Have an Idea, Then Go to Sleep

We are told that patents protect the creator. I think they do the opposite. A patent doesn’t protect the person who creates; it rewards the person who stops creating. Have an idea, file it, and go to sleep: for the next twenty years the state will stand guard over your idea while you rest. Meanwhile, everybody else who could have improved it has to wait.

What really protects a creator is something else entirely: competition, the obligation to keep doing things better than everyone else. That, in two paragraphs, is why I’m against patents. Not only the software kind — all of them.

I believe in private property. I think it’s one of the foundations of civilization: the house you built, the land you work, the machine you paid for. What I don’t believe in is owning something you can’t touch. I’ve thought this way since 1993, when I started at the ETSIT in Madrid and first used free software in the school’s labs; by 1994 I was installing GNU/Linux at home, amazed that a few thousand strangers could build an operating system by simply sharing what they knew. Years as a Debian developer, and later as a business owner, have only made me more sure.

You can’t touch an idea

Property over material things makes sense because they’re scarce. If I take your bicycle, you no longer have it. Two people can’t ride the same bike to two different places at once, so we need a rule that says whose it is. That’s what property is for: settling conflicts over things that can’t be shared.

An idea is nothing like a bicycle. If I learn how you built yours, you still have your bicycle and your knowledge. Nothing has been taken from you. Stephan Kinsella built his whole case against intellectual property on this: property exists because of scarcity, ideas aren’t scarce, so treating them as property is a category mistake. (Fittingly, he waived all copyright on that essay.)

So what does a patent actually protect? Not anything of yours. It restricts what other people may do with their own property — their workshop, their materials, their computer — because they had the same idea you did, or learned it from you. That is what I mean by unnatural: it isn’t ownership of a thing, it’s a claim on everyone else’s hands.

Thomas Jefferson put it better than anyone, in a letter to Isaac McPherson in 1813:

He who receives an idea from me, receives instruction himself without lessening mine; as he who lights his taper at mine, receives light without darkening me.

And then, flatly: “Inventions then cannot, in nature, be a subject of property.” (full letter)

To be fair to Jefferson, he wasn’t an abolitionist. As Secretary of State he sat on the first US patent board, and in the same letter he accepts that society may grant inventors an exclusive right if it finds that convenient. But look at what that concedes: the man who helped run the first patent system didn’t think patents were a natural right. He thought they were a privilege handed out by the state, to be kept only if they did more good than harm.

A patent was never the opposite of a monopoly. It is one. The modern patent was born in England with the Statute of Monopolies of 1624: Parliament abolished the monopolies the Crown had been selling to its friends, and kept one exception — inventors could still get up to fourteen years of exclusive rights. Patents are the monopoly that survived.

And monopolies have a well-known effect on whoever holds them: they make effort optional. The reward of a patent doesn’t depend on what you do next. It depends on having been first to the patent office. Once you hold it, the easiest way to make money is not to keep improving your invention but to stop anybody else from improving it, and to charge rent to those who want to use it. The extreme form of this has its own name — the patent troll, a company that makes nothing at all and lives entirely on licences and lawsuits. That’s not an abuse of the system. It’s the system working exactly as designed, with the inventing part removed.

The economists Michele Boldrin and David Levine tell this story through James Watt, in their book Against Intellectual Monopoly (2008). The textbook says Watt’s patents rewarded the genius who gave us the steam engine. They read it the other way round: Watt was one of many clever people improving steam power, and once he secured his 1775 patent extension, much of his firm’s energy went into defending it, while competing designs — including the high-pressure engine — sat idle until the patent expired in 1800. Then steam power took off. Not every historian accepts this reading; George Selgin and John Turner published a rebuttal arguing the record is more nuanced. I find Boldrin and Levine more convincing, but read both.

Meanwhile, the rest of us wait

The cost of that sleep isn’t paid by the patent holder. It’s paid by everyone else: by the inventors who can’t build on the idea, by the companies that can’t use it, and by all of us who get the improvement years later than we could have.

