« Reforming the Patent System Toward a Minimalist System-Becker | Main | Will U.S. Economic Growth Slow? Posner »



Feed You can follow this conversation by subscribing to the comment feed for this post.

Hanna Griesbach

I always like to read the views you both have on issues. There are a lot of complex issues going on with patents and for them to cover so many different industries. I think of the baby industry and all the products in that category alone! Thanks for a great read and thought provoking ideas.



Let's not forget the operational triumverate of Intellectual Property Law, "Patent, Copyright, and Trademark". All three operate in unison to protect not just individuals and companies financial and intellectual investments, but also National Economies and Trade Balances. Without which, huge International Blackmarkets develop piggybacking themselves to open freetrade avenues. Much like that which have developed to date in countries that do not have a legal history of Patent, Copyright and Trademark or the currently developing trade in cheap knockoffs that are now threatening various producers, damaging National productive capacity and National Trade Balances which are increasingly controlled by International Organized Crime Syndicates.

Perhaps from viewing the problem from this perspective, tighter Legislation is warranted. Not less.

Ralph Koettlitz

The theory that copyrights giving authors an incentive is a theory but not very good support by history. In many countries they can't publish their books without signing a total buyout contract. The consequence is, that many authors can't live from their books even with a good amount of sold books.

Eckhard Höffner presented in "Geschichte und Wesen des Urheberrechts" a good point, when he compared the history of copyright in England and Germany in the year 1800, when in Germany were every year published more than 4000 new books despite a non existing copyright, while in England the amount of published books was at 700. England had a copyright since 1710. Eckhard Höffner observed that the marked concurrency between the publishing houses were positive for the authors, because they wanted publish the books.

A future copyright should be a right of the author and not of a publishing house. Every copy is to be paid by a publishing house or a internet customer to the author. In our world of a global internet, this not really a problem. But the huge amount of orphaned books, films and other works of art is a shame for the western civilization. So many art has been forgotten, because of the unsure copyright situation.


Judge Posner, I am sure by now you have read Richard Epstein's response to your recent ruling: http://www.hoover.org/publications/defining-ideas/article/123926

Although Epstein voices his disagreement with your judicial opinion and personal views at a somewhat general level, it is plausible that if one zooms in on details, his argument may be solid.

It might very well be the case that without patent protection, it would be difficult to attract the various investors to a particular firm aiming to make and market a product. The counter-argument to that, of course, is that once the playing field is level - patents are either wholly eliminated, or eliminated from some fields - investors would have no choice but to invest in firms and products based not on the attractiveness of patents, but based on the marketing expertise and potential of a given firm.

However, this imposes its own costs on the system as a whole. It would surely create a state of flux, because uncertainty of marketing is still greater than a legal uncertainty of patent protection.

In sum, I think it's worth to examine potential costs on all sides of the equation and this is a debate that should be taken seriously not just by scholars, judges and other experts in the field, but by US Congress.

Yegor Jbanov

Biggy82, in software the problem is reversed. What scares away investment is fear of litigation, not fear of ability to protect your invention. It's not the idea that's hard to come up with, it's executing on that idea. If someone else is able to execute (code, test, deploy) on the same idea without direct copying of source code, kudos to them. You, as an investor, shouldn't fear that someone else will implement the same idea. It's hard. As a thought experiment, think of how hard would it be to create iPhone, Google Search, or Microsoft Office. Even with access to all necessary patents. Software startups fail not because someone steals their idea, but because of poor execution, poor quality or simply bad idea.

In pharmaceutics the hard part is to get the formula for the drug, not producing the drug. In software the formula (i.e. a specific algorithm, idea or technique) is the easy part, but producing a working piece of software is hard (see the ratio discussion in the blog post). Software is already protected by copyrights, trademarks and code obfuscation techniques. So it's hard to replicate software even without patents. You cannot (or not allowed to) copy source code directly. Yet these protections do not block competition or create entry barriers. Patents, however, take this protection way too far and are primarily used to:

- Eliminate competition - big monopolies, such as Apple, Microsoft, and Oracle do this to eliminate competitors to their own products.
- Extract licensing fees - patent trolls buy IP for the sole purpose of extracting licensing fees while not innovating themselves.

It's these two things that hurt competition and investment. Not lack of protection. But, patent holders, obviously, have an incentive to argue otherwise.

Disclaimer: I've had a quite successful 10-year software engineering career so far.

