The Direct Costs from NPE (Patent Troll) Disputes
papers.ssrn.com
papers.ssrn.com
First, just yesterday, I was at an event where a patent lawyer declared (I kid you not):
"You should be careful how much you innovate on top of existing products, because you might end up falling under someone else's patent."
Seriously. I couldn't believe my ears. Incentivising innovation fail.
The other soundbite came from an entrepreneur at a TechHub meeting in London, about the Hargreaves Report on IP, last year. He said:
"I wish I hadn't come tonight. The more I learn about IP, the less I want to start a business."
I think that pretty much sums up how well the patent system incentivises innovation.
Much of the time lawyers are there to present a worst case scenario to push you into using their services. I wouldn't place much store in what they say (speaking as a lawyer myself).
In terms of the entrepreneur discouraged from starting a business, I'd be interested to hear about what the business ideas were.
The fact of the matter is that a lot of people paint this idealized picture of the morally virtuous techie, where the only people who own patents are big bad patent trolls, and the only people who get sued are innocent innovators who independently arrive at an invention.
That's not the actual story, or at least that's not the whole story. Like in every other field, there is a gaussian distribution of business ethics at various companies. Look at how many companies get nailed for using GPL-ed software without releasing source. You think those same companies don't also copy competitors' designs wholesale? When our company was doing business with an American telecoms supplier, a manager there mentioned that a Chinese telecoms supplier was ripping off their designs wholesale, down to the silk-screening on the PCBs.
There is a difference between "let's innovate on top of this idea" and "oh this idea is good, let's just copy it."
Now, NPE's in their current form are clearly just rent-seeking. However, I think there is a role for NPE's that create value by serving to create a market for patents. Not every innovator wants to get into the business of making products. ARM wants to design microprocessors, they don't want to design and manufacture iPads. Patents facilitate that division of labor, and NPE's can make the whole system more efficient by creating markets for patents.
Instead software patents seem to be mostly used to sue people who independently invented the same thing. They enable rent-seeking and provide no educational value to spread innovation. (Which naturally arises and spreads in our industry on a time scale that is too short to get patents assigned.)
How do I know this? Because the one has happened based on my work, and the other came close to happening. At one of my employers, a rogue employee cloned the code base and then took it to some Eastern European programmers to reimplement the website. It was litigated based on copyright and trade secrets - no patents were available for that lawsuit.
However a different employer of mine took out two patents based something I did when I was there. Since then a number of companies have infringed on those patents, and could be sued for large sums. By luck the patents have wound up owned by an organization that is unlikely to ever let them go to a patent troll. But I've spent years fearing that that work will get used to stifle independent innovation.
Only if software is the product you're selling. I worked at two companies that had patents on software algorithms, but our software was just a reference implementation. Explain to me how a company like ARM would operate relying on just copyright and trade secrets. I think ARM represents a great and valuable business model, and patents enable that business model.
> Instead software patents seem to be mostly used to sue people who independently invented the same thing.
I don't think you can point to any statistic to back this up.
>I don't think you can point to any statistic to back this up.
I think that was already clear from the word 'seem'. To me, the same thing seems to be true, actually. That /might/ be caused by the number of software-patent cases I see that fall into this category, or are about trivial inventions, while I simply don't see the cases that do have some merit.
On the other hand, I've never yet seen a software patent related case where I could actually relate to the suing party.
Having money in the bank means that can afford lawyers, but it also means that you are a more attractive target for litigation as you are then worth suing (if they are after cash).
[NB The startup I co-founded years ago was involved in deeply unpleasant litigation soon after getting our first round of VC investment. Fortunately our lawyers made them go away.]
There are too many unscrupulous, unethical people about, out to make lots of money for zero real creativity, value, and effort. They are forever inventing new artifices towards this end, be it indulgences (sin forgiveness chits), fractional banking (credit out of thin air), software patents, CO2 indulgences, etc. and then getting fat on trading them.
We are so far gone that many people can no longer even discriminate between real value and these things; sinking as we all are in this sea of BS.
Actually it also made me click through to the paper, which also has this title. This is much better.
That struts and frets his hour upon the stage. And then is heard no more: it is a tale. Told by an idiot, full of sound and fury, Signifying nothing.
