The paper says a few more good things than that:) You cite extensively but only focus on one point in summarizing anyhow. Merges emphasizes how market forces provide mechanisms such as SSOs as balancing factors to increased patenting, thereby obviating the need for silly things like eliminating certain classes of patents entirely. Anyone who has read Merges knows him as a fairly free market thinker who has increasingly warmed to software patents over the years after initial skepticism. Glad to see you writing about him.
Tim Lee
· 3 years ago
Noel,
I don't understand what's "silly" about excluding categories of inventions from patentability. As I understand it, there are a variety of things, such as boat hull designs and fashion, that are currently not eligible for patents. Are those restrictions "silly?" If not, why is it any sillier to say that software and business methods should be on the non-patentable side of the line?
Noel Le
· 3 years ago
By "eliminating certain classes of patents entirely" I meant excluding subject matters CURRENTLY patentable.
Noel Le
· 3 years ago
By "eliminating certain classes of patents entirely" I meant excluding subject matters CURRENTLY patentable.
MikeT
· 3 years ago
Noel, so business processes should be patenable as well? I'm not talking about manufacturing techniques, but simply ways of organizing a business and offering products to a customer. That strikes me as an old English crown trade monopoly, not a legitimate market incentive.
Steve R.
· 3 years ago
I find Robert Merges thesis to be "stretched" and of dubious applicability. One can view this as an apple vs. oranges type of comparison. If I have this correct, in the past patents were issued for tangible products/devices. Today, the concept of the patent is being aggrandized to include abstract concepts such as methods and processes. Allowing patents for abstract concepts makes it difficult to discern a property boundary. Hence we have a plethora of patent infringement lawsuits.
Additionally, we are seeing that patents are increasing being used as a toll both for exacting revenue. The owner of a piece or property has a right to use that property to make money; however this must be accomplished within the rules. In this case patent law. It appears, from reading many articles, that companies are acquiring portfolios of patents for the purpose of acquiring property. The purpose of patent law is to allow a patent owner a limited (in terms of time) financial return for inventing a product that fosters scientific progress. The accumulation of patent portfolios to simply operate a toll both runs counter to this concept therefore, I would advocate, that no property right has been established since the toll both is not akin to a research laboratory.
Tim Lee
· 3 years ago
So would it have been "silly" if I had argued, back in the 1980s, that software shouldn't be patentable? And once something is made patentable, does it instantly become "silly" to argue that that decision was a mistake?
Obviously, it's important that patent rules not be changed abruptly or without good reason. But if the patentability of software is detrimental to technological progress, as many programmers believe it is, then it can't possibly be "silly" to suggest changing the rules.
There's nothing sacred about the patent rules that happen to prevail at the moment. The rules have been changed repeatedly in the last decade to expand the scope of patentability. It's hardly radical to suggest that the process has gone too far.
Noel Le
· 3 years ago
I guess my biggest argument about eliminating software patents is how supporters of this move sound like the demise of the software industry will occur tomorrow. The software industry is doing pretty well, and if changes are needed, there should be clear evidence.
Critiquing one patent a week won't convince me to eliminate software patents. Half the time I agree with you that some patents are bad But bollions of dolars in licensing exchange hands every year in the software industry. How do you account for that?
How do you propose eliminating software patents by the way? You sound like you support abrupt changes, as you don't really discuss gradual means such as changing various elements of patenting standards or administration.
MikeT, you should look at this testimony by Professor Dreyfuss from NYU where (I vaguely remember that) she argues against current interpretations of State Street, saying that it should have allowed software but not business method patents. I'd send to you if I could find it.
Tim Lee
· 3 years ago
Critiquing one patent a week won't convince me to eliminate software patents. Half the time I agree with you that some patents are bad
I would be very curious to know which of the patents I've reviewed you think are not bad.
But bollions of dolars in licensing exchange hands every year in the software industry. How do you account for that?
Taxi medallions in New York City fetch hundreds of thousands of dollars at auction. Yet I don't know of any free-market economist who thinks the New York taxi monopoly is good public policy. The fact that a government-granted monopoly is valuable doesn't prove that it's economically worthwhile.
