> A more nuanced approach would be to stop software patents and lawsuits brought on by non-practicing entities.
ARM, MIPS, and many research universities would count as non-practicing entities. As they are, they only produce "intellectual property", and sell licenses to it to fund their operations. They would be forced to find an alternative business model. Universities have alternative funding sources, and they might not be affected too heavily.
Whether this is a good or a bad thing is open to debate, but it's clear to me that non-practicing entities that produce valuable results do exist.
Good point, and IANAL etc, but since we as humans can more or less clearly see the difference between patent trolls and such institutions, there has to be a way to encode those nuances in legalese.
But I guess the real problem isn't whether the patents usually involved will hold up in court, but the costs involved in setting up a defense and disproving the accusations.
The nuances are in the patents themselves. Are they non-trivial? Are they innovative? Did they require any significant amount of effort to come up with?
I'm of the opinion that two measures will help fix the patent system, and this applies to most fields.
First, the duration of patents needs to be cut. 5 or so years would probably suffice.
Second, there needs to be an incentivized prior art busting effort. Perhaps the party filing the patent could put $1000 or so into escrow for a 1 month waiting period, and if someone manages to find valid prior art, they get the money.
You know, escrow sounds like a great idea if it can reduce applications by the same amount it increases prior art submissions. It would make it harder for individuals, though.
From a lot of what I've read about different NPEs, it sounds like they're already on the cusp of being vulnerable to racketeering / fraud charges. It might not take that much to amend the existing laws to add some language about shell companies and patents that clarifies that this is illegal.
Another thing that could help a lot is minimizing the extent to which patents can be sold. One of the biggest problems with the patent system that I see is that someone can invent something that sounds worthless, sell the patent for $10,000, and then the company that buys the patent can rake in millions in royalties. In general no one should be making money hand over fist for things that they neither invented nor produce.
I agree with you on the first point, but completely disagree on the second. If the patent is valid to begin with, it doesn't really matter who owns it. It's true that a crap patent owned by a deep-pocketed litigant can be more dangerous than the same patent owned by a poor inventor. But the problem is the crap patent, not the deep pockets.
I really think it does matter who owns it. If most of the profit from patents is the result of speculation, I think the system is fundamentally failing to do the job of rewarding innovators. It doesn't matter if the patent is a good one - if it's primarily rewarding hedge fund managers with no interest in research, the incentives are screwed up.
> there has to be a way to encode those nuances in legalese.
How about: more than 50% of your employees are lawyers, or the lawyers in your organization make more than 50% of the money (to prevent a few lawyers from "hiring" a dozen or so highschoolers at minimum wage.
ARM, MIPS, and many research universities would count as non-practicing entities. As they are, they only produce "intellectual property", and sell licenses to it to fund their operations. They would be forced to find an alternative business model. Universities have alternative funding sources, and they might not be affected too heavily.
Whether this is a good or a bad thing is open to debate, but it's clear to me that non-practicing entities that produce valuable results do exist.