OK, first of all, I’m not sure whether this even qualifies as a GQ or should go into Great Debates, but here goes.
I was reading some comments on an online column wherein a commenter asserted that there was “fairly strong evidence” that patents – actually, the column was about copyrights, but the way the comment read I think he was talking about patents – did not foster innovation but instead discouraged it. Does anyone know what this “fairly strong evidence” might be?
I’d ask the guy myself, except that:
a) The discussion had degenerated into a big ol’ mess I don’t want to read; and,
b) I can’t remember where it was, anyway.
RR
Perhaps this person has a novel theory of patent law. But I won’t believe it until I hear it.
Until then: patents encourage innovation because the exclusive rights a patent offers to inventors encourages research and development by promising a temporary monopoly.
Patents also encourage innovation when inventors “design around” existing patents, and thereby create a new and different product or process that fufills the need of the original product or process.
Patents also encourage innovation because they require that the details of the invention be published. If inventors were not given a temporary monopoly in exchange for publication, one way of protecting some kinds of inventions would be to keep them secret. Publication of inventions helps the process of research for further inventions, and also makes it easier for others to use the invention once the patent has expired,
In my field, software development, it is a very mainstream opinion that patents do more harm than good, for the following reasons:
Patent offices routinely grant patents for things that are obvious or non-novel. Thereby, it allows the patent holder to prevent other people from developing innovative software using a particular technique. Having to spend time on finding silly workarounds because the obvious solution to a particular problem has already been monopolized by someone else, does not foster innovation, it just makes it more difficult.
Patents are often granted (or interpreted afterwards) unreasonably broadly, so that the parent holder gains a monopoly not only on his innovative solution to a given problem, but also on all other programs which may solve the same problem in a different way. Example: patenting the generic concept of “selling pet food over the Internet”, rather than patenting a specific technical implementation of that concept.
The 18-year monopoly granted for a patent is felt by many people to be unreasonably long in today’s world, especially for software. Even if you are the legitimate inventor of a given technique, it is very unlikely that it would not have been independently re-invented many times over during the next 18 years.
“Submarine patents” are situations where somebody patents an idea, then sits on it silently while waiting for someone else to do the hard work of turning it into a marketable product, and then finally starts threatening lawsuits after the product has become popular and a lot of people have become dependent on it. Famous example: the Unisys GIF patents.
The fact that just about every common programming technique is covered by multiple patents, and there’s no way to find out whether a particular technique is safe to use, makes software development a risky business. In a world without patents, any smart teenager could develop a new product using just their brains and a cheap PC. But patents enormously raise the barrier-to-entry, threatening to turn commercial software development into a game that only large corporations can play.
It is by no means certain that much, if any, innovation in the software world is driven by the lure of patents. Generally, innovation happens in the course of developing a product, and the reason for investing in the innovation is the hope of selling that product. Patents are often more of a distraction, something you need to do because otherwise your competitors will use their patents to drive you out of the market. But I’m not aware of any software product which would not have been developed if the patent system had not been there to motivate the developers.
Like any other kind of monopoly, a patent is likely to make the monopoly-holder lazy and complacent, encouraging them to sit on their laurels and stop investing in new innovations, because they can use the legal system against their competitors, rather than being forced to engage them in an innovation arms-race which would be beneficial to consumers.
There’s some trouble with patenting genes and life forms.
IIRC, the Salk institute has brought lawsuits in order to stop development of some diagnostic tests which use genes they hold the patent to. No doubt, googling could turn up links to such actions by Salk, or others.
However, history shows that most of the time, innovations do not happen in a vacuum: once the time is right for a particular development, it is usually the case that multiple people are working in that area and come up with the invention more or less simultaneously.
E.g. let’s say that persons A and B are both working on a new type of battery. A finishes his work a few days earlier than B, so he wins the race to the patent bureau. He can now prevent B from working on this type of battery for the next 18 years, or demand arbitrarily high fees for it. So now, instead of A and B competing with each other in the production of these batteries, and continuously improving their products in quality or production costs, there is just A. A now has a monopoly on this type of battery, so he can raise his prices and can stop investing in improving his product because nobody else is allowed to produce similar batteries without his consent. (ETA) B has lost his entire investment, despite having done the same research & development work independently.
