondary definition (obvious; plain), I believe the original purpose of patents was to promote progress by discouraging secrets and thus encouraging the sharing of ideas.
The sharing of how inventions work, not of mere ideas. Have a look at this presentation, especially slides 4 till 6.
However, your basic idea is indeed correct: they're supposed to promote the sharing of information, in return for a monopoly on the concrete application of this information (i.e., on building inventions based on whatever is described). The problem with software patents, is that the application of this information still is information (a computer program). So you are getting a monopoly on the information itself, not on some concrete and physical thing/process.
It's as if a patented engine would result in a prohibition for anyone else to write and sell technical manuals on how to construct such an engine, because if the reader follows those instructions, he would be violating the patent. So the patent is not promoting the spreading of information, but inhibiting it.
Traditionally, the physical invention is on the monopoly side and information on the disclosure side, but with software patents both are pure information. As such, you indeed end up with patents on ideas (information) instead of on inventions. The result is that you get much more broad monopolies, where the advantage of disclosure no longer weighs up to the disadvantages of the granted monopolies.
It really isn't a problem of the patent offices, it's simply the patent principle and patent law which aren't fit to protect advances in logical reasonings. Especially read the discussion between the programmer and the UK Patent Office Deputy on that page (first point).
If you'd invoke them, it would probably on a state level: after all, you are challenging the validity of other laws. For example regarding TRIPs, only a state can file complaints regarding breaches, and that must be done that the WTO. So we'd have to find a nation willing to do that. And I'm not sure which state would be willing to do that, as it would result in a head-on collision with the US.
Then again, recently a country (forgot which one) successfully won such a complaint against the US regarding online gambling, so it's definitely not impossible. I don't know who and where one can complain about infringements to the Berne convention.
The whole "TRIPs requires software patents" fallacy is covered in great detail by FFII.
Let's look at those costs and see how many of them are actually required and how many could be done by the applicants/OSS community:
EPO Fees: 4300 - Required by the EPO
Professional Representation: 5500 - Optional, the cases could be filed pro se for free
This is extremely discouraged. Even someone from the European Commission whom I talked to (the person who is handling the software patents directive in the Commission) admitted you have very little chance to get your application approved if you don't have any experience with patent law.
These people are programmers, not lawyers!
Translation: 11500 - Optional, could easily be done by bilungual volunteers
Not optional, it must be translated into all official languages of the countries where you want your patent to be valid. And you don't have to be just a "bilingual volunteer", you have to be a "bilingual volunteer with lots of time and who knows the patent jargon in both languages".
National Renewel Fees: 8500 - Required by the EPO
So, in the end with a little work the cost of a patent for a 10 year term becomes EUR 12800 or EUR 1600 per country. I really don't think this is excessively high for anyone serious about OSS projects, epecially since it could be raised by funding drives.
A "little work"? EUR 12800 not excessively high? And why on Earth should only "serious" OSS projects be allowed to be viable? It's like saying that only people who write for a living should be allowed to publish; the rest only does it as a hobby and thus should put up or shut up.
FWIW, I personally work on an open source project, which exists already for about 13 years (see url in my info). We have over 10GiB of downloads per day on our main ftp site alone (I don't have statistics on the mirrors). It's used by several companies and universities all over the world.
We do not have EUR 12800 or even EUR 1000. Why should we and other people start to have to pay and spend time on learning patent jargon and translating it so that we can continue our hobby project (which happens to be useful for other people)? What does society as a whole gain from this extra burden?
And what can you do with a single patent when a company like IBM or even Microsoft attacks you? Make paper planes to throw at them in a the court room, in a lawsuit that costs another EUR 50,000 ($1 500 000 if you're unlucky and sued in the US), and which requires time which you would normally spend doing your day job earning money that allows you to spend time on your hobbies?
Besides, there's much more than open source projects. If you take an SME, then there are even no imaginary volunteers which are ready to do whatever you ask. They have to pay all those things by themselves. It's them who will be hit the worst. They most certainly do not have EUR 12800 to spare, much less EUR 30000 (or EUR 50000 according to the European Commission -- google cache because original site seems to be down).
Since you didn't bother to link to whatever FTC study it is you are talking about, I can't really respond to it. Although just on the basis of what you said I'd question if they were studying the national effects of patents on the economy or the individual effects on a company/inventor.
