Home
> Uncategorized > Patent trolls
Patent trolls
March 6, 2015
This morning I’m preparing for my weekly Slate Money podcast by trying to learn all about patents and patent trolls. To tell you the truth, so far I don’t know why patent trolls are all that bad, besides the fact that they obviously have a terrible sounding name. It seems like the patent system works in many ways for good, at least when there’s no weird extensions of the time limitations and the original patent was valid. Feel free to disagree, though, and tell me why.
Also on the slate (harhar): Amanda Palmer, Wu-Tang Clan, and Lumber Liquidators. Another typical week in the world of podcasting, in other words.
My number of the week is going to be 2063, by the way, but I’m not saying why.
Categories: Uncategorized
mathbabe 
One issue with patent trolls is that patents have been issued for things that many people think should not be patentable. This is mainly confined to software-related patents issued for obvious things that everyone needs to do. In the early 2000’s, this involved Unix and Linus issues, but that has all died down since then.
LikeLike
Journal of Economic Perspectives published an issue in 2013 discussing the pros and cons of patents. Timothy Taylor links to the articles here
LikeLike
The podcasters I listen to say they are being shaken down by people who claim to have patented vague things like “using the internet to put out episodes of things.” They send letters to the podcasters threatening legal action unless they pay fees directly to them, and because of court costs it’s often cheaper and easier for the podcaster to pay those fees. The troll then wins and gets more resources for the next round. One proposed remedy, as I understand it, is a proposal that if the people bringing claiming patent infringement lose, they have to pay (some? all?) of the court costs. This would at least discourage people claiming to have patents that would be unlikely to survive an actual legal challenge.
LikeLike
Great point!
LikeLike
So, if Noble Little Guy Inventor sees Big Bad Corporation infringing, goes to court and loses (I am both a patent author and have worked as an expert witness and can tell you that the result of a patent lawsuit is not just about the merits of the patent.) then the Noble Little Guy is out all his savings, his business, his house, etc. A patent case can easily run into the millions for each side.
LikeLike
The “loser pays” proposal really only doubles the cost of losing a court case compared to the status quo. Remember, under the current system, Noble Little Guy is already on the hook for their own costs if they lose, and as you say, those costs easily run into the millions. I don’t think the propsect of risking $10 million instead of “only” $5 million is really going to change Noble Little Guy’s risk-reward calculation.
LikeLike
I was unaware DJ that one party in a suit has control over how much the other side spends.
LikeLike
Regardless of how much money the other side spends, a successful patent lawsuit requires millions of dollars of expenditure on your own side. I just don’t see any difference for Noble Little Guy in either scenario. The difference between risking $5 million and risking $100 million is just not going to affect the little guy. Ironically, big corporations would care about that kind of risk.
LikeLike
According to the law, patents are not supposed to be granted for obvious ideas. Unfortunately, the reality is that many patents are granted for obvious ideas. The patent office has a strong financial incentive to encourage over-patenting, since 100% of their revenue comes from patent applications and fees.
In the 1990s and 2000s, you could add “on the internet” to anything and patent it. In the 2010s, you can add “on a mobile device” to anything and patent it. These patents do not contribute to society. They are the major reason why patent trolls are a problem. (The problem goes beyond patent trolls — big companies with legitimate business dealings also abuse patents.)
Patent reform entails enforcing the part of the law that says patents must be non-obvious, changing the law to include more serious penalties for abuse, and funding the patent office using funding sources that are independent of the rate at which they approve patents. These are not big changes. They’re just common sense.
LikeLike
None of this is true. The patent office gets most of its fees from application filng fees and not from issuing patents. In fact, the patent office often has run a budget surplus that Congress illegally takes.
The requirement that a patentable invention be “non-obvious” is always a judgment call, and there always will be some bad calls. But Sumpreme Court decisions over the past several years have ended any serious chance of getting a patent by just adding “on the internet”, and this was not really the case anyway.
Defining “abuse” of a patent is not easy; it’s always an abuse when you’re on the wrong side of a suit. I offered some other ideas below.