Tim Lee
· 3 years ago
Also, you didn't answer my question: what exactly is "silly" about suggesting that software shouldn't be patentable? Was it silly to take that position in the 1980s?
Noel Le
· 3 years ago
I'm not going to read over your interpretation of patent claims, unless you cite more in your summaries. Remember, you're simplifying a patent claim to prove its obviousness. Your approach itself may be accomplishing your overall goal. If you cite more, I'll have time to review. I'm not dismissing your approach, just that I'd be careful about what policy considerations to draw from it.
I'm not aware of the NY Taxi thing, but if you're calling patents a "monopoly" analogous to organizational or economic monopolies, we should talk further:)
By silly, I meant *eliminating* software patents altogether. The whole issue is kind of "for the history books" as Lemley would say. I view it as a rigid and inflexible way of affecting the software industry, with many unknown implications. Patents have many benefits, only one of which is their doctrinal justification.
Remember, our disagreement with software patents is that you don't think they should be in the system to begin with (my view is that if they meet high standards of patentability, then keep them). How then would you identify a "software patent." I haven't seen a uniform definition- thats why I think simply eliminating them would be difficult. Look at the Professor Mann article I write about here where he discussed 4 or 5 ways of defining software patents: http://weblog.ipcentral.info/archives/2006/06/p....
Steve R.
· 3 years ago
I find Robert Merges thesis to be "stretched" and of dubious applicability. One can view this as an apple vs. oranges type of comparison. If I have this correct, in the past patents were issued for tangible products/devices. Today, the concept of the patent is being aggrandized to include abstract concepts such as methods and processes. Allowing patents for abstract concepts makes it difficult to discern a property boundary. Hence we have a plethora of patent infringement lawsuits.
Additionally, we are seeing that patents are increasing being used as a toll both for exacting revenue. The owner of a piece or property has a right to use that property to make money; however this must be accomplished within the rules. In this case patent law. It appears, from reading many articles, that companies are acquiring portfolios of patents for the purpose of acquiring property. The purpose of patent law is to allow a patent owner a limited (in terms of time) financial return for inventing a product that fosters scientific progress. The accumulation of patent portfolios to simply operate a toll both runs counter to this concept therefore, I would advocate, that no property right has been established since the toll both is not akin to a research laboratory.
Tim Lee
· 3 years ago
I have no idea what you mean by "cite more in your summaries." The patents are often quite short, and I provide a handy link to them. If you don't have time to read a 10-page patent, then why are you spending so much time expressing opinions about them?
By silly, I meant *eliminating* software patents altogether. The whole issue is kind of "for the history books" as Lemley would say. I view it as a rigid and inflexible way of affecting the software industry, with many unknown implications.
That's ridiculous. We know exactly how the software industry would work without software patents: the same way it worked for the first 30 years of its existence. Whether it's politically feasible to eliminate them, and how best to manage the transition, are of course difficult questions. It may be that for practical reasons eliminating them is too difficult to be attempted.
But that is an entirely separate question from the empirical one that interests me: do software patents promote the progress of science and the useful arts? I contend that they do not, and I have yet to find anyone who makes a coherent and well-informed argument to the contrary.
Taxi medallions are a government-granted monopoly (well, cartel, technically) on operation of taxicabs in a specific geographical area. A patent is a government-granted monopoly over the use of a specific technology. The fact that holding such a monopoly is valuable tells us nothing about whether granting them is good public policy.
Noel Le
· 3 years ago
Steve look at Lemley`s article w Cohen on the judicial history behind software patents. "Patent Scope and Innovation..." Tim if you do this patent of the week thing for ten years you will not even have touched on 1% of current software patents (Lemley estimates at least 100k of them). How innovation worked 30 years ago in the 1970s seems to be your goal? Do you know how innovation works? Forty economist gathered at the FTC 3 years ago and agreed they didnt. But they knew the industry and its policies have evolved since the `70s and they proposed changes that worked w the existing policy landscape ... And the industry is doing fine so why change things when your only point of reference is innovation in the 70s. Public policy should consider the way things are and what has actually transpired, not just how things should be. I haven`t found anyone to make the case against software patents and I`ve looked at 95% of the literature. There is one person very very persuasive against software patents, but Ill leave OSS supporters with their current cast of experts (Moglen, Lessig, Benkler:):) PS your monopoly analogy is OK, but I would apply "monopoly" only to what courts would enforce in the claim, not to the revelant technology or product/consumer market. And what is thing of being worked up about software patents. All of this fear of litigation wars which haven`t surfaced, and if it has its because of patent trolls.
