Regarding an answer to that blog, see http://wiki.ffii.org/Epstein050310En. At least 11,500,000 European SMEs disagree with you. And the summary of the study at http://www.isi.fhg.de/ti/Soft_summary.pdf by the Fraunhofer Max Planck institutes also literally states "The strategic benefits of patents in international competition is obvious, but concentrated on very few large companies" (point 2 page 12).

Next, most of the opponents of software patents do not think that all software should be spread free of charge (where did you get this idea?). Free software people want to be able to spread their own copyrighted works free of charge, but I haven't seen (m)any which think that everyone else has to do the same thing with their own independent creations.

Obtaining software patents and dedicating them to the public seems to be an awfully wasteful procedure. Patent pooling and negotiating too. The goal of those projects would presumably be to negate the negative effects software patents have for those small companies, but then they are paying to get a situation which not even as safe for them as when there are no software patents.

The high transaction costs of the patent system are now already one of its main weak points. Encouraging companies to add to those costs without having any additional benefits compared to no software patents seems to be unproductive.

In so far "making the distinction" is concerned: there will undoubtedly always be a grey zone somewhere (note that even the US statutes mention patentable subject matter, and that e.g. a "purely economic effect" cannot be patented in the US either).

The question is whether you want this grey zone to be between more or less anything that is computer-implementable and "pure economic effects" (US), between processing "physical" data (such as image data) and processing "non-physical" (such as financial) data (as is currently the case at the European Patent office), or somewhere between psycho-acoustic audio processing and other (plain mathematical/logic) audio processing (FFII's position).

You make several false statements in this opinion piece.. mostly about the idea that we think that just because software patents all protection of software is bad and that patents are the only protection available.

'they tend to argue that the software should be freely distributed.' This has nothing to do with software patents or protection of intellectual property.

I write software that is protected in various ways - Most of my employers software is kept under lock and key (without the use of patents) as it contains or describes our and our clients business information and processes. This software is protected by copyright, trade/business secrets and confidentiality and contracts. The rest of the software that I write (both at work and at home) is available with source UNDER LICENSE so that it is protected by copyright and contract.

Both are protected and remain my intellectual property - nobody can copy or steal the ideas without breaking several laws. They can distribute the source code within the stipulation of the license. This protects my ideas but does not hamper others who may have similar or better (or worse) ideas, only my implementation. This is similar to how the implementation of a patented circuit or device works.

'Rather than protesting and complaining, take the bull by the horns and use technical creativity and business ingenuity to use the system to your advantage.'

I'm sorry - I have a day job, and it takes plenty of my spare time lobbying politicians and opinionaters on the dangers of this law.

You clearly have no idea how large the problem is - you can't just build a bigger pool of patents than IBM, Samsung or Microsoft. There are hundreds of companies employing hundreds of lawyers to ensure the problems and flaws in the market place are never fixed. The same people employ a huge number of lobbiests and PR flunkies to ensure that politicians protect their interests. Just look at how contreversial progress of the CII directive in the EU - lots of backroom deals and visits from MS and IBM.

You can't just nullify all the bad and trivial patents, there are too many and for every patent removed, several fill its place. Its like plugging the holes in a dyke with matchsticks.

You have to fix the system to ensure that ideas have the APPROPRIATE protection, and patents are not appropriate for mathematics, algorithms, software and business processes. Patents are for inventions.

Copyright and business secrets protect software and algorithms. They have done successfully for 50 years and will continue to do so. GPL (free software) has taken large companies to court and won or got settlements based on copyright and licenses.

Aaron:

"'Rather than protesting and complaining, take the bull by the horns and use technical creativity and business ingenuity to use the system to your advantage.'

I'm sorry - I have a day job, and it takes plenty of my spare time lobbying politicians and opinionaters on the dangers of this law."

You miss my point. My point is that your efforts in lobbying politicians on the alleged and ultimately unfounded 'dangers' would be better spent actually utilizing the patent system to your advantage. When all is said and done, the anti-software arguments are untenable, both on legal and technical grounds.

With the patent system, you can get your ideas into the public domain so that others can build on them, but -here's the key- you can also _prevent_ other people from patenting the idea in the future. With all the talk about the big bad corporations, why doesn't the anti-software people seek to beat the big guys at their game?

You are close on your analysis of protection. Mathematics are not protectable, but when the math is put to a useful function (utility), whether in the form of a mechanical machine or one that is electronically configurable (such as a general purpose microprocessor), that math becomes an invention that is protectable by patent.

You are right that business secrets protect software and algorithms. In almost all cases, I counsel my clients not to patent their software because it is better kept as a business or trade secret. However, copyright protection is much, much weaker than you probably think.

How can a small business have even ONE patent? Estimates of lifetime cost for a european patent vary from 23,000 euro to over 100,000 euro. And free software enthusiasts (as opposed to large corporations using free software) certainly cannot afford ONE european patent. Even if they may be able to afford a few UK patents.

