6 ms·

Some confusion here about government research and patents. As a government researcher with patents, I will weigh in with an approximation of our patent criteria

by cmsmith 10y ago

Some confusion here about government research and patents. As a government researcher with patents, I will weigh in with an approximation of our patent criteria: 1. Increases the potential for commercialization of the technology. If you have an invention which will take effort to commercialize, companies are less likely to spend that effort without any patent protection. Offering an exclusive license on the patent can increase the chance of anyone doing anything with it. 2. Fulfill the goals of a collaborative research agreement between the government and a private company. 3. Further US manufacturing. The taxpayers pay us to do research to fuel industry in the USA, not Germany or China. 4. May lead to commercialization license. This is way down on the list, and total patent royalties amount to <0.1% of our budget. I would guess that the NASA patent set represents cases where the patent did not end up fulfilling one of those goals, and they are releasing it into the wild to see if it does better on its own.

sitkack 10y ago
> 1. increases the potential for commercialization I don't get this. If an idea is good, monopoly is icing, but not required for its adoption. If say NREL comes up with a something that increases efficiency of something by 2%, I'd want all of America and potentially the world to use it. If NIST finds a cheap way to make welds more robust and void free, I'd want every bridge, building and car to use these new techniques, not trapped in a patent monopoly. > 3. Further US manufacturing Do US Gov patents require that manufacturing be in the US? Is government sponsored research an R&D arm of industry or is it to undertake projects that _don't_ have a direct economic payback or are too large or too risky? What you say might be the "correct answer" but it isn't one I find palatable. The majority of government sponsored research should be open to all. In some circumstances I can see limiting it to people in the US and under even narrower circumstances, limiting it via a patent.
- maxerickson 10y ago
  If an idea is good What if an idea is merely a small incremental improvement with a clear but somewhat expensive path to market?
  - sitkack 10y ago
    Then is it good? Patent the innovations in bringing it to market, or keep them as a trade secret. Other players shouldn't be punished because someone is both inefficient and has the money to spend on acquiring a patent. Government patents look like an excellent way to transfer publicly developed technology to the corporations with the access and cash to purchase them (GE, Boeing, etc).
    - maxerickson 10y ago
      I guess I didn't explain my scenario well enough. NASA has some idea. People that work in the market see the idea and think "Oh, gee, that would make our products better, but it would take $10 million to figure it out". With no patent they then think "too bad it isn't worth developing". Maybe that's an overly constructed scenario, but tech transfer isn't just about getting people to use great ideas (a better welding technique is an example of a great idea), it's also about commercializing ideas that would not necessarily get commercialized in the absence of the patent.
- tgb 10y ago
  If the new welding method is developed enough to patent, but possibly not enough to mass produce new welding equipment, then it will require an investment from someone to get it into use outside the laboratory. Being able to license them a patent provides a greater incentive to make that investment. It's possible for an idea to be simultaneously an improvement, cheap in the long run, and also require an up-front investment that might never be made back without a guarantee of monopoly. I can't judge how frequently this occurs, but it seems reasonable.
  - mindcrime 10y ago
    Being able to license them a patent provides a greater incentive to make that investment. In theory. But that's not the only way things can play out. It's just as likely that if such an innovation is offered freely to everyone, that everyone will feel compelled to adopt it, exactly because of the fear that their competitors will. And nobody wants to get left behind. Given that it's basically impossible to know how a given scenario will play out ahead of time, I don't buy this whole "grant patents on government sponsored developments and then sell them" based on some hypothetical scenario that lacks any real justification.
- rayiner 10y ago
  In engineering, there is a phrase "the last 10% is 90% of the work." Much government R&D (which is more "R" than "D") falls into that category. In order to commercialize it, someone has to come along and do that remaining 90%. The patents create an incentive for someone to do that.
- cmsmith 10y ago
  >If an idea is good, monopoly is icing, but not required for its adoption. I would amend this to say that if an idea is obviously very good, and has an obviously large market, then monopoly is icing. The inventions that you described fit this description, but 99% of inventions do not. It's more likely that your invention is either a small part that might contribute to a big improvement in a major industry, or a complete invention that could make a big difference in a market that doesn't exist yet. In either case, a company might not want to invest the resources to find out if your invention is useful just to have their competitors copy them once the proof-of-concept is available. >Do US Gov patents require that manufacturing be in the US? Not by statute that I'm aware of. But again, if you've invented some esoteric device that might have a $50M market one day, nothing makes government patent review boards happier than licensing that to a small business in Minneapolis who will hire a few dozen engineers and maybe even get the thing built in the US.
peter303 10y ago
Also a non-monetary incentive for scientist types. It shows your peers you have accomplished something and could be useful for future employment. Not everyone is motivated to start a business. When I was in private industry I was someimes allowed to patent something before publishing it.

