Hacker News Reader

Comment by krab | Hacker News Reader

In the case of drugs, it works. The day a patent expires, you have generics that instantly lower the market price. Replicating a drug from a patent is not free but you know that the compound works and can be synthesized. It's much cheaper than doing original research where a drug candidate can fail after you invested a lot into it.

View on HN

For drugs, patents:

1. Successfully provide a profit

2. Are necessary to provide said profit

3. Are long enough to provide enough profit to pay back the investment

But it's not clear that they're short enough to encourage further rapid improvements. Like, suppose a patent lasts 20 years but would still be worth pursuing even if it lasted only 5 years; that's 15 years of public development opportunity lost, for each patent.

Currently, the system is that the government declares "X years" by fiat. If it's too short, then corporations can lobby for longer patent durations. If it's not too short, then corporations can lobby anyway. This is a functional but mediocre system, and there's clearly room for improvement.

I agree. This might be worth optimizing. In case of drugs, you can easily spend the first ten years in research and testing, while paying the patent fees. Software or manufacturing patents, on the other hand, might get different time frame.

That's generally because other companies figure out the synthesis themselves in that 20 years timespan, not because they got accurate synthesis instructions.

Yes, that's what I'm saying. But the generics researchers know exactly what they're trying to make. And they don't have to prove that their drug is better than the standard of care. Just that they synthesized the right thing.

Reply on news.ycombinator.com