That's indeed what they were studying. The original study and an extract of all software patents related stuff (it's on a page of FFII UK, but page numbers are given and if you compare it to the original, you'll see it's uncommented and really does contain all software
The problem with software patents is *NOT* the concept of patents, but the inability of governments to effectively administer the systems. here are some of the issues:
Please read this page, especially the discussion between the Deputy Director of the UK Patent Office and a programmer to see why that assumption is incorrect. There are however also some practical programs, as you point out:
Hard to do prior art: I see patents all the time for which I know prior art existed in the 70's. The problem is that it is difficult or impossible to find the prior art references.
This is one of the reasons the patent system is unfit for software. It starts from assumptions which are untrue in that field.
What can be patented: We've had the US Patent office grant a huge number of stupid "business process patents" and patents on all kinds of silly and obvious stuff. it is clear that they don't have any sense and even less software knowledge or expertise.
This is addressed on the earlier referenced page. It is very hard to prove that even the most trivial algorithm does not contain anything new or a (very small) inventive step. Patent law does not contain any way to distinguish between a "small" and a "big" inventive step. Novelty and inventive step are simply filters intended to take out most "bad" patents. They know they can't catch them all, but with those rules they have determined they can catch enough to make the patent system have a positive effect on the whole (whether or not this is the case, is another discussion; I don't know).
Anyway, those filters don't work nearly as well for advances in the abstract (maths, business methods, rules for organisation and presentation of information,... and their notation in a form that computers can understand: computer programs) as they do for inventions related to applied natural science. The reason is that they were never intended to work for those things, which is exactly why the European Patent Convention excludes all those things from patentability.
To sum it up: patents are an archaic tool totally unfit to protect abstract knowledge, and were never intended to do that in the first place. If we think copyright is not enough protection, then we will have to devise a new "sui generis" ("one of its own kind") protection for software which is specifically adapted to it (just like there is one for chip design, which are not protected by copyright -deemed not creative enough- nor by patents -not inventive enough).
Ah yes, the old "patents are too expensive to file" argument, complete hogwash.
The average cost of a granted European Patent (valid in 8 countries) is EUR 30 000 (yes, 30 000, not 3000).
As far as enforcement, they wouldn't have to actively enforce their patents, they could use them strictly as defensive patents.
What's so great about a system that requires people to invest money with as sole purpose to protect themselves from that system? Especially if it's extremely hard to find any macro-economical advantages (and many hints at disadvantages) of this system (cf. FTC study).
Well, actually one of the biggest proponents of software patents is the UK -- both in the European Parliament and on national level. Even in the courts, (much) more software patents have been upheld in the UK than in the rest of Europe.
Newsflash #2: It is the Council of Ministers that is pushing this decision through. Guess what that is? The EU member governments elected by the people on national level.
The main problem is that the position of the Council has been written by the Council working group on intellectual property (patents). The Council itself barely understands the issues. Now who's in that working group? The patent administrators of the national and European patent office(s)! It's very convenient for those people to be able to rewrite a directive which was drafted in response to their lawlessness...
So in total, the current Council's effect is completely undemocratic: the Parliament is being overruled by a bunch of patent administrators who don't like being criticised.
US Company : Let's expand to the E.U.
US Company : Oh hot diggity-dang! Will you look at that! Nobody here patented it, 'cos they can't, we won't have any trouble competing on a 'level' playing field!:D
And this a problem how exactly? That precisely how the free market is supposed to operate!
EU Company ; Let's expand to the U.S.
EU Company : Aww shite. Wtf? We have to license this if we want to operate in the U.S. ?
Or: we don't want to operate in the US and it's not yet patented there: let's just patent it in the US and sue all US companies to death! They can't do anything to us, as we don't sell anything there. Watch out for the European Eolas coming after you.
EU Company : Eh.. wtf ? Some US Company just started business here, and is using our ideas that we couldn't patent!
US Company : Haha. Silly Europeans - either they pony up for a license, or they just don't get to enter the U.S. market. Win/win for us!:D
I don't understand this one. If an EU company can't get software patents in Europe, a US company can't either. So the European company doesn't have to pay for any license. Whether it also wants to operate in the US, is completely independent from this. European software patents are software patents in Europe, not software patents for European companies. There's a very big difference.