[edit] also thinking about it, is a reference to god as well, as the tale is life, but the teller is the idiot.
I am not saying anyone supports patents. I'm saying hackernews should be more consistent - or otherwise admit that "creating value" isn't an end in itself. It's the wrong question.
Edit: I think the current title is fine. ("Patent trolls created $29B extra cost in the U.S. last year") I just want us to be clear that one guy's cost is another guy's "value".
I would love to see more cooperative-optimistic-tit-for-tat solutions to the prisoner's dilemma, so that instead of just saying "creating value" we are more specific about the whole system.
This is what (an unnamed restaurant review site) gets wrong and Google gets right, in my opinion. Google adds value whereas (unnamed) removes it.
Of course in a naive sense (unnamed) is also "creating value". Just as these patent trolls are.
We don't condone it, but let's be consistent.
Patent trolls who serve no purpose other than to litigate don't create value. And if we take one in particular Lodsys. They had the potential to destroy a lot of value when they sued small app developers many of whom simply didn't have the resources to defend themselves.
If an enterprise shrinks the size of the economy by $1b but in the process gets people to pay it $100m, then by getting people to make that payment it has "created 100m of value." Nevermind the effect on the rest of the system.
Please be consistent here.
The simplest way to demonstrate it is to look at the industrial revolution. Vast amounts of people were put out of work because automation made the economy more efficient. $1bn industries, as you say, were permanently shrunk to $100m industries, left right and center.
All those workers displaced eventually found employment elsewhere, and it turned out to be one of the largest economic booms in history.
Would you rather we still weave our sheets by hand employing 100 times as many workers as the modern textile industry, each one costing $1000?
Because that's exactly the consequence of your line of argument.
In the typical entrepreneur sense, you are selling/renting something that the other people want or need without any coercion whatsoever.
I bet the people being sued didn't want or need to pay lawyers and settlements, short of being forced by the system.
I'm disagreeing with the definition of "value" used around here - for normal companies, not patent trolls - and to show my disapproval with this notion I insist that the word value be applied here as well.
And when I said system, I meant the law forcing the companies to pay, not the troll using the law as a tool to create profit.
We could enter murky territory on whether other forced forms of income (taxes or fines, for example) provide value, they are forced but on the other hands they are supposedly used to common benefit, instead of private profit.
A good way to put it would be: "Patent laws and companies created a market for Patent Trolling that is worth $29B per year."
By your standard, Bernie Madoff created $18B in value.
HN as a group may have many contradictions and blind spots in its general wisdom, but you haven't identified one of them.
Give me your guess, if you think that the article quotes the systemic cost and not the part that is booked as revenue.
Wealth transfer is not necessarily value creation. This is obvious, because if we all pay everyone else on the planet $100 a day we don't magically have an econmy worth $1.3 × 10^24.
Wealth transfer often destroys value. Wealth unjustly taken away from value creators becomes a disincentive to produce.
You have to look at wealth creation. Are the patents involved in litigation honest to god inventions that made the world a better place? Wealth was created. Eg, the first retrovirus medication.
Are the patents involved non inventions that were of no use, but are broad enough that patent trolls can, without merit, start seeking rent on products from companies they no part in creating? No wealth was created. Eg Lodsys.
In general, different segments of HN disagree over which patents fall into which categories, but the vast majority agree, myself included, that software patents basically all fall in bucket #2.
This is my last reply to you, as I'm pretty sure by now that you're just trolling, and HN doesn't need to be polluted with this thread.
Nobody blames landlords of costing businesses billions of dollars a year - they own property and businesses pay rent. Nobody builds a business using a landlords property without paying rent, then plausibly claims their legal costs defending themselves against landlords asserting their rights are somehow an unexpected and unfair imposition. (Well, actually food carts probably do exactly that, but nobody writes academic papers claiming billions of dollars worth of cost for them…)
I haven't read the whole paper, but I wonder about two other numbers - 1) how much money did the NPE's pay to the inventors of the patents they're defending (who under the current system have every right to choose "selling their patents/IP" as their means of monetizing their inventingwork)? and 2) how much would licensing the existing patents (instead of running up legal bill later) have been, compared to the $29B losses described here?