Ideas in the public domain theoretically constitute "prior art"; if I have implemented a given algorithm before a patent is filed on it by somebody else, then _in theory_, that constitutes prior art and can be used to void the patent. In practice it costs at least 300,000 euros to dispute a patent like this. A sum which no small business can afford. In the US it's even worse.

An SME cannot sue or threaten Microsoft or IBM for infringing its single patent, because if it does, and if it actually produces any products, it is nearly certain that the SME infringes on one of the patents of the large corporation. It would probably be ignored because of administrative overhead if it hadn't demanded royalties for its patents, but if it does...

And as far as open source goes, no, we can't beat them at their own game. Because of the cost. Because of the trivial patents constantly granted - http://webshop.ffii.org/ has some examples, and there are more at http://wiki.ffii.org/?UkWebPatentsEn - and because we don't particularly _want_ to exclude proprietary software from implementing ideas we happen also to have implemented. We want to compete with it on a level playing field. We cannot produce 10,000 patents overnight and we certainly cannot pay for them. It may be that if we lose the political game, we will have to resort to such measures - but it will only work if IBM and others decide to do it.

Matthew:

The cost of the patent varies dramatically with its purpose. In the US, for $500 an independent inventor can file his own patent that will be published in 18 months and serve as prior art for any future patents. If you want a professional to write the patent for you, it may cost $5,000 to $15,000.

One of the biggest problems with the patent system right now is that there is so much prior art that is not searched. This is because the various patent offices do not have the ability to search it. Thus, the best way to make sure they search your invention is to put it in their hands.

I am aware of the yearly maintenance fees that are due in Europe for patents. If the patent is not worth the continuing cost, don't pay it. Certainly if the technology has been eclipsed by something new, the original publication prevents others from patenting the technology.

IBM used to have some kind of public disclosure system that they used specifically for defensive publications. FFII and company could be putting together a searchable database that would prevent bad patents from being issued, if they were not so busy trying to recruit protesters.

Can you beat them at their game? I think you could if you pooled your creative resources and formed a patent pool that could be licensed or cross licensed with the alleged big bad corporations. Then, you have a forum for reasonable discourse and economic and political clout. The costs are not enormous, especially when you learn how to write your own patents or if the money spent hiring protesters was used for patent filing fees. It certainly could be done, but it would require a change of focus which may be seen as 'selling out' to the patent system.

The fact that FFII does not go down that path implies that they are not really about software patents, but actually have an anti-capitalism and anti-business long term goal, of which software patents are one small step in that campaign.

When you boil it all down, the argument seems to be one of capitalism vs socialism.

I'd like to see a point by point rebuttal by Russ Krajec of Jonas Maebe's document http://wiki.ffii.org/Epstein050309En

Also, I can't believe that Russ has resorted to emotive attacks to the tune of "all anti-software patent people are socialists." That is simply not the case and I hope that at heart Russ really knows that. It's like when people call free software advocates communists. It's completely ridiculous. That kind of false argument weakens your case and your reasoning.

It is really sad that the only recourse you seem to have left is trying to paint the FFII, PricewaterhouseCoopers, Deutsche Bank, organisations representing over 11,500,000 SMEs, over 390,000 petition signers etc as "communists". I think the next time you should rather paint us as terrorists, that's much more contemporary. And before I forget, of course we kick small puppies as well.

Looking at the puppy-kicking FFII as objectively as possible, the underlying principle is that the 'ideas' embodied in 'software' should be 'free'. This is the standard borne by the anti-software patent proponents.

The patent system for inventors, in a similar way to the copyright system for artists (though sadly, not programmers) allows the inventor to have limited control over his invention, which is often the toil of his hard work or sometimes his good fortune. When you strip that control from the inventor and give it to the masses, whether he be typing his invention on a keyboard or machining it from metal, it looks more like socialism than a free market economy. In a free market, you keep what you earn, but in socialism, you must give away your earnings for the social good. The patent system is a balance between the two: encouraging the patentee to disclose his invention for the common good, but giving him limited rights in exchange.

Am I not correct, or am I missing something?

BTW, I started working my way through Jonas's Epstein rebuttal in today's posting.

Not that I expect you to understand any better now than the previous times:

1) patents -> monopolies
2) free market != monopolies

1) & 2) -> patents != free market

Please keep your name calling up to date:
Today's ennemies are *terrorists* not communists, your mudslinging is sooo 70's...

I think the formula really is:

patents -> you control what you earned
you control what you earned -> free market

Please be more accurate in your description of patents. They are _NOT_ monopolies, there are a _limited_ right of exclusion. Misconstruing patents as monopolies is inflammatory rhetoric.