by cmsmith 10y ago

sitkack 10y ago

> 1. increases the potential for commercialization I don't get this. If an idea is good, monopoly is icing, but not required for its adoption. If say NREL comes up with a something that increases efficiency of something by 2%, I'd want all of America and potentially the world to use it. If NIST finds a cheap way to make welds more robust and void free, I'd want every bridge, building and car to use these new techniques, not trapped in a patent monopoly. > 3. Further US manufacturing Do US Gov patents require that manufacturing be in the US? Is government sponsored research an R&D arm of industry or is it to undertake projects that _don't_ have a direct economic payback or are too large or too risky? What you say might be the "correct answer" but it isn't one I find palatable. The majority of government sponsored research should be open to all. In some circumstances I can see limiting it to people in the US and under even narrower circumstances, limiting it via a patent.

maxerickson 10y ago
If an idea is good What if an idea is merely a small incremental improvement with a clear but somewhat expensive path to market?
- sitkack 10y ago
  Then is it good? Patent the innovations in bringing it to market, or keep them as a trade secret. Other players shouldn't be punished because someone is both inefficient and has the money to spend on acquiring a patent. Government patents look like an excellent way to transfer publicly developed technology to the corporations with the access and cash to purchase them (GE, Boeing, etc).
  - maxerickson 10y ago
    I guess I didn't explain my scenario well enough. NASA has some idea. People that work in the market see the idea and think "Oh, gee, that would make our products better, but it would take $10 million to figure it out". With no patent they then think "too bad it isn't worth developing". Maybe that's an overly constructed scenario, but tech transfer isn't just about getting people to use great ideas (a better welding technique is an example of a great idea), it's also about commercializing ideas that would not necessarily get commercialized in the absence of the patent.
tgb 10y ago
If the new welding method is developed enough to patent, but possibly not enough to mass produce new welding equipment, then it will require an investment from someone to get it into use outside the laboratory. Being able to license them a patent provides a greater incentive to make that investment. It's possible for an idea to be simultaneously an improvement, cheap in the long run, and also require an up-front investment that might never be made back without a guarantee of monopoly. I can't judge how frequently this occurs, but it seems reasonable.
- mindcrime 10y ago
  Being able to license them a patent provides a greater incentive to make that investment. In theory. But that's not the only way things can play out. It's just as likely that if such an innovation is offered freely to everyone, that everyone will feel compelled to adopt it, exactly because of the fear that their competitors will. And nobody wants to get left behind. Given that it's basically impossible to know how a given scenario will play out ahead of time, I don't buy this whole "grant patents on government sponsored developments and then sell them" based on some hypothetical scenario that lacks any real justification.
rayiner 10y ago
In engineering, there is a phrase "the last 10% is 90% of the work." Much government R&D (which is more "R" than "D") falls into that category. In order to commercialize it, someone has to come along and do that remaining 90%. The patents create an incentive for someone to do that.
cmsmith 10y ago
>If an idea is good, monopoly is icing, but not required for its adoption. I would amend this to say that if an idea is obviously very good, and has an obviously large market, then monopoly is icing. The inventions that you described fit this description, but 99% of inventions do not. It's more likely that your invention is either a small part that might contribute to a big improvement in a major industry, or a complete invention that could make a big difference in a market that doesn't exist yet. In either case, a company might not want to invest the resources to find out if your invention is useful just to have their competitors copy them once the proof-of-concept is available. >Do US Gov patents require that manufacturing be in the US? Not by statute that I'm aware of. But again, if you've invented some esoteric device that might have a $50M market one day, nothing makes government patent review boards happier than licensing that to a small business in Minneapolis who will hire a few dozen engineers and maybe even get the thing built in the US.

peter303 10y ago

Also a non-monetary incentive for scientist types. It shows your peers you have accomplished something and could be useful for future employment. Not everyone is motivated to start a business. When I was in private industry I was someimes allowed to patent something before publishing it.