Account Deleted

When it comes to textbooks you might be ignoring the benefits ("one has to consider both the cost of inventing and the cost of copying"). After all nothing is more important than education,long term.
About movies that cost hundreds of millions, one can argue that the cost is a direct result of the current rules of the game, in a more reasonable world where nobody gets 20 millions for 3 months of work movies would cost far less.Also such movies are positive from an economic point of view but the tragic lack of quality might be detrimental to society.
Copyright also feeds a lot of leechers,while that has a positive economic impact,starving them would not harm creativity long term.
There are also alternative ways to create revenue that should be factored in.
Also a fundamental problem with copyright is that it's a monopoly in itself.

About patents,while i agree that there is a need for a reform,i will point out that an idea is not worth just what was invested in it, an idea is creativity, an idea generates revenue,those factors should matter.
Maybe one way to do it is for all patents to to considered essential and force companies to license them at a fair price and/or an expiration after 2 years from first revenue or 5y from filling.Forcing big farma to license at a fair price would not hurt either,after all it's not only about the money.
Guess in the end the calibration of the solution depends on one's humanity or ... political orientation.

Jonathan Brown

The primary purpose of patents is not to protect inventors, corporate or individuals or those who invest in them, but is for the preservation and self interest of the issuing State. Looking at patent law in abstract from the main purpose of them as instruments of state control is where much of the confusion about patents starts.

In a world that is increasingly technology dependent, global in trade and political influence, national interest and economic survival may have ever widening interpretations and consequences.

Patent laws remain parochial in form, archaic in structure and are used to underpin protectionism.

Przemysław Lib

Biggest problem of US patent system.

Nobody measure cost of checking if product infringe any patent.

This cost can only rise (more patents are being granted not less, and that is goal as patents == innovation in USA legal system :( ).

But nobody check if cost of checking if your product is on the green side is still affordable!

So we have all those patent reporting sites who warn readers that in given post they will talk about some patent! So people who my have products that infringe can IGNORE such text, and look other way.

And those triple damages. If nobody knows if cost of checking is still economical, then how anyone know if triple damages have effect they should have?

And best of all.

Triple damages for trying and failing, when succeeding may be bluntly IMPOSSIBLE.

Judge Posner if you want to start serious talk about patents talk about:

1) Costs of checking if product infringe
2) Costs of defending litigation based on invalid patent (defending fees vs license fees on INVALID patents)

While talking about influence of lifespan of patent on innovation pace in current industry is good. Its not the biggest problem by far.

Stefan Wundrak

Why not make the granted time during which an invention is protected directly dependent on the investment that was necessary for this invention?

Elliott Chenoweth

The only fair implementation of copyright and patent law is none at all.


Both copyright and patent laws are being abused by players that specialize in extorting money from consumers and corporations.

Software Patents exists on scale largely because IBM (which used to be known for strong arm tactics using patents against rivals in the 70s), filed a whole bunch of them before patent officers had any experience of software at all. Later, as software industry matured, it seems like the law was reformed to justify IBM's patents in hindsight, with the patent lawyers advising. I can't imagine there were any software developers or entrepreneurs present, so the patent law framework became a jobs program for lawyers.

Patent protection of drugs make sense, but not to this extent. Recovering 100s of millions of invested R&D is not the same as making 100 Billion from a lucky break. Drug research is mostly about trial and error and gambling, backed up by some clinical grunt work. They should get paid back, but not have 7 years to hold the world hostage if they get lucky and find a cure for cancer. The investors in that company would insist on full Patent protection and then sell the drug for the equivalent of what the patient has in saved assets.
Let drug companies recover proven R&D expense + a 30% premium and then put drugs in public domain.

Copyright should be applied to creative artwork where there is an individual or a limited number of defined authors. If there are many, maybe the time length of protection should be divided by number of participants. The law no encourages the media industry to create any kind of industrially engineered garbage concoction in search of a hit, and then when a hit is found then the world is milked and milked and milked for money. To give machine generated output full copyright protection is a biz odd.

The intellectual property (IP) industry has become an extortion racket where we fist sell people copy tools like the computer and the copy machine (and charge copyright and patent fees) but then sue them when they do copy... It seems like the aims of the IP lobby is to get all of the household budget that people have left after paying for rent and food.



Just to add another comment about software patents since I have worked in the industry since 1983.

The software industry was extremely vibrant and competitive in the period 1983-93 during which there was literally no patent protection. Products adopted ideas from each other left and right, and nobody thought it was unfair. Innovation was thriving and the players all made good amounts of money.

Then when MS was about to split from the joint venture with IBM around 91-92, I guess the IBM lawyers got involved and as part of the divorce deal wanted a patent swap. We (I was there and reviewed some IBM patents for Word) at MS said "holy shit we don't even have any", and MS started a massive filing campaign, filing for every imaginable software feature we could think of.

