3 ms·

Your entire first paragraph is straw men. It doesn't matter how much any individual pays, it doesn't matter what any individual's sentiment is. If Google pays a

Your entire first paragraph is straw men. It doesn't matter how much any individual pays, it doesn't matter what any individual's sentiment is. If Google pays a researcher to do some research, they own it, regardless of how much any individual shareholder paid of that amount or what any individual's sentiment about that research is. Those things are irrelevant. It's not as easy as "publicly funded research should be free", I agree. As I said, it's that funded research should benefit the funders. Startups don't benefit from patents. There is plenty of (probably publicly funded) research to show that. Here's a recent paper on patents and innovation: https://www.aeaweb.org/articles.php?doi=10.1257/jep.27.1.3 https://www.aeaweb.org/articles.php?doi=10.1257/jep.27.1.3, or read Joel Mokyr's "The Enlightened Economy" or "The Lever of Riches" and see how James Watt held up the widespread commercialization of the steam engine by refusing to allow his patents to be used downstream (or read the NBER financed study on the impact of patents on downstream innovation: http://www.nber.org/papers/w20269 http://www.nber.org/papers/w20269). This paper (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2029098 http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2029098) is probably the most targeted towards whether patents help startups or not. The conclusion they came to was "the patent system was working 'neither well nor poorly for [startups]' and tended to favor larger companies." And last, if it benefits "society", that is, all people, how can there be free riders? If you mean it benefits non-taxpayers as well as taxpayers, well, that is one of the points of taxes (as opposed to fees.)

tzs 12y ago
> It's not as easy as "publicly funded research should be free", I agree. As I said, it's that funded research should benefit the funders. It does benefit the funders in the case of government grants. The NSF gives a researcher a grant to scientifically study some specific thing. The researcher makes scientific discoveries about the thing. The researcher publishes the research, and also (possibly in collaboration with people who were not involved in the research) thinks about how the new science can be applied to produce real world products. The research gets patents on some of those real world applications of the research results. Note that the public that contributed to funding the research did benefit. The research was released, and they can use it as the basis of further research, or think up useful applications for it (and patent those if someone hasn't already). I think people tend to confuse the research (which is often paid for by government grants, at least in part), and development of practical commercial applications for the results of the research, which is generally not paid for by government grants. Note that if the researcher had not worked on coming up with practical applications and patented them, but instead just published the research and moved on to the next project, that doesn't mean there would not be patents. Anyone would be free to take the published research, figure out how to apply it to practical problems, and likely get a patent on that use of the research results. Their patents would not cover the underlying research--just the use of it as described in their patent.
- ganeumann 12y ago
  Fair enough in theory. But in my experience both the research and the commercialization efforts leading to patents are paid for via grants. Perhaps that would be a fair trade, since the University provides the opportunity for all of this to happen, except that the University usually also taxes the grants to pay overhead. They have their cake and eat it too.
throwawaykf05 12y ago
> If Google pays a researcher to do some research, they own it, regardless of how much any individual shareholder paid of that amount... Do you assume you are Google rather than the shareholder that you get any say in what happens with that research? > Here's a recent paper on patents and innovation... I would take with a jarful of salt anything Boldrin and Levine say. This is a rehash of their "Against Intellectual Monopoly" book, and they do the same mischaracterizing of other research, mis-portraying of actual events and claiming outright falsehoods. Case in point, the myth that Watt's patents held up steam development (which you too repeated): https://news.ycombinator.com/item?id=6865980 https://news.ycombinator.com/item?id=6865980. What's really egregious is they refused to admit the truth when called out on it. Another comment pointing more falsehoods on their part: https://news.ycombinator.com/item?id=7890441 https://news.ycombinator.com/item?id=7890441 And before anyone else mentions it, the "Wright patent held up US aircraft industry" thing is a myth too: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2355673 http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2355673 The startup / patents paper indeed is interesting, thanks, but the results are more nuanced than the quote you selected: patents do help startups with financing and making successful exits, but they cannot be leveraged effectively because enforcing them is more expensive. That is something that can be fixed if people focused on it rather than the "patent troll" bogeyman. The "Cumulative Innovation" paper is new to me, but a quick glance suggests some problems with it right away: 1. They derive a model to evaluate their claims, and unless backed up by solid empirical proof, this can be made to support arbitrary hypotheses (see this exchange on an unrelated field: https://news.ycombinator.com/item?id=9082932 https://news.ycombinator.com/item?id=9082932). 2. Their final data set is very small (1200 or so patents out of millions) and highly biased, since they relate only to litigated patents, and even then, on those they deem to be the "stronger" ones. Indeed, they admit that most other patents being invalidated has no effect whatsoever on their metric for downstream innovation. Even if valid, it is impossible to generalize these results to all patents as a whole. On the other hand, there are many, many papers that point out the benefits of patent systems that you may not have considered. Yet another comment of mine mentioning some: https://news.ycombinator.com/item?id=8155043 https://news.ycombinator.com/item?id=8155043 The number of links to my own comments should give you a hint as to how often I debate this topic on HN :-) As usual, I'll wrap up by saying it's a complex and nuanced topic, and point to a meta-study that looks at a large number of other studies that explore both the pros and cons of current patent systems: "Recent Research on the Economics of Patents" - Bronwyn H. Hall, Dietmar Harhoff, (Google for PDF.)

