[wongarsu]

A copyright term of 30 years after first publication, or 30 years after creation if no publication happens in this period. Ideally add some provision that ensures works are actually available after that period (similar to how many countries require all printed books to be submitted to the national library, but extend it to all media that achieves some benchmark of significance). Patent duration adjusted on a per-industry basis. Trademarks are fine as is

[jacobr1]

The best model I've seen for patents is increasingly escalating renewal fees for periodic terms. Such as get the first 5 years for very limited fees, every 5 years (up to some max such as 30) require 2-5x to renew or the patent expires. The exponentially increasing fees limit the hoarding of patents without direct economic benefits, but the high cost means you are able to provide offsetting social benefits while still providing incentives for innovation.

[aezart]

Another model I've seen for copyright is that you declare a "buyout price" for the copyright, the price at which you will release the work into the public domain. You pay 1% of that price every year as a renewal fee, with a 10 year limit.

Don't know if it would work but it's interesting.

[tivert]

> The best model I've seen for patents is increasingly escalating renewal fees for periodic terms. Such as get the first 5 years for very limited fees, every 5 years (up to some max such as 30) require 2-5x to renew or the patent expires. The exponentially increasing fees limit the hoarding of patents without direct economic benefits, but the high cost means you are able to provide offsetting social benefits while still providing incentives for innovation.

So, basically, long-term patents should be a privilege reserved for the rich (people and corporations). Everyone else should GTFO? I'm sure FAANG would still be totally able to massively horde patents with whatever fee structure you propose.

FFS, Amazon developed literally dozens of commercial properties in my area, and then let them sit vacant for literal years because they had some change in strategy. They're perfectly happy to burn money, since they have so much.

So, your idea doesn't solve the problem you're trying to solve, and it would make things worse besides.

[gbear605]

If the scale is exponential, it becomes billions of dollars to keep a patent after a couple decades. Then trillions, at some point after which they’ve decided to give up the patent. Many companies are rich, but they aren’t infinitely rich.

[freejazz]

Patents already have a limited term, why would we want this in a policy?

[jacobr1]

The term is too long. This changes it from a fixed term, to a use-it-or-lose-it oriented policy.

[freejazz]

Patent term is 20 years from the date of filing, and doesn't begin to run until the patent issues, which is usually years after it was filed. It's incomprehensible to me how you can seriously argue that the US patent term is too long but then should be replaced with a system that allows for an indefinite term.

Do you know what a trade secret is? Trade secret already has an indefinite term. There's no reason why patent terms should be indefinite, it's antagonistic the concept of disclosure that is required to obtain a patent in the first place.

[jerojero]

Current term is something like 70 years after the death of the author.

It's simply too long for a world that moves this fast.

[freejazz]

You are thinking about copyright, which the fact that you can't discern between the two should probably be a sign to bow out of dictating IP policy.