LikeLike
Obviously, if the patent office issues fewer patents, then people will be less inclined to pay those application filing fees. I maintain that the patent office’s funding is directly and 100% dependent on over-aggressive patenting. We need to fund the patent office independently of acceptance rate. This is what I said originally and I stand by it. Nothing you said changes this stance.
LikeLike
I’m not sure it’s that obvious. The real question is the expected value of the issued patent. During the many decades when courts did not look kindly on patents, the amount of patenting was relatively steady. When Congress reorganized the patent courts and made patents more enforceable (and therefore valuable) then applications surged. But application have accelerated faster than issued patents; so I think your thesis is not so air-tight. See here.
LikeLike
Application numbers since 1990 have increased only slightly faster than the number of patent grants. The important distinction is that applications and grants have both soared enormously since 1990, the year when Congress made the USPTO self-sufficient. My thesis is consistent with the statistics. We are issuing way too many patents compared to the optimum. Of course there may be other confounding factors.
LikeLike
Having practiced patent law at various law firms and pharmaceutical companies for many years, I’ve heard a lot of nonsense about trolls and patents. Generally, I find economists completely ignorant about the patenting system and how patents are really used. Many business commentators are usually grinding an axe for some benefactor.
The US patent system was created by Thomas Jefferson to encourage the commnication of new ideas and inventions to the public by giving some degree of economic protection to inventors who would otherwise keep their inventions a trade secret. This has often worked well. In the modern age, where capital industries have become increasingly capital intensive and require lots of invenstment up-front, patents are often the only asset an inventor or small business has to offer as collateral.
Contrary to what most economists think, patents do NOT create a monopoly by themselves: Patents define “inventions”, which are legal abstractions not physical products; monopolies require physical products. Often, and this is especially true in the high-tech and biopharma fields, multiple patents cover various aspects of a product, which require cross licensing and often does lead to collaboration. So, in the end, if everyone acts reasonably they get a cut of the action.
Over the past couple of decades, the rise of finanacializtion has affected patents too. Traditionally, the vast majority of patents had little to no real market value, if only because many inventions don’t give rise to valuable products. But starting in the ’90s groups of investors thought they could do better by buying large portfolios of patents and doing the necessary ground work to identify licensing plays. Their models were the venerable US high-tech titans IBM and Texas Instruments, both of which found huge profits in licensing fees from the computer and electronics industries. But both of those companies had strong experience and focused patent portfoilios. Many troll plays don’t have either the experience or focus and haven’t done so well.
But of course, if you’re on the wrong end of a bogus claim it’s still expensive. Even with the recent changes to the patent law, it’s hard to quickly and cheaply fend of an weak or baseless claim. Some lay the blame on the patent office for issuing “bad” patents. Given the literally hundreds of thousands of applications filed each year, it’s not surprising that some bad ones will get through. (And don’t forget that some deserving application are erroneously rejected too.) Frankly, I can’t think of a patent I’ve worked on that wasn’t reasonable given the available evidence.
To me, the real objection is the idea that people who have no interest in bringing real products to the market are using patents as a tax on those who make things. So, why not focus on that by hitting these rentiers where they live? Here are a few ideas:
(1) Increase maintenace fees, i.e., fees paid to the patent office to keep issued patents in force, so that it’s expensive to simply hold on to patents waiting for a target.
(2) Provide a “use-it-or-lose-it” affirmative defense. If a defendant in an infringement suit can show the ower hasn’t made any substantial effort to product a product covered by a patent, either directly or by direct investment, then the court can rule the patent invalid or assign a nominal licensing fee.
(3) Provide a “loser pays” provision for plaintiffs that are not working their patents.
(4) Tax policies that encourage real investment and not bullshit finanacialization schemes for our rentier class.
LikeLike
Wow, amazing comment, thank you so much!
On Fri, Mar 6, 2015 at 9:13 AM, mathbabe wrote:
>
LikeLike
I think your comment about Thomas Jefferson goes right to the heart of why people in today’s tech industry don’t like patents: they no longer serve (in that industry) to encourage the communication of new ideas. It’s pretty much always faster to just solve a problem yourself than it would be to read a (no doubt totally inscrutable) patent and turn it into a practical solution. Most patent infringement in the industry occurs when two (or several) companies simultaneously innovate in roughly the same way and one files for a patent first.
