Hacker News Reader

Comment by Symmetry | Hacker News Reader

First: if an inventor comes up with some idea but decides that instead of producing products based on their invention waits for other people to come up with the same idea and get money from them (or has someone do it by proxity), then is that actually something we want to encourage?

Second: both the times an entity I was part of was on the receiving end of a patent troll, there was at least circumstantial evidence that the troll hadn't invented anything. The first time was when one of the students at my dorm came up with the idea of hooking up our laundry machines to the Internet for a senior project. Shortly after they got Slashdoted someone filed for a patent on the idea. The second time was when the place I worked went to some outside contractors, and they tried to file patents based on the designs we gave them.

View on HN

> First: if an inventor comes up with some idea but decides that instead of producing products based on their invention waits for other people to come up with the same idea and get money from them (or has someone do it by proxity), then is that actually something we want to encourage?

The idea is that the original inventor publishes the secret sauce via the required public disclosures and people either license the technology from him, he sues people who copy the technology without licensing, or he just cashes out and sells his whole interest to some other entity that does any of the above.

If someone else independently comes up with the same idea within a relatively short time frame, well in that case I don't think the patent should be granted. I think the legally sensible thing to do is to move patents in the direction of being an unfair competition type claim (unfair because you simply copy someone else's technology instead of developing your own).

I find it extremely unlikely that someone can conceive of and take the time patent something so novel that someone, somewhere, hasn't already tried it at least once.

In those rare cases, it should be a requirement to produce a detailed description of the method by which the desired result is achieved, and that the patent would cover that specific method.

"Attaching a computer to a database" is not a method. A very specific driver format employing particular protocols is.

That's at the heart of what it is to be a patent. It is neither an idea, nor a specific device. Its supposed to be a technique or method or something equally slippery - thus all the lawyers have something to argue about.

> it should be a requirement to produce a detailed description of the method by which the desired result is achieved

In theory, that's how it is. In order to receive a patent, you must first "reduce it to practice". That doesn't mean you have to build it yourself, but it does mean that you have to describe it in sufficient enough detail that a person of "ordinary skill in the art" should be able to actually build it.

In practice, that doesn't happen. Part of that is that a lot of really general patents got through in the 90s, as the article states. Another part is that patents are written in such thick legalese that you almost need formal education just to make sense of the language, and most people in software won't even bother.

That's exactly what a patent is! You don't patent "communicating wirelessly between two devices." That's a result. Instead, you patent something like OFDM, a particular, novel, way of achieving that result. Infringement is only found if someone else not only achieves the same result, but does so in substantially the same way. Even further, the scope of that way is cabined by prior art--you only get to claim protection over the parts of that mechanism that are actually novel.

...in theory...

Reply on news.ycombinator.com