[fauigerzigerk]

>1. Patents usually involve a lot of research — and existed as a way to ensure that competition couldn't imitate your product without also making that investment, licensing it from you, etc.

I agree with the gist of your comment, but a fundamental issue with patents is that your investment in research can be destroyed if someone else independently did overlapping research and filed first.

[marcinzm]

Which you mitigate by keeping an eye on competitors and avoiding overly saturated market spaces. This in turn incentives a broader research focus and also incentives publishing paper in saturated areas to make sure there's documented prior art. There's also an incentive for multiple smaller inventions spread over time. This all seems like a positive and not a negative for society as a whole.

[fauigerzigerk]

I think it can be a positive as long as the threshold for what counts as an invention is high enough. If the threshold is too low then patents become an instrument for harassment and rent seeking, slowing down industrial progress.

We have to find the right balance. If a large number of patents sound like a sad joke to most professionals in a field, then something has gone very wrong.

[kiba]

We have to find the right balance. If a large number of patents sound like a sad joke to most professionals in a field, then something has gone very wrong.

That presume that we have the ability to finetune patent laws as needed and also presume that world changing inventions are necessarily novel enough to professionals in a given field. It has often happened commonly enough that inventions will occur to several innovators at once as the next step. It is questionable that it would be fair to penalize an inventor just because another was first to file a patent, potentially driving these inventors out of the industry.

It is also pushing against the reality that people are always building on the works of others. Patents are by themselves monopolies. That is how they work. Thus, people had resorted to arrangements to avoid constant patent lawsuits, which had happened in the past.

[nness]

Yes, that's an interesting problem — the more capital you have to invest, conceptually, the more you can invest in talent and the faster you can research compared to smaller competitors.

Great for the rate of technological advancement, not so great for fairness.

[marcinzm]

Now image if there weren't patents. The larger entities would simply wait for their smaller competitors, copy their products and then use their capital to more efficiently/cheaply manufacture/market them.

edit: While also maintaining massive private documentation stores that further give them a moat. Patents are public so you at least know what someone did 20 years ago even if you're a one person shop. Without them you'd be in trouble unless you had access to a large companies massive and very private internal invention documentation store.

[kiba]

Larger entities have lawyers and can afford more of them. It doesn't protect the little guys so much as allow larger entities monopoly power.

Patents are by themselves not a business model. They merely buy a seat at the negotiation table. Leverage is another matter entirely.

[Leherenn]

It feels like you could overcome this by awarding a patent to both, if they can prove the research was independent?

[fauigerzigerk]

Yes, but how do you prove that it was after one of the teams publishes their work? It seems nearly impossible.

[spion]

You can use the "dropping a hash" method (patio11 has used this). As you progress your work, you publish a hash of it using a service trusted by everyone to correctly identify the researchers and timestamp their hashes.

If someone publishes the paper, you can then publish the files that resulted with the dropped hash and therefore prove you had certain progress of the work at a certain time.

[arthurcolle]

4d merkle trees etched on a proton or something

[fauigerzigerk]

Yes you can provably document your progress, but it doesn't solve the problem that someone else may have published overlapping ideas earlier and you can't prove that you didn't read what they published.

[Leherenn]

I was mostly thinking in the case of overlapping research as mentioned in the parent comment, not independent research years after the publication (where indeed it sounds impossible to prove you didn't read it).

Basically, if you could show that you had already made significant progress before the publication date, then you might be granted the patent as well.

[datascienced]

a use for blockchain at last?