Then soon after MS got hit with a $100m jury verdict for patent infringement on STAC's compression technology for DOS files. The jury and judge were all asleep during the proceedings as lawyers debated the meaning of machine language instructions...but decided that infringement had occurred.

After this, the patent war was on, and every tech company has been filing for patents on every imaginable line of code they write.



This article wrote, “Further impediments to effective patent policy in the software industry include a shortage of patent examiners with the requisite technical skills…”

Isn't this the same thing as saying that the Highway Dept can't hire enough patrolmen, so we ought to abolish speed limits and other laws subject to judgement, such as “weaving” or “distracted driving?”

The laws are meant to be enforced; if too many patents end up in the courts, we have merely chosen a rather expensive option. And that choice, too, was made by Congress, which is (still, last time I looked) responsible for writing the law.

Martin Langhoff

As a software and hardware practitioner with many years in the industry, thank you for such a clear and succinct exposition of what is wrong with patents in the technology world.

The current approach of USPTO of defaulting towards approval and letting the courts settle the matter makes the situation much worse; piling a huge uncertainty on anyone working in this industry. Yes, you can be sued anytime, and it will cost millions to sort it out; even if you prevail in court the lawyers' bill will likely sink your venture.


Here in the USA, we live in an environment where traditionally there have been patents and copyrights. This legal tradition has encouraged a culture where innovation is valued, because innovators have reaped the rewards of innovation. In cultures where innovation was not rewarded, innovation was not valued in the same way. As a result of the culture established here, due to strong intellectual property protection, the USA has traditionally had the best science and engineering in the world.

Software people cite the early progress in the software industry, when patent protection was either not available or not sought, as evidence that patent protection is not necessary. This is a fallacy.

First, those software people worked in this culture that had engrained encouragement for innovation that came from our history of intellectual property protection.

Second, there was quite early on a strong move for at least copyright protection to help out those innovators. Early copyright cases quickly started talking about protecting "structure, sequence, and organization," to try to extend protection as broadly as possible. This was a bit of a stretch in the law, but it showed a recognition of the importance of rewarding those who created economic benefit to the country.

Problems with software patents have continued because at first the United States Patent and Trademark Office refused to hire patent examiners with a computer science background, so poor art searches were performed.

More problems have been created because the original Supreme Court case on this topic, Gottshalk v Benson, has severe logical flaws and a nonsensical result. I encourage people to read my brief in the Bilski case for more discussion of that. My brief is up on my blog.

The nonsensical opinion in that first case has resulted in extensive legal uncertainty and much litigation, leaving the entire field of patent protection for software unsettled for almost half a decade.

Have nots always want to take from haves. People who don't have money want to take money from people who do. People who don't have houses want to take housing from people who do. People who do not have intellectual property want to take it from those who do. I find this ethos repugnant.

Patents need to be whole heartedly endorsed by statute and the entire line of cases stemming from Gottshalk v Benson needs to be overturned.

Moreover, the idea that mathematics is not an invention also needs to be overturned.


I am a software developer from New Zealand, working alone on my project for 3 years, and just prior to release I was sent a patent infringement threat by a US company for technology that is in the public domain since 1998.

I have consulted an attorney and he tells me their patent would not stand, but I have no funding for such a legal battle. My software is very useful for the scientific community and general data analysis, and contains many improvements over other similar software which I am quite proud of. I now fear 3 years of my work have gone down the drain for nothing, and I feel powerless and desperate. It is very unfair. How am I supposed to start a business at all?

Perhaps non-US software developers should avoid selling software to US clients at all, I know some researchers in this general area are starting to put up notices that they won't sell software to US clients, due to fear of patent infringement. Unlike promoting innovation, software patents seem to be preventing innovative solutions from reaching the US.

Thank you for your blog post, hopefully a few policy makers will consider your views.


Although I agree with the general sentiment here, I feel there's not enough examples of how the current patent system goes wrong on a day-to-day basis. Only the big stuff makes the news, like Apple, Motorola, etc.