Usually in the case of two big tech companies, they won’t actually sue each other for infringement because each knows that the other also holds a ton of patents that they are infringing on, and so while there’s a lot of time wasted applying for patents that will at “best” be useful in bullying startups, at least they usually don’t have to waste huge amounts of money going to trial. With patent trolls though, there’s no infringement countersuit to threaten, so it’s either pay up, or pay even more going to trial even if you win.
It’s not clear that software patents could ever be beneficial to the world, but one reform that might have a chance in that direction would be to add cheap compulsory licensing and greatly increase the disclosure requirements: upon being granted a software patent, the patent-holder should be required to disclose the full source code (fully documented) of their invention (with free copyright licensing and only the compulsory patent licensing rates to pay to use it).
LikeLike
As a counterpoint to some of moosesnsquir’s comments and as someone who holds half a dozen patents and has only benefited from the US patent system, I can state unequivocally that, at least in the software industry, patents are a net drain on the industry and society as a whole. Companies are trapped in an arms race that forces them to spend time and energy filing patents solely for the purpose of building a patent portfolio that they can use to protect themselves from competitors (and trolls). And software developers are encumbered by the task of avoiding stepping on existing patents (it’s essentially never worth licensing the patent, its easier to just find a different way of solving the same problem, but when the obvious solution has been patented, one has to expend effort coming up with a sufficiently different less obvious solution, which while entertaining as an intellectual exercise is a colossal waste of time).
The reality is that nearly all software patents are either obvious (and should never have been granted in the first place) or obsolete by the time they are granted and available for licensing.
The industry would be net better off without a patent system at all — it would spur innovation and lower the cost of development, which would be better for everyone. But there is unfortunately no obvious way to get there from here.
Of course the legal profession has done quite well, and is one of the primary beneficiaries of the patent system (something to keep in mind when you hear lawyers defending it).
I actually do agree with some of moosesnquir’s proposed remedies, but they don’t go far enough. The current system is seriously broken and patent trolls are just one symptom of the problem.
LikeLike
I agree 100%. I also hold a half dozen software patents. (More precisely, I am an inventor; the patents are assigned to my employer.) I have worked in the software industry. The lawyers have their perspective, which as you rightly point out is colored by their own self-interest. What they don’t see is the software engineers in the trenches, who are actually all too cognizant of the dangers of the patent minefield. The wasted effort expended by the low-level grunts to work around broken patents is completely invisible to anyone not working with the technical people day in and day out. The damage is measured not just in high-profile court decisions, but also in the unheralded daily costs, energy drain, and death by a thousand cuts.
I don’t claim any expertise in big pharma or biotech. But in software, I see no societal value in software patents. At all.
LikeLike
Well, of course only software counts! 😉
But I’m old enough to remember practicing when software was not considered patentable. When the courts allowed patenting software in the mid-’90s, the UPSTO was flat-footed but also hampered, severely IMHO, by the lack of the sort of coherent literature base (academic and patent) that exists in area such as engineering and the sciences. The ambiguitites inherent in describing an claiming software also pose huge problems.
Since I’ve spent a lot of my career in the biopharma area, whether or not software was patentable was not a huge deal for me. And my comments were about patents in general. I still believe that on balance patenting is valuable. But we have to always be clear about the social functions of patenting. If we can’t make a particluar area of technology work within the boundaries of those functions, then Congress should act accordingly.
However, I think it should either to create a patenting system that does justice to software or ban software patents. I don’t surpport separate “software patents” that work on different rules from the other technologies. And the same comments can be applied to business method patents too.
LikeLike
Count me as another named inventor on several software patents, who thinks that making software patentable probably does net harm to the overall productive value of the software industry. (At least mine mostly have the virtue of implicating some hardware — I was working at the interface between text entry software and the physical keys, for small-market phones. So we were trying to figure out how to build cheap phones on which you could write in Thai, Tamil, Hebrew, etc.)