I worked for a company that sold domain names, and the VP started driving us to identify code we had that could be patented. The software was used directly in our ecommerce system but none of this code was part of a commercial product offering. It was an attempt to use patent law as a legal tool to deal with what otherwise would be legal competition. I would not say that any of these techniques were obvious, because we were the first company doing what we were doing, but all us technical guys thought the whole patent idea was a little silly. This was a little like Amazon patenting their One-Click buying, which I think they may have actually done.
I just wanted to correct a couple of factual errors in this list. I also disagree with it philosophically, but I think that’s outside the realm of GQ.
Patents do not last (and have never lasted, at least in this country) 18 years. The term used to be 17 years from date of issuance, and is now 20 years from date of filing (which generally works out to be shorter for all but the most trivial applications).
Submarine patents are applications which are filed and not prosecuted (or, more accurately, prosecution is delayed by various tactics) so that they do not issue for many years, but pop up as surprises once the industry has taken off. They no longer exist in most circumstances, as most patent applications are published 18 months after filing. (There is an exception for patents where the inventors declare at time of filing that they will not file in a foreign country.) Once the idea is patented (and usually earlier if it is published), the patent is freely available for anyone to read, including those considering developing in the technology area.
That’s not quite the way it works. If they are, indeed, that close, then B will finish his work between the time A files his patent and the time it’s granted (I’ve filed three-one took 18 months to be denied, one took 5 years of wrangling to be granted, the third is still pending after 2 years). Under those conditions, B would be able to continue using the technology, and might even be able to present enough prior art through engineering notebooks to be able to prevent granting of A’s patent entirely.
U.S. patent law specifically requires that the invention be novel and non-obvious. A patent for an invention that is obvious and non-novel will not stand up to a legal challenge.
I can’t see any argument for patents standing in the way of innovation at the conceptual level - the idea is essential. Patent law, and how patents are applied for, granted and administered - the nuts and bolts of the process in other words, gives more scope for argument. Some good posts on the software field and molecular biology already, I’m more familiar with the pharmaceutical industry patent area, and there are some controversial areas to do with innovation. A good example is that you don’t want to be the first person to discover a drug, you want to be the second.
It cost’s millions upon millions of dollars to invent a new class of drug from scratch, have it work, and most importantly of all, prove that it is safe. It costs considerably less for your competitors to read the patent and figure out ways of breaking it, coming up with their ‘fast follower’ drug. Lipitor, the world’s biggest selling drug ($10 billion pa for Pfizer) is an example of this phenomenon. This idea could be argued in terms of stifling innovation, both in terms of the copycats, who aren’t really inventing anything new, and in more general terms: Why try and show your ass by striking out into the great unknown when you can just come up with proton pump inhibitor #7 (for stomach ulcers and acid reflux), market it out the bazoo and profit?
Another example is dodgy granting of patents in the first place; the HIV medicine AZT has been fiercely debated in this context. IIRC it was discovered in the 60s at the NIH (ie with US tax-payers money), and patented in the 80s for its ‘new use’ by the pharma company BurroughsWellcome. I need to refresh my memory on this one, but I know it has been held up by advocates of patent law reform as a cardinal example of abuse of the patent system, and by extension the inventive process as a whole
Increasingly inane software patents are becoming a pretty widely-despised trend among software engineers and IT professionals. If you want a some highly sensationalized hyperbole on the matter, just read Slashdot on a regular basis. It’s not very objective (understatement of the year), but it should give you some idea of the range of attitudes generally held among the “innovators” in the field.
Like ENugent suggested, though, anything further should probably be taken to GD.
If the system worked perfectly, obvious and non-novel patents would not be granted in the first place. Legal procedures are expensive, even when you expect to win, especially when you’re an individual or a small company going against a large corporation. And even when the obviousness of a patent is clear to the people who work in a given field of expertise, it may be difficult to convince a judge or lay jury of that. So there’s definitely a barrier to innovation there, even if it’s not an insurmountable one.
In case it isn’t obvious, I do not claim to be a patent lawyer or legal expert. However, in answer to the OP’s General Question, I believe I have given an overview of the most commonly given objections to the patent system that are regularly discussed by software developers. Without a link to the column referenced in the OP, there is of course no way to know which specific objection was being discussed there.