- As a computer programmer, I'm scared to start my own business developing software. You might say, "Oh, but look at Facebook and Google, who got started in the same atmosphere as you." That's true, but they're just outliers. Most software companies don't make it, and I'm terrified of getting my ass sued.
- I once worked for a huge corporation that did lots of legal publishing. There were rewards for doing something that led to a patent ($5000 for each patent, IIRC), but we were not allowed to discuss old patents in writing. It was all about plausible deniability, in case the patent went to court. Very sad...
- It's not as theoretical as "should patent protections be stronger or weaker?". There's a lot of crap patents out there, especially in the tech world. The Apple vs. Motorola, etc. news has kinda hit that point home. But there's even patents on swinging (think kids on a playground) and on peanut butter and jelly crustless sandwiches; thankfully it was overturned /six years/ after it was granted. Now just imagine what kind of crap a PhD could think of that the USPTO would overlook.

- http://www.google.com/patents?vid=6004596
- http://www.msnbc.msn.com/id/7432980#.UGsNGPk-uAQ


Software patents are pure spam for software developers.

You cannot tell a software developer what to write and what not to write as software via patents, since his speech is protected by freedom of expression and the first amendment.

And software developers HAVE TO CONSTRAIN their speech if they want to respect patent law.

Sorry, this is not consistant.


Diehr v Diamond should be overturned.

Algorithms should not be patentable - they are part of mathematics.

Mathematics wants to be free ...
Software wants to be free ...
Ideas want to be free ...

It is only concrete representations that should be subject to patents and/or copyrights.


Four aspects of patents you didn't touch on are: obviousness, prior art, disclosure, and working models.

The current interpretation of obviousness is: nothing is obvious and everything is patentable even though it's painfully obvious to an ordinary practitioner of the art. This is a consequence of something you DID touch on: the inexperience of patent examiners in the software field.

If I were a patent examiner (and I doubt the USPTO's salary range overlaps mine, so that ain't gonna happen), then anything I could google for would be obvious. Anything taught in colleges would be obvious. Anything you can find in a textbook would be obvious. This is not currently the case.

The next is prior art. There are several patents which were issued that overlap with previously issued patents. If the patent system cannot even find prior in its own database of ideas, how can it possibly keep up with the entire world? Basically, prior art is ignored as a criteria for unpatentability. That means that you can write a program, publish it as open source software, and somebody can use it to write a proprietary program, and patent your prior art. This is not a hypothetical -- it happened to me.

In addition, there is also the issue of disclosure. The patent system rewards disclosure with a temporary monopoly. And yet software can be both patented and copyrighted. This makes no sense. How do we allow people to claim a copyright on something which must be disclosed, and disclosed so that after the patent expires, people can reuse the idea.

And then we also have working models. People used to have to submit a working model, illustrating the operation of the patented device. It's possible to get a software patent which doesn't even work! A software patent MUST include a working stand-alone implementation which is in the public domain.

All of these problems should be fixed, or, better yet, just make software unpatentable.


@WaltFrench "Isn't this the same thing as saying that the Highway Dept can't hire enough patrolmen, so we ought to abolish speed limits"

No, this is a trite and ignorant analogy. There are laws which protect people from actual harm and of course these should be enforced. Then there are laws which setup rules with the intent of balancing the mechanics of various activities and these laws are utterly amoral. IP law is supposed to strike a balance, but as it currently stands it fails to achieve its intent and needs to be scrapped.

In fact, as Judge Posner clearly understands, there is zero evidence that patent law promotes innovation at all outside of the pharmaceutical industry. It is human nature to innovate, we don't need pieces of paper from the government to encourage it.


While digging into these so called "inventions" by Samsung ( and undoubtedly many many others) I can only conclude it does not concern inventions at all. Like their description of signaling in a mobile communications device. I mean comon guys, EVERY university student in ICT or communications technology has done the same thing YEARS before Samsung or any other company did it. These are not even inventions but simple ways of getting things done. It is how every graduate would do it if you asked him to write software for a cell phone. If I look at what is patented, I might as well patent "the technique and apparatus to connect two or more electrical devices by means of a medium that transports electricity above or sub terrane" In other words patent the way you would dig a frakkin hole in the ground to throw in a damn electricity cable to get light in your wooden cabin in the back yard! Is that truly an invention???

Phillip Helbig

"You cannot tell a software developer what to write and what not to write as software via patents, since his speech is protected by freedom of expression and the first amendment."

Really? Where did you get that idea? Even Stallman, whom I loathe, can and does distinguish between "free speech" and "free beer". The idea that copying something one is not allowed to copy is somehow allowed because it is free speech is so stupid that it wouldn't be worth commenting on were it not for the fact that many people seem to be so deluded as to actually believe it.


@Phillip Helbig

If you are right, then programmers have to refrain from writing certain instructions?



Do you write code, or do you write patent claims?

Maybe you should spend more time learning how to write code.

The comments to this entry are closed.

Become a Fan

May 2014

Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31