4 ms·

> I’d see it as the same situation as Camera makers when digital was coming at the horizon. They only needed to be there when the tech is ready, and had no need

> I’d see it as the same situation as Camera makers when digital was coming at the horizon. They only needed to be there when the tech is ready, and had no need to bet the farm on making it a reality sooner than necessary. I kind of hate to say it, but isn't this why patents exist? The first company to make a product with this tech will have exclusivity for 20 years. Shouldn't that be a sufficient incentive?

makeitdouble 5y ago
Patents only cover processes I think ? For instance Kodak was first on the ground with their digital camera, but we all know how it went. The same way pharmaceutical companies patent drugs but it doesn’t stop a rival company from doing their own development cycle and coming up with a competing drug based on the same molecule.
- patmorgan23 5y ago
  Patents cover devices too. The cotton gen, dish washer, and toilet paper holder were a patented at one point.
caslon 5y ago
It's a demonstration that the modern patent system fails at that.
sterlind 5y ago
I think if patents worked like that, there'd be a thriving market for patents on emerging technologies. VCs would fund research groups to do basic science, and their deliverables would be successful experiments and prototypes, turned into patents. Instead, patents are used defensively, for obvious things and rent-seeking. Why is that? I'm not sure, but I can imagine a combination of: * You don't have to prove your invention works. * Patents can be vague and broad, and the non-obviousness bar is depressingly low. Low effort patents have more power than real advances. * It's hard to enforce patents on in-house processes and systems. * Workarounds can be found after you've done the hard work. * Basic research (e.g. algorithms) can't be patented, and basic research is the hardest part. We really do need a system that makes basic research with high eventual impact profitable. Brilliant scientists are getting paid peanuts.
- Wowfunhappy 5y ago
  I'm trying not to show my ignorance by saying something terribly wrong, but my sense is that the patent system works better in pharma than in software. Pharma companies don't need to patent algorithms, but—this is where I hope I'm not saying something wrong—I think they can patent e.g. molecules? Certainly, the first company to the market with a drug is usually rewarded pretty heavily, until generics come online 20 years later.
  - sterlind 5y ago
    they can patent molecules and synthetic routes. but once one company shows that targeting a particular receptor works, or binding to some site on a protein, other companies can find drugs that so the same thing but look different. for instance, Prozac was a breakthrough (the first SSRI), but it was easy for other companies to make their own SSRIs.

makeitdouble 5y ago

Patents only cover processes I think ? For instance Kodak was first on the ground with their digital camera, but we all know how it went. The same way pharmaceutical companies patent drugs but it doesn’t stop a rival company from doing their own development cycle and coming up with a competing drug based on the same molecule.

patmorgan23 5y ago
Patents cover devices too. The cotton gen, dish washer, and toilet paper holder were a patented at one point.

caslon 5y ago

It's a demonstration that the modern patent system fails at that.

sterlind 5y ago

I think if patents worked like that, there'd be a thriving market for patents on emerging technologies. VCs would fund research groups to do basic science, and their deliverables would be successful experiments and prototypes, turned into patents. Instead, patents are used defensively, for obvious things and rent-seeking. Why is that? I'm not sure, but I can imagine a combination of: * You don't have to prove your invention works. * Patents can be vague and broad, and the non-obviousness bar is depressingly low. Low effort patents have more power than real advances. * It's hard to enforce patents on in-house processes and systems. * Workarounds can be found after you've done the hard work. * Basic research (e.g. algorithms) can't be patented, and basic research is the hardest part. We really do need a system that makes basic research with high eventual impact profitable. Brilliant scientists are getting paid peanuts.

Wowfunhappy 5y ago
I'm trying not to show my ignorance by saying something terribly wrong, but my sense is that the patent system works better in pharma than in software. Pharma companies don't need to patent algorithms, but—this is where I hope I'm not saying something wrong—I think they can patent e.g. molecules? Certainly, the first company to the market with a drug is usually rewarded pretty heavily, until generics come online 20 years later.
- sterlind 5y ago
  they can patent molecules and synthetic routes. but once one company shows that targeting a particular receptor works, or binding to some site on a protein, other companies can find drugs that so the same thing but look different. for instance, Prozac was a breakthrough (the first SSRI), but it was easy for other companies to make their own SSRIs.