Patent Trolls are Only the Symptom

I was going to write a post on how trolls aren’t the fundamental problem with the patent system but Timothy Lee has it covered:

…trolls aren’t the primary problem with the patent system. They’re just the problem Congress is willing to fix. The primary problem with the patent system is, well, the patent system. The system makes it too easy to get broad, vague patents, and the litigation process is tilted too far toward plaintiffs. But because so many big companies make so much money off of this system, few in Congress are willing to consider broader reforms.

A modern example is Microsoft, which has more than 40,000 patents and reportedly earns billions of dollars per year in patent licensing revenues from companies selling Android phones. That’s not because Google was caught copying Microsoft’s Windows Phone software (which has never been very popular with consumers). Rather, it’s because low standards for patents — especially in software — have allowed Microsoft to amass a huge number of patents on routine characteristics of mobile operating systems. Microsoft’s patent arsenal has become so huge that it’s effectively impossible to create a mobile operating system without infringing some of them. And so Microsoft can demand that smaller, more innovative companies pay them off.

… In effect, the patent system is acting as an innovation tax, transferring wealth from companies that are creating successful technologies today to companies that acquired a lot of patents a decade ago.

A more fundamental change would be to offer patents of varying length, say 3, 7, and 20 years with the understanding that 3 year patents will be approved quickly but 20 year patents will be required to leap a high hurdle on non-obviousness, prior art and so forth. See my paper Patent Theory versus Patent Law.

My video on patents is a quick and fun introduction.


Creative people will continue to create even if their creations aren't protected by patents, but rich people will not continue to work if their incomes are subject to progressive taxes. I understand that a foolish consistency is the hobgoblin of little minds, but are creative people and rich people that different? And if they are, does that mean rich people aren't creative?

You have this weird misconception that the recipients and beneficiaries of patents are "creative people."

At least when it comes to low quality patents, the only creativity is in the knowledge of the filing process, and the legal persistence, nothing to do with invention.

Setting the bar higher for patents would probably hurt a few creative people, but the vast majority of people hurt would not be inventors are all, and other creative people would be helped far more.

"Creative people will continue to create even if their creations aren’t protected by patents"

What you're arguing is that "creative people will continue to create, even if they can't benefit from their creations because someone who never created anything already has a patent on the concept (even with no way of implementing it.)"

The current patent process on the whole makes it harder for creative people to benefit from their creations, at least if by creative you mean people who actually invent and do, as oppose to people who know how to file and lawyer.

You missed the part where the patent system *is a tax* on creativity.

Isn't this the most bizarre shit you've ever heard? Comparing people who are compelled to express themselves with people who compare rates of return versus a new yacht? Rayward: You aren't being compensated for creating this garbage, so what drive you?

I don't know what to say other than "He's right". Software patents are ridiculous in the way they are granted and enforced. For example, Apple winning against Samsung for violating a 'slide to unlock' patent was crazy. 'Slide to unlock' is a software simulation of a slide switch used on countless devices (my coffee maker has one) -- something like that should *never* have gotten past the obviousness test let alone be the basis for an enormous judgement.

So the basic problem identified here is Congress (government).
Interesting observation. Wonder if this problem extends to other aspects of society?

Has anyone ever noticed this phenomenon before?
Surely Jeb or Hillary could fix it all for us.

ur dum

Michael Foody wants to impose his vision of society on you. In the interim he's willing to accept gross violations of rights and norms.

How would you deal with the backlog of existing patents? Grandfather? Automatically time limit? I assume you could not go back and re-judge them all.

Also, does your proposal require a working prototype or would you still let people patent ideas they have yet turned into reality?

Grandfather, because they will all expire.

As RoyL says, patents expire so grandfathering them in does limited damage. You can also make it a bit easier to challenge trivial patents, even ones that have been grandfathered in. The valuable ones will not be affected by this process.

It's not that hard to challenge trivial patents. No patent is granted immediately upon application by fiat; it is displayed on the USPTO website for a sizeable amount of time (could be anything from a year to a decade). Check out Joel Spolsky's article on his successful attempt to challenge a patent and get it rejected.

