Categories
Uncategorized

How to fix the patent system in 10 easy steps

1. Software cannot be patented. If there is any doubt as to whether a patent is a software patent, it is. 2. Business methods cannot be patented. As before, if there is any doubt as to whether or not it is a business method patent, it is. 3. Obviousness and novelty should be decided by […]

1. Software cannot be patented. If there is any doubt as to whether a patent is a software patent, it is.
2. Business methods cannot be patented. As before, if there is any doubt as to whether or not it is a business method patent, it is.
3. Obviousness and novelty should be decided by experts in the field who, as much as possible, have no stake in the matter.
4. You cannot re-patent a new application of the same idea in solely order to extend the life of the patent.
5. You must include a working model with your patent.
6. The patent office gets the same fees whether they grant patents or not.
7. The patent office employees get a bonus every year. For every overturned patent in court, the legal fees of the winner come out of the bonuses of those who were involved in granting the bad patent. (Bonus can be 0, but not negative).
8. Funding for the patent office stays with the patent office and cannot be diverted by congress or others for other purposes. Excess funding beyond a working threshold goes towards bonuses.
9. Patents are limited to 5 years.
10. You cannot patent life, or life processes.

22 replies on “How to fix the patent system in 10 easy steps”

Bravo! I really like many of your suggestions. The five-year limit may be good for technology-related patents, but for traditional items, the old limit may still work. I especially like #10, because so many pharma companies are trying to profit off medical treatments! Withholding treatment for profit is just not right!

#3 sounds easy to implement, but unfortunately peer review does have its own set of problems. I work in science research, so I know.

Wish it were the way you suggest! Not very likely though. Patents, which are supposed to encourage innovation, are in fact hindering innovation. There is hardly anything you can do which doesn’t infringe on multiple patents. Apparently, 20% of human DNA has been patented. I’ve had my DNA for over 50 years, which should give me prior claim. How long before you have to pay royalties because you fathered (or mothered) a child?

Pleasantly idealistic and wholly impractical.
Reads like one of those chain letters that go around claiming to be from Carlin or someone similar that tell how to fix the country.

Why shouldn’t be be able to patent software? I’ll give you that, say, Amazon’s one-click patent seems mighty obvious. But, suppose I spend twenty years doing research and I discover an general algorithm to solve NP-complete problems in polynomial time. Why shouldn’t I be able to patent that and charge purveyors of traveling salesman scheduling software to use my algorithm? Why should they profit from 20 years of my work just because my “machine” isn’t physical and is instead a set of configuration parameters for an existing, generalized, machine?

Other than “5 years”, agree with everything. Five years is too short for things that require lots of investment and risk, and are sold to John Q Public and not to the Pentagon. New drugs for example.

Nice idea… incomplete in practice.

First, even nowadays, 5 years is not enough. The original 17 years is short enough as it is to commercially exploit a normal invention. The problem with patents is WHAT is patented, and patent extensions… the standard patent length is fine.

#10 should be put up to #3. It’s the same concept as #1 and #2.

#7 sounds nice, but it would never work in practice. It can be MANY years before a patent gets overturned… many patent employees have left by then. This would ONLY be an incentive for the long timers, and a negative incentive at that… yuck. Negative incentives almost never motivate people to perform better. Heck, often positive incentives don’t either… see Joel’s comments on Incentive Pay: http://www.joelonsoftware.com/articles/fog0000000070.html

“5. You must include a working model with your patent.”

What about designs for what is either not yet technologically feasible or too expensive to manufacture? Do these ideas not deserve to be patented?

Bah, you can fix the patent system in one easy step. Abolish it.

Is there any proof, hell any evidence whatsoever, that the patent system encourages innovations? No, there is not. In fact, there is plenty of counter-evidence.

If you want to know what really encourages innovations, you might want to bother reading Sources of Innovation by Eric von Hippel.

Is there any evidence that patents even work to protect innovations? Yeah right. The only patents that work are chemical patents, period. (Biochemical patents are a subset of chemical patents.)

