I find this article weak in places. Software patents are already on the rocks - if you look at cases like re: Bilski, it is clear that many software patents do not meet the new requirements for process/business method patents.

I am continually surprised the hacker community cares less about reform and more about abolition. Not only is there a higher chance of success passing reform, but the legislation can be reworked to support the inventor / the hacker / the startup once again.

Patents should be allowed, but the laws much tighter, and the benefit to the public in the long term much clearer.

- Fees should be higher for independent prior arts searching

- Community involvement in prior art search before granting of patent

- Use it or lose it

While he is at it :), there should be a proof of concept demonstrated in some way, again increasing costs. I'm sick of time machine and cancer cures being patented.

There should also be a requirement of legibility, so that people unskilled in patent law can replicate the invention.

I would also add:

* If you're going to sue someone for patent violations, you have to go after the largest company in your industry violating your claims.

* Big companies don't get to sue smaller companies

Agree on the first. The second would be easy to get around (subsidiaries).

I've thought of that but I think you could structure the regulation to account for those situations.

Agreed with the first - to some degree. But I can too easily envision a company spending millions on some technology and never benefitting from it before it's undercut by a swarm of small companies.

> Use it or lose it

Hmm. Suppose that I invent a better CPU branch predictor.

I can't afford to design/build a CPU, and even if I could, I can't afford the legal costs to get Intel to license the x86 instruction set patents.

How can I profit from my invention?

Note that Intel won't agree to "I've got this great idea to tell you about but I won't show it to you unless you agree to pay me lots of money if you like it." What should I be able to do to profit from my invention?

Agreed. Im wrong - this wouldn't work in practice. There must be a way to prevent misuse without hurting the little guy though.

> There should also be a requirement of legibility, so that people unskilled in patent law can replicate the invention.

There is. Since you feel that the results are inadequate, you need to explain what a good requirement consists of. You don't get to talk about the benefits of having a good requirement or the output of a good requirement - it's all about implementation.

I did. People unskilled in patent law should be able to replicate the invention. The case law would clarify what this meant in practice.

The reason is that even software patents are so cryptic that it's almost impossible to replicate them sometimes. I have a couple and have tried to replicate many. I couldn't believe how the patent lawyers turned a my simple ideas into something so hard to understand. This isn't helping the general public in the long run.

> I did. People unskilled in patent law should be able to replicate the invention.

And, as I said, that's what the current requirement is.

> The case law would clarify what this meant in practice.

But it hasn't.

I'm not disagreeing with that evaluation of the status quo. I'm pointing out that the proposed "solution" is how we got there. Given that, it's unlikely to be the way out.

Note that "should be able to replicate" is a description of a desired outcome. As current experience demonstrates, it doesn't produce said outcome. (It's like the difference between requirements and implementation.)

> I couldn't believe how the patent lawyers turned a my simple ideas into something so hard to understand.

Did you ask them why they did what they did in your name?

> I'm sick of time machine and cancer cures being patented.

Why do you care? (The patent office is profitable.)

Patents have a requirement of utility. If the invention is impossible or claims to do something it can't this should defeat it's utility.

In a perfect world, claims should be backed by evidence.

If the invention is impossible or claims to do something that it doesn't do, it won't be used in an infringment action.

Why does the existence of a patent that can't be used against anyone bother you? It's just an expensive (to them) note on the inventor's resume, a note that says "bozo" to folks competent enough to evaluate it.