5 ms·

Amen. The most important part about patents to remember why they exist. Why would congress grant a legal monopoly to anybody? The reason is (or should be) to

Amen. The most important part about patents to remember why they exist. Why would congress grant a legal monopoly to anybody? The reason is (or should be) to protect any investment that went into the invention; for the benefit of society as a whole - to make sure expensive research is still going to happen. An idea does not cost anything. It might have some value if nobody had that idea before, but the idea would have been had anyway regardless of any legal protection. It might suck for the individual having the idea, but society is not served by protecting ideas that came without cost. Copyright on the other hand is useful and needed, it protects a specific expression of an idea. If you translated software patents to books, a patent would be like a legal monopoly on "science fiction stories", or "crime stories". Whereas copyright protects a specific store - as it should. Other folks are free to write other science fiction or crime stories. Anyway... I am not holding my breath. Also there are so many patents already out there, these will likely not be invalidated retroactively.

grandinj 13y ago
Patents were not there to protect investment. That is a post-hoc reason. The original justification was to encourage inventors to share their ideas instead of hiding them. This was at a time when reverse engineering something was a to harder, so it actually made sense.
- alexqgb 13y ago
  Um, right. And how did the law intend to "encourage inventors to share their ideas"? That's right, by protecting the investments made in their origination. That's it.
- Hermel 13y ago
  In that case, one should require working source code to be filed along with software patents.
  - axman6 13y ago
    Why? A patent lacks full disclosure if it doesn't have enough information for a person skilled in the art to reproduce the invention; it must be enabling. Patents are not about specific implementations of an idea, they are about the idea itself. IMO, patents are much more accessable when they are written in plain language than in a specific source language because the ideas may be understood at the higher level that the patent protects and allows for implementation in any language. Do you think that pharma companies should have to provide working versions of their drugs along with their applications? They usually do not have a fully working implementation at the time of application, and won't have anything they can prove works for many years after the necessary regulatory testing is finished. The same for huge mining equipment, it might take years to actually manufacture something that works, but the invention exists once it has been invented, not produced. Why should software be any different? We've grown used to a world where many people freely share their software, but that is their choice, not a moral obligation, just like any field of science. Lastly, one thing that really irks me about the software patents debate is the constant dissing of patents because the idea is obvious. Of course it's obvious once you know the solution, but that's ex post facto analysis. Invention comes from discovering the solution, not from the complexity of the solution itself. I personally feel that things like the slide to unlock patent are more deserving of an innovation patent than a full patent, so there are definitely cases where things have gone wrong. But I do not believe the majority of cases fall into this category.
Snail_Commando 13y ago
> The reason is (or should be) to protect any investment that went into the invention; for the benefit of society as a whole - to make sure expensive research is still going to happen. This is critical. Not just to enable expensive investment, but to create an incentive structure that encourages scientific research that remains in the public domain. This is a critical point that people often miss in these arguments. In computerland, we don't come up with new computer science every time we program an application. In pharmaceutical land, you literally have to come up with new science to make a new drug. Let's not forget that, as software developers, our perspective and opinions on patents are not fungible across all domains. The patent system in the pharmaceutical industry enables a healthy drug research and design market with proper incentive structures. Abolishing patents in the pharmaceutical industry would retard research, skew incentives, and encourage trade secrets. A patent free system would most likely result in a system of trade secrecy. Companies would not share their results, they would keep the secret formula to themselves. It would result in perpetual proprietary lockdowns. After all, why make public the concept that you just spent billions of dollars and years of opportunity cost on? So that competitors can manufacture the drug at essentially zero cost? Nah, we'll just keep that to ourselves, thank you very much. Trade secrets are a threat to science. You cannot build a foundation of science on a platform eaten away by trade secrecy. Trade secrecy will deprive humanity of the positive externalities of a significant chunk of pharmaceutical research. As developers, we see patents as a threat. A sharp implement of the greedy. In the pharmaceutical world, patents are a necessary rung in the ladder of human advancement. Yes, it doesn't always feel good to climb this ladder (consumer drug prices) but without it- there would just be a scrum of companies scrambling to climb up the research pole, each trampling each other's investment returns (incentive) to zero. They'd get bruised and go find easier poles to climb up. Yes, the patent system in the pharmaceutical industry has flaws. But it is the best thing we have given the alternatives. If you are interested, please see see my other comment: https://news.ycombinator.com/item?id=7497872 https://news.ycombinator.com/item?id=7497872
- linuxhansl 13y ago
  Thanks, and I completely agree with you. It wasn't clear from my post, but patents are obviously important, just not to protect everybody's random ideas. That is why it so important to make the point about the investment. In the pharmaceutical industry (and others) there is upfront cost and that should be encouraged and subsequently protected for some time. I still have some philosophical reservations about patenting facts i.e. "this gene does that"; yet I think that is not avoidable. The "Is society as a whole served" question, to me is the paramount criterion. How to establish that for a particular invention, I do not know.
peterashford 13y ago
My understanding was that the original idea of patents was to prevent inventions being lost to mankind due to their being held secret. Patents allow inventors to have a short term monopoly in return for societies long term intellectual gain. Clearly, this is NOT occurring with today's software patents. (1) Most s/w patents are not for 'inventions' that are non-obvious and therefore not even capable of being kept secret in the first instance, and, (2) So many patent applications are so broad as to not actually describe a system in an implementable way. End result: we offer up monopolies, benefiting the 'inventor' but end up with no knowledge added to society. A situation that impoverishes the many to pay the few for pretty much nothing.
- axman6 13y ago
  I'd love to see some evidence for your claims, because they're pretty ridiculous. Arithmetic coding is now something that anyone can use because the patent applying to it has reached the end of its term. If IBM had instead decided to keep the discovery as a trade secret, and no one else had figured out how to perform the same method of compression, we wouldn't to this day have access to it. But we do, this is clearly a benefit to society because we can now produce compression software with much closer to optimal compression rations than we could before. So there goes (1). Would you also say that arithmetic coding is obvious? It sure is, once it's explained to you, but discovering the idea took a lot of effort, and was definitely not obvious given the knowledge that existed before. So there goes (1). All that I would say is that I think the term for software protection is too long given the current pace of development in the software space. This will not be true forever though, like all other technologies, things will begin to slow, and discoveries will be even more valuable once the space of computer science is more broadly mapped.
  - peterashford 13y ago
    I said 'most' not 'all'