Until, of course, it comes up again next year, and the year after that, and the year after that, until finally some group of people who are stupid/corrupt/bribed vote to pass it.
It takes a lot more than a year to prepare a directive. If they fail now, you can bet they will take a while before attempting again. They'll probably not even try the directive-way anymore, but try to directly re-negotiate the European Patent Convention (which happens immediately between states, without that pesky European Parliament that can interfere). However, it would be a significant victory in the sense that a lot more people will start the debate better informed (the problem really is mainly ignorance and not necessarily malevolence).
No, my understanding is that the directive would be at the phase of a "second reading" when it returns to the EU Parliament this time. For a second reading, the Parliament's options are much more restricted
No, the original poster is right. The decision Wednesday was in the Committee of Permanent Representatives (COREPER). It's a group of civil servants cf. ambassadors representing all members states doing the bulk of the negotiations. Nothing decided there is really official however, that only happens after the ministers confirm it (which would be on 17/18 May in this case).
If they confirm it, only then we are in the situation you describe.
Trully innovative software patent (innovative algorithm) are not necessarly bad. The problem with the situation in the US is that obvious stuff is being patented (and not just software).
Without backup, that's like saying patents on truly innovative story lines are not necessarily bad. And what is your definition of a "truly innovative algorithm"?
Further, please also read this page, especially the first section (A problem of law, not of patent examination). The "newness" and "inventive step" conditions of patent law are completely unfit for software. Using patents for software is like using jet engines to power your bike. Sure, you may be able to get it to work, but it's a very dangerous and destructive combination, and there are much better solutions available.
I bet you just made that up so you get modded up and present an entirely false argument to readers.
He's quite correct actually.
What happened was:
* the concept of excluding "computer programs as such" has been refined over time:
... by the patent administrators of the European Patent Office, in contradiction with several treaties and laws (TRIPS article 10, EU Software Copyright article 4, Berne Convention,...) and against most economical studies that software patents are bad for the economy. Hence, his comment that it's a wishlist by the patent offices is correct.
software patents are allowed _only_ if they should a concrete "technical effect" (in the US, it is different: they only need to be "useful and produce a tangible result") by case law and appeal board hearings;
Too bad that technical in patent jargon is completely undefined and as such means squat, and has over time been interpreted by the EPO to mean basically anything useful. Read this page by a colleague of yours, especially under the heading "When is something technical", and especially the second and third point.
Making a program run faster or use less memory is interpreted as a technical effect on the way the computer works. And using a computer to run a business method is technical, because it "provides surprising speed or economy of scale benefits"
* over time, the national courts have adopted subtly different principles;
A few national courts actually, and mostly the UK one at that
.
* to resolve this, the EU directive is proposed that enforces a common approach across the EU (and, upon any members of the EU that are part of the EPC).
The Commission and the Council promote the approach of the European Patent Office, which has no safeguards against software patents or business software patents. The Council's version was even written by people from the patent offices. You know, the people that have to abide by it, but haven't until now, got a chance to rewrite their own laws. Quite convenient, no?
Let me re-iterate:
* if this EU directive does not go through, then _software will still be patentable_ -- do not labour under the false impression that this directive is able enabling software patentability: software is _already patentable_, this directive _does not change that_ (in fact, it actually restricts it slightly).
Software patents are only enforceable in a few countries currently, and not even that always pans out evenly. The originally proposed directive by the Commission and the new proposal by the Council want to put it black on white that software patents are enforceable. That is a change. It's less of a change to render them unenforceable, as until now they've been unenforceable in more cases than where they were when considering the EU as a whole.
* thus, software will remain patentable, but not in a consistent way: this is even worse because some member states may actually take a very relaxed approach and allow patentablity of software.
Please stop your lies. The Irish proposal literally states:
A claim to a computer program, either on its own or on a carrier, shall not be allowed unless that program would, when loaded and executed in a computer, programmed computer network or other programmable apparatus, put into force a product or process claimed in the same patent application in accordance with paragraph 1.
When removing the double negation, we get:
A claim to a computer program is allowed when running that program on a computer does something t
And the same goes if you are working in the steel industry on creating new steel production processes - you would have to keep up with the state of the art to ensure that you don't infringe a patent. I don't see how software is any different and why
Software is different because you don't have to build a steel new mill to use some new technique, or even to begin producing. All you need is a computer and a programmer. By introducing software patents, you introduce an extra entry barrier. Of course, some big established companies would love that: it allows them to better control who can enter the market.
