Hacker News Reader

Another potential reform (from the book "Radical Markets") - charge a (say 1%) property tax on the patent based on self-assessed value, with the caveat that you MUST sell the patent to anyone willing to pay your claimed value.

This basically destroys patent trolling, which only works because of the asymmetrical difference in cost between obtaining a patent and defending a patent lawsuit. If the troll prices the patent too low, the person they are suing just buys it, if they charge too much, they have a massive recurring tax bill (and an incentive to settle quickly, not to drag it out in the courts for years).

Doesn't this work against small-time inventors? Price it too high, and they can't afford to keep it (don't have time to sell it or build a business around it); price it too low, and someone can just buy it off them. Perhaps it could lapse sooner if you aren't paying the tax (to discourage people from just sitting on patents), but that doesn't solve patent trolling.

Remember, the rationale for the existence of patents is to "promote the Progress of Science and useful Arts", not to reward people with monopolies, which is the mechanism.

This is one of several related reasons why I hate the term "intellectual property". Ideas are not property, and shoving them into a property-rights framework and thinking of them as such leads to all sorts of perverse results.

It seems fair that you should have to pay some kind of fee to reserve useful ideas. If you don't have the ability to turn ideas in to reality then they should be given to someone else who can pay the purchase price.

It would be possible to, say, make it royalty free until either you tried to use it in litigation or, say, five years had passed.

If you can’t make the thing event slightly useful in five years, society doesn’t gain much for you having an exclusive monopoly on the thing. So long as the tax is low percentage wise, the yearly cost to keep it as your own would be well worth it.

Doesn't work very well for e.g. drugs, where you may have the first 10 years of the patent term effectively useless because of approval cycles, etc.

Add on a "use it or lose it" policy and I'm onboard. If the patent holder isn't building products with their IP or licensing it as the original author than they shouldn't be allowed to hold it.

It would disable "defensive" patent strategies where companies don't patent their actual technology, but the path to it, which is against the spirit of patents, and disable patent acquisition as an IP strategy.

Meanwhile companies like ARM, which live off IP licensing but wouldn't be punished because they're the original patent authors.

What about just losing software patents altogether? Sorry, but from my perspective on the other side of Atlantic, it looks like a yet another example of "there's no way to avoid this - says the only country where this happens".

They have software patents in many (most? all?) EU countries. The only thing they don’t have is patents on software code itself, which the US doesn’t really have either. See: https://www.ipwatchdog.com/2018/11/15/epo-publishes-revised-...

The main criteria in the EU is whether the method embodied in the software is “technical.” (Image recognition as opposed to say a business method.) See: https://mewburn.com/resource/patentability-of-business-metho...

In the US, under Alice, Section 101 has morphed into a very similar analysis.

I'd prefer just upping the potential consequences for patent trolling - heavier fines for standard violations and the option, if a violation can be shown to demonstrate clear intent, to jail the offender.

Seriously, want to solve all white collar crime ever? Put some jail time on the table and watch everyone straighten right out.

So they'd have one company for every patent. It won't work.

> In this scope they could bring in previous litigative history of the plaintiff (or any related entities, piercing the corporate veil).

Transparent shuffling of money and legal responsibilities don't have to be an effective shield if we follow the money.

Plus if you toss lawyer in jail and disbar them you will run out of lawyers to sac pretty quickly.

Or alternatively you'd manage to take down all of microsoft's patents for that one time they thought a paper clip telling people what to do was innovative... Less jokingly, I'm sure microsoft has a dumb patent or two kicking around, that doesn't mean their entire suite is BS.

There's a difference between having a handful of dumb patents among a sea of valid ones v. having a portfolio composed entirely of dumb patents.

There's also a difference between sitting on a handful of dumb patents v. most or all of your patent litigation cases being over dumb patents.

There isn't a difference unless that's outlined in the law or it's interpretation and I don't know how you'd sanely define things to disallow abuse here.

I'd go a step further. The people behind this lawsuit, and anyone else owning a software patent, should be in jail for fraud.

The loosing side having to pay the legal fees of the winner is an often used mechanism to deter frivolous lawsuits. Patent troll cases I've seen tend to stick to places of trial/judges without those, so could be another area to improve...

AFAIK that was what Facebook tried with their modified BSD that granted patents unless you sued Facebook.

Though that doesn't help against NPE.

Sounds like Anti-SLAPP with IP characteristics.