4 ms·

It'd be extremely foolish for any advanced economy to give up on intellectual property, including software patents. Which is why it's not going to happen.

deogeo 7y ago
Foolish for the companies currently on top, yes. Not necessarily foolish for consumers, that are harmed by competition being stifled by patents.
- jacques_chester 7y ago
  By itself, knowledge is a public good. It's not excludable (I can't block non-payers from getting it) and it's not rivalrous (your use doesn't reduce its value for someone else). Public goods are chronically undersupplied in a normal market situation. Providing them imposes a cost on the provider, but because the good is non-excludable you're relying on the goodness of consumers to pay for it. Some will, some won't, but the net amount recovered will be less than it would be otherwise. So what do we do about the undersupply of public goods? There are a variety of options. Taking knowledge as a particular example, one option is the patent system, which grants a temporary legal monopoly on the claims in the patent. It creates an artificial excludability that allows the provider of the knowledge to recover more of the cost of providing it. Overall that should raise the production of knowledge. But, you say, that reduces competition. And sometimes it will. But many ideas can be substituted and so the competitors will patent different things. In fact the monopoly of the competitor gives them an incentive to find new ideas that the competitor hasn't found yet, an incentive which was weaker in the original situation. The alternative is provision through a central supplier, typically: government. But it turns out that the costs of developing knowledge still remain, so how are they paid for? Taxes, essentially. Generally speaking the costs work out to be about the same, but become differently distributed. Instead of knowledge being created by active, incentivised parties who can then get a return from those who value the knowledge most, the incidence of cost falls on everyone. Someone who doesn't value that knowledge gets ripped off, and someone who values it very highly gets subsidised. That said: pure research almost always needs to be publicly subsidised because, by very sensible policy, the laws of nature can't be patented. And representatives passing public budgets can give consideration to which public goods ought to be subsidised, beyond the cold light of economics, as a principle of fairness or morality or furthering humankind. But most knowledge is not pure research. So either we pay for it with taxes, or we pay for it with patents, or we pay for it by getting less common knowledge. Nothing comes for free. Not even ideas.
  - deogeo 7y ago
    A very nice, very theoretical answer. But how are software patents actually used in practice? As drivers of innovation, or as a way to use not-too-clever-but-essential ideas as a moat to keep out competitors?
    - jacques_chester 7y ago
      Absolutely as a moat, that's what makes them valuable. The point is that the moat becomes a public good once the patent expires. The moats would exist without patents, but there would be less total knowledge and less competition for an indefinite period, rather than a fixed one. That greed is a motivation for obtaining a patent doesn't make them bad, any more than my decision to care about my wages makes me bad. Greed will exist either way, the question is how best to direct it towards maximising the total good.
      - theamk 7y ago
        > moat becomes a public good once the patent expires I have heard this argument a lot of times, but I don't see how it applies to software patents in particular. At one of the previous jobs, I was doing some research, and there were lots of useful inputs: conference/journal papers, technical reports, thesises, and so on. The only exception are software patents -- I have never learned anything from them.
        jacques_chester 7y ago
        Since we're take anecdotes as the standad: I've done research and read patents too. I found clever ideas in all of them.
        theamk 7y ago
        In all of them? I doubt it. I can believe that some software patent somewhere have a clever idea that is not described elsewhere -- but I still have to find such patent. Let's take for exa/mple first google hit for "software patent", which for me is https://patents.google.com/patent/US6353926B1/en https://patents.google.com/patent/US6353926B1/en "Software update notification". In my opinion, it is completely useless. - It has nothing particularly innovative. It describes a specific Windows update mechanism in needless details -- while it is a solid design, most engineers, given the same problem, would come up with similar design. I know I did, and I did not know anything about OSD, CDF or windows update subsystem at all. - It does not present information clearly. The text is hard to read and needlessly complicated. There are better sources of information about this topic -- I am sure there is a MSDN article and a dozen of blog posts which describe this much better. - It did not even motivate the inventor! I am almost sure that whatever motivation Microsoft had for promoting Open Software Description standard, it was not to get this patent. So what good is this patent? As far as I can see, there is no public benefit at all. The only value is for Microsoft, so they can threaten others and stifle competition.
Beldin 7y ago
Note that international patent law isn't national patent law. Sure, it's possible to have your patent valid in multiple areas. But e.g. Chinese patent practice isn't quite in line with EU or US patent law. Secondly: patents are inherently a form of publishing your ip. The "giving away" part is enforced by laws, but the whole concept of patents is to make knowledge publicly available. Typically, "giving away" is only restricted to commercial uses - anyone who duplicates the patented invention for personal use is fine. You may argue that that is foolish for an advanced economy, but that is what most of them do.
kstrauser 7y ago
Everything I've ever read on the subject says that China is far and away the greatest infringer of IP. American companies have almost zero capacity to prosecute infringement claims in China, but Chinese companies operating in America have no particular barrier to suing. Frankly, I'm amazed that American companies haven't banded together to call for the end of software patents to erase that enormous and expensive trade deficit.
- bakuninsbart 7y ago
  IP infringement is one half of the coin, industrial espionage the other. Historically, some of the very earliest documented cases of industrial espionage were performed against China by european powers. - The US is not innocent in that regard either, but probably much better at hiding their activity. We needed Edward Snowden to substantiate what people in the industry here in Germany had been suspecting for decades [0]: I've heard of numerous cases where german companies are developing new tech, and shortly before or after release an american company with close ties to the government releases an equal product. [0] https://www.spiegel.de/international/germany/german-intelligence-agency-bnd-under-fire-for-nsa-cooperation-a-1030593.html https://www.spiegel.de/international/germany/german-intellig...
- jacques_chester 7y ago
  It's unlikely that China will remain so in the long run, as their own R&D output continues to rise.

