Software Patents are Not Good Property Rights

A good system of property rights establishes clear borders. Clear borders reduce disputes, encourage investment and promote efficient trade. Software patents, however, often fail to define clear borders. I am one of the amici in a amici curiae brief to the Supreme Court (regarding Alice Corp. v. CLS Bank) on software patents that makes this point:

Such abstract claims as “displaying data in
frames,” “recommending media based on past choices,” “reproducing information in material objects at a
point of sale,” or, as in the present case, using “a third
party . . . to eliminate ‘counterparty’ or ‘settlement’
risk,” simply cannot be reliably construed to define a
reasonable area of covered technology. See Wang, 197
F.3d at 1379; Interactive Gift, 256 F.3d at 1323; Pinpoint, 369 F. Supp. 2d at 995; cf. CLS Bank Int’l v.
Alice Corp. Pty. Ltd., 717 F.3d 1269, 1274 (Fed. Cir.

A general counsel at a technology startup
would be hard-pressed to describe any concrete
bounds or permissible follow-on innovations to her
fellow engineers in the face of such claims. Any
software that resulted in a similar functional result
could be construed as infringing, and any investment
in the commercialization of those technologies could
inevitably carry liabilities, risks, and costs whose
magnitudes are impossible to predict in advance.
Thus, the property system that ostensibly exists to
assure investors that long-term rents are secure does
the very opposite, casting a pall of uncertainty over
the viability of any commercial product that happens
to be adjacent to a lurking abstract claim.

Eli Dourado and I note that the Federal Circuit seems to have quite willfully disregarded the intent of the Supreme Court regarding patents on abstract ideas and I think this case may provide further pushback from the SC.


I suspect regulatory uncertainty may be a big part of the problem. Take a new bundled technology that requires 100 sub technologies. Any one of these can potentially take all your revenue under the sort of plausible argument that you couldn't do it without them. Will you bet against a court doing this? Now put all the patents in possession of lawyer trolls rather than inventors. Solve for the equilibrium.

Patents are not being used by honest inventors to protect their brilliant innovations against copycats. They're being used by big companies to tie up obvious solutions to narrow problems so that they can edge their competitors out of the market.

Patents are not being used by honest inventors to protect their brilliant innovations against copycats

How do you know that?

Look at any of the parade of high-profile lawsuits between various phone vendors. For example, Apple vs. Samsung: sued them (and won) for using a "rounded rectangular shape" in their phones. (There is nothing innovative about a rectangle.)

No, it wasn't a patent on "rounded rectangles." Lots of people make things with rounded rectangles and aren't sued. If you are saying that's what it was, you are either incredibly dense or being deliberately misleading to others.

For anyone else who cares, there was a long list of things that led consumers to confuse the Samsung device for the Apple device. Including an oblong and proprietary charger, which is an anti-feature -- there is no way any engineer would say "customers really want something that doesn't fit any of their other USB devices but looks a lot like Apple's proprietary connector."

Wait, why did you change the subject from software patents to design patents? Are you gish galloping or mood affiliating?

Damn, I even let you distract the issue even more from my question:

How do you patents are not being used by honest inventors to protect their brilliant innovations against copycats?

You post your argument that they are being used illegitimately, but you haven't said how you know that they aren't being used by legitimate inventors?

Patents are no doubt used by legitimate inventors to protect their ideas.

But face facts. They are also used by patent trolls to intimidate small companies into settling non-meritorious case because litigation costs are so high. Patent cases are among the most expensive types of litigation.

Further, small companies who sell to big ones are sometimes dragged into cases because they have warrantied that their software product is non-infringing.

Part of the problem - a big part - lies in the patent office - which seems far too willing to grant patents. One-click shopping? Really? That was "non-obvious?" But part lies in the costs, and I'm not sure a loser pays rule will solve that.

Oh, I can think of a number of reforms for software patents. They are by no means an unalloyed good.

But I also don't want them to disappear completely. I won't form any coalitions to reform patents with someone who thinks they are an unalloyed bad. If those people can be kicked to the curb we can get to the hard work of establishing differences between the trolls who have no intention of ever practicing with those who need patent protection to create their work.