I like your idea of having patents of varying length with varying thresholds for patentability. This also has the neat effect of giving patent trolls something to live on instead of yanking the carpet out from under them entirely, and is thus less likely to arouse staunch opposition. The people writing software patents can just go for the 3 or 7 year patents which might be enough to keep their clients happy, especially if they think they are more likely to get them that way.

My idea: patents are supposed to promote investment in innovative by granting temporary monopoly on a to reap the returns on said investment, right? Then, if you want a patent, prove you are investing in the innovation. The length of the patent depends on how much you are sinking into it.

Oops, it removed my angle bracketed: "innovative (something) ... on a (something)"

Disagree strongly. This would bias the patent process against small companies and individuals. A massive corporation can easily sink millions into developing a new drug, but they don't deserve a patent proportionally more than a individual working in his garage.
Also the value or innovativeness of new inventions isn't proportional to how much you spend on them.
You'd basically be saying to the guy that built something in his garage that he only gets a patent for a year, because he didn't spend that much money on it, but the massive corporation that bought a bunch of expensive equipment and paid ten engineers to do the same thing would get their patent for 20 years.

You seem to think the purpose of a patent is to reward someone for having a good idea.

The PTO is a prime example of regulatory capture. Look at the announcements that come out of the PTO; "shareholders" refers to large corporations and IP attorneys rather than the public. The PTO's fee structure is built so that allowances pay the bills, and there are perverse incentives for examiners to allow patents. And never during the current "quality initiative" does anyone suggest giving examiners more time to examine applications, because the "shareholders" don't actually want better patent examining. Variable length patents won't do anything to fix that.

"That’s not because Google was caught copying Microsoft’s Windows Phone software."

No, but Google obviously copied a number of concepts from the older Windows Mobile/PocketPC/Windows CE platform.

Yeah, I'm not a big fan of the patent system, but this is a poor choice of example. With respect to smartphones, Google is a big 'ol thief, and Samsung is a big 'ol thief. Sun Microsystems, Microsoft, and poor little Apple have been robbed blind.

I am complaining about Alex here, not RR.

Ho-hum, AlexT trolling his readers again on the patent issue. I guess it's a slow news week. As I stated to Dr. Tabarrok by email, and he agrees, we need in the USA a reform of the patent system so we have two patent systems, as they do in Germany and Japan: one for trivial "industrial design" patents and the other for more fundamental patents, with differing levels of protection and examination.

But more to this post, who cares that there are industrial design patents that accrue to large corporations, or that patents for obvious stuff goes to large corporations? Nike has design patents on the non-function designs made by the undersides of sneakers--but who cares? In trademarks, an ordinary word or phrase or even color (pink in the case of Owens Corning fiberglass insulation, made famous by their Pink Panther (TM) adverts) can become exclusive property of another company by nature of "secondary meaning", meaning heavy advertisement. In patents, analogously, there are 'secondary considerations' that can make an obvious invention patentable (too complicated to specify here, but suffice to say an 'obvious' invention can become patentable if it is a commercial success--patent law recognizes this). Does this mean the colors of the rainbow will be exhausted as insulation companies patent every hue of color? Unlikely. That sneaker undersoles will run out of designs? Unlikely. That obvious look-and-feel stuff will be foreclosed from being manufactured except by mega-corporations with big advert budgets? Unlikely. More likely is that consumer welfare is advanced in each of these cases, hand-in-hand with corporate welfare. Off soapbox now...

In effect, the patent system is acting as an innovation tax, transferring wealth from companies that are creating successful technologies today to companies that acquired a lot of patents a decade ago.

The patent system also enables the real innovators like Apple and VMware to benefit from their expensive high-quality research rather than having the likes of Xiaomi and Huawei run off with all the benefits (and thus discouraging future research).

the real innovators like Apple

Didn't Apple basically steal all it's ideas from other people?

You have the direction wrong on smartphones. Google and the Android ecosystem stole from a bunch of ideas from precursors and Apple. Apple is a successful innovator in this space. Microsoft and Sun were unsuccessful innovators.

