4 ms·

> If you invent something fantastic, say a way for a self driving car to perfectly sense its surroundings That sounds like a highly constructed example, and ev

> If you invent something fantastic, say a way for a self driving car to perfectly sense its surroundings That sounds like a highly constructed example, and even if we take it for bare value, makes a lot of assumptions, for example that whatever you invent is patentable. The way you phrase it also makes it sound like patents are all about individual contributions, when in reality the era of lone genius inventors is long over (if they ever existed). In the US, it's practice for your funding institution to keep all the rights to your inventions, so if you invented the perfect sense for self driving, it's highly likely that a company like Ford already has the patent, because they quite literally own your intellectual output. Imo, I feel the same way like OP, there needs to be a use it or loose it doctrine, just like there is with trademarks. Overall the legislation around intellectual property and copyright feels to be in favor of big corps at the moment and should be heavily castrated.

willsmith72 3y ago
Doesnt sounds constructed to me. The "you" in that case could very easily be a startup
- rpnx 3y ago
  The number of patents owned by startups is greatly dwarfed by the number owned by large companies. Furthermore, the competition opportunity for small businesses to offer similar products at lower prices greatly outweighs the value of patent invention. Some people would say large businesses can make things cheap in a way that small businesses can't compete with. However, that turned out to not happen in practice as big companies chase huge profit margins, supported only by the government. China doesn't have this problem the US does because they have weaker patent laws. Patents need to be eradicated ASAP or we will not be globally competitive. We are also harming humanity as a whole as the inventive utility of patents simply doesn't scale in proportion to the harms, with the large populations we have now, they are clearly harmful on balance. The fundamental problem with patents is that the benefit has lower asymptotic complexity with respect to population size than than the harm does. When you exceed some population size, patents become harmful. The utility of copying is N^2 since you have O(N) copiers and O(N) inventions to copy from. The harm of patents is therefore O(N^2) since this is the copying that patents prevent. What we find is that the benefits of patents seem to scale at most around some O(N /log N) ish metric. Doubling the population size increases the number of inventors, but the chance an invention was already invented by someone else increases as population size increases. Hence the benefits of patents scale worse than O(N). Applying that to the harm and we still get O(N(N/logN)) for harm and O(N/logN) for benefits. Clearly patents do not scale. * Here I am using Log N as a substituite for the difficulty increase of finding an invention not already invented. This exact measure is difficult to estimate.

willsmith72 3y ago

Doesnt sounds constructed to me. The "you" in that case could very easily be a startup

rpnx 3y ago
The number of patents owned by startups is greatly dwarfed by the number owned by large companies. Furthermore, the competition opportunity for small businesses to offer similar products at lower prices greatly outweighs the value of patent invention. Some people would say large businesses can make things cheap in a way that small businesses can't compete with. However, that turned out to not happen in practice as big companies chase huge profit margins, supported only by the government. China doesn't have this problem the US does because they have weaker patent laws. Patents need to be eradicated ASAP or we will not be globally competitive. We are also harming humanity as a whole as the inventive utility of patents simply doesn't scale in proportion to the harms, with the large populations we have now, they are clearly harmful on balance. The fundamental problem with patents is that the benefit has lower asymptotic complexity with respect to population size than than the harm does. When you exceed some population size, patents become harmful. The utility of copying is N^2 since you have O(N) copiers and O(N) inventions to copy from. The harm of patents is therefore O(N^2) since this is the copying that patents prevent. What we find is that the benefits of patents seem to scale at most around some O(N /log N) ish metric. Doubling the population size increases the number of inventors, but the chance an invention was already invented by someone else increases as population size increases. Hence the benefits of patents scale worse than O(N). Applying that to the harm and we still get O(N(N/logN)) for harm and O(N/logN) for benefits. Clearly patents do not scale. * Here I am using Log N as a substituite for the difficulty increase of finding an invention not already invented. This exact measure is difficult to estimate.