tzs 12y ago

> It's not as easy as "publicly funded research should be free", I agree. As I said, it's that funded research should benefit the funders. It does benefit the funders in the case of government grants. The NSF gives a researcher a grant to scientifically study some specific thing. The researcher makes scientific discoveries about the thing. The researcher publishes the research, and also (possibly in collaboration with people who were not involved in the research) thinks about how the new science can be applied to produce real world products. The research gets patents on some of those real world applications of the research results. Note that the public that contributed to funding the research did benefit. The research was released, and they can use it as the basis of further research, or think up useful applications for it (and patent those if someone hasn't already). I think people tend to confuse the research (which is often paid for by government grants, at least in part), and development of practical commercial applications for the results of the research, which is generally not paid for by government grants. Note that if the researcher had not worked on coming up with practical applications and patented them, but instead just published the research and moved on to the next project, that doesn't mean there would not be patents. Anyone would be free to take the published research, figure out how to apply it to practical problems, and likely get a patent on that use of the research results. Their patents would not cover the underlying research--just the use of it as described in their patent.

ganeumann 12y ago
Fair enough in theory. But in my experience both the research and the commercialization efforts leading to patents are paid for via grants. Perhaps that would be a fair trade, since the University provides the opportunity for all of this to happen, except that the University usually also taxes the grants to pay overhead. They have their cake and eat it too.

throwawaykf05 12y ago

> If Google pays a researcher to do some research, they own it, regardless of how much any individual shareholder paid of that amount... Do you assume you are Google rather than the shareholder that you get any say in what happens with that research? > Here's a recent paper on patents and innovation... I would take with a jarful of salt anything Boldrin and Levine say. This is a rehash of their "Against Intellectual Monopoly" book, and they do the same mischaracterizing of other research, mis-portraying of actual events and claiming outright falsehoods. Case in point, the myth that Watt's patents held up steam development (which you too repeated): https://news.ycombinator.com/item?id=6865980 https://news.ycombinator.com/item?id=6865980. What's really egregious is they refused to admit the truth when called out on it. Another comment pointing more falsehoods on their part: https://news.ycombinator.com/item?id=7890441 https://news.ycombinator.com/item?id=7890441 And before anyone else mentions it, the "Wright patent held up US aircraft industry" thing is a myth too: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2355673 http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2355673 The startup / patents paper indeed is interesting, thanks, but the results are more nuanced than the quote you selected: patents do help startups with financing and making successful exits, but they cannot be leveraged effectively because enforcing them is more expensive. That is something that can be fixed if people focused on it rather than the "patent troll" bogeyman. The "Cumulative Innovation" paper is new to me, but a quick glance suggests some problems with it right away: 1. They derive a model to evaluate their claims, and unless backed up by solid empirical proof, this can be made to support arbitrary hypotheses (see this exchange on an unrelated field: https://news.ycombinator.com/item?id=9082932 https://news.ycombinator.com/item?id=9082932). 2. Their final data set is very small (1200 or so patents out of millions) and highly biased, since they relate only to litigated patents, and even then, on those they deem to be the "stronger" ones. Indeed, they admit that most other patents being invalidated has no effect whatsoever on their metric for downstream innovation. Even if valid, it is impossible to generalize these results to all patents as a whole. On the other hand, there are many, many papers that point out the benefits of patent systems that you may not have considered. Yet another comment of mine mentioning some: https://news.ycombinator.com/item?id=8155043 https://news.ycombinator.com/item?id=8155043 The number of links to my own comments should give you a hint as to how often I debate this topic on HN :-) As usual, I'll wrap up by saying it's a complex and nuanced topic, and point to a meta-study that looks at a large number of other studies that explore both the pros and cons of current patent systems: "Recent Research on the Economics of Patents" - Bronwyn H. Hall, Dietmar Harhoff, (Google for PDF.)