It is my goal through this blog to give the small business owner, independent inventor, or manager in a large corporation the tools to effecively use the various IP protections. The strategies may change for various situations, but they all focus on the here-and-now business realities of today's laws. Where the patent or copyright laws are not beneficial, I advocate improving them. However, just because I don't like the hypothetical outcome of an extreme situation, I do not advocate carving out an arbitrary and very, very, very poorly defined section of the patent system, which is what FFII is advocating.

In the end, FFII's proposals make the law murkier, which leads to more lawsuits and more cost to everybody. How about putting some of that creativity to work to make the law clearer?

You probably say 'Well, if we eliminate software from patent coverage, it would be clearer.' Since there is no concrete definition of what is 'software', there will be plenty of litigation, especially when there is so much money involved.

If you take the stance that the pure 'software' is not patentable, is that software on an FPGA 'software' or 'hardware'? Is a program on a CD-ROM unprotectable 'software'?

These are just examples of the murkiness. If you say the CD-ROM is not protectable by patent, but the FPGA is, guess what? Not much software will come via CD-ROM, and much more will come via FPGA or some other means. In the end, you are chasing a shadow that is constantly moving. I think there is a better, more logical, and in the end, less costly way to improve the situation that we both think can be improved.

The only logical extension to the FFII position to eliminate the murkiness is to eliminate patents altogether, or strip the guts out of the patent law so that they are irrelevant.

Oh no. You said "patents -> you control what you earned you control what you earned -> free market"

This is kitchen economics. In fact patents restrict the free market per definition. Therefore market radical economists hate the patent system at large. As patents interfere in the free market you have to justify the introduction of patent law in any field by economic advantages.

A simple two scenario model
a) free market in that specific field for all players
b) patent law as an incentive system installed in that specific field for all players.

Simple test: a better b --> install patent law.

You further said "Since there is no concrete definition of what is 'software', there will be plenty of litigation, especially when there is so much money involved.
If you take the stance that the pure 'software' is not patentable, is that software on an FPGA 'software' or 'hardware'? Is a program on a CD-ROM unprotectable 'software'?"

Of course there always will be a gray layer, so we have to implement strong, well defined natural borders, tests.

And you wrote about your intentions: "It is my goal through this blog to give the small business owner, independent inventor, or manager in a large corporation the tools to effecively use the various IP protections. The strategies may change for various situations, but they all focus on the here-and-now business realities of today's laws."

The discussion is about normative justification of patent law in a specific field, not explaination of positive today's law. Of course you have to get a gun when you are on the streets of Bagdad.

"Please be more accurate in your description of patents. They are _NOT_ monopolies, there are a _limited_ right of exclusion."

For economists it is of no interst that patents give you no positive right to use the invention. The result is a market monopoly and an associated wellfare loss. It is of no importance to this analysis that you can license the use of the rights for a fee.

Andre:

You claim that you want to "implement strong, well defined natural borders, test."

Could you please elaborate? I have been looking for such a test, and having various conversations with FFII members, and _nobody_ has ever offered a test.

The biggest hole in the FFII position is that such a test does not exist, or at least nobody seems willing to explain what it is.

You are right that you have to get a gun when you are on the streets of Bagdad. Why doesn't FFII take up arms and fight using the patent system and their creativity, rather than putting all that energy into protesting and trying to change the playing field.

In the end, there will still be big companies who will find other means for mastering their market share. The patent system is one of the few mechanisms for leveling the playing field for the small players, if the small players know how to use it properly.

Without the patent system, the little guys will be at an even bigger disadvantage.

"Could you please elaborate? I have been looking for such a test, and having various conversations with FFII members, and _nobody_ has ever offered a test."

http://swpat.ffii.org/papers/bgh-dispo76/

Translated into law: the positive and negative definitions from the European Parliament mentioned at http://swpat.ffii.org/papers/europarl0309/cons0401/#tech

I doubt you'll like 'em though, despite the fact that they still leave up the grey zone I mentioned before (e.g. between psycho-acoustic audio processing and other (purely mathematical/logic) audio processing)

"Why doesn't FFII take up arms and fight using the patent system and their creativity, rather than putting all that energy into protesting and trying to change the playing field."

Again: we are not changing the playing field in Europe. The Commission proposes to do that (see also the study mentioned at the end of this reaction). And we are opposing the extension of patentability because we are convinced it will make the playing field less efficient and less open for everyone to compete.

See slide 10 and onwards of http://www.ffii.org/~jmaebe/conf0411/tue/Jim%20Bessen.pdf

You cannot argue numbers with gut feeling or derogatory terms.

"In the end, there will still be big companies who will find other means for mastering their market share. The patent system is one of the few mechanisms for leveling the playing field for the small players, if the small players know how to use it properly."

Can you please finally provide some evidence which shows this? Of course there is anecdotical evidence, but that goes for both sides of the argument. Macro-economically, all evidence I have seen points to patents being generally more advantageous to large than to small companies.