It's also different because a programmer does tens of small "inventive steps" per day while programming. Just like the author of a story makes tens of small "inventive steps" per day. It's the normal course of action when doing something based on creativity and will be realised much more as opposed to when you are constrained by physical properties of materials. You're hampered more or less only by your imagination and creativity, instead of by the physical world.
This means that the "inventive step" condition of patent law is totally unfit for creations in the realm of abstraction, as it's passed by too many things. The newness and inventive step (or usefulness, in the US) conditions are merely filters to try to make sure that most granted patents are of good quality. They fail to achieve this in the software world, as they were devised for the physical world. It's a completely outdated concept when applied to today's information society, which is much better served by copyright (encourage as many separate works of the same ideas as possible to increase competition, but inhibit plagiarising or plain copying of other people's work).
In the physical world, laws of nature are not patentable. In software patents, there are no equivalents of laws of nature. The patented technique doesn't have to be able to run on a computer of today, it just has to be representable in the mathematical Turing model, that's all. That's one reason why so many software patents seem so basic: there is nothing basic that is unpatentable.
Finally, patent law is an economical law (devised in the 15th century, no less). You introduce patents in a field because you have good indications it will improve innovation and competition in that field. So before introducing it in software, you should first research whether they will have the same effect. Thanks to the US, we have already several economical studies, including the one by the FTC, that indicate they don't have positive effects, and even hamper progress.
One problem is that Irish Presidency is simply pushing its own interests. In Ireland, there's a 0% tax on patent revenues. So the more patents a company has, the more interesting it is for that company to have its official base in Ireland.
Slashdot Mirror
You seem to think that Europe only has communist or fascist governments which decide everything on their own without any possibility to influence them.
First of all, as member of the EU, Germany has to comply with EU directives that are passed. Next, WIPO does not only not require software patents, it even forbids them (just like TRIPS).
The excuse used by software patent proponents regarding TRIPs, is article 27:
This text however explicitly uses terms which are defined nowhere else in the treaty (like "invention", "field of technology" and "inventive step"), so that signing members can define these terms themselves in such a way that they fit best in their existing laws.
According to article 52 of the the European Patent Convention, a computer program can never constitute an invention. And in the Parliament proposal of the directive, "field of technology" is defined in such a way that computer programs, maths, business methods etc do cannot belong to one (even if they're executed on a computer).
And on top of that, there's articles 7 TRIPs which is interpreted by the WTO as that the measures as implemented must ...
Most evidence points to the contrary as far as software patents are concerned.
So TRIPs does not require software patents, how does it forbid them?
Article 10 of the TRIPs treaty states:
As opposed to what a first reading would suggest, namely that this simply means that copyright protection must be available for computer programs, this article goes further. The WTO states on its website regarding article 10.1:
Since patent protection is unavailable for literary works, it can't be available for computer programs either according to TRIPs. Proponents of software patents often counter this using their interpretation of "computer program as such", which turns "computer programs with a further technical effect" into "computer-implemented inventions", which in turn would supposedly not be affected by this exclusion.
This interpretation is however invalid due to article 4 of the EU Software Copyright directive from 1991. This article states that a computer program as literary work includes the following (emphasis mine):
The WIPO Copyright Treaty also contains applicable clauses (article 10):
And it's also not like Apple won't develop any new UI paradigms anymore if they can't get monopolies on them.
That's a big problem with UI (and business method) patents: the "inventors" get a monopoly in exchange for doing something they would do anyway (publish their innovation).
Do non-US citizens count?
The original owners of the patent that's now asserted against JPEG also promised they would not sue anyone over it. Then they sold all their assets, Forgent got hold of the patent and did start to assert it. I'm not sure if a promise (or even a written license) from Cisco would still be valid if they were to sell that patent to someone else.
The only thing that's different, is that US companies already have this problem immediately in their home country, while a European company's home base is currently a safe haven. I honestly still fail to see how introducing software patents in Europe would help European companies.
Of course, most of the people in the Parliament directly responsible for this directive were pro-software patents (as the whole purpose was to legalise software patents, and not "clarification" and "harmonisation" like the Commission claims). Let's hope indeed the next Parliament will be ready to show its teeth if we can't get the Council to reconsider.
