As someone with a Philosophy BA and a keen but very amateur interest in copyright law, this development fascinates me and I desperately want to understand it. Can anyone give me an example of where this fallacy would lead to unfairly issuing or failing to issue a patent?
Is
> For an invention to be patentable, the core inventive concept must be beyond the grasp of those with average powers of imagination.
Really the case? I had always interpreted the obviousness test as more ‘an average person familiar with the state of the art would not immediately think of the invention’, not that they never would.
Yeah. It’s more a case of “average person would only be able to come up with the invention after months/years of R&D”. What’s being protected is the effort (and corresponding investment of resources), not the intellectual prowess of the inventor.
IMO the OP has not understood the obviousness test. Its not that a person "skilled in the art" would never make the invention, but that it would require many attempts, thus justifying the reward of a patent for their efforts. Thus it's not
"obvious to a person skilled in the art" which construction will succeed.
These days, of course, most software patents should fail this test.
no, the patent is a reward for the disclosure of it to the public (and thus supposedly for sparing the others from the "many attempts" thus speeding up the public progress) no matter how much effort you spent coming up with the invention - be it half your adult life and the life savings or you just invented it in 10 minutes. If there is no value in patent disclosure, then it shouldn't be granted. That basically means that patents are obsolete these days as there is no value in the public disclosure anymore as anything invented by a person A right now, if not disclosed, will be invented half an hour later by a person B and person C and a bunch of others. The only thing patents do these days is preventing the persons B, C and the others from using the idea that the A just happened to file first thus blocking their progress - completely opposite to the original idea that the disclosure by A would speed up the progress of the others. I mean have you ever seen somebody saying "Thanks to the person A for their patent, now i know how to accomplish that thing and will happily pay to the A the license fee as their patent saved me time and effort." ?
Maybe for software, but I bet there are industrial processes which would be significantly harder.
In any case, even if inventions would not be lost, there still needs to be some kind of protection to allow profiting from invention, for some period. Otherwise the competitive advantage in making progress would be severely diminished.
Musk and SpaceX for example is a counter example to your argument. Anybody is free to repeat what the SpaceX is doing. And many is trying to repeat what the Tesla is doing. Seems Musk isn't enforcing any patents.
And what a remarkable boom and progress we have in deep learning, LLM, etc - everybody copies and improves upon everybody's work - precisely because various key items like transformers are either not patented or the patents aren't enforced by their owners.
>Otherwise the competitive advantage in making progress would be severely diminished.
The making progress is the main competitive advantage today.
>protection to allow profiting from invention
that means blocking others which means slowing down public progress these days - why should we do that? The patents were to the public benefit in 19th century (and even then the patent value was already starting to be of the perverse kind - ie. Diesel invented his engine to circumvent Otto's patent. Unfortunately for us the 0.9 or 1.1 click of button seems to be physically impossible.) when trained engineers and scientists were rare and expensive, and thus their work had outsized value. Today we have crowds of them.
Sure, there are maybe some things which don't need protection. The existence of counter examples doesn't invalidate the general case.
Clearly, building reliable rockets is hard and takes a long time to get to profitability. Tesla is losing market share rapidly.
> The making progress is the main competitive advantage today.
I'm not sure what point you are making here. The competitive advantage is largely happening in a system that uses patents. My point is if you take that away, it is by no means certain that progress would remain a competitive advantage.
>that means blocking others which mean slowing down public progress
Only if you discount the observation that without some protection, the progress may not happen at all.
Yeah. I'm not aware of any IQ test designed to measure inventiveness.
And I don't think it's about rewarding above-average inventors, but above-average inventions. That is, not every incremental change should grant an exclusive marketplace for over a decade.
> Thus, the statement ‘An invention shall be taken to involve an inventive step if it is not obvious to a person skilled in the art, having regard to any matter which forms part of the state of the art […]’ (s 3 Patents Act 1977) is logically equivalent to ‘An invention shall be taken to involve an inventive step if the inventor’s powers of imagination are above those of a person skilled in the art’.
Um, no? First of all, if A is "this invention is not obvious", and B is "the inventor has higher-than-average powers of imagination", then the first sentence is A -> B, and the second sentence is B -> A. That is, if A is not obvious, it may be that the inventor is more imaginative; but just because the inventor is imaginative doesn't mean that every idea she comes up with is non-obvious.
Kind of stopped reading after that.
In any case, just because "this invention is not obvious" doesn't mean it can only be seen by people with higher-than-average powers of imagination; you can hit upon things either by thinking hard about the problem, or by just trying thousands of things until you hit upon one. However it's done, the idea is that 1) someone has invested a lot of effort to find a new technique 2) this technique is easy to copy once you've been told about it. We want to encourage people to do this work and then share it with others, so we reward them with a patent.