You can claim that this is because the large majority of small companies don't know how to use it properly, but this fallacy is debunked in every study which looks into it. Another one was published last Thursday, see http://wiki.ffii.org/Rossi050310En

As an aside, I have often felt that FFII is somehow distorting facts in their website. As an example, your first link to the German court opinion says "FULL TEXT", when in fact it contains several edited portions (note the '...' in several places). It would be more intellectually honest to state "Excerpted Text". But moving on...

If I understand the language of BGH 1976-06-22: Dispositionsprogramm, the essence of a 'technical invention' is one which requires using 'controllable forces of nature', as opposed to 'human mental activity' that does not 'control the forces of nature.'

'Controlling the forces of nature' must be a key component of the 'technical' invention, not just one method of implementing.

I hope this is a fair statement of the test.

In the US, we have a similar test for patentable subject matter.

In order to be patentable under the test above, you need to show that the algorithm performed on a computer is different from doing the algorithm in one's mind alone. Thus, the computer implentation of the idea has a different benefit and utility than merely doing the task manually.

Is this a fair statement so far?

The problem I see is that almost all of the utility of software is the fact that it is executed on a computer much faster than can be done by the human mind, especially when there are repetitive steps. In other words, I can make the argument that almost any piece of software gets its utility through the computer and not through 'human mental activity.'

I think the position of FFII is to exclude any claimed process that may be performed by a human mind. However, at least under US law (I don't know about European law), if you show special utility for the algorithm if it is executed on a computer, it becomes patentable subject matter.

So how do I get a patent on software? I merely add some computer hardware to the claims. This expressly eliminates the human mental version of the algorithm from patent coverage and only claims the computer version. This eliminates the pure algorithm argument.

I don't see why it is worth all the effort on the part of FFII to cause a few words to be added to software patent claims. In the end, the patents will still be granted and you will have expended all that effort for nothing. The test you propose will have very little, if any, effect on the harms you want to alleviate.

Regarding the small business owner's use of the patent system, the point is that the patent system is one very effective weapon in the business's arsenal, which includes marketing, sales, management, etc. In the US, patents are weilded very effectively by independent inventors and small businesses to bring products to the market and get some return on their investment.

For $5-10K, an inventor can get a patent in the US. I understand that the costs are orders of magnitude higher to get pan-European coverage. This alone is much more of a burden on an independent European inventor than a US one.

One of the problems at the US PTO is that the fees are so high that it generates a surplus which our Congress has been more than willing to spend on other programs. If the same is true in Europe, why not try to get the fees lowered over there?

One problem with FFII is that they have painted themselves in a corner. By advocating the elimination of 'software' patents completely, they seem to have shut the door to being able to improve the patent system, both by lowering the costs and improving the quality. Why don't we all work together in improving what we have, rather than attempting to tear down what already exists?

"If I understand the language of BGH 1976-06-22: Dispositionsprogramm, the essence of a 'technical invention' is one which requires using 'controllable forces of nature', as opposed to 'human mental activity' that does not 'control the forces of nature.'"

No, the mere usage of controllable forces of nature is not enough, because in that case, as you remark correctly further on, all you have to do is spice up your patent application with some technical-sounding terms. That is in fact quite close as to how the European Patent Office operates today (and whose practice the Commission wants to "harmonise" throughout the courts of Europe).

The BGH's judgementl based on the so-called "core theory". This amounts to the fact that you do not consider the invention to be "the claims as a whole", but that you only consider that part of each independent claim which is actually being added to the prior art (i.e., the part which is new, non-obvious and industrially applicable).

If that part is patentable subject matter (i.e., it isn't a mental rule, mathematics, ..., but instead a teaching about a the use of controllable forces of nature to achieve predictable results), then a patentable invention is present the claim is admissible.

"I think the position of FFII is to exclude any claimed process that may be performed by a human mind. However, at least under US law (I don't know about European law), if you show special utility for the algorithm if it is executed on a computer, it becomes patentable subject matter."

The European Patent Office currently uses a similar doctrine, but it's not codified in the law. That's why they are pressing so hard for the software patents directive to get through. You can read how to patent computer-implemented business methods at the EPO here (a website by a Dutch patent attorney who works for Philips): http://www.iusmentis.com/patents/businessmethods/epc/

You can find many other interesting texts on his site, including how to interpret the current laws so that they do allow software patents (making the exclusions in the law null and void, as explained in the study at the end of my previous post).

Not all courts follow that reasoning yet though (and in many member states of the EU, no software patent trial has even ever been held), so they now want to "harmonise" this reasoning throughout the whole EU.

"In the US, patents are weilded very effectively by independent inventors and small businesses to bring products to the market and get some return on their investments"

Again, please show some numbers. From the earlier mentioned study, footnote 114 says: "See the empirical study of the Canadian and U.S. software industry by Chabchoub and Niosi [2004], finding that firm size is strongly related to patent propensity: very few SMEs request and obtain patents and 90 per cent of U.S. software patents is currently held by a handful of firms."