PS: Here's the whole codecision procedure in pseudo-java. I wonder whether this means that the "underlying principles and processes" of it should be patentable as well...
If the Council destroys the Parliament's amendements, the Parliament can only restore the amendements they approved in first reading (i.e., they can't add anything new) in second reading, but only with absolute majority this time... So it becomes much harder.
Anyway, you're right on one count: the way an invention is implemented is irrelevant. Let me give two examples to illustrate this.
This is perfectly patentable under the directive that the parliament voted, because the invention is not the fact or how you changed those parameters in your program, but how and when you have to rotate those knives. In this case, you also get a patent on the turning of those knives, not the way you implemented it in software. As such, even if someone would do that using pure mechanical means, they'd still violate your patent.
The reason is that an invention must be new and non-obvious (and industrially applicable) to be patentable. Those sensors, the display and the connections are not new (nor non-obvious, probably). The only new thing is your software and the presentation of information it does. As such, there is no (physical/material/technical/...) invention present, and you cannot get a patent.
However, your basic idea is indeed correct: they're supposed to promote the sharing of information, in return for a monopoly on the concrete application of this information (i.e., on building inventions based on whatever is described). The problem with software patents, is that the application of this information still is information (a computer program). So you are getting a monopoly on the information itself, not on some concrete and physical thing/process.
It's as if a patented engine would result in a prohibition for anyone else to write and sell technical manuals on how to construct such an engine, because if the reader follows those instructions, he would be violating the patent. So the patent is not promoting the spreading of information, but inhibiting it.
Traditionally, the physical invention is on the monopoly side and information on the disclosure side, but with software patents both are pure information. As such, you indeed end up with patents on ideas (information) instead of on inventions. The result is that you get much more broad monopolies, where the advantage of disclosure no longer weighs up to the disadvantages of the granted monopolies.
It really isn't a problem of the patent offices, it's simply the patent principle and patent law which aren't fit to protect advances in logical reasonings. Especially read the discussion between the programmer and the UK Patent Office Deputy on that page (first point).
If you'd invoke them, it would probably on a state level: after all, you are challenging the validity of other laws. For example regarding TRIPs, only a state can file complaints regarding breaches, and that must be done that the WTO. So we'd have to find a nation willing to do that. And I'm not sure which state would be willing to do that, as it would result in a head-on collision with the US.
Then again, recently a country (forgot which one) successfully won such a complaint against the US regarding online gambling, so it's definitely not impossible. I don't know who and where one can complain about infringements to the Berne convention.
The whole "TRIPs requires software patents" fallacy is covered in great detail by FFII.
This is extremely discouraged. Even someone from the European Commission whom I talked to (the person who is handling the software patents directive in the Commission) admitted you have very little chance to get your application approved if you don't have any experience with patent law.
These people are programmers, not lawyers!
Not optional, it must be translated into all official languages of the countries where you want your patent to be valid. And you don't have to be just a "bilingual volunteer", you have to be a "bilingual volunteer with lots of time and who knows the patent jargon in both languages".
A "little work"? EUR 12800 not excessively high? And why on Earth should only "serious" OSS projects be allowed to be viable? It's like saying that only people who write for a living should be allowed to publish; the rest only does it as a hobby and thus should put up or shut up.
FWIW, I personally work on an open source project, which exists already for about 13 years (see url in my info). We have over 10GiB of downloads per day on our main ftp site alone (I don't have statistics on the mirrors). It's used by several companies and universities all over the world.
We do not have EUR 12800 or even EUR 1000. Why should we and other people start to have to pay and spend time on learning patent jargon and translating it so that we can continue our hobby project (which happens to be useful for other people)? What does society as a whole gain from this extra burden?
And what can you do with a single patent when a company like IBM or even Microsoft attacks you? Make paper planes to throw at them in a the court room, in a lawsuit that costs another EUR 50,000 ($1 500 000 if you're unlucky and sued in the US), and which requires time which you would normally spend doing your day job earning money that allows you to spend time on your hobbies?
Besides, there's much more than open source projects. If you take an SME, then there are even no imaginary volunteers which are ready to do whatever you ask. They have to pay all those things by themselves. It's them who will be hit the worst. They most certainly do not have EUR 12800 to spare, much less EUR 30000 (or EUR 50000 according to the European Commission -- google cache because original site seems to be down).