I worked in software for 25-30 years before retiring. We copyrighted our work. To my knowledge we didn't patent at any shop I was in. I was aware though that we were skirting some "patterns." Consider plug-ins. Someone in the deep past discovered that you could expand a program at runtime by pulling in a bit of script or code. We know it was in Emacs way back, right? And yet the USPO accepted an asinine argument that each time the plug-in pattern was used, it was a novel invention and patentable.

That's not the worst of it, it is really just a special case of a broader problem. In the years I worked, 1980 to 2000 in particular, a lot of new hardware was released. The first programmer handed new hardware did a lot of simple and straightforward things. But guess what? Because he was the first one with the hardware, each and every simple and straightforward thing he did was also novel and patentable. It was insane. And it continues a bit as the first programmers given new phone hardware do natural and straightforward solutions for them.

The ONLY defense for this is that in the long run patents are dead, but I agree with what I think is Alex's core premise ... that we are leaving a lot of innovation on the table by blocking 2nd entrants from doing simple and straightforward things.

I should say "And yet the USPO accepted an asinine argument that each time the plug-in pattern was used [in a new problem domain], it was a novel invention and patentable."

Dan, what is an example of a "good" software patent?

I can name one that you might like. At first glance the FAT32 file system looks like a novel invention, right? But I'm going to say it's not. When FAT32 was invented file system design was well understood. No one was making novel improvements. In fact FAT32 is just a random configuration in a known idea space.

Yet MS has used the patent to block competitors from communicating with their systems, right?

The patent has failed in Germany - 'A Microsoft storage patent that was used to get a sales ban on products from Google-owned Motorola Mobility in Germany has been invalidated by the German Federal Patent Court.

Microsoft's FAT (File Allocation Table) patent, which concerns a "common name space for long and short filenames" was invalidated on Thursday, a spokeswoman for the Federal Patent Court said in an email Friday. She could not give the exact reasons for the court's decision before the written judicial decision is released, which will take a few weeks.'

If we'd had that kind of strong willed patent system in the US, that was willing to call average work "invalid" I'd be more on board with it.

(I think the current lay of the land though is that the USPO will never tell a US major corporation that their work is average, and invalid.)

The first programmer handed new hardware did a lot of simple and straightforward things. But guess what? Because he was the first one with the hardware, each and every simple and straightforward thing he did was also novel and patentable

"Novel" is required but not sufficient for patents. Someone may have told you "this is new so it is patentable" but, on the books, it's also required to be non-obvious. (And "useful" which is a term of art meaning it does something.)

RSA is a good example. You need the algorithm to be publicly viewable so that academics can debate and evaluate how well it works, and it wasn't going to be recreated by some guy in his mom's basement as the obvious way to do things.

The first programmer handed new hardware did a lot of simple and straightforward things. But guess what? Because he was the first one with the hardware, each and every simple and straightforward thing he did was also novel and patentable.

New hardware is part of it. More potent hardware is another. That means you start to develop new applications, because it suddenly becomes practical to do certain things on a computer. And while it may be difficult, in the sense of taking time to write the code and get it working, there is a difference, to my mind, between that and "non-obvious."

Apologies, byomtov, that paragraph of my comment should have been in italics. I was quoting a previous commenter on this page.

It's not the case that all software patents are accepted. I personally have witnessed some fail (that I'm under NDA about). Companies tend to not apply for things that won't make it merely out of self-interest to not waste their time and money.

Dan, I gave you the plug-in patents and FAT patent as examples of how bad things have been.

It took 10 years and the deepest pockets possible to finally defeat Eolas.

"Hey look, I put a plug in in a browser" should have been laughed out of the USPO at first sight when submitted in 1999.

"Plug-ins appeared as early as the mid 1970s, when the EDT text editor running on the Unisys ..."

EOLAS is an example of a high profile patent that should have failed on the basis of obviousness and prior art.

The NTP v. RIM litigation should have been dismissed on the grounds of obviousness and prior art.