"For $5-10K, an inventor can get a patent in the US. I understand that the costs are orders of magnitude higher to get pan-European coverage. This alone is much more of a burden on an independent European inventor than a US one."

Obtaining a patent is not enough, you also have to be able to enforce it (and to defend yourself against patents of others). Court cases cost an order of a magnitude more than obtaining them (in the US at least, in Europe that's a little "cheaper": between 50,000 and 500,000 euro, as opposed to between $500,000 and $2,000,000 in the US).

"One of the problems at the US PTO is that the fees are so high that it generates a surplus which our Congress has been more than willing to spend on other programs. If the same is true in Europe, why not try to get the fees lowered over there?"

Parts of the EPO fees are indeed also diverted to member states and used for other purposes. But it's not like the EPO is short on cash. Have a look at the first paragraph of this document: http://www.sipf.se/epi/lugano51/Ag%2013-7.pdf

Besides, lowering those fees would not mean that much, those fees only account for about 4O% of the total cost of a patent, and then there's still the transaction costs when using the patent (or when patents are used against you).

"One problem with FFII is that they have painted themselves in a corner. By advocating the elimination of 'software' patents completely, they seem to have shut the door to being able to improve the patent system, both by lowering the costs and improving the quality."

We are simply convinced that the paradigm of the patent system is inherently unsuited for monopolising assets in the information economy, and are supported in this belief by dozens of economic studies and opinions of many prominent players in the economy (like Deutsche Bank).

I think it's the proponents of software patents who painted themselves in a corner by taking as starting point "allowing more kinds of patents is always good". Why limit yourself to looking at nothing but patents, instead of looking at the economy and innovation at large?

To quote David Martin of m-cam.com: they're losing the argument around the tree while letting the forrest burn.

"Why don't we all work together in improving what we have, rather than attempting to tear down what already exists?"

We don't have any problem at all with improvements to the patent system. However, we do not represent people with a vested interest in the patent system. The people advocating strengthening the patent system are implicitly advocating to make e.g. copyright and time to market less valuable as business tools.

It doesn't help to have the copyright on your program to sell it, if someone else can stop you with a patent. It also doesn't help to be the first on the market with a product if someone else can then stop you with a patent until they've caught up (or make you pay to help them catch up).

Patents on average only protect 15% of all investments (more in certain sectors, less in others -- such as software and semi-conductors). If you strengthen/extend patents, that percentage obviously goes up, but at the same time the value of other means goes down.

It's not a free choice you have. If there are software patents, you will be affected by them. Either because you have to get a defensive portfolio, or because you have to pay for patent licenses, or because you have to start filing oppositions at the patent office (which almost no one does in Europe, for the record) etc.

It's not just a business tool you have in your arsenal, but which everyone else has as well. And if you are small, the chance that the overall effect on you will be negative is a whole lot bigger than the chance that you will profit from it.

If I summarize FFII's position on inventions, the claimed invention is _not_ considered as a whole for the purposes of being a 'technical' invention, but only the improvements are considered. Thus, if I have the original computer and software invention, it would be patentable, but any improvement on the computer and software invention would not be patentable if the substantial changes were only on the software side.

Is this a correct summary?

This legal theory is a clever way to eliminate any software patents that run on pre-existing computers, but it fails the basic premise of the US definition of patentable subject matter.

In very broad strokes, the US test is that after the invention is operated, has the physical world changed? If the physical world has changed, then it is patentable subject matter. If the physical world is unchanged, then no. A purely mathematical formula does not change the physical world, but that formula executed by a computer may.

In the case of software, when a program runs, there is generally some change of physical state within the bowels of the computer. Thus, software is generally patentable.

Your legal theory breaks down immediately when a only a software element is claimed from an otherwise acceptable independent claim. Do you suggest that a software-only dependent claim would also be unpatentable?

On the subject of costs: if you have a valid case in the US, you can always find an attorney who will take the case on contingency and not require huge up front fees. I don't know if this is true for the various countries in Europe.

One of the fundamental, overarching problems with FFII and the anti-software patents movement is that they suggest copyright as if it is sufficient protection.

Copyright is intended for artists and literary writers for the protection of aesthetic and artistic works, whereas software is neither. Software is a utilitarian, (I might say 'technical'), engineering product designed to solve a specific problem. The programmer is arguably less of an artist than a mechanical engineer.

The US courts are starting to look at copyright protection for software with extreme scrutiny, and finding no copyright protection is allowed for the meat-and-potatoes portion of software, and only in the extraneous portions.

Thus, someone can potentially copy the meat of a program almost in its entirety and dress it up a little to get out from under copyright protection. It may be sufficient to rewrite only the comments in source code. The limits to this are not exactly clear from US case law, but it is heading in this direction.

I would think that programmers would value their ingenuity, expertise, and competence enough to see the difference between patents and copyright and the importance of the underlying idea. If my experience as a product design engineer is any measure, the problem with engineers/programmers is that they have never been educated on what is protectable, how it is protected, and why it is important to protect it. I think that if they were educated on both sides of the argument, the number of FFII members would dwindle.