grandinj 13y ago

Patents were not there to protect investment. That is a post-hoc reason. The original justification was to encourage inventors to share their ideas instead of hiding them. This was at a time when reverse engineering something was a to harder, so it actually made sense.

alexqgb 13y ago
Um, right. And how did the law intend to "encourage inventors to share their ideas"? That's right, by protecting the investments made in their origination. That's it.
Hermel 13y ago
In that case, one should require working source code to be filed along with software patents.
- axman6 13y ago
  Why? A patent lacks full disclosure if it doesn't have enough information for a person skilled in the art to reproduce the invention; it must be enabling. Patents are not about specific implementations of an idea, they are about the idea itself. IMO, patents are much more accessable when they are written in plain language than in a specific source language because the ideas may be understood at the higher level that the patent protects and allows for implementation in any language. Do you think that pharma companies should have to provide working versions of their drugs along with their applications? They usually do not have a fully working implementation at the time of application, and won't have anything they can prove works for many years after the necessary regulatory testing is finished. The same for huge mining equipment, it might take years to actually manufacture something that works, but the invention exists once it has been invented, not produced. Why should software be any different? We've grown used to a world where many people freely share their software, but that is their choice, not a moral obligation, just like any field of science. Lastly, one thing that really irks me about the software patents debate is the constant dissing of patents because the idea is obvious. Of course it's obvious once you know the solution, but that's ex post facto analysis. Invention comes from discovering the solution, not from the complexity of the solution itself. I personally feel that things like the slide to unlock patent are more deserving of an innovation patent than a full patent, so there are definitely cases where things have gone wrong. But I do not believe the majority of cases fall into this category.