We don't need software patents to encourage innovation and the need to license dozens or hundreds of patents for a new product is a serious impediment to progress. Whatever benefit we get from software patents is outweighed by the cost. The golden age of software progress took place before software patents were recognized - that should tell you something.

"I put a plug in in a browser”

I agree with you. In the early 1990s I bought an SDK from Lotus for 123 to develop "add-ins" for 123.

The license specifically prohibited using the SDK to develop add-ins so databases could be viewed through the 123 interface. That would be a plug in and no different from adding a plug in to a browser to permit viewing of remote data. Around the same time I purchased Magellan which had resource files that permitted the user to view a wide range of file types through Magellan. If you wanted to add a new file type you added a new resource file telling Magellan how to read and display the new file type..

Noooooo!!!!!!!!!!!!!!!!!!!!!!!!!!! This Alex amicus is WRONG WRONG WRONG! How can anything be so WRONG?

Points that are wrong: too many to even count. But I'll just highlight four [two] (btw I'm not a lawyer, but I probably know more patent law than Alex):

1) the amicus brief fails to mention copyrights, specifically, non-literal copyright infringement. What this law says is the "look and feel" of a copyrighted program can be protected. And since source code can be impossible to decipher, and in fact the Copyright Office last I checked only wants the title page and the last page of source, what this "non-literal copyright infringement" law says (and btw the influential 2nd Circuit of NY state ascribes to this copyright law; the Boston patent people who wrote the amicus should know about it) is that any software developer who has been exposed to the source of a program, even if that source is thought to be 'open source' (which nowadays is easy to do, check out the Michael Lewis piece on this: and if then that s/w developer develops a program that copies the "look and feel" of the copyrighted program (yes, 'look and feel' is not defined very well), that dev is guilty of "non-literal copyright infringement" even if the dev did not copy and paste the entire source, and in fact even if that dev independently derived their program. So essentially copyright law is 'worse' than software patent law. At least software patents give better notice to potential infringers than a copyrighted program, especially the way the Copyright Office works with source code (see above).

Question for AlexT: why didn't your legal eagles of Boston U mention this non-literal copyright infringement law? As a fallback to software patents that is worse than software patents? Remember, the purpose of an amicus is to aid the court, as a 'friend of the court'! With friends like you...who needs enemies?

Second, the 'troll' argument AlexT makes is expected to be moot, when and if, this year, Congress passes what is known as the "English Rule" where the loser of a patent litigation has to pay for the winner's legal fees. That means if a troll does not have a strong case, they'll be wiped out by this expected new rule.

Question for AlexT: why didn't your legal eagles of Boston U mention this new, expected to pass, "English Rule" that will make trolling moot? Remember, the purpose of an amicus is to aid the court, as a 'friend of the court'! With friends like you...who needs enemies?

Third, AlexT and pals make the howler along these lines: 'since we have had innovation even when software patents were weak, we don't need stronger software patents'. Howler! A grandmaster overlooking a mate in one. I guess I could say since slavery has been around since age immemorial, why do we need to abolish it? Besides, if they were not slaves they'd have been killed in battle, wasn't that the rationale behind slavery? Howler!

Fourth, the signatories to the amicus are a bunch of nobodies. Never heard of any of these people. If you wanted anti-patent law professors you should have gone to Berkeley, but the only person from there is: "Brian W. Carver, Assistant Professor, University of California, Berkeley, School of Information". Say what? School of what? Info? Assistant Prof? What is this, a teaching assistant? Lame!

I'm sick. Physically sick. No I'm not. I've seen this kind of anti-patent nonsense before. I just hope that the Supremes don't fall for it. It's lame stuff equivalent to a freshman in Econ 101 saying: 'property is theft!' and 'free trade is lifeboat economies that retards developing countries!' (there's a small grain of truth in both these statements, but they are essentially fundamental errors, along the lines AlexT and company make with their software patent arguments).

Since software is so much easier to bring to market than, say a semiconductor or a new tire tread design. The period of copyright and/or patent protection should at the very least be shortened.

When SCO bought the rights to 45-year-old Unix technology and tried to sue corporate Linux users, that was a gigantic, expensive, waste of everyone's time. That kind of garbage needs to be barred from our legal system.

