Hacker News new | ask | show | jobs
by dissenter 6613 days ago
There are two common ways to justify patents:

1. They help the public by encouraging inventors to publicly document their inventions

2. They protect and encourage inventors by giving them control over intellectual property

You've taken Position 1, which I find to be a very weak argument in favor of patents.

The world owes a great debt to Einstein and Godel, for relativity and incompleteness, but it is widely recognized that without them the same results would still have been achieved only a few years later. That's the way science works. It progresses along a linear track, and the best minds are only a few years, and more often a few months, ahead of the pack. The literature is rife with examples, but here are a few more good ones to chew on: Newton-Leibniz, Cook-Levin, Knuth–Morris–Pratt, Williamson-Diffie-Hellman, and Cocks-Rivest-Shamir-Adleman.

Though Williamson and Cocks are special cases because of the secrecy granted to government institutions, the point is clear: it is almost certain someone else would've arrived at the same ideas around the same time. Ideas are precipitated by conditions. These citations make this explicit with conflicting dates. You usually don't get this type of evidence because you usually don't get to publish the same idea twice.

For Position 1 to make any sense, it would have to take me longer than 20 years to reverse engineer your idea. Most patentable ideas could be reverse engineered in a day, and more to the point, most patentable 'flashes of genius' are repeated by other people in a vacuum, that is, without any knowledge of the original. Conditions are such that nowadays most people have the same idea within weeks of each other.

With an environment like that, patents are not working to bring new ideas to the public. The public is already having these ideas. What the patent is doing is preventing competitors from entering the market, when competition is almost universally good for the public. You can make the argument that the patent serves the interest of the patenter, and that would be true, but any argument that makes the claim that patents support the public---except in the circuitous way that they help inventors---is specious at best.

I know a lot of people who've dealt with the patent system (software). I'm in the process of filing two myself. I don't know anyone who looks at the patent database as a 'treasure trove' of ideas they can put to use. Most people think of it as a 'minefield' of ideas that are already taken. The patent database is boring, and all of its contents are written specifically to hold up in court battles, not to explain ideas to people who might put them to use. That's a pretty good empirical argument that the patent system is not working according to the guidelines in Position 1.

You take a very sympathetic view of Intellectual Ventures, but to me they don't look like a particularly benevolent company. If they're so intent on licensing new inventions to the public, why isn't there any link to their inventions on their website? Why aren't they pushing a product? There isn't even a basic search link to the USPTO that fills in their name in the assignee field.

The fact that this company raised "hundreds of millions of dollars" indicates to me that a lot of very intelligent people view this as a very profitable field. It is worth keeping in mind that Intellectual Ventures, as a company, aims to produce no product. They only seek to amass a portfolio of rights granted by the US Government to seek injunctions against others using ideas to which they've laid claim.

Now I suppose it is possible that Intellectual Ventures could come up with useful inventions and license them to the public at a fair price, but that is not the impression I get in their case. It seems to me that, their contributions to national defense notwithstanding, they've created a humongous patent troll, and that they intend to front-run intellectual development in this country. They only have to pay a stable of experts to stay a few months ahead of the state of the art, and they don't have to pay a dime for product development. Whether or not they turn into a patent troll, this is in fact what they are incentivized by the market to do. Companies with unregulated, government-granted monopolies rarely charge fair rates for their services.

Public understanding will almost certainly be advanced short-term by Intellectual Ventures, but the question is whether they are deserving of this money, producing no product themselves, and whether they won't harm us in the long run by stymieing the progress of someone who might. Remember, without the protections of patent law, this company wouldn't exist. My concern is not with the people who start a company like Intellectual Ventures, but with the patent laws that make it so enticing for a company like Intellectual Ventures to behave in a counterproductive way.

3 comments

Great reply.

My invocation of Position 1 was somewhat whimsical, and I'm not inclined to defend it much further. I'm not convinced that these Intellectual Ventures guys can justifiably be called "patent trolls" -- I think it's more likely that they're fishing for research grants, or slices of other people's research grants, using many, many pieces of bait -- but I'm not sure they're necessarily doing something useful. Based solely on the article, they look like theorists to me -- long on ideas, short on implementation, lacking in (and perhaps ignorant of) good statistics, in love with the backs of their own envelopes, and apt to gloss over any pesky details with Olympic-level handwaving.

I will take note of this line of yours:

I know a lot of people who've dealt with the patent system (software)... I don't know anyone who looks at the patent database as a 'treasure trove' of ideas they can put to use. Most people think of it as a 'minefield' of ideas that are already taken.

I wonder whether your opinion of the patent system is colored by your experience with software patents -- a modern abomination that threatens to bring down the whole system. When you say "most patentable ideas could be reverse-engineered in a day", you sound like a justifiably frustrated software engineer.

But I think patents make more sense when applied to physical machines and gadgets that are harder to get working, harder and more expensive to reverse-engineer, and harder to cover with enormously wide blanket patents (perhaps because the prior art is so much richer, the phase space so much larger, and the inspectors so much better trained).

"The world owes a great debt to Einstein and Godel, for relativity and incompleteness, but it is widely recognized that without them the same results would still have been achieved only a few years later. "

This is hardly widely recognized.

1.The mathematics that sort of fell into and formed the structure of Einsteins work was about 50 years old when he used them. I am talking about the attempt to prove Euclids 5th axiom using the first 4 and the subsequent evolution of hyperbolic geometry.

2.Godel was working on the problem that was proposed by Hilbert (Google for Hilbert's List Of Problems). At the time when Godel was working on it, it was actively pursued by prominent math dudes of the era;

One connection that I find in both these cases is this: They were both trying to prove something is not possible. And whenever that happened in Science - it usually causes a huge splash. Another example is Group Theory: fell out of the proof for the fact that a 5th degree equation cannot be solved using algebra.

Actually, I believe group theory emerged very gradually from many simultaneous developments. Other sources: permutations, symmetries of crystals and linear transformations, quaternions and matrices (though theory of equations was indeed the main driver). I believe it wasn't until after a bunch of these had reached a high level of development (around 1880) that the abstract group concept began attracting attention as a field of research, because then people could see its use for unifying all those others. Abel's proof for 5th-degree polynomials came out in 1824, and Galois' general proof was published in 1846, so this was slow going, even by nineteenth-century standards.

BTW, I thought I read that Gödel worked pretty much as a lone wolf on his stuff, as most mathematicians weren't into it. And the trend has continued: foundations of mathematics has gotten research attention from surprisingly few people. I'm not 100% sure of this, though.

But anyway, yeah, proving that something can't be done is definitely a big deal. Interesting observation. I will percolate on that.

as someone else wrote here, einstein's general relativity would have been discovered much, much later if he hadn't; and many people believe we might not even have GR at all today if it weren't for him. it's certainly the greatest leap that i am aware of