Please do not bolster your point by suggesting that one large company or another is supporting FFII. Remember that IBM used to be against the concept of software patents when it was convenient for them, back in the 1960's, because they saw themselves as a hardware vendor.

Many supporters of one cause or another, and probably supporters on both sides of this argument, have a very short sighted view of what should be done and why. I would rather find a tenable long term solution instead of undermining the entire patent system because someone said that Microsoft will sue Linux developers and I happen to like Linux right now, which seems to be the big fear at the moment and drives people to FFII.

FFII could be amassing a deadly amount of patents to fight Microsoft if it wished, but sadly it is going for another route.

Personally, I am inventor and entrepreneur. I happen to write patents for other people to put food on the table for the time being, but I always have several business ideas cooking. My personal long term interests are not to write more patents, but to get some of my ideas into the marketplace. The patent system is but one mechanism for doing so, one that works very well in the US.

Mr Krajec, one of us does not understand what patents are about. It is my understanding that patents can by granted for inventions as defined by subject matter of patent laws. I also understand that patents are *not* about problems or solutions, but only about inventions solving some problems.

So I have absolutely no trouble in considering that patents can by granted for some hardware implementation while software implementation solving the same problem would be denied.

I really cannot understand the argument "But we can have patents for hardware performing the same function!"...
So what ?
Patents never were about functions, only about inventions performing some functions so the argument is completly irrelevant.

For PLD where you fuse logical gates, the resulting hardware could be patented, for simulator where the virtual circuit is only in some computer memory, you would not get a patent.
I do not see what is wrong with that.
Yes they perform the same thing, no they are not the same thing, so if patent are granted (or not) according to the nature of the so-called "invention" and not according to the nature of performed task, there is absolutly no contradiction.

Russ Krajec says:
"Thus, someone can potentially copy the meat of a program almost in its entirety and dress it up a little to get out from under copyright protection. It may be sufficient to rewrite only the comments in source code. The limits to this are not exactly clear from US case law, but it is heading in this direction."

I don't believe this. Could you please give some references to back those extravagant claims of yours ?

Russ Krajec says: "A purely mathematical formula does not change the physical world, but that formula executed by a computer may.

In the case of software, when a program runs, there is generally some change of physical state within the bowels of the computer. Thus, software is generally patentable."

Are you aware that thinking is a electro-chemical process of your brain ? That mathematical formula is surely changing some physical state when you think about it. Surely it must be patentable then ...

Bernard:

For a good example of the limits of copyright, see Manufacturer’s Technologies, Inc. v. CAMS, Inc 706 F. Supp 984, 994 (D. Conn 1989)

In this case, the meat of the program was held to be unprotected by copyright. The only part held to be copyrightable was a small portion that 'confused the user' and thus was not filtered out by the abstraction-filtration-comparison test.

If an FPGA is 'hardware' and could be patented when the 'software' is loaded, could the same be said for a read-only optical disk, an EEPROM, a magnetic hard disk, or RAM? Where do you draw your line?

"This legal theory is a clever way to eliminate any software patents that run on pre-existing computers, but it fails the basic premise of the US definition of patentable subject matter."

Your summary (which I deleted) is correct, but I don't see a problem with the fact that our definition conflicts with US law. This is Europe we are talking about. Everyone has to work in his own judicial system to obtain the desired policy goals.

"In the case of software, when a program runs, there is generally some change of physical state within the bowels of the computer. Thus, software is generally patentable."

That's extremely similar to what the European Patent Office says. Proponents of software patents in Europe are generally however extremely aggressive in denying that they are basically introducing the US methodology in Europe, because it is associated with things like business method patents, which everyone is (or at least claims to be) strongly against over here.

That's why we have this whole mess about "computer-implemented inventions" in Europe, simply because people do not want to admit they want to introduce software patents. This makes it both easier (to show contradictions) and harder (little or no discussion on substance possible) for us.

"Your legal theory breaks down immediately when a only a software element is claimed from an otherwise acceptable independent claim. Do you suggest that a software-only dependent claim would also be unpatentable?"

If you are talking about computer program product claims (e.g. "a computer program, characterised in that when loaded on a computer, a process according to any preceding claim is realised"), then the answer is simply that we do not allow those. I don't see how it breaks down the theory.

Supplying a computer program in order to realise an otherwise perfectly technical invention which has been patented, and clearly for that purpose, can still be attacked on the basis of contributory infringement. We do realise there are limitations to that approach, but we think those should be addressed at that level, and not by making the distribution of software a direct infringement.

You can find more on that, based on a real world patent application, at http://wiki.ffii.org/DilabProg04En

"On the subject of costs: if you have a valid case in the US, you can always find an attorney who will take the case on contingency and not require huge up front fees. I don't know if this is true for the various countries in Europe."