That English law rule could actually make things worse. Not all software companies have millions hanging around to fight patent trolls . If you can't afford to defend yourself do you not lose by default?

Yes, while the American system has lots of problems, the British system has different problems.

fyi, the nobodies Ray says are signatories include Bronwyn Hall, Eric von Hippel and Nobel prize winner Eric Maskin.

Oh yeah, my wife and I were just talking about the three of them over dinner last night. I think they were on the front page of People.

You would be more impressed if J Lo were a signatory? :)

Thanks Alex for that amplification; though my "nobodies" barb was a sort of cheap shot and the weakest of my three points. As for not mentioning non-literal copyright infringement, I guess your Boston U friends can say this copyright law is sparingly applied and is a sort of paper tiger, that looks ferocious but usually does not prevail. Here is a scholarly article from 2006 on it that's probably still current today:

Still, the alternative to patents is trade secret and copyright, incl. non-literal copyright infringement, which is arguably worse than patents (that is, getting rid of software patents will merely resurrect non-literal copyright infringement claims IMO). Also the fact that everybody who programs (including me) copies other people's stuff is not IMO a good argument against software patents. A better solution might be to make software patent examination more rigorous, or abolish software patents but have the government offer a cash prize to worthy software inventions that are then dedicated to the public free of any license. I agree with you that the current software patent system is broken, and have said as much myself.

The fundamental problem of look and feel IP is this "new hardware" problem I mention above, or to put it differently, new territory.

Say you give me the first prototype autoteller box, and I program it in an average and straightforward way. Our patent system says I must be a genius, and must have just made a great invention, because no one has ever programmed an autoteller box the way I just did.

My company may now block 2nd entrants from programming their autoteller box in the straightforward way, and must now force them into their own "novel" solutions. Repeat, and software gets worse and worse.

@ john personna--you are behind the times, you are referring to "novelty" as a sole grounds for invalidity. That _used_ to be the law, but since the Supreme Court case of KSR vs Teleflex (, the "obviousness" test has gotten stronger, and patents are being invalidated more and more on this standard. But you do have a point though, since most people don't try and invalidate patents on obviousness since it's hard to do, so you are kind of correct even today, from my understanding of the law (again I'm not a lawyer but I know a lot about this area).

I hope I am behind the times, and that something like real invention is required these days. I did see that the Apple attempt at a multi-touch pinch patent went down, as it well should have done, with multi-touch and pinches going back decades ... but how close a thing was it? And how many of these dueling phone UI patents are in fact obvious?

US Patent 8,451,232 looks like a classic example of my "new territory" scenario. Since Apple got into multi-touch a little earlier than other commercial vendors, they could just go out and design a bunch of obvious junk, and then seek patents. Wherever they extended from the fairly well trod but core and general academic work on multi-touch they were mining a "new territory" and could stake claims with wild abandon.

"It isn't a problem because look what we did to address the problem."

I had to laugh at Tyler's use of "her fellow engineers" in his brief. Anyone in technology knows the odds of every saying that in real life is fairly low. But, if you're trying to tickle the fancy of the two lesbians on the court, it makes sense.

If the title didn't make clear who wrote this post, the byline should have.

The word "pedantic" comes to mind, but that's probably too generous. "Obtuse" is a better term. But, missing the point is something you good at so you have that.

@Z -do you dream of tickling the fancy of two unattractive lesbians? Then there's a porn-o movie for you, on Piratebay, that can be downloaded for free. Search for "mature lesbians".

Being asinine is a not the same thing as being witty. Maybe someday you'll understand that.

I've never understood what was so wrong with "their.". However computer software development industry " engineers" are not exactly the same as engineer. I know several females working in development.

Maybe Alex was using "fellow" to mean "dudes".

A I hate software patents (see above), and business methods patents, I wonder if there is some clean line we could make.

Can we drop all abstract patents? Should patents only be for physical devices?

That might force us to free algorithms ... but that might be an acceptable, even beneficial, change.