The Amazon One-Click Purchasing patent may be the all-time facepalmiest example of something that should not have been patentable. Should some random catalog retailer be able to patent “keeping the customer’s address and credit card number on file” (so that they can make a new purchase by just phoning it in or mailing in an order, without re-providing the billing/shipping info). For that matter, should a random retailer be able to patent “selling to trusted clients on credit”? Should Joe Schmoe’s Saloon be able to patent “opening a tab”, letting customers make subsequent purchases with less friction?
LikeLike
I’ve heard that Thomas Edison had a patent fight over the light bulb. Someone had already been granted a patent on the idea of generating light by putting electricity through a filament even though they had not made it work in practice.
LikeLike
This 2011 episode of This American Life paints a disturbing portrait of the patent lawsuit landscape.
http://www.thisamericanlife.org/radio-archives/episode/441/when-patents-attack
LikeLike
Patent trolls are companies that produce nothing, using patents to extort money from productive companies and succeeding because it’s usually cheaper to pay them than to fight them to court. Trolls are often shell companies, with a small group of lawyers using a variety of shells to harass and sue different groups of victims while hiding their identities.
While the patent office may have become stricter about “on the Internet” style patents, there are plenty of such patents still being used to demand payment from companies. Soverain sued just about every online retailer using its shopping cart patents. Amazon paid them, but Newegg won in court: http://arstechnica.com/tech-policy/2013/01/how-newegg-crushed-the-shopping-cart-patent-and-saved-online-retail/. Newegg is well known for fighting patent trolls in court instead of paying up.
Another patent troll scam is a group of trolls who demand $1000/employee for using scanners with scan to email technology. You can read about them at https://www.techdirt.com/articles/20130102/08174721543/patent-troll-shell-companies-shake-down-small-businesses-1k-per-employee-using-network-scanner.shtml. They annoyed enough people to get sued by state governments, with mixed results. More about that at http://arstechnica.com/tech-policy/2014/09/scan-to-email-patent-troll-loses-its-lawsuit-against-ftc/.
LikeLike
You might want to browse the transcript of the patent episode that This American Life did back in 2011. They did some great investigative journalism on patent trolls. http://www.thisamericanlife.org/radio-archives/episode/441/when-patents-attack
Looking forward to hearing the podcast!
LikeLike
I don’t think patent trolls are any more than a symptom. Non-transferrable property is not really property at all. If you agree with the idea that the govt can turn ideas into legal property then I don’t see a coherent argument for why the owner of the property must be the one that manufactures something. Insisting on that just comes from an attempt to fix the symptom, not the cause. Inventing a great solution, disclosing it and charging others for the use is the system working as intended. Selling on the rights to someone else because you don’t want to be in the business of patent enforcement is also perfectly legit. It frees up to inventor to go back to inventing the next great thing.
The cause is granting large numbers of terrible patents. Software and business these days are successful based on recognising a problem that others didn’t notice. The solution is often trivial once you state the problem. Then they try to protect that with patents that are often patents on the _problem_ not on the solution. So, e.g. you notice that everyone wants to pay with 1 click, you want to patent 1-click shopping.
Now you want a patent, but your solution is trivial, so you have no choice but to try make as broad a patent as possible, so that you actually cover the entire problem-space (rather than just the narrow solution you have found) and sadly the PTO grants it.
LikeLike
I don’t believe that patents should be considered property. The US Constitution does not declare that patents are property. Rather, the US Constitution states patents are RIGHTS. A right cannot normally be sold away. E.g. we don’t permit contracts for murder or slavery, even between willing parties.
The phrase “intellectual property” really upsets me, because this phrase requires that patents be viewed as property, and closes off all possible debate.
LikeLike
I agree that the phrase “intellectual property” is often used to allow some spurious analogies with physical property and to justify some positions on that basis. That’s not what I’m doing.
I’m just saying that if you don’t fix bad patents then you still have a major problem. Microsoft is a patent bad-actor but it does not fit the definition of troll because it makes lots of stuff. So just getting rid of trolls would still leave MS bullying their competitors with a massive portfolio of terrible patents. Why is that OK?