Whether the effort was purely mental, or a mix of mental and experimental, or pure brute-force experimental doesn't really matter.
The "non-obvious" test isn't a test of the inventor's imagigination, but a test of the trviality of the inventive step. And it isn't a particularly high barrier to overcome in most cases.
For example, one might invent a novel metal alloy, which would be patentable. However, "inventing" objects commonly made from metal (e.g. a knife) now to be made from this novel alloy would not be a creative step, and so not patentable.
Are patents even worth today other than for the medical sector?
Like, lots of things can be covered by copyright anyway. For example there is absolute 0 point in software patents, software is covered by copyright and society does not benefit for covering with patent law the parts that are not covered by copyright as they would be invented anyway, and a patent on non-copyrightable software stuff is the same as not being disclosed to the public as by the time the patent expires, it will most likely be too late, only thing it does is prevent other people from inventing it again, which I think is a worse consequence.
The only sector where I see that patents still make a lick of sense for society is in the pharmaceutical sector as medicines are very expensive to develop and would not be properly covered by copyright.
Maybe it is just me but I just fail to see the point of patents for anything else in the world of today, they are a net negative for society.
> Maybe it is just me but I just fail to see the point of patents for anything else in the world of today, they are a net negative for society.
It's not just you. Most people agree that the current patent system stifles innovation instead of promoting it. They are borderline unenforceable for small players, but make an effective moat (through sheer numbers, not quality) for the incumbents. And let's not even mention patent trolls. The only positive thing about them is that they allow people with strict NDAs to talk about their work, but this is hardly what patents are supposed to be for.
They can theoretically do some good, but the standards would have to be raised by a factor of maybe 100 for that, maybe more.
This article is confused about the obviousness standard and the meaning of the phrase "person having ordinary skill in the art" (PHOSITA). The phrase is a legal concept, kind of like the "reasonable person" standard in other areas of law.
Think about the number of things in your job that you use uncritically. Maybe they could be optimized with a lot of effort, but none have a specific, obviously busted thing worth stopping to fix. That's what "non-obvious" means in the patent sense. It's not that PHOSITA are less creative or intelligent, it's that the ordinary practice don't require applying that creativity and intelligence to the problem the patent solves. The whole idea behind patents is that it incentivizes people to invest time in improving workflows.
There's still subjectivity, but if there's an improvement over the state of the art that no one has made before and there's no hard evidence suggesting that everyone in practice was about to make the improvement, then most patent offices will consider that improvement non-obvious.
Or, stated in the European way: obviousness isn't about whether people of ordinary skill could come up with the invention, it's whether they would come up with the invention had it not been patented (this is the "could-would" standard).
The canonical cases—both of which the Supreme Court found to be obvious—are a 1999 patent application for replacing the hydrolics connected to a car's gas petal with a microcontroller, and a 1950 patent for adding shock absorbers to a plow (just as tractors hit the general market). In both cases, the Supreme Court said those patents were obvious, but the fact those cases made it so far suggests they're right on the line.
22 comments
[ 2.8 ms ] story [ 51.9 ms ] threadReally the case? I had always interpreted the obviousness test as more ‘an average person familiar with the state of the art would not immediately think of the invention’, not that they never would.
These days, of course, most software patents should fail this test.
no, the patent is a reward for the disclosure of it to the public (and thus supposedly for sparing the others from the "many attempts" thus speeding up the public progress) no matter how much effort you spent coming up with the invention - be it half your adult life and the life savings or you just invented it in 10 minutes. If there is no value in patent disclosure, then it shouldn't be granted. That basically means that patents are obsolete these days as there is no value in the public disclosure anymore as anything invented by a person A right now, if not disclosed, will be invented half an hour later by a person B and person C and a bunch of others. The only thing patents do these days is preventing the persons B, C and the others from using the idea that the A just happened to file first thus blocking their progress - completely opposite to the original idea that the disclosure by A would speed up the progress of the others. I mean have you ever seen somebody saying "Thanks to the person A for their patent, now i know how to accomplish that thing and will happily pay to the A the license fee as their patent saved me time and effort." ?
You are in danger of throwing the baby out with the bathwater here.
Surely we don't want to return to the days when all invention had to be kept secret in order to profit from it (and potentially lost)?
I do agree that trivial patents are indeed a blocker to progress, and there are far too many of them.
There is no such danger as no invention/technology on the market can be kept secret these days - anything can be reverse engineered :)
In any case, even if inventions would not be lost, there still needs to be some kind of protection to allow profiting from invention, for some period. Otherwise the competitive advantage in making progress would be severely diminished.
