Comment by dave_sullivan | Hacker News Reader

Comment by dave_sullivan | Hacker News Reader

Parent Full thread

> being second to market by a few weeks costs far less than the years of R&D required to build a truly revolutionary product

Can you point to a time where this has turned out to be true by your estimation? An example of a product where there was a tremendous amount of R&D required and the only thing saving the company from being destroyed by competitors beating them to the market 2 weeks later was a strong patent portfolio?

Even in the original iPhone presentation, there's a point where Steve is like, "Oh, and you bet we patented it, lol." But was that why the iphone succeeded? Was that why it could exist in the first place? Were it not for those patents, would Apple have been crushed by competitors mimicking them?

The pro-patent arguments are always expressed as hypotheticals, whereas there are numerous specific examples of patents hurting smaller companies unable to afford the cost of defending themselves. I'd be very interested in situations where the upshot is, "Oh boy, good thing for science and art that those patents existed."

A huge company like Apple is a bad example. The best example of patents working well are when they are used by startups formed from university research groups.

You have a group of people that has performed basic research with minimum funding and decides to commercialize an idea. Since the group's work is already published in conferences and/or journals[1], it is extremely easy for a huge company to just throw some money at the idea and come out with a commercial implementation. But with patents in hand, the research group and the university can decide to either sell the patents to the highest bidder, or continue developing the product if they see the potential. And since the idea is protected by the law, no amount of money can kill the company.

I actually believe that the US patent system is one reason behind the success of advanced tech startups here.

[1]: Publication of results is a requirement for virtually all types of research funding. In other words, if you don't patent the outcome of your research, it's essentially in the public domain.

That sounds pretty reasonable, but what's a specific example of this happening? One that comes to mind is CRISPR, and the patent dispute over that has literally held the field back. CRISPR is the thing that's going to cure cancer, and it's being held back because of patent nonsense. I can't prove it, but if patents didn't exist, I'm pretty sure CRISPR (and the specific nuances at the heart of the actual patent dispute) would nonetheless have been discovered.

Meanwhile compare to something like deep learning, which is uniquely unencumbered by patents for several reasons. Yet somehow corporations seem to be interested and the pace of progress is rapid.

The massive data required for machine learning presents an opportunity to create a moat around those businesses. Releasing the algo but not the data doesn't help competitors much. There's also often a first-mover advantage in data collection: operating a service can give you a unique opportunity to collect relevant data, making it very hard for competitors to catch up. The canonical example is Google search: a competitor can scrape the web (at great cost), but they can't get Google's result click data until they're competitive with Google. And without the data it's damn hard to be competitive.

If a university research group has public funding -- then what? My tax dollars go to support a company that then makes billions from a technology that wouldn't be possible without my involuntary investment.

The university takes a cut, so your money contributed to the US education system, and probably ended up funding more research endeavors.

Pharmaceuticals are a great example. The actual chemical compounds are mostly trivial to manufacture. This is the central reason generic drugs are so cheap. Once you know a particular chemical is useful as a drug and someone else has done the legwork of proving it to regulators, copying that drug is simple and cheap.

The R&D process for pharmaceuticals is long, hard and expensive because it needs to pass regulations. And it kinda needs to be like that to keep the safety of people. But most areas of invention are not like that, there are no such strict regulations, or the need for them. So an idea would be to remove patents as a general invention-protecting mechanism, and replace it with a protection tailored specific to the regulations that cause the problem to be so significant, like in pharmaceuticals. Maybe the FDA could even administer/enforce it.

Reply on news.ycombinator.com