There will be no clean line as long as the Federal Circuit has its current composition of judges. As Alex notes, the Federal Circuit is quite willfully disregarding Supreme Court rulings on the matter. Looking at the backgrounds of those judges, this should come as no surprise. The court is packed with patent lawyers.

After all, "It is difficult to get a man to understand something, when his salary depends upon his not understanding it!"

I guess I'm really saying I want a new patent bill in congress ... but that would require a bit of a more workmanlike congress than we have now.

@Komori--the real money in patent law is not prosecution, which is what you are referring to, but rather patent litigation: it's over 100x more profitable in the latter, literally. Statistics tell us that 40-50% of patents are invalidated, and 50-60% of patents are upheld. Regardless however, both sides in litigation will be paid. Unless you are proposing that we change the constitution and abolish all patents, you will still get patents litigated and that's where the real money is at. So your theory fails. In partial defense of what you are trying to say (I think) is that long term it has been said that what drives lawyers to create law is uncertainty. Lawyers hate 'bright line' tests, meaning it's easy for a lawman to figure out who is right and who is wrong, so you don't need a lawyer. Hence the laws, written by lawyers, become more and more convoluted, not producing a 'clear result'. In any event, abolishing patents or making them stronger will not change this trend towards complexity and obscurity. What you need is what they have in Germany, a cheap and fast court to decide patent matters. The USA needs such a 'rocket docket' type system, and I'm not talking about Texas and Virginia, which have a reputation as being fast and cheap. We need a really order of magnitude better court system. Case closed, I have spoke.

@Ray Lopez responding to Komori--I see you are not talking about patent prosecutors, who get patents procured, but rather patent judges, sorry my bad, so in fact you are making the same point I make: it's been said patent judges make the law complicated so they are guaranteed to have a job. Anyway a faster, cheaper 'rocket docket' court would help, even if the law was complicated, since in the USA 'discovery' (pre-trial legal maneuvers) is where money is wasted during litigation. So a faster trial date would cut down on this discovery.

I wasn't referring just to prosecutions. I'm well aware the only definite winner in any given paten litigation are the patent lawyers (on both sides). But the current court is made up of people who filled that role, and who made huge amounts of money in that role, and see nothing wrong with that role. It'd be nice to have a few patent skeptics on the scene.

The other side of the problem, of course, is the patent office. Since companies can refile as often as they want until some patent examiner lets it through, and in patent trials the burden is on the defense, the system is designed to fail. A system like Germany's would help, but given US history, it'd be another ten or twenty years before we'd end up right back where we are now. Unless there's some real penalties codified into law for bad (obvious, non-novel) patents, both to the filer and the patent office, this is just not going to get better.

I don't think we need to get rid of all patents. But I think the only way we can possibly realistically reform the system in a way that would work for any significant duration is to flatly (and legislatively) deny patents on anything non-tangible. As in, if you can't deliver a physical model to the patent office, you don't get a patent (and if you deliver a physical model, that's all that's covered). Anything less will be gamed, and our government has shown a great willingness to participate in the gaming.

Yeah I've said the same thing about patent models, which were abolished in the 19th century, and if re-instituted would cut down on so-called paper patents. I am not against paper patents however, but I would have a clause that a paper patent cannot be enforced against somebody who actually reduced to practice the invention embodied in the paper patent. Thus there's an incentive to actually be the first to actually physically make the paper patent invention.

Intellectual property in any form is never reasonable.

Agreed. Suggest that you cast a critical eye over the Trans Pacific Partnership (which will increase copyright in some cases to 125 years).

You know-the trade pact Tyler Cowen was schilling for 24 hours ago:


You know absolutely nothing about the real world or patents.

First of all, most technology startups have no general counsel - they use outside counsel on an as need basis.

Second, the GC is usually untrained in the area of patent claim construction. They have to a hire an experienced patent attorney or agent to explain the patents.

Third, most of the key R&D people already know what is out there and what is new.

Fourth, you can hire a patent expert to spend one or two days performing a search and they'll tell you everything relevant that has already been published - which is usually done well before a new product is launched.