deogeo 7y ago

Foolish for the companies currently on top, yes. Not necessarily foolish for consumers, that are harmed by competition being stifled by patents.

jacques_chester 7y ago
By itself, knowledge is a public good. It's not excludable (I can't block non-payers from getting it) and it's not rivalrous (your use doesn't reduce its value for someone else). Public goods are chronically undersupplied in a normal market situation. Providing them imposes a cost on the provider, but because the good is non-excludable you're relying on the goodness of consumers to pay for it. Some will, some won't, but the net amount recovered will be less than it would be otherwise. So what do we do about the undersupply of public goods? There are a variety of options. Taking knowledge as a particular example, one option is the patent system, which grants a temporary legal monopoly on the claims in the patent. It creates an artificial excludability that allows the provider of the knowledge to recover more of the cost of providing it. Overall that should raise the production of knowledge. But, you say, that reduces competition. And sometimes it will. But many ideas can be substituted and so the competitors will patent different things. In fact the monopoly of the competitor gives them an incentive to find new ideas that the competitor hasn't found yet, an incentive which was weaker in the original situation. The alternative is provision through a central supplier, typically: government. But it turns out that the costs of developing knowledge still remain, so how are they paid for? Taxes, essentially. Generally speaking the costs work out to be about the same, but become differently distributed. Instead of knowledge being created by active, incentivised parties who can then get a return from those who value the knowledge most, the incidence of cost falls on everyone. Someone who doesn't value that knowledge gets ripped off, and someone who values it very highly gets subsidised. That said: pure research almost always needs to be publicly subsidised because, by very sensible policy, the laws of nature can't be patented. And representatives passing public budgets can give consideration to which public goods ought to be subsidised, beyond the cold light of economics, as a principle of fairness or morality or furthering humankind. But most knowledge is not pure research. So either we pay for it with taxes, or we pay for it with patents, or we pay for it by getting less common knowledge. Nothing comes for free. Not even ideas.
- deogeo 7y ago
  A very nice, very theoretical answer. But how are software patents actually used in practice? As drivers of innovation, or as a way to use not-too-clever-but-essential ideas as a moat to keep out competitors?
  - jacques_chester 7y ago
    Absolutely as a moat, that's what makes them valuable. The point is that the moat becomes a public good once the patent expires. The moats would exist without patents, but there would be less total knowledge and less competition for an indefinite period, rather than a fixed one. That greed is a motivation for obtaining a patent doesn't make them bad, any more than my decision to care about my wages makes me bad. Greed will exist either way, the question is how best to direct it towards maximising the total good.
    - theamk 7y ago
      > moat becomes a public good once the patent expires I have heard this argument a lot of times, but I don't see how it applies to software patents in particular. At one of the previous jobs, I was doing some research, and there were lots of useful inputs: conference/journal papers, technical reports, thesises, and so on. The only exception are software patents -- I have never learned anything from them.
      - jacques_chester 7y ago
        Since we're take anecdotes as the standad: I've done research and read patents too. I found clever ideas in all of them.
        theamk 7y ago
        In all of them? I doubt it. I can believe that some software patent somewhere have a clever idea that is not described elsewhere -- but I still have to find such patent. Let's take for exa/mple first google hit for "software patent", which for me is https://patents.google.com/patent/US6353926B1/en https://patents.google.com/patent/US6353926B1/en "Software update notification". In my opinion, it is completely useless. - It has nothing particularly innovative. It describes a specific Windows update mechanism in needless details -- while it is a solid design, most engineers, given the same problem, would come up with similar design. I know I did, and I did not know anything about OSD, CDF or windows update subsystem at all. - It does not present information clearly. The text is hard to read and needlessly complicated. There are better sources of information about this topic -- I am sure there is a MSDN article and a dozen of blog posts which describe this much better. - It did not even motivate the inventor! I am almost sure that whatever motivation Microsoft had for promoting Open Software Description standard, it was not to get this patent. So what good is this patent? As far as I can see, there is no public benefit at all. The only value is for Microsoft, so they can threaten others and stifle competition.

Beldin 7y ago

Note that international patent law isn't national patent law. Sure, it's possible to have your patent valid in multiple areas. But e.g. Chinese patent practice isn't quite in line with EU or US patent law. Secondly: patents are inherently a form of publishing your ip. The "giving away" part is enforced by laws, but the whole concept of patents is to make knowledge publicly available. Typically, "giving away" is only restricted to commercial uses - anyone who duplicates the patented invention for personal use is fine. You may argue that that is foolish for an advanced economy, but that is what most of them do.

kstrauser 7y ago

Everything I've ever read on the subject says that China is far and away the greatest infringer of IP. American companies have almost zero capacity to prosecute infringement claims in China, but Chinese companies operating in America have no particular barrier to suing. Frankly, I'm amazed that American companies haven't banded together to call for the end of software patents to erase that enormous and expensive trade deficit.

bakuninsbart 7y ago
IP infringement is one half of the coin, industrial espionage the other. Historically, some of the very earliest documented cases of industrial espionage were performed against China by european powers. - The US is not innocent in that regard either, but probably much better at hiding their activity. We needed Edward Snowden to substantiate what people in the industry here in Germany had been suspecting for decades [0]: I've heard of numerous cases where german companies are developing new tech, and shortly before or after release an american company with close ties to the government releases an equal product. [0] https://www.spiegel.de/international/germany/german-intelligence-agency-bnd-under-fire-for-nsa-cooperation-a-1030593.html https://www.spiegel.de/international/germany/german-intellig...
jacques_chester 7y ago
It's unlikely that China will remain so in the long run, as their own R&D output continues to rise.