Tim Lee
· 3 years ago
How innovation worked 30 years ago in the 1970s seems to be your goal?
I think you need to read what I write more carefully. You claimed that eliminating software patents would have "many unknown implications." I was just pointing out that we have plenty of experience with a patent-free software industry, so there's little reason to worry about unintended consequences. And software patents were introduced in the 1990s, not the 1970s.
There are lots of people who have made the case against software patents. Here is a succinct case against them that's been around for 15 years. Can you point me to any articles or studies that specifically respond to these arguments? If there are, I have yet to find them.
Similarly, if software patents were a good idea, someone should be able to point me to at least one example of an indisputably meritorious software patent. Yet after months of looking at patents, I have yet to find one. Nor have the software patent advocates I've encountered been able to point one out. It seems to me that suggests there might be a problem.
Noel Le
· 3 years ago
Wow Tim knows how innovation works... Contact IBMs licensees about patents. Eliminating patents now would not simply rewind time BTW. You disagree? Ask IBM, oh wait where do they get money to invest in OSS.
Doug Lay
· 3 years ago
On a slightly tangential note - here's a really good article about IP and how overrated it is in most cases as a driver of business success. Note that it appears in a business magazine.
Noel: Unfortunately, I cannot stay up as late as I used to, but I did peruse the paper you referenced. My summary opinion is that they did not present any persuasive arguement and, I believe, some of the cases they cited in support of their position were actually bad judicial decisions.
1. the number of patents being issued is not an indication of innovation.
2. Cohen and Lemley imply that patents foster innovation yet the examples seem to prove the opposite, large corporations quashing upstarts. For example "Microsoft has asserted patents on a popular video file format to block distribution of an open source version developed by reverse engineering."(Page 21)
3. I particulary disliked their recommendation for limited reverse engineering Without reverse engineering we would not have the competition between Intel and AMD.
4. If I understand the phrase "refuse finding of equivalence" a whole new form of frivilous lawsuits would be allowed. Lets assume that the light switch has not been invented yet. John invents the light switch. Now Jane invents another form of light switch using a different technology. The court however finds that Jane infringed on John's patent since it functionally accomplishes the same equivilant action of turning on a light. I hope I misunderstood this statement.
5. My belief is that patents are no longer actually about fostering innovation rather they are about claiming a piece of turf for purposes of extorting revenue.
Noel Le
· 3 years ago
Steve I referred you to the paper not to side you w software patents (or me:) but to show you one perspective on them. And as someone who argues his side, I didnt mind referring a paper I agree 50% with as you seemed like you would be be interested in its narrative of judicial interpretation based on your comments re Merges. Email me to continue this discussion. I dont like following these long threads.
Noel Le
· 3 years ago
Doug I did a review of that Inc.com article yesterday on IPcentral. Personally the argument against IP could have been better (and has been in other publications) but I felt it hit on too many obvious issues. It amazes me these impractical views. Soliciting a readership is one thing, but dont pass it off as informative to public policy.
Doug Lay
· 3 years ago
Noel:
I started to read your response but didn't finish it. You lost me at the beginning of the third paragraph.
The Inc article is aimed at businesspeople, not policymakers. The current generation of policymakers is virtually hopeless with regard to IP reform. The best we can hope for from these clowns is gridlock.
MikeT
· 3 years ago
Noel, where is your proof that ending software patents would, as you have said, "rewind time?" Software patents do not seem to have any meaningful impact on the advancement of the IT industry. Most of the tricks done in C# and Java, for example, could or were first done in languages like Smalltalk back in the late 1970s and 1980s. Microsoft is no discovering functional programming in the form of its F# language which is based on OCaml and the paradigm goes back to LISP which was first worked on in 1958! (Arguably LISP is the most powerful language ever put into mainstream use by a wide factor)
You worry about things being "rewound back to the 1970s," but I'm here to tell you that there really haven't been many advances in core areas since then. It's not even been that long that ethernet hardware has been a common, needed feature of PCs and that technology is now about 27 years old. Your fear is already happening, even with a robust patent system.