Last, as far as your examples of "abstract" ideas, there is nothing abstract about "displaying data in frames" (think Oracle, Microsoft, and so on) or "recommending media based on past choices" (Tivo, Netflix and many others do it). Only a very lazy, non-technical patent-hating person would call it abstract. And you're taking bits and pieces out of the entire patent claim, so your snippets are also misleading.

In those thumbnails they aren't even inventions, only goals.

And I get that the patent system is being use to block others from pursuing the same goals.

Kaldırma ve Taşıma Ekipmanları sektöründe hizmet vermek üzere kurulan firmamız, kurucuların meslek ve bilgi alanları nedeniyle ağırlıklı olarak Kaldırma ve Taşıma Ekipmanları, Çelik Halat, Zincir ve Deniz sektöründe siz değerli müşterilerimizin ihtiyaçlarını karşılamak amacıyla Kocaeli / Gebze bölgesinde kurulmuştur.

Akülü İstif Makinası Trans Paletler Zincirli Vinç adı altında yayın hayatına başlayan internet sitesine hoşgeldiniz. Bu internet sitesi, KT LİFT KALDIRMA VE TAŞIMA SİSTEMLERİ Firması tarafından hazırlanan bir sitedir. Sitede, firmanın sağlamış olduğu tüm hizmet ve ürünler hakkında detaylı bilgiler elde edebilirsiniz.

Akülü istif makineleri, Zincirli vinçler, Ceraskallar ve çok daha fazlası transpaletler KT Lift Kaldırma ve Taşıma sistemleri tarafından müşterilerine sunulmaktadır. Tüm bu ürünlere ait çok sayıda resime ve teknik verilere sitesi sayesinde ulaşabilirsiniz.

domuz arabası, KT Lift firmasının online mağazasıdır. Bu mağazada, firmanın tüm ürünleri sergilenmektedir. Ürünler hakkında daha detaylı bilgiler elde etmek istiyorsanız, aşağıdaki telefon numaralarını arayarak müşteri temsilcileriyle irtibata geçebilirsiniz.


I read the first 20-25 pages of your brief. It sucks. Here are a few comments.

1. What is your definition of "abstract", "software" and "abstract software"?

2. Where is your evidence that "abstract software patents discourage innovation"? You cited two Supreme Court decisions, seriously? That would get you an 'F' in my class. Where is the evidence or data for this assertion? It's not obvious to me.

3. Why the time period of 1984-2000 in that table regarding patent lawsuits filed within 4 years of issue? Is there a reason you left out the past 14 years? Does it hurt your argument? How many of those cases were actually filed by the "defendant" to invalidate the patent? Perhaps the uncertainty over the validity of the property right is generating more litigation? Ever thought of that?

4. You're an economist and you're bashing these Patent Assertion Entities for not making a product? Are you against specialization? division of labor? comparative advantage? You don't like specialized intermediaries? That sounds hypocritical to everything else posted on this blog.

5. You make the argument that any software product is covered by so many incremental patents that you're bound to infringe. Why is this a software problem? Automobiles, airplanes, and many other "complex" devices have hundreds of patents covering the product, yet we don't hear complaints about these type of inventions? Why is that? You may argue the fixed cost aspect, but that's a separate issue...and not all software inventions have low fixed costs.

Your brief is a joke. And I'm being kind about it.

why do you have to use "abstract" in your argument? That's weak. It's like saying I'm against bad drugs or costly services or unfair systems, it just begs the question and it's a weasel word.

You're like the March of Dimes after the polio vaccine, you're looking for a new problem to cure that isn't there.

Well, abstract is the whole key to the problem. A patent can be so narrow as to be useless, or so abstract as to stop anyone from approaching the problem space.
What is in the middle? Pretty much nothing. Which is why you will find relatively few developers that actually want software patents.

The Supreme Court is trying to create a concrete definition of the word "abstract". Good luck with that.

In 1951, ahead of the peaceful liberation of Tibet, the U.S. plotted and supported the reactionary forces of the Tibetan upper classes to resist the People's Liberation Army to liberate Tibet. While its plot to help the Dalai Lama flee Tibet failed, it said.

Comments for this post are closed