I can testify however, that as a software engineer I have more than once found myself in a situation where we were asked to implement some feature, and after just a few minutes of brainstorming we came up with an obvious and straightforward way to do it, only to learn afterwards that our “innovation” had already been patented. Obviously, in such a case, the role of the patent system as an encouragement to innovation is zero, while on the other hand the hassle of having to deal with a possible legal threat to our work is very real. If we have to go with an inferior solution because we cannot use the obvious approach which we discovered independently within five minutes, how does anybody benefit from that?
That happened to us, too, but in one situation the patent was made by our arch-competitor, and in such a way that it would have killed our business model. I can’t get into the details for obvious reasons, but in this case we did take them to court and the patent was overturned on the grounds that there was prior art – in other words, their idea wasn’t new.
I worked for a Research Scientist for two years who had 20 patents with a major medical equipment supplier. Only two of his patents actually made money for the company (major money). When he would drink he would brag that the other 18 were “defensive” patents and say stuff like “Just think how much I’ve saved the company by stopping other people from working on this stuff!”. And he believed this was true. Just saying . . .
So how does that work in practice? Let’s say I have developed a new video compression technique which I believe to be original and superior to existing methods. What’s my next step?
If I just go ahead and attempt to productise my invention, it is virtually certain that I will be sued by a large number of people who hold patents on video compression, even if there is only a little overlap between my method and theirs. So I first need to search for patents I might be unknowingly violating. Of course, patents are written by lawyers, for lawyers; they are not intended to be understood by technical people. So I’m going to have to hire a patent lawyer, explain my invention to her, and then pay her a lot of money to spend a lot of time searching through the many thousands of data compression patents and investigating every one which could plausibly be stretched to cover my invention.
That’s a lot of time and energy I’m not spending on improving my product. In fact, I may even decide to give up on the idea altogether and go find another hobby. Can you see how one might have some doubts about whether this system actually encourages innovation?
Now, in a world without patents, I could turn my invention into a product suitable for end-users, which is usually the difficult part anyway, and try to make money off it in whatever way I saw fit. Simply keeping the source code secret would discourage casual copying, copyright would prevent people from blatantly stealing my product, and by the time my competitors had reverse-engineered my work and developed their own “clean-room” implementation, I would be ready to hit the market with version 2.0 of my product. Doesn’t that sound like a more fruitful climate for fostering innovation and benefiting consumers?
Also, in the case of the GIF patent, it was actually well-known that Compuserve had a patent on the compression technique. But they never tried to enforce it, and in fact did everything to encourage the adoption of GIF as a de-facto open standard. But then after many years, Unisys, the new owner of the patent, suddenly announced that they were going to take a hard line on “protecting their intellectual property”. In other areas of civil law, I believe (but IANAL) that this is called estoppel: if you allow something to happen openly and with your tacit approval, you cannot then suddenly turn around and sue for damages. But apparently, either the rules are different in the field of patents, or maybe Unisys would have eventually lost a court case but the mere threat of legal action was enough to cause some serious headaches for everybody involved in the field of image manipulation software.
That’s great in theory, but when it takes years of legal work and hundreds of millions of dollars to get a patent overturned, while in the meantime your business is shut down because you’re in violation of that patent, this is a rather cold comfort. To paraphrase, “I was right” makes for a really shitty epitaph.
Oh, yeah, no problem, just go read all seven million patents that the USPTO has issued and decide whether your product violates any of them.
There exist companies whose entire business model is either licensing patents they own or suing companies who break them. This is not fostering innovation.
The League for Programming Freedom has many links to articles and information about how patents can stifle innovation more than they encourage it.
In particular, I like this essay by Richard Stallman, in which he takes a typical software patent and shows how it does not really contain any technical inventions, but has been obfuscated with legalese to make it seem complex and innovative. He also makes the point that once a patent has been granted, there is a strong presumption of validity: the courts are reluctant to overturn them on the basis of obviousness.
Here (warning: PDF) is a paper by two economists from MIT, showing that in dynamic industries such as software and semiconductors, strong patent protection will stifle innovation rather than encourage it.
It’s not very objective (understatement of the year), but it should give you some idea of the range of attitudes generally held among the “innovators” in the field.