Noel Le
· 3 years ago
Doug, nobody ever said policy or law makers craft law by sitting down, having friendly chats and leaving with everyone happy. But hey, thats democracy.
Steve, Lemley and Cohen propose a bold step, as it calls for an "assertive" right" that would rid any interpretative ambiguity that might deter reverse engineer patented software. Mike, I stated that eliminating software patents "would *not* simply rewind time," as that would be impossible. Of course, if we get rid of software patents, we'll end up with a patent policy similar to that we once had. Yet the unintended consequences I speak of after eliminating software patents doesn't just relate to how industry would innovate without these patents, but to the transition of over 100,000 patents suddenly having no status. The most difficult aspect of Napsterizing software patents, besides IBM losing mass royalties, are whether other forms of IP or contracts can sustain cross-licensing agreements, and whether the USPTO can identify exactly what are software patents.
***We know exactly how the software industry would work without software patents: the same way it worked for the first 30 years of its existence.*** Again, I'm enthralled by anyone who knows how innovation works. The full-fledged private technology sector didn't thrive before the 1970s. A lot of funding came from the US Govt through procurement and federal contracts. I'd take a later time period, perhaps the 1980s, as a milestone on which to reference a point of time where the software industry thrived without patent protection. We can speak more about this later...
***Whether it's politically feasible to eliminate them, and how best to manage the transition, are of course difficult questions. It may be that for practical reasons eliminating them is too difficult to be attempted.*** Tim, this is Lemley's point, and one I believe we agree on but in different terms.
***do software patents promote the progress of science and the useful arts? I contend that they do not, and I have yet to find anyone who makes a coherent and well-informed argument to the contrary.*** Tim, often policies serve functions for which they were not originally intended. Sometimes their value lies elsewhere than why they were conceived and implemented. To propose eliminating software patents because you can't find proof of how well they serve their doctrinal justification, without considering their secondary benefits and without direct proof of negative consequences leaves a lot unconsidered.
If you've not found any justification for software patents, also consider that you aren't providing any arguments against them. I'm more inclined to look at MikeT's arguments, because he speaks from the perspective of a programmer with valid concerns. Since you're on the public policy side, you have to make coherent "policy" arguments.
Steve R.
· 3 years ago
Noel: Please define, tell me, what does "assertive right"??? Usually when I hear this type of phraseology, the actual intent is to diminish an existing right.
Steve R.
· 3 years ago
Pardon me, I was called away so I did not properly proofread.
Noel: Please define, tell me, what does the phrase "assertive right" mean??? Please translate the phrase "refuse finding of equivalence" into something I can understand. Usually when I hear this type of phraseology, the actual undiscolosed intent is to diminish an existing right. I hope you will help me with understanding what Lemley and Cohen are proposing.
All Comments
I don't understand what's "silly" about excluding categories of inventions from patentability. As I understand it, there are a variety of things, such as boat hull designs and fashion, that are currently not eligible for patents. Are those restrictions "silly?" If not, why is it any sillier to say that software and business methods should be on the non-patentable side of the line?
Additionally, we are seeing that patents are increasing being used as a toll both for exacting revenue. The owner of a piece or property has a right to use that property to make money; however this must be accomplished within the rules. In this case patent law. It appears, from reading many articles, that companies are acquiring portfolios of patents for the purpose of acquiring property. The purpose of patent law is to allow a patent owner a limited (in terms of time) financial return for inventing a product that fosters scientific progress. The accumulation of patent portfolios to simply operate a toll both runs counter to this concept therefore, I would advocate, that no property right has been established since the toll both is not akin to a research laboratory.
Obviously, it's important that patent rules not be changed abruptly or without good reason. But if the patentability of software is detrimental to technological progress, as many programmers believe it is, then it can't possibly be "silly" to suggest changing the rules.
There's nothing sacred about the patent rules that happen to prevail at the moment. The rules have been changed repeatedly in the last decade to expand the scope of patentability. It's hardly radical to suggest that the process has gone too far.