Palm Pilots and PDAs (Personal Digital Assistants) were around since the late 90s, and they had similar form factors and user interfaces to the smart phones that started to emerge in the mid-2000s. I could argue that adding telephony capability to a PDA (like the Compaq IPAQ I owned 15 years ago) and using touch instead of a stylus (more a personal preference than a revolutionary advance in UI in my opinion) converts a PDA to a smart phone; a rather trivial and obvious conversion. The reason Apple convinced everyone it created something revolutionarily innovative with the iPhone is that is is good at making business deals (like with AT&T) and is brilliant at marketing.

The big innovation in the iPhone is the lack of hardware keys, and the many small design decisions that make it usable anyway.


Yeah, I never understand all these claims that use American companies as examples.

The reality is that without IP protections, China would run rampant.

Chinese copycats could already beat Samsung to market with a Samsung designed product. (Samsung even tried to hire these companies - they refused.)

Also, there is a major bias to look at large firms and software. Sure, they are important. But IP really helps small firms. There is a kick-starter recently of an innovative new beekeeping system from Australia.

Remove patent protection or only give them 3 years and you've just wasted their time. China will have a copycat product out within 3 months.

Though, to be fair, patent lawsuits are already so expensive as to make IP useless unless the market is in the multi-million dollar range.

We had someone infringe on a product that had around $500,000 FOB sales...not enough for a lawyer to even consider on contingency.

Now thanks to Amazon, we can sell retail, so maybe its up to 1,000,000 in sales. Probably still not enough for a lawsuit, though lawyer's letters threatening a lawsuit does help.

I don't get much of the criticism of the patent system. It's one of the most essential and important component of a modern economy. It's property rights, and yet somehow the main criticism from "the right" or "libertarians" seem to hinge on to things:

1) Supposedly it prevents innovation

2) It's not the most perfect system in the world!

The first criticism which is usually used by hard core libertarians is nonsense. That's the equivalent of saying that the fence around my house limits my neighbor's economic activity. It's supposed to.

But the second criticism, the one used here by also nonsense. Patenting is exceptionally hard business. You have patent examiners who are trying t use their knowledge to figure out whether or not something can be patented, whether it is an actual innovation, and whether it infringes on anyone else's innovation.

I.e., it is a very costly undertaking, and in any costly undertaking...there will be inefficiencies. I don't think the inefficiency is, however, in allowing Microsoft to have so many patents or whatever. If these are...legitimate...patents, then of course they should have them.

Now you may say, maybe they are not "legitimate" patents. But that's the issue!!! This has nothing to do with Congress or the political system. This is a technical problem of evaluating such complex information. There is no solution to this issue, since this is a subjective cal by a human...subjective in the sense that it takes years of evaluation to reach a conclusion.

And yet this immense and costly undertaking is somehow reduced to "meh, it's not perfect, so lets pass a law by congress to fix it!"

What is there to fix? This is a basic economic problem...the more complex something becomes, the more difficult it becomes to evaluate, the less efficient evaluations you are going to get. But that's not something you can fix!

I don't see many people advocating abolishing patents entirely. The criticism, as stated above, is that it's too easy to get patents on inventions that are not terribly novel and the bar needs to be set higher. Also the way the system works creates biases towards granting patents that shouldn't be granted.

It's a pointless criticism. And it certainly isn't solved by an "act of congress".

This is purely an issue of complexity...and inability for individual examiners and parties involved to create "perfect" property rights.

But the lack of existence of "perfect" property rights, isn't much of a critique of the patent system. That phenomenon exists in every area of life.

You also forgot the part about how IP is infinitely copyable property, not "real" at all. Additionally, you've forgotten how IP represents a regulatory capture. But, whatever, I guess everyone in the comments should start pushing out troll patent applications from their home office, because, why not? I may not have invented it, but I can certainly profit off of it. Thanks, AIG!

Most of the world works on the concept of "I may not have invented it, but I can certainly profit from it".

When you buy a liquidated property or equipment or whatever from another source, you may not have "invented" it, you may not have "build" it...but you bought it in a secondary market because the person that did invent it or build it no longer had a use for it.

I.e., patent trolls are nothing more than secondary markets for ideas. Why this should be a "good" thing for every other good, but not not quite clear.

You obviously don't work in software.