That's indeed what they were studying. The original study and an extract of all software patents related stuff (it's on a page of FFII UK, but page numbers are given and if you compare it to the original, you'll see it's uncommented and really does contain all software
Anyway, those filters don't work nearly as well for advances in the abstract (maths, business methods, rules for organisation and presentation of information, ... and their notation in a form that computers can understand: computer programs) as they do for inventions related to applied natural science. The reason is that they were never intended to work for those things, which is exactly why the European Patent Convention excludes all those things from patentability.
To sum it up: patents are an archaic tool totally unfit to protect abstract knowledge, and were never intended to do that in the first place. If we think copyright is not enough protection, then we will have to devise a new "sui generis" ("one of its own kind") protection for software which is specifically adapted to it (just like there is one for chip design, which are not protected by copyright -deemed not creative enough- nor by patents -not inventive enough).
Well, actually one of the biggest proponents of software patents is the UK -- both in the European Parliament and on national level. Even in the courts, (much) more software patents have been upheld in the UK than in the rest of Europe.
So in total, the current Council's effect is completely undemocratic: the Parliament is being overruled by a bunch of patent administrators who don't like being criticised.
If they confirm it, only then we are in the situation you describe.
Further, please also read this page, especially the first section (A problem of law, not of patent examination). The "newness" and "inventive step" conditions of patent law are completely unfit for software. Using patents for software is like using jet engines to power your bike. Sure, you may be able to get it to work, but it's a very dangerous and destructive combination, and there are much better solutions available.
You're an IP lawyer. So what?
He's quite correct actually.
Too bad that technical in patent jargon is completely undefined and as such means squat, and has over time been interpreted by the EPO to mean basically anything useful. Read this page by a colleague of yours, especially under the heading "When is something technical", and especially the second and third point.
Making a program run faster or use less memory is interpreted as a technical effect on the way the computer works. And using a computer to run a business method is technical, because it "provides surprising speed or economy of scale benefits"
A few national courts actually, and mostly the UK one at that
The Commission and the Council promote the approach of the European Patent Office, which has no safeguards against software patents or business software patents. The Council's version was even written by people from the patent offices. You know, the people that have to abide by it, but haven't until now, got a chance to rewrite their own laws. Quite convenient, no?
Software patents are only enforceable in a few countries currently, and not even that always pans out evenly. The originally proposed directive by the Commission and the new proposal by the Council want to put it black on white that software patents are enforceable. That is a change. It's less of a change to render them unenforceable, as until now they've been unenforceable in more cases than where they were when considering the EU as a whole.
Please stop your lies. The Irish proposal literally states:
When removing the double negation, we get:
It's also different because a programmer does tens of small "inventive steps" per day while programming. Just like the author of a story makes tens of small "inventive steps" per day. It's the normal course of action when doing something based on creativity and will be realised much more as opposed to when you are constrained by physical properties of materials. You're hampered more or less only by your imagination and creativity, instead of by the physical world.
This means that the "inventive step" condition of patent law is totally unfit for creations in the realm of abstraction, as it's passed by too many things. The newness and inventive step (or usefulness, in the US) conditions are merely filters to try to make sure that most granted patents are of good quality. They fail to achieve this in the software world, as they were devised for the physical world. It's a completely outdated concept when applied to today's information society, which is much better served by copyright (encourage as many separate works of the same ideas as possible to increase competition, but inhibit plagiarising or plain copying of other people's work).
In the physical world, laws of nature are not patentable. In software patents, there are no equivalents of laws of nature. The patented technique doesn't have to be able to run on a computer of today, it just has to be representable in the mathematical Turing model, that's all. That's one reason why so many software patents seem so basic: there is nothing basic that is unpatentable.
Finally, patent law is an economical law (devised in the 15th century, no less). You introduce patents in a field because you have good indications it will improve innovation and competition in that field. So before introducing it in software, you should first research whether they will have the same effect. Thanks to the US, we have already several economical studies, including the one by the FTC, that indicate they don't have positive effects, and even hamper progress.
One problem is that Irish Presidency is simply pushing its own interests. In Ireland, there's a 0% tax on patent revenues. So the more patents a company has, the more interesting it is for that company to have its official base in Ireland.