That doesn't help much if you have to defend yourself...

"Copyright is intended for artists and literary writers for the protection of aesthetic and artistic works, whereas software is neither. Software is a utilitarian, (I might say 'technical'), engineering product designed to solve a specific problem. The programmer is arguably less of an artist than a mechanical engineer."

All this discussion about "idea" vs "expression" and "utilitarian" vs "artistic" matter little if anything from a business perspective. We do no argue for copyright because we consider programmers to artists creating useless expressive works. We argue for it because It Just Works(TM) for the people and businesses involved.

"Thus, someone can potentially copy the meat of a program almost in its entirety and dress it up a little to get out from under copyright protection. It may be sufficient to rewrite only the comments in source code. The limits to this are not exactly clear from US case law, but it is heading in this direction."

That sounds like a quite worrying evolution.

"I would think that programmers would value their ingenuity, expertise, and competence enough to see the difference between patents and copyright and the importance of the underlying idea."

They certainly do. But in general, they value the ability to freely sell there eventual creations (and to freely use all ideas out there) a lot more than the possibility to monopolise one or other underlying innovation of their own (no matter how inventive you are, you're always going to use a lot more existing stuff from other people).

"If my experience as a product design engineer is any measure, the problem with engineers/programmers is that they have never been educated on what is protectable, how it is protected, and why it is important to protect it. I think that if they were educated on both sides of the argument, the number of FFII members would dwindle."

In my experience, it's the other way round. You're right there are many people who don't know the difference between patents and copyright. That's why you often see remarks that we supposedly want everyone to give their software away for free. Once you explain it to them, they much more often than not agree with us.

"Please do not bolster your point by suggesting that one large company or another is supporting FFII. Remember that IBM used to be against the concept of software patents when it was convenient for them, back in the 1960's, because they saw themselves as a hardware vendor."

I don't think Deutsche Bank aspires to ever become a software vendor. In fact, they have said that they currently have a software development department (to develop internal software for ATMs and things like that), but don't have a patent department.

If software patents come through, they feel they will have to create a patent department, just for that reason (even they don't feel like doing that at all). From this the so-called Deutsche Bank test evolved: if a directive would force Deutsche Bank to establish a patent departement, it's a bad directive.

"Many supporters of one cause or another, and probably supporters on both sides of this argument, have a very short sighted view of what should be done and why."

Everything the FFII does is supported by a large number of macro-economical studies. I really don't see how you can describe that as "short sighted". What else are we supposed to do? Base our position on gut feeling or on how one system or another might work in a perfect world which may or may not be realistic to achieve?

"I would rather find a tenable long term solution instead of undermining the entire patent system because someone said that Microsoft will sue Linux developers and I happen to like Linux right now, which seems to be the big fear at the moment and drives people to FFII."

Well, the guy who insinuated that Microsoft might sue Linux users and developers was Steve Balmer, so you'll have to complain to Microsoft for that. But that's not exactly our driving force (though I grant you it does push free and open source software supporters more to our side)

"FFII could be amassing a deadly amount of patents to fight Microsoft if it wished, but sadly it is going for another route."

We don't really care about Microsoft in particular, nor are we open source or Free Software advocates. Our goal is not to destroy Microsoft. Our goal is simply to have the healthiest possible environment for innovation and economy, enabling fair and free access to the market in order to stimulate competition.

We do not believe software patents contribute to this (i.e., while they can stimulate competition, they can stifle it as well, and we think they stifle it more than that they help).

"Personally, I am inventor and entrepreneur. I happen to write patents for other people to put food on the table for the time being, but I always have several business ideas cooking. My personal long term interests are not to write more patents, but to get some of my ideas into the marketplace. The patent system is but one mechanism for doing so, one that works very well in the US."

It's true that patents provide for a way to get technology transfers going. This whole argument with pro's and con's is nicely elaborated on at pages 35 and following of the paper at http://www.epip.ruc.dk/Papers/ROSSI_Paper.pdf, under the title "Patents, copyright and markets for technology".

Jonas:

One of my biggest worries about the anti-software movement is that the unnoticed erosion of copyright protection, coupled with an erosion of patent rights, may lead to virtual anarchy in the software field. It will be like the bazaars that I have visited in China and Eastern Europe where any piece of software is available for $1.00 on pirated CD's.

At that point, programmers will throw up their arms in disgust because they cannot make a living and move on to other fields. When the market degrades to that level, innovation quits and cannot be revived.

Both the patent and copyright systems are artificial government stimuli of the marketplace, and they work when there is an understandable and stable balance. Tipping the balance one way or another is always going on, but tremendously upsetting the balance can have disasterous consequences. The FFII position may have such an effect, especially when copyright protection is not the panacea alternative to patents that it is often touted to be.

"One of my biggest worries about the anti-software movement is that the unnoticed erosion of copyright protection, coupled with an erosion of patent rights, may lead to virtual anarchy in the software field."