Critiquing one patent a week won't convince me to eliminate software patents. Half the time I agree with you that some patents are bad But bollions of dolars in licensing exchange hands every year in the software industry. How do you account for that?
How do you propose eliminating software patents by the way? You sound like you support abrupt changes, as you don't really discuss gradual means such as changing various elements of patenting standards or administration.
MikeT, you should look at this testimony by Professor Dreyfuss from NYU where (I vaguely remember that) she argues against current interpretations of State Street, saying that it should have allowed software but not business method patents. I'd send to you if I could find it.
I would be very curious to know which of the patents I've reviewed you think are not bad.
But bollions of dolars in licensing exchange hands every year in the software industry. How do you account for that?
Taxi medallions in New York City fetch hundreds of thousands of dollars at auction. Yet I don't know of any free-market economist who thinks the New York taxi monopoly is good public policy. The fact that a government-granted monopoly is valuable doesn't prove that it's economically worthwhile.
I'm not aware of the NY Taxi thing, but if you're calling patents a "monopoly" analogous to organizational or economic monopolies, we should talk further:)
By silly, I meant *eliminating* software patents altogether. The whole issue is kind of "for the history books" as Lemley would say. I view it as a rigid and inflexible way of affecting the software industry, with many unknown implications. Patents have many benefits, only one of which is their doctrinal justification.
Remember, our disagreement with software patents is that you don't think they should be in the system to begin with (my view is that if they meet high standards of patentability, then keep them). How then would you identify a "software patent." I haven't seen a uniform definition- thats why I think simply eliminating them would be difficult. Look at the Professor Mann article I write about here where he discussed 4 or 5 ways of defining software patents: http://weblog.ipcentral.info/archives/2006/06/p....
Additionally, we are seeing that patents are increasing being used as a toll both for exacting revenue. The owner of a piece or property has a right to use that property to make money; however this must be accomplished within the rules. In this case patent law. It appears, from reading many articles, that companies are acquiring portfolios of patents for the purpose of acquiring property. The purpose of patent law is to allow a patent owner a limited (in terms of time) financial return for inventing a product that fosters scientific progress. The accumulation of patent portfolios to simply operate a toll both runs counter to this concept therefore, I would advocate, that no property right has been established since the toll both is not akin to a research laboratory.
By silly, I meant *eliminating* software patents altogether. The whole issue is kind of "for the history books" as Lemley would say. I view it as a rigid and inflexible way of affecting the software industry, with many unknown implications.
That's ridiculous. We know exactly how the software industry would work without software patents: the same way it worked for the first 30 years of its existence. Whether it's politically feasible to eliminate them, and how best to manage the transition, are of course difficult questions. It may be that for practical reasons eliminating them is too difficult to be attempted.
But that is an entirely separate question from the empirical one that interests me: do software patents promote the progress of science and the useful arts? I contend that they do not, and I have yet to find anyone who makes a coherent and well-informed argument to the contrary.
Here are details about the taxi medallion system:
http://www.city-journal.org/html/12_2_how_to_fi...
Taxi medallions are a government-granted monopoly (well, cartel, technically) on operation of taxicabs in a specific geographical area. A patent is a government-granted monopoly over the use of a specific technology. The fact that holding such a monopoly is valuable tells us nothing about whether granting them is good public policy.
I think you need to read what I write more carefully. You claimed that eliminating software patents would have "many unknown implications." I was just pointing out that we have plenty of experience with a patent-free software industry, so there's little reason to worry about unintended consequences. And software patents were introduced in the 1990s, not the 1970s.
There are lots of people who have made the case against software patents. Here is a succinct case against them that's been around for 15 years. Can you point me to any articles or studies that specifically respond to these arguments? If there are, I have yet to find them.
Similarly, if software patents were a good idea, someone should be able to point me to at least one example of an indisputably meritorious software patent. Yet after months of looking at patents, I have yet to find one. Nor have the software patent advocates I've encountered been able to point one out. It seems to me that suggests there might be a problem.
http://www.inc.com/magazine/20060801/let-your-g...
Also, here's some info about a much-loathed software patent that I don't believe you've written about in the past:
http://noedupatents.org/
1. the number of patents being issued is not an indication of innovation.