And what a remarkable boom and progress we have in deep learning, LLM, etc - everybody copies and improves upon everybody's work - precisely because various key items like transformers are either not patented or the patents aren't enforced by their owners.
>Otherwise the competitive advantage in making progress would be severely diminished.
The making progress is the main competitive advantage today.
>protection to allow profiting from invention
that means blocking others which means slowing down public progress these days - why should we do that? The patents were to the public benefit in 19th century (and even then the patent value was already starting to be of the perverse kind - ie. Diesel invented his engine to circumvent Otto's patent. Unfortunately for us the 0.9 or 1.1 click of button seems to be physically impossible.) when trained engineers and scientists were rare and expensive, and thus their work had outsized value. Today we have crowds of them.
Clearly, building reliable rockets is hard and takes a long time to get to profitability. Tesla is losing market share rapidly.
> The making progress is the main competitive advantage today.
I'm not sure what point you are making here. The competitive advantage is largely happening in a system that uses patents. My point is if you take that away, it is by no means certain that progress would remain a competitive advantage.
>that means blocking others which mean slowing down public progress
Only if you discount the observation that without some protection, the progress may not happen at all.
Also it seems that prooving inventor capacity in this way is better than IQ test, suggested in the text.
And I don't think it's about rewarding above-average inventors, but above-average inventions. That is, not every incremental change should grant an exclusive marketplace for over a decade.
But the key idea stands: an inventor who invents something rare and non-obvious should be rewarded, supported, and encouraged to invent again. Or not?
Um, no? First of all, if A is "this invention is not obvious", and B is "the inventor has higher-than-average powers of imagination", then the first sentence is A -> B, and the second sentence is B -> A. That is, if A is not obvious, it may be that the inventor is more imaginative; but just because the inventor is imaginative doesn't mean that every idea she comes up with is non-obvious.
Kind of stopped reading after that.
In any case, just because "this invention is not obvious" doesn't mean it can only be seen by people with higher-than-average powers of imagination; you can hit upon things either by thinking hard about the problem, or by just trying thousands of things until you hit upon one. However it's done, the idea is that 1) someone has invested a lot of effort to find a new technique 2) this technique is easy to copy once you've been told about it. We want to encourage people to do this work and then share it with others, so we reward them with a patent.
Whether the effort was purely mental, or a mix of mental and experimental, or pure brute-force experimental doesn't really matter.
The "non-obvious" test isn't a test of the inventor's imagigination, but a test of the trviality of the inventive step. And it isn't a particularly high barrier to overcome in most cases.
For example, one might invent a novel metal alloy, which would be patentable. However, "inventing" objects commonly made from metal (e.g. a knife) now to be made from this novel alloy would not be a creative step, and so not patentable.
The only sector where I see that patents still make a lick of sense for society is in the pharmaceutical sector as medicines are very expensive to develop and would not be properly covered by copyright.
Maybe it is just me but I just fail to see the point of patents for anything else in the world of today, they are a net negative for society.
It's not just you. Most people agree that the current patent system stifles innovation instead of promoting it. They are borderline unenforceable for small players, but make an effective moat (through sheer numbers, not quality) for the incumbents. And let's not even mention patent trolls. The only positive thing about them is that they allow people with strict NDAs to talk about their work, but this is hardly what patents are supposed to be for.
They can theoretically do some good, but the standards would have to be raised by a factor of maybe 100 for that, maybe more.
Think about the number of things in your job that you use uncritically. Maybe they could be optimized with a lot of effort, but none have a specific, obviously busted thing worth stopping to fix. That's what "non-obvious" means in the patent sense. It's not that PHOSITA are less creative or intelligent, it's that the ordinary practice don't require applying that creativity and intelligence to the problem the patent solves. The whole idea behind patents is that it incentivizes people to invest time in improving workflows.
There's still subjectivity, but if there's an improvement over the state of the art that no one has made before and there's no hard evidence suggesting that everyone in practice was about to make the improvement, then most patent offices will consider that improvement non-obvious.
Or, stated in the European way: obviousness isn't about whether people of ordinary skill could come up with the invention, it's whether they would come up with the invention had it not been patented (this is the "could-would" standard).
The canonical cases—both of which the Supreme Court found to be obvious—are a 1999 patent application for replacing the hydrolics connected to a car's gas petal with a microcontroller, and a 1950 patent for adding shock absorbers to a plow (just as tractors hit the general market). In both cases, the Supreme Court said those patents were obvious, but the fact those cases made it so far suggests they're right on the line.