On the other hand, if you fix the bad patents problem then you fix everything. You may still be left with companies which make nothing and wield a portfolio of patents they have bought, but what’s wrong with that (assuming you agree with patents to start with)? The original inventors have been rewarded for their invention, their patent really did add to the store of useful techniques and the people using it derived real benefit from its being patented. It shouldn’t really matter whether they are paying the original inventor or someone else who paid the original inventor. I don’t see how it benefits anyone to force inventors to become patent enforcers (and I can also imagine that if patents were non-transferrable, there would be clever ways of effectively selling the exclusive right to act on behalf of the patent holder so that everything ends up just the same anyway).
I just think that trolls are a convenient bad guy that allows MS, Oracle etc to continue to benefit from a system that is fundamentally broken while looking like they’re doing something to fix it.
To be clear, for me, fixing the bad patent problem means disallowing business method and software patents entirely (or at least radically shortening their lifespan and raising the bar on novelty). For other patentable fields, I have no experience.
LikeLike
Yes, I agree, we should fix the underlying problem, which is bad patents. But I also think patent transfers are a (lesser) problem, which should also be fixed. I certainly don’t want to use patent transferability as a starting point for arguing against bad patents. Even if the conclusion is correct, I disagree with the premise of the argument.
It’s really not that hard to make patents non-transferable. You just make the law say that any exclusive licensing deal is void. If someone wants to license, or even delegate enforcement, fine, let them do that — but contract or no contract, they can change their mind later, and there’s nothing anyone can do about it. “Moral rights” under existing European law are close to what I have in mind.
If I had to pick only one thing, yes, I’d fix the bad patents problem first. (Sadly, the reality is that nothing will happen.)
LikeLike
I don’t think we should make transfers illegal. The people who invent things are often not good business people. They should be able to sell the patents after they get them.
On Sat, Mar 7, 2015 at 9:47 AM, mathbabe wrote:
>
LikeLike
To be honest, I’d rather keep out the business people. The negative externalities of patents are so crushing that I’d like as little activity around patents as possible.
Practically speaking, though, one can outsource management of business dealings without transferring ownership. The owners of sports teams do this all the time and in fact it is the normal mode of operation for them.
LikeLike
You can start by reading this: http://levine.sscnet.ucla.edu/general/intellectual/against.htm
LikeLike
Frivolous lawsuits by patent trolls have created a significant drain on the tech economy.
http://arstechnica.com/tech-policy/2011/09/study-patent-trolls-have-cost-innovators-half-a-trillion-bucks/
“Three Boston University researchers have produced a rigorous empirical estimate of the cost of patent trolling. And the number is breath-taking: patent trolls (“non-practicing entity” is the clinical term) have cost publicly traded defendants $500 billion since 1990. And the problem has become most severe in recent years. In the last four years, the costs have averaged $83 billion per year. The study says this is more than a quarter of US industrial research and development spending during those years.”
LikeLike
Do they give an estimate of the cost of frivolous lawsuits and stifled innovation due to practicing entities holding patents that should never have been granted? E.g. Apple’s rounded corners.
LikeLike
There is a lot of baseless assumption on this board. Like Bayes’ theorem, many on here are accusing a system on the basis of one type of empirical observations of the system, but there is no objective attempt to analyse the system in the absence of any empirical observations. In other words, what would things be like in a world devoid of patents? Big companies have already gotten to where they are on the back of ownership rights to significant IP. They have been able to keep small inventors and/or startups from entering into competition by illegitimately stealing their ideas. Keep in mind that if a big company is accused of infringing a patent, one of the best defenses available to them is to show that the patent is invalid because the idea is either not novel or obvious. A jury of laypersons get to decide this. Not lawyers. If, after paying, $500 billion, the big companies are unable to win, whose fault is that? Yet, when they are sued, they cry foul and use their considerable size and influence to cannibalize public conversation using pejorative terms like troll” and “non-practicing entity”. Naturally, because of their size and ability to pay, they have fostered (festered?) a patent litigation industry around the idea of defending against “trolls”.
In a word without patents, we would need to reset by stripping the big companies of all ill-gotten gains from the patent system. Then, assuming everybody begins from zero, let’s ask the right questions. how should Joe be able to protect his investment in developing a technology? How would he stop a competitor from simply reverse engineering his ideas and using them in their products?
LikeLike