And the reason why they work is because nobody really understands chemistry, so you can’t engineer your way around those patents. And also, because patents covering chemicals never cover one chemical but literally thousands of “analogues” with unknown properties.

But again, if you bother to read Hippel’s books, it’s in there.

And as for the notion that patents reward innovators? Well, they’d first have to WORK in order to be effective!

As to the fool who asks “why should people profit from 20 years of my work”….

Well, why should people profit from General Relativity without having paid Einstein for it?

Why should people profit from Quantum Mechanics without paying royalties?

Why should people profit from DNA without paying the estates of Watson and Crick?

You don’t understand what drives creation and you will never create anything. Why should the whole world be held up to the standards of a desolate and impotent mind?

I wish everybody good health and zero intake of pharmaceuticals. Now, should any of us really need a good therapy for disease X, belive me that nothing beats good old greed as an incentive for research. I do dislike this situation, but it’s what we’ve got.

Yahoo for you, another patent rant that made it to the top of Reddit.

So, what’s the next step? How do we actually DO something about this, how do we make it so? How dow we protect our futures as independent software developers so not to be devoured by large corporations assuming ownership of simple ideas?

Seriously, it’s time to stop crying about the damn problem, and it’s time to start acting. Protests, patitions, letters, organized efforts to spread the word and fear of how patents can halt innovation and put the little guy out of business.

Has America just gotten lazy? Let’s organize, let’s fight this with more than our blogging.

There is no way the patent office is going to read a blog and start shaking in their boots. They just laugh at how small we are to them and how we don’t actually do anything but hide behind our keyboards and expect our “voice” to be heard.

We punish theft, don’t we? We don’t just demand that the thief return the money?

So I would suggest

11. Whenever a patent owned by an organization or individual is overturned in court, as punishment for theft of public intellectual resources, two more patents with the same owner, chosen by random lottery, are also immediately released into the public domain.

I think you overlook an important function of patent law – the disclosure of inventions to the public.

I agree that there are a great number of problems with the patent system which require improvement. However, firms should be encouraged to give up their ideas to the public, rather than keep them secret.

This extends to software and business methods and biotechnology. I will admit that certain inventions like Amazon’s 1-click are self-disclosing, so it would be impossible to keep that invention secret once commercialized. It would seem that the disclosure of that technology in a patent is rather worthless, because the invention is obvious on its face.

But there are countless other examples of technology that we could never appreciate or use unless published in some form or other. Academia or Open source are examples of disclosure forums. Patents is another.

Indeed, one of the goals of the patent system is to organize scientific information. This goal has not been emphasized well. Patents need to be written more straightforwardly, and addressed to real people, not lawyers.

I avoid the sticky question of moral judgment on the technology itself. But I pose the question of whether it is better to know about a life form technology with strings attached (in the form of a patent), or to keep the scientific knowledge private to the researchers?

The system is not perfect. It is improving, though, I think. Better search engines and user interfaces might help to improve the disclosure function of patent law. What do you think?

Yuri, one of the PUTATIVE goals of the patent system is to disclose information. But does it actually function that way? No, it does not.

The last place any innovator, user or manufacturer looks for inspiration is patent databases. Furthermore, any innovation that can actually be hidden is kept secret and protected by trade secret law. Those that are patented are those that CANNOT be kept secret.

Given these simple facts, the only possible response to your claim that patents is a forum for disclosure is *cough* bullshit!

Ben, what you should realize is that patents, except for chemical patents, really don’t matter because they really don’t work. So in a very real sense, the patent office doesn’t matter either. It only matters for chemical patents, which work.

The solution to chemical patents is to advance the state of chemical knowledge to the point where any bright chemist *can* engineer around a patented chemical. That would make chemical patents ineffective and the entire patent system irrelevant.

Richard, I take your point. But I’m not stating opinions, just asking questions.