Boldrin and Levine made the broader case in “The Case Against Patents” (Journal of Economic Perspectives, 2013): there’s no solid empirical evidence that patents increase innovation or productivity — unless you measure productivity by the number of patents granted. Patents and their legal protection have grown enormously over the last decades, and the burst of innovation that was supposed to follow hasn’t. They call it the patent puzzle. Their recommendation is to abolish patents, or at the very least to cut them back hard.

As an engineer, I see the problem most clearly in software. A program is mathematics written down in a particular way. Any non-trivial program today touches hundreds of patented “inventions”, many of them obvious to anyone who has worked in the field for a week. The result isn’t more innovation; it’s a tax on everyone who writes code, collected by whoever has the biggest portfolio and the most lawyers.

Europe had a chance to settle this. On 6 July 2005 the European Parliament rejected the software patents directive by 648 votes to 14, after years of campaigning by developers and small companies. And yet the European Patent Office keeps granting patents on “computer-implemented inventions”, because the European Patent Convention only excludes programs “as such”. The door was closed at the front and left open at the back.

The saddest example: medicines and the poor

Software patents are a tax. Patents on medicines, in the wrong place and at the wrong time, cost lives.

At the end of the nineties, the triple antiretroviral therapy that had turned AIDS from a death sentence into a chronic illness cost between $10,000 and $15,000 per patient per year in the United States and Europe. In southern Africa, where the epidemic was worst, that price meant one thing: almost nobody got treated.

In 1997 South Africa passed a law allowing the government to import cheaper versions of patented medicines and to promote generics in a health emergency. The response was a lawsuit: 39 pharmaceutical companies, counting their trade association, took the South African government to court to block it. The case dragged on for three years, while the dying went on.

Then, in February 2001, the Indian generics maker Cipla offered the same triple combination to Médecins Sans Frontières for $350 per patient per year — less than a dollar a day, provided it was given free to patients — and $600 to governments. Cipla could do it because, at the time, India didn’t grant patents on medicines themselves. Same molecules, same pills, a few percent of the price. The only thing that had changed was that somebody was now allowed to compete.

Two months later, under a storm of public outrage, the companies dropped the case unconditionally. In November that year, the World Trade Organization’s Doha Declaration had to state the obvious in writing: that patent rules should not prevent countries from protecting public health, and that they may issue compulsory licences to do so.

I know the counterargument, and it’s the strongest one patents have: developing a new drug is enormously expensive, and without exclusivity, who would pay for it? It deserves a serious answer, not a slogan. But notice what the AIDS story shows. The research had already been done and paid for. The patent wasn’t protecting the next discovery; it was protecting the price of the last one, against people who could never have paid it anyway. Nobody in Rome or Boston would have bought one box less if a clinic in Soweto had been allowed to buy Indian pills. The only thing the monopoly achieved there was to make sure those people got nothing.

Competition is what protects a creator

“But how will creators make a living?” I hear this a lot, usually from people who know I run companies. My answer is the thesis of this whole post: competition is what protects a creator. Not a monopoly granted by the state, but the daily pressure of having to do things better than everyone else — and the reward of winning when you do.

A creator who has to compete keeps creating. They execute better, faster and more reliably than anybody else, and then they have the next idea, and the one after that. If someone copies them, they’re already one step ahead, because they know the work from the inside and the copier only knows the result. That’s the only protection a creator really needs, and unlike a patent, it also makes the world better: every competitor who copies and improves pushes the creator to improve again. A patent puts the creator to sleep. Competition keeps them awake.

A name is a different matter. My projects carry a trademark, which protects something you can point at — a name — so nobody can sell their stuff pretending to be me. That’s protection against fraud, not a monopoly on an idea. The code itself is open, under permissive licences, and anyone can take it.

And if you really don’t want something copied, there’s an honest way: don’t publish it. Keep it secret, protect it with whatever technical means you have, the way streaming services protect their catalogues. That’s your right, and it costs nobody else anything. What you can’t do, in my view, is give an idea to the world and then ask the state to stop the world from using it — while you go to sleep.

You can’t touch an idea