26 ms·

> Only six patent holders on their list have sent more than two demand letters. The point is that monopolies on ideas are fundamentally unfair, and can only be

by jeswin 11y ago

> Only six patent holders on their list have sent more than two demand letters. The point is that monopolies on ideas are fundamentally unfair, and can only be justified if they provide substantial value over the norm. That might be true in some industries like Pharma, but it is provably false in Software. When has something (in software) been invented by building on some text in a patent filing?

studentrob 11y ago
> When has something (in software) been invented by building on some text in a patent filing? Hah! That's a great way to put it. Patents are supposed to be an exchange of (1) opening the idea to the public for (2) protection. Software patents, to my knowledge, do not do (1). Arguably, other industries suffer the same trouble. Patent filers these days go to great lengths to disguise their meaning so that they are ambiguous enough to cover many things, and presumably, difficult to actually understand and borrow ideas.
throwawaykf05 11y ago
> The point is that monopolies on ideas are fundamentally unfair... I could just as easily say "free-riding off somebody else's ingenuity is fundamentally unfair". > ... but it is provably false in Software. Citation needed? Many of the sources cited in OP have flaws and/or have been refuted elsewhere, especially studies by Bessen & co. Many of their other sources are from people who don't know the first thing about patents. They cite that "This American Life" episode where the clueless dude from M-CAM suggested there's a patent on toast, something that you'd know is BS with a glance at the actual claim and 5 minutes of research on Google. On the other hand, there are studies showing that having patents increases the chances of software startups get VC funding and successful exits by a factor of four. How much innovation happens without funding? Even the majority of open source is financed by various industry sources. > When has something (in software) been invented by building on some text in a patent filing? And how often is something in software not been built by simply copying somebody else's ideas? Some ideas are harder to generate (e.g. RSA) than others (e.g. online pet store) and deserve protection.
- phaemon 11y ago
  > I could just as easily say "free-riding off somebody else's ingenuity is fundamentally unfair". You could, but it would be ludicrous. The very existence of human civilisation is due to "free-riding off somebody else's ingenuity". > And how often is something in software not been built by simply copying somebody else's ideas? Who cares. Not copying ideas is not the purpose of patents, it's the exact opposite. > On the other hand, there are studies showing that having patents increases the chances of software startups get VC funding Again, irrelevant. Getting startup funding is not the purpose of patents. If you want the role of patents to be to help startups then that should be their stated role. In which case, you'd probably only allow companies <5 years old to hold a patent, they'd only be valid for 5 years, and ownership could not be transferred. Of course then you have something very different from our current system, but frankly it sounds a lot better.
- loup-vaillant 11y ago
  > free-riding off somebody else's ingenuity is fundamentally unfair Ideas are not a rival good. When you share an idea with someone, you don't lose half of it. You both get the whole idea. So it's not a zero-sum game. As long as the genius gets rewarded, free riding is fair. > How much innovation happens without funding? […] Some ideas are harder to generate (e.g. RSA) Well, the best and hardest ideas tend to be state-funded anyway. RSA for instance was funded by MIT salaries¹ (Rivest, Shadir, and Abelman worked there). ARPANET (now the internet) was basically a DOD project. You're right, inventions need funding. Patent just aren't the only way —nor, I would argue, the most efficient. [1] https://en.wikipedia.org/wiki/RSA_%28cryptosystem%29#History https://en.wikipedia.org/wiki/RSA_%28cryptosystem%29#History

by jeswin 11y ago

studentrob 11y ago

> When has something (in software) been invented by building on some text in a patent filing? Hah! That's a great way to put it. Patents are supposed to be an exchange of (1) opening the idea to the public for (2) protection. Software patents, to my knowledge, do not do (1). Arguably, other industries suffer the same trouble. Patent filers these days go to great lengths to disguise their meaning so that they are ambiguous enough to cover many things, and presumably, difficult to actually understand and borrow ideas.

throwawaykf05 11y ago

> The point is that monopolies on ideas are fundamentally unfair... I could just as easily say "free-riding off somebody else's ingenuity is fundamentally unfair". > ... but it is provably false in Software. Citation needed? Many of the sources cited in OP have flaws and/or have been refuted elsewhere, especially studies by Bessen & co. Many of their other sources are from people who don't know the first thing about patents. They cite that "This American Life" episode where the clueless dude from M-CAM suggested there's a patent on toast, something that you'd know is BS with a glance at the actual claim and 5 minutes of research on Google. On the other hand, there are studies showing that having patents increases the chances of software startups get VC funding and successful exits by a factor of four. How much innovation happens without funding? Even the majority of open source is financed by various industry sources. > When has something (in software) been invented by building on some text in a patent filing? And how often is something in software not been built by simply copying somebody else's ideas? Some ideas are harder to generate (e.g. RSA) than others (e.g. online pet store) and deserve protection.

phaemon 11y ago
> I could just as easily say "free-riding off somebody else's ingenuity is fundamentally unfair". You could, but it would be ludicrous. The very existence of human civilisation is due to "free-riding off somebody else's ingenuity". > And how often is something in software not been built by simply copying somebody else's ideas? Who cares. Not copying ideas is not the purpose of patents, it's the exact opposite. > On the other hand, there are studies showing that having patents increases the chances of software startups get VC funding Again, irrelevant. Getting startup funding is not the purpose of patents. If you want the role of patents to be to help startups then that should be their stated role. In which case, you'd probably only allow companies <5 years old to hold a patent, they'd only be valid for 5 years, and ownership could not be transferred. Of course then you have something very different from our current system, but frankly it sounds a lot better.
loup-vaillant 11y ago
> free-riding off somebody else's ingenuity is fundamentally unfair Ideas are not a rival good. When you share an idea with someone, you don't lose half of it. You both get the whole idea. So it's not a zero-sum game. As long as the genius gets rewarded, free riding is fair. > How much innovation happens without funding? […] Some ideas are harder to generate (e.g. RSA) Well, the best and hardest ideas tend to be state-funded anyway. RSA for instance was funded by MIT salaries¹ (Rivest, Shadir, and Abelman worked there). ARPANET (now the internet) was basically a DOD project. You're right, inventions need funding. Patent just aren't the only way —nor, I would argue, the most efficient. [1] https://en.wikipedia.org/wiki/RSA_%28cryptosystem%29#History https://en.wikipedia.org/wiki/RSA_%28cryptosystem%29#History