Snail_Commando 13y ago

> The reason is (or should be) to protect any investment that went into the invention; for the benefit of society as a whole - to make sure expensive research is still going to happen. This is critical. Not just to enable expensive investment, but to create an incentive structure that encourages scientific research that remains in the public domain. This is a critical point that people often miss in these arguments. In computerland, we don't come up with new computer science every time we program an application. In pharmaceutical land, you literally have to come up with new science to make a new drug. Let's not forget that, as software developers, our perspective and opinions on patents are not fungible across all domains. The patent system in the pharmaceutical industry enables a healthy drug research and design market with proper incentive structures. Abolishing patents in the pharmaceutical industry would retard research, skew incentives, and encourage trade secrets. A patent free system would most likely result in a system of trade secrecy. Companies would not share their results, they would keep the secret formula to themselves. It would result in perpetual proprietary lockdowns. After all, why make public the concept that you just spent billions of dollars and years of opportunity cost on? So that competitors can manufacture the drug at essentially zero cost? Nah, we'll just keep that to ourselves, thank you very much. Trade secrets are a threat to science. You cannot build a foundation of science on a platform eaten away by trade secrecy. Trade secrecy will deprive humanity of the positive externalities of a significant chunk of pharmaceutical research. As developers, we see patents as a threat. A sharp implement of the greedy. In the pharmaceutical world, patents are a necessary rung in the ladder of human advancement. Yes, it doesn't always feel good to climb this ladder (consumer drug prices) but without it- there would just be a scrum of companies scrambling to climb up the research pole, each trampling each other's investment returns (incentive) to zero. They'd get bruised and go find easier poles to climb up. Yes, the patent system in the pharmaceutical industry has flaws. But it is the best thing we have given the alternatives. If you are interested, please see see my other comment: https://news.ycombinator.com/item?id=7497872 https://news.ycombinator.com/item?id=7497872

linuxhansl 13y ago
Thanks, and I completely agree with you. It wasn't clear from my post, but patents are obviously important, just not to protect everybody's random ideas. That is why it so important to make the point about the investment. In the pharmaceutical industry (and others) there is upfront cost and that should be encouraged and subsequently protected for some time. I still have some philosophical reservations about patenting facts i.e. "this gene does that"; yet I think that is not avoidable. The "Is society as a whole served" question, to me is the paramount criterion. How to establish that for a particular invention, I do not know.

peterashford 13y ago

My understanding was that the original idea of patents was to prevent inventions being lost to mankind due to their being held secret. Patents allow inventors to have a short term monopoly in return for societies long term intellectual gain. Clearly, this is NOT occurring with today's software patents. (1) Most s/w patents are not for 'inventions' that are non-obvious and therefore not even capable of being kept secret in the first instance, and, (2) So many patent applications are so broad as to not actually describe a system in an implementable way. End result: we offer up monopolies, benefiting the 'inventor' but end up with no knowledge added to society. A situation that impoverishes the many to pay the few for pretty much nothing.

axman6 13y ago
I'd love to see some evidence for your claims, because they're pretty ridiculous. Arithmetic coding is now something that anyone can use because the patent applying to it has reached the end of its term. If IBM had instead decided to keep the discovery as a trade secret, and no one else had figured out how to perform the same method of compression, we wouldn't to this day have access to it. But we do, this is clearly a benefit to society because we can now produce compression software with much closer to optimal compression rations than we could before. So there goes (1). Would you also say that arithmetic coding is obvious? It sure is, once it's explained to you, but discovering the idea took a lot of effort, and was definitely not obvious given the knowledge that existed before. So there goes (1). All that I would say is that I think the term for software protection is too long given the current pace of development in the software space. This will not be true forever though, like all other technologies, things will begin to slow, and discoveries will be even more valuable once the space of computer science is more broadly mapped.
- peterashford 13y ago
  I said 'most' not 'all'