As far as I can tell, Libertarians are over-represented in the software industry, and anyone who isn't a rent-seeker in the industry loathes patents. In the software field, they very much do prevent innovation. It's far too easy to get a vague patent on an entire idea, rather than a specific implementation of it (which is what the patent is supposed to cover), and completely block anyone else from doing anything in that area without payouts. I have never, ever seen a software patent that included enough information to actually implement the patented idea (again, this is supposed to be a no-no for a patent, but the patent office doesn't care), with the possible exception of the RSA patent (4405829).

The system, moreover, is designed to facilitate these kind of horrible patents. The patent office gets their filing fees, so they encourage refiling patents. Refile a bad patent often enough and it'll get through. Once it's granted, it's horrendously expensive to challenge, no matter how bad it is. It's a huge dead-weight on the industry.

A simple fix for software patents would be to require all software patents to actually include an implementation of the patent in source code. Then that implementation will be patented, and people who can figure out other ways to do the same thing don't have to worry about it (as is supposed to happen with patents). Of course, the people who oppose patent reform never propose any such thing, because that would harm their rents just as much as eliminating software patents would, which really tells you what the basis to the objections is all about.

BTW, can you name any other field besides software that is covered by both patents and copyright?

I don't know much about software or software patents, that's for sure. But clearly the world of patents is much bigger than just software.

If there may be issues with software patents, again this brings us back to what I said earlier:

1) The solution isn't to scrap the system, nor is there are reasonable "reform" which jumps out which would alleviate the "problem", if one exists.

2) The issue is again one of complexity. The more complex the implementation of the idea becomes, or the idea itself, the more likely you are to get inefficient property rights allocations. This is purely a...physical human problem...not one to be solved by politicians.

Most patents do not provide sufficient information to implement an idea, be it in software or in ANY field. And the reason for that is obvious: you don't want to disclose too much information to make it easier for your competitors to figure out exactly HOW to circumvent your property.

You want to disclose enough information to signal that your idea is indeed novel and deserving of a property right, but not enough to allow someone to know exactly how you're doing what you're doing.

And again...this is a feature of almost all economic activity. Which is why...THERE ARE NEVER EFFICIENT PROPERTY any field.

A basic economic concept...which most people here seem to be forgetting/ignoring.

I started working at Microsoft in the mid 90's on their smartphone projects, back when it was the PDA space. There was indeed a lot of early innovation. And much came because Microsoft was there early. It was greenfield. Your filesystem on a phone used to live in battery backed RAM. Do you remember having to change batteries on a PDA in 1 minute, otherwise you lost all your data? That's because the SRAM used at the time was held up by a capacitor. When you removed the main AAA batteries, the cap couldn't last too long preserving the memory, then poof it was gone.

The solution, of course, was obvious for mobile: Store the data in flash. But flash systems at the time were what was called NOR flash. It was a low density flash that operated a lot like RAM. NOR was too expensive to hold large amounts of data, though it was fine for phones of the 90s. So, we had to migrate everything to new flash tech called NAND, and this had huge implications for the mobile OS of time. This required a journaled file system (a filesystem that doesn't get screwed up if you yank the battery at any time), a mobile OS that knows how to demand page from that file system, compact silicon that can support all this (which didn't exist at the time), software that knows how to load level NAND (because NAND eventually wears out) and on and on.

Slowly but surely, all the pieces came together. But the summary is this: The act of simply moving large data storage from SRAM to NAND required enormous amounts of change to the industry. From silicon providers to software writers. And by the time Apple and Android came around, that work was all done. It was part of all the silicon, and the basic framework for how it should all work were figured out and done, largely by Microsoft. Instead of spending 3 years wondering "what's the best way to do this" the Android folks knew, for example, they needed these 8 pieces put together just so and then it would work

From 1995 to 2005, I'd estimate Microsoft had between 500 and 1000 people working on mobile. Easily several billion $ were spent in early innovation.

In short, there was enormous innovation that took place in the phone space in the 90's, long before the giants of today entered the segment. It's not at all unreasonable for MSFT to recoup that. In the last 24 months, Apple has run a commercial showing an appointment being changed on a phone and it updating on the ipad. ActiveSync was doing that in 2000 over a circuit switched network. And yes, it was impressive as hell at the time, and definitely worthy of patents.