It surprises me to hear you talk as if we should accept this weakening of copyright as if it is something inevitable. I would suggest we fight against that. I don't see the use of letting the legal system decide how software writers can and cannot defend their works, against all economic evidence showing which is more used and effective.

Law is not the ultimate truth, it's just a bunch of rules we make up to achieve the goals we want.

"It will be like the bazaars that I have visited in China and Eastern Europe where any piece of software is available for $1.00 on pirated CD's."

I doubt literal bit copies will ever be exempted from copyright (unless Greg Aharonian would win his silly lawsuit... You should be riling against him for trying to undermine the software companies' IP).

"At that point, programmers will throw up their arms in disgust because they cannot make a living and move on to other fields. When the market degrades to that level, innovation quits and cannot be revived."

Please realise that overdoing it completely destroys your credibility :)

"Both the patent and copyright systems are artificial government stimuli of the marketplace, and they work when there is an understandable and stable balance. Tipping the balance one way or another is always going on, but tremendously upsetting the balance can have disasterous consequences. The FFII position may have such an effect, especially when copyright protection is not the panacea alternative to patents that it is often touted to be."

Claiming that the FFII is undermining some balance is silly, since we're trying to keep the status quo.

Jonas:

I don't know that you can take the high ground in your argument by saying that you want to preserve the status quo.

By FFII's own admission, there are tens of thousands of software patents already issued in Europe and many more in the US. Seems to me that the status quo is to allow software patents. Generally people do not, at least in this country, protest and march in the streets to keep the status quo. Please be more intellectually honest.

You are looking to change the current thinking on 'software' patents in a wholesale fashion. This is far from the 'status quo'. And you do this after suggesting that I don't have credibility.

It is this type of half truth political rhetoric that bothers me. Please be honest when you make your arguments. Save that kind of talk for recruiting protestors and work on legal, technical, and factual arguments for here.

Greg Aharonian's lawsuit seeks to clarify whether software should be protected by copyrights or patents. It is something that should be resolved, one way or another. Since copyright does not serve software very well at all, the patent regime is much better suited.

I am afraid that FFII will be standing on the dock, having cut the patent ship free, only to look to the copyright ship which sailed long ago. In the end, you will have ensured that everybody's software is freely and easily copied, including that of your constituents. And who will win when the software industry dries up?

"By FFII's own admission, there are tens of thousands of software patents already issued in Europe and many more in the US. Seems to me that the status quo is to allow software patents."

In the US: yes. In Europe: no. The software patents are here, because the European Patent Office is an organisation which is completely independent of any law. It can do as it likes. And it likes to hand out software patents.

These software patents have to be enforced in national courts, however. These national courts are not ruled by the EPO (although they of course can take its case law into account), and still to this day reject (although sometimes also accept) software patents from being enforceable.

The European Patent Convention states that "computer programs as such are not inventions". The Commission and Council proposals would effectively delete that exclusion (http://wiki.ffii.org/Rossi050310En). We want to reinforce it.

"You are looking to change the current thinking on 'software' patents in a wholesale fashion. This is far from the 'status quo'."

You're talking about a different status quo than I am. I'm talking about the European juridical situation. You are talking about the US situation, large companies who would like to see the US practice come to Europe and the practice of the European Patent Office.

"Greg Aharonian's lawsuit seeks to clarify whether software should be protected by copyrights or patents. It is something that should be resolved, one way or another. Since copyright does not serve software very well at all, the patent regime is much better suited."

Then how come everyone in the software industry (even those also in favour of software patents) is in favour of copyright? I'd love to see you argue against Microsoft that copyright for software is worthless and should be abolished.

"I am afraid that FFII will be standing on the dock, having cut the patent ship free, only to look to the copyright ship which sailed long ago. In the end, you will have ensured that everybody's software is freely and easily copied, including that of your constituents. And who will win when the software industry dries up?"

I really wish you would finally learn something about this industry. Have you ever written a non-trivial computer program? Have you ever worked in a software company? Have you ever performed algorithmic research?

It would also be nice if you'd put just as much energy in promoting copyright so it can keep its rightful place, as opposed to deriding it as much as possible in order to promote patents as the Holy Grail.

It may also be useful to know that the number of companies which makes money by producing software is dwarfed by the number of companies which develop software for internal use. I.e., for many more companies software is a cost center than it is a profit center. That's one of the reasons that open source/free software are gaining popularity in the business world as well.

And I really think it is strange you do not care at all about the macro-economic implications of software patents. You cannot ignore the economy by focussing on case law. In real life, you can't ignore what you don't like or what doesn't fit your views (just like we can't ignore the EPO's practice and are trying to deal with it).

A small clarification regarding my previous post: the European Patent Office is bound by the European Patent Convention, but in practice it interpretes that treaty at will. It does not fall under the jurisdiction of the European Union, and only partially and indirectly under the administration of the subscribing countries.