5 ms·

I feel like in order to patent a thing, you should need have to have a credible implementation of the thing being patented. An implementation of this clearly re

I feel like in order to patent a thing, you should need have to have a credible implementation of the thing being patented. An implementation of this clearly requires at least level 4 autonomy[0], which AFAIK, nobody has demonstrated. This is only slightly closer to a real-world implementation than a patent for asteroid mining, never mind commercial viability. Mind you, I don't consider that a bad thing; this is pretty dystopian, but still. [0] https://www.sae.org/blog/sae-j3016-update https://www.sae.org/blog/sae-j3016-update

Gorbzel 4y ago
> I feel like in order to patent a thing, you should need have to have a credible implementation of the thing being patented. Reduction to practice is absolutely a requirement of patent law in the US.
- mauvehaus 4y ago
  That seems at odds with TFA itself, which claims "Ford itself recently announced that it was giving up on its goal of developing full self-driving technology, at a cost of $2.7 billion." Perhaps the standard for "reduction to practice [0]" is a little looser than I'm imagining. I can't even see how this would be covered by sufficiency of disclosure[1] since the apparent non-existence of level 4 or 5 autonomy suggests that no person skilled in the art[2] yet exists. If it isn't obvious at this juncture: IANAL. [0] https://en.wikipedia.org/wiki/Reduction_to_practice https://en.wikipedia.org/wiki/Reduction_to_practice [1] https://en.wikipedia.org/wiki/Sufficiency_of_disclosure https://en.wikipedia.org/wiki/Sufficiency_of_disclosure [2] https://en.wikipedia.org/wiki/Person_having_ordinary_skill_in_the_art https://en.wikipedia.org/wiki/Person_having_ordinary_skill_i...
- lolinder 4y ago
  Yes, but it doesn't mean you have to actually have built anything or even be able to build anything. The filing of the patent application is considered a constructive reduction to practice: > Reduction to practice may be an actual reduction or a constructive reduction to practice which occurs when a patent application on the claimed invention is filed. The filing of a patent application serves as conception and constructive reduction to practice of the subject matter described in the application. Thus the inventor need not provide evidence of either conception or actual reduction to practice when relying on the content of the patent application. https://mpep.uspto.gov/RDMS/MPEP/e8r9#/e8r9/d0e207753.html https://mpep.uspto.gov/RDMS/MPEP/e8r9#/e8r9/d0e207753.html
danbruc 4y ago
Isn't this perfect then? Once someone can actually implement it, the patent will have expired.
- JohnFen 4y ago
  Which means that everybody and their dog will implement it, instead of just the patent holder.
  - danbruc 4y ago
    Which they could also do if there was no patent to begin with. If you want different outcomes for different patents, then the point from the original comment will not help. If you think some patent blocks everyone from implementing a good idea, then you would prefer the patent being filled early without a working implementation so that it is closer to expiring when you can actually implement it. For a patent that you consider a bad idea and that no one should be able to implement on the other hand you would prefer that there has to be a working implementation so that it blocks everyone else as long as possible from also implementing it.
luckylion 4y ago
Isn't a large problem with FSD to do it at speed? If you never go above 5mph, a lot of things suddenly get much simpler because the amount of time you need to stop the car is very close to zero, and you probably don't care how long it takes for your possessed car to make it back to your show room.
- jaywalk 4y ago
  I feel like a car driving at 5mph on public roads will absolutely be a problem.
  - ronsor 4y ago
    Not just a problem, but actually illegal in most places.
- saghm 4y ago
  "Why is there so much traffic today?" "A few cars got repossessed and everyone got stuck behind them"
- georgemcbay 4y ago
  > and you probably don't care how long it takes for your possessed car to make it back to your show room. But you probably do care that your self-repossessed car is now impounded for driving dangerously slowly on public roads.
  - luckylion 4y ago
    Might be cheaper to pay the fine and get your cars by the truck load while the local law enforcement takes care of transport and security. But then again, I'm sure you can increase the speed to the average careful elderly driver.
ramraj07 4y ago
For the sake of parenting I think the level of autonomy displayed today is sufficient; as a matter of fact it is likely quite possible even today, in a controlled model city for example.
onion2k 4y ago
An implementation of this clearly requires at least level 4 autonomy Not if you want to have a quiet chat with the person who's been missing their payments..
SrslyJosh 4y ago
> This is only slightly closer to a real-world implementation than a patent for asteroid mining I think we actually have a better idea of how to mine an asteroid.
e_i_pi_2 4y ago
Can I reapply for a patent every week until I can prove that someone else demonstrated the tech, then claim their developments? We already have patent trolls, I don't think this would necessarily help. I think the better solution would be to say "you registered the patent, you get credit for the idea, but whoever makes it first gets the legal rights to sell it" - then people get credit for idea creation and also for making it a reality. I'd ideally add that they have to publicly disclose the info as well but idk how that would affect R&D