2. Cohen and Lemley imply that patents foster innovation yet the examples seem to prove the opposite, large corporations quashing upstarts. For example "Microsoft has asserted patents on
a popular video file format to block distribution of an open source version developed by reverse engineering."(Page 21)
3. I particulary disliked their recommendation for limited reverse engineering Without reverse engineering we would not have the competition between Intel and AMD.
4. If I understand the phrase "refuse finding of equivalence" a whole new form of frivilous lawsuits would be allowed. Lets assume that the light switch has not been invented yet. John invents the light switch. Now Jane invents another form of light switch using a different technology. The court however finds that Jane infringed on John's patent since it functionally accomplishes the same equivilant action of turning on a light. I hope I misunderstood this statement.
5. My belief is that patents are no longer actually about fostering innovation rather they are about claiming a piece of turf for purposes of extorting revenue.
I started to read your response but didn't finish it. You lost me at the beginning of the third paragraph.
The Inc article is aimed at businesspeople, not policymakers. The current generation of policymakers is virtually hopeless with regard to IP reform. The best we can hope for from these clowns is gridlock.
Noel, where is your proof that ending software patents would, as you have said, "rewind time?" Software patents do not seem to have any meaningful impact on the advancement of the IT industry. Most of the tricks done in C# and Java, for example, could or were first done in languages like Smalltalk back in the late 1970s and 1980s. Microsoft is no discovering functional programming in the form of its F# language which is based on OCaml and the paradigm goes back to LISP which was first worked on in 1958! (Arguably LISP is the most powerful language ever put into mainstream use by a wide factor)
You worry about things being "rewound back to the 1970s," but I'm here to tell you that there really haven't been many advances in core areas since then. It's not even been that long that ethernet hardware has been a common, needed feature of PCs and that technology is now about 27 years old. Your fear is already happening, even with a robust patent system.
Steve, Lemley and Cohen propose a bold step, as it calls for an "assertive" right" that would rid any interpretative ambiguity that might deter reverse engineer patented software.
Mike, I stated that eliminating software patents "would *not* simply rewind time," as that would be impossible. Of course, if we get rid of software patents, we'll end up with a patent policy similar to that we once had. Yet the unintended consequences I speak of after eliminating software patents doesn't just relate to how industry would innovate without these patents, but to the transition of over 100,000 patents suddenly having no status. The most difficult aspect of Napsterizing software patents, besides IBM losing mass royalties, are whether other forms of IP or contracts can sustain cross-licensing agreements, and whether the USPTO can identify exactly what are software patents.
***We know exactly how the software industry would work without software patents: the same way it worked for the first 30 years of its existence.***
Again, I'm enthralled by anyone who knows how innovation works. The full-fledged private technology sector didn't thrive before the 1970s. A lot of funding came from the US Govt through procurement and federal contracts. I'd take a later time period, perhaps the 1980s, as a milestone on which to reference a point of time where the software industry thrived without patent protection. We can speak more about this later...
***Whether it's politically feasible to eliminate them, and how best to manage the transition, are of course difficult questions. It may be that for practical reasons eliminating them is too difficult to be attempted.***
Tim, this is Lemley's point, and one I believe we agree on but in different terms.
***do software patents promote the progress of science and the useful arts? I contend that they do not, and I have yet to find anyone who makes a coherent and well-informed argument to the contrary.***
Tim, often policies serve functions for which they were not originally intended. Sometimes their value lies elsewhere than why they were conceived and implemented. To propose eliminating software patents because you can't find proof of how well they serve their doctrinal justification, without considering their secondary benefits and without direct proof of negative consequences leaves a lot unconsidered.
If you've not found any justification for software patents, also consider that you aren't providing any arguments against them. I'm more inclined to look at MikeT's arguments, because he speaks from the perspective of a programmer with valid concerns. Since you're on the public policy side, you have to make coherent "policy" arguments.
Noel: Please define, tell me, what does the phrase "assertive right" mean??? Please translate the phrase "refuse finding of equivalence" into something I can understand. Usually when I hear this type of phraseology, the actual undiscolosed intent is to diminish an existing right. I hope you will help me with understanding what Lemley and Cohen are proposing.