I'd argue that patents are doing precisely what they should be doing here. Microsoft generated enormous early innovation--things we all take for granted today--and spent billions to develop that innovation. For other reason they never were successful in the space, but it makes a lot of sense they can still derive a revenue stream from those innovations.

+1, although Sun Microsystems also deserves a huge amount of credit in this space as well. Google probably owes more to Sun for enabling its Android products than to Microsoft, in terms of intellectual debt.

+1. I had a Compaq Ipaq running Windows Pocket PC for 3-4 years in the early 2000s, and loved it. I used it on a more frequent basis than I use my iPhone today (for the same purposes, to read ebooks, play games, listen to music, watch videos, etc.); the only thing lacking was seamless connectivity on the road. Microsoft had smart people then (as it does now as well) who can create stuff from clay, but the political culture of the company kills innovation; I know from personal experience, having worked there for a couple of years.

totally agree. Significant r&d expense, coupled with real tech breakthroughs, should warrant a patent (so firms can have a limited monopoly to garner profits from.) No new pharmaceuticals would exist if this weren't the case. But the problem is that under current patent law firms are encouraged to patent anything and everything, regardless if a significant expense was occured in development, or if they actually are using the patented technology.

Patents were designed to encourage innovation, but trolls, and large firms that practice defensive patenting, do the opposite.

Very weak arguments by you and Tim Lee.

1. It's now almost impossible to get a patent on anything that smells like software, and previously granted software patents are probably unenforceable. You guys are behind at least 10 years in your thinking.

2. Big software companies like Google and Microsoft are behind the push to weaken the patent system so they can restrict competition and maintain their dominant market positions. Funny, they got slapped down recently for their policy of not to recruit or hire each other's employees. Socialism for them, free markets for everyone else.

3. Knowledge is not required to infringe a patent, unlike copyrights, so do a prior art search and get a legal opinion before launching a product.

4. Litigation is not tilted towards plaintiffs. This is flat wrong. You can attack a patent so many ways. Rather than take a license and pay royalties up front, big companies will spend easily a million dollars to invalidate a patent.

5. Different terms for different levels of examination? What the hell does that supposed to mean? Either have a registration system that grants patents on every application or maintain the current system with better trained examiners - though they do a better job than people think, for the most part.

It’s now almost impossible to get a patent on anything that smells like software,

Even though the post-Alice landscape is not clear, this is a broad overstatement.

A solution in search of a problem.

Is lack of innovation really a problem in smart phones or computer technology in general? I do not think so. New products and features and apps are constantly arriving and at good price points imho.

So what if Google has to pay Microsoft or Apple for use of vague patents (or vice versa)? Its just one huge company versus another. So long as the spice flows to the consumer, who cares if the Guild or Dune get the money?

All true and a powerful reason to oppose the TPP.

How would real estate development work if you had no chances to know who owns the land you want to build on before starting building?

Alex, you say that too many bad patents are issued. What is the optimum level of bad patents, considering the costs of examination and the imperfections of human patent examiners? It can't be zero, because the PTO will always make mistakes. It's actually very difficult to figure out how many patents are bad, because most patents are never litigated to a final judgment and many, many patents are licensed without being litigated to a final judgment. If we look at patents that are litigated to final judgment, it's about 1-2% of all patents, and only about half of those are invalidated. So, we only know that about 1% of patents are bad, we can speculate that more are bad, but we don't really know (remember, also, the selection effect that weak patents are more likely to be litigated and invalidated). Even if we knew how many patents were bad, I am not sure how you could compare that number against some Platonic ideal ratio of good/bad patents, keeping in mind that there will always be errors. You might read Lemley's Rational Ignorance article, if you haven't already.

The purpose of the patent office is "Through the preservation, classification, and dissemination of patent information, the Office aids and encourages innovation and the scientific and technical advancement of the nation."

Take note that even the patent office acknowledges in their mission statement that innovation is NOT the same as advancement. They are two separate things. More importantly though, the purpose is not to protect the individual or encourage innovation and gain by individuals, it is there to encourage the advancement of the country, not the advancement of science or the individual.

Comments for this post are closed