Ask why so many large companies such as IBM, Apple, CA and so on are making donations of patent portfolios to Open Source. They are not keeping enabling source code secret. Is it because patent rights are strong? I think maybe: it is a credible argument that patents encourage disclosure of ideas, even to open source. The disclosure seems to be getting better, with good documentation sometimes included.

While I don’t doubt that there is a profit motive in there somewhere, the end result is disclosure of ideas, which is what I care about.

Yuri, not only is it not a “credible” argument that patents encourage disclosure of ideas (did you read anything I said?) but it is LOGICALLY ABSURD and EMPIRICALLY FALSE. If you start by assuming blatant lies, you will never understand the world around you.

Why is it that large companies are donating their patent portfolios to open source foundations? That’s really simple to understand *IF* you understand that patents don’

Yuri, not only is it not a “credible” argument that patents encourage disclosure of ideas (did you not read anything I said?) but it is LOGICALLY ABSURD and EMPIRICALLY FALSE. If you start by assuming blatant lies, you will never understand the world around you.

Why is it that large companies are donating their patent portfolios to open source foundations? That’s really simple to understand *IF* you understand that patents don’t work and what they’re *actually* used for.

Patents are legal weapons. That’s what they are, that’s all they are. Large company A has a warchest of patents which it uses against large companies B and C which have their own warchests. If startup company D doesn’t have any patents, then companies A, B and C can clobber it to death. Exactly like the USA with its vast military arsenal can clobber third world countries like Vietnam, Nicaragua, Korea and now Iraq.

Is that the reason why large companies have patents, to clobber poor dumb schmucks? Actually no. Large companies have patents as a defensive arsenal against other large companies. And just like any military or nuclear arsenal, patents waste a lot of resources to maintain and do nothing productive whatsoever.

If you understand that, then it’s trivial to understand why large companies are donating patents to open source foundations. In effect, they’re signing mutual defense treaties, in the hopes of cutting the costs of military expenditures in hot and cold wars (corporate lawyers and threatened lawsuits), possibly even in the hopes of cutting their military assets (patents).

Companies don’t WANT patents, but they have to have them as a cost of doing business. Now they’re doing something to try to get rid of patents and you talk about how patents must be a good thing? Your logic is tantamount to “everyone wants to give up their guns to make a professional police force, so guns must be a good thing, nevermind the murders”. Bah!

Yuri, if you’re interested in the disclosure of innovations, you might want to read Eric von Hippel’s paper on trade secret trading. (http://web.mit.edu/evhippel/www/papers/Knowhow%20Trading.pdf) Trade secrets are things that you actually CAN withold disclosure of, unlike innovations covered by patents.

And a word of advice, the next time you have a good indication that the argument you want to make falls in the ‘plausible but likely false’ category, you might want to disclaim it. You could have prefaced your argument with “Everybody says,” or “The party line is,” or some other such.

Richard, thanks for your interesting comments.

I would hardly categorize my ideas as “plausible but false”; instead, I would say they are open questions going to the heart of knowledge transfer theory.

You say that the claim that patents encourage disclosure is “logically absurd” but offer no proof. Instead, you refute 1 example and conclude the entire claim is false. I question the rigor of this approach.

What I was trying to say is that software firms are going above and beyond the disclosure required in a patent and donating enabling source code to Open Source (remember, source code is not required to get a patent). I theorize that this may be because of strong patent rights.

If patent rights were weak (in terms of enforcement), there would be a reluctance to give up enabling source code to anyone, because it could be used by any competitor once disclosed.

Thus, in this case, strong patent rights encourage disclosure to Open Source. If a firm knows that a court will readily enforce its patent, it is much more likely to donate it to a certain segment (Open Source) knowing that it is protected against commercial competitors.

Your view that firms are donating to disarm competitors’ patents is confusing. IBM is not donating its patents to Microsoft or vice versa. There is a phenomenon of cross-licensing of patents between large firms, but that is more likely anti-competitive behaviour of oligopolistic firms.

I welcome your comments, cheers,

Leave a Reply

Your email address will not be published. Required fields are marked *