Gorbzel 4y ago

> I feel like in order to patent a thing, you should need have to have a credible implementation of the thing being patented. Reduction to practice is absolutely a requirement of patent law in the US.

mauvehaus 4y ago
That seems at odds with TFA itself, which claims "Ford itself recently announced that it was giving up on its goal of developing full self-driving technology, at a cost of $2.7 billion." Perhaps the standard for "reduction to practice [0]" is a little looser than I'm imagining. I can't even see how this would be covered by sufficiency of disclosure[1] since the apparent non-existence of level 4 or 5 autonomy suggests that no person skilled in the art[2] yet exists. If it isn't obvious at this juncture: IANAL. [0] https://en.wikipedia.org/wiki/Reduction_to_practice https://en.wikipedia.org/wiki/Reduction_to_practice [1] https://en.wikipedia.org/wiki/Sufficiency_of_disclosure https://en.wikipedia.org/wiki/Sufficiency_of_disclosure [2] https://en.wikipedia.org/wiki/Person_having_ordinary_skill_in_the_art https://en.wikipedia.org/wiki/Person_having_ordinary_skill_i...
lolinder 4y ago
Yes, but it doesn't mean you have to actually have built anything or even be able to build anything. The filing of the patent application is considered a constructive reduction to practice: > Reduction to practice may be an actual reduction or a constructive reduction to practice which occurs when a patent application on the claimed invention is filed. The filing of a patent application serves as conception and constructive reduction to practice of the subject matter described in the application. Thus the inventor need not provide evidence of either conception or actual reduction to practice when relying on the content of the patent application. https://mpep.uspto.gov/RDMS/MPEP/e8r9#/e8r9/d0e207753.html https://mpep.uspto.gov/RDMS/MPEP/e8r9#/e8r9/d0e207753.html

danbruc 4y ago

Isn't this perfect then? Once someone can actually implement it, the patent will have expired.

JohnFen 4y ago
Which means that everybody and their dog will implement it, instead of just the patent holder.
- danbruc 4y ago
  Which they could also do if there was no patent to begin with. If you want different outcomes for different patents, then the point from the original comment will not help. If you think some patent blocks everyone from implementing a good idea, then you would prefer the patent being filled early without a working implementation so that it is closer to expiring when you can actually implement it. For a patent that you consider a bad idea and that no one should be able to implement on the other hand you would prefer that there has to be a working implementation so that it blocks everyone else as long as possible from also implementing it.

luckylion 4y ago

Isn't a large problem with FSD to do it at speed? If you never go above 5mph, a lot of things suddenly get much simpler because the amount of time you need to stop the car is very close to zero, and you probably don't care how long it takes for your possessed car to make it back to your show room.

jaywalk 4y ago
I feel like a car driving at 5mph on public roads will absolutely be a problem.
- ronsor 4y ago
  Not just a problem, but actually illegal in most places.
saghm 4y ago
"Why is there so much traffic today?" "A few cars got repossessed and everyone got stuck behind them"
georgemcbay 4y ago
> and you probably don't care how long it takes for your possessed car to make it back to your show room. But you probably do care that your self-repossessed car is now impounded for driving dangerously slowly on public roads.
- luckylion 4y ago
  Might be cheaper to pay the fine and get your cars by the truck load while the local law enforcement takes care of transport and security. But then again, I'm sure you can increase the speed to the average careful elderly driver.

ramraj07 4y ago

For the sake of parenting I think the level of autonomy displayed today is sufficient; as a matter of fact it is likely quite possible even today, in a controlled model city for example.

onion2k 4y ago

An implementation of this clearly requires at least level 4 autonomy Not if you want to have a quiet chat with the person who's been missing their payments..

SrslyJosh 4y ago

> This is only slightly closer to a real-world implementation than a patent for asteroid mining I think we actually have a better idea of how to mine an asteroid.

e_i_pi_2 4y ago

Can I reapply for a patent every week until I can prove that someone else demonstrated the tech, then claim their developments? We already have patent trolls, I don't think this would necessarily help. I think the better solution would be to say "you registered the patent, you get credit for the idea, but whoever makes it first gets the legal rights to sell it" - then people get credit for idea creation and also for making it a reality. I'd ideally add that they have to publicly disclose the info as well but idk how that would affect R&D