I'm not saying the current system is "right", but it is reasonably well understood. I can only assume the $29B cost is accepted as a "cost of doing business" by companies who know (or even just suspect) they may be infringing patents? (or perhaps it's not an assumed "cost", and many businesses choose to gamble that they'll never be held accountable?)
If the invention is junk (like the Lodsys patents), it doesn't matter how wealth has been transferred around; no value has been created. Any rent sought on those patents is an economic distortion and (in my opinion) immoral.
I didn't even start this off to take a side (though I guess I have by now) - I was just pointing out a fallacy in GP's thinking.
Perhaps patents should require proof of use- If not by the inventor, then by the patent owner within a probation period of 3 years or so.
Now don't get me wrong, I love small tech companies. The engineering companies I have worked at have been exclusively small tech companies. They're great. But most in the industry have constructed this idealization of the morally virtuous startup founder. Everyone likes free things, sure, but big companies at least seem to acknowledge the fact that they're inflating their profits at the cost to someone else. Small company culture doesn't do that. The AirBnB debate recently was a great example. People were decrying regulations that could undermine AirBnB, but it didn't even register to people that a lot of the "value created" by AirBnB was actually just costs that were shifted to the other residents of the building/surrounding community.
I see the same sort of thinking when patents come up on HN. There is this presumption of the morally virtuous innovator. Therefore, every $1 he must spend paying NPE's must be $1 of dead weight loss to society. There is no conception of the fact that some part of that $1 represents money he would rightfully have had to pay to the original patent holder, because (statistically!) he's not morally virtuous 100% of the time.
However, if we are talking about people who are just shifting money around without creating anything other than unnecessary work, then they are a cost to the wider economy, they are not increasing it's value.
Assuming it only costs 10c to print a $20 bill then the cost to society of the act was a meagre $5000. I imagine the entertainment value provided would been worth more than $5000. Thus the act could have been a net gain for society.
The other $995,000 would have been redistributed through-out the money supply in the form of deflation.
Imagine two possible worlds. In the first world, when you -- as an engineer -- have a particularly difficult problem in front of you, you first go to a catalogue of solutions, look up the problem, and see if there is a solution already. If so, you pay a nominal fee to the solver, incorporate it into your design, and move on to unsolved problems.
In the second world, you have the same particularly difficult problem, but there is no catalogue of solutions, so you buckle down and solve the problem yourself. Later, someone looks at your solution and declares that a portion of it is similar to something they wrote down in the big pile of disorganized ideas. You didn't know this, because looking in the big pile of disorganized ideas is a lot of work and rarely pays off, because most of the ideas are half-assed and don't really tell you anything you didn't already know. An exorbitant fee is then charged.
There may be some industries living in the first world; I can imagine that in -- say -- chemical engineering, it is very easy to organize the ways to produce chemicals by the formula of the desired end product, and that the standard for describing how to produce a chemical leads to a pretty useful solution.
The software industry, however, isn't.
The patent system is too broken. Anyone who writes software should hate patents. Anything you write will infringe on some patent out there. If all you want is money then writing software isn't the best way to go about getting it.
A better system would be one where patents do not give protection to ideas, but do give minor tax incentives to the people who own and use a related patent in their business. Then their ideas are shared, but not using the patents gives them nothing, and they cannot block others from using the all too often obvious ideas.
I write software (open source on GitHub) and don't hate patents. I hate bad patents, the kind you can infringe upon just by writing software, but lots of patents aren't bad patents. E.g. IBM has a number of patents on compiler optimizations. Am I going to stumble upon Steensgard's pointer analysis by accident? No, the only reason I know about it is that IBM let him publish a paper based on the research they paid him to do at Watson.
Yes! Yes, you could stumble on Steensgard's algorithm. Like convergent evolution in nature, it's easy for requirements to shape implementation, and for multiple implementations to naturally overlap. There are so many software patents out there- it is a minefield.
Additionally, engineers at large companies are discouraged from doing their own patent research, because of the risk of losing a "sorry, we created this organically in good faith" defense in litigation.
Publication of patents does not offset the chilling effect and legal overhead on our industry.
A (neo-)liberalist would not approve the difference, as long as money is made, but the patent system is government regulation of trade, so he would not approve of the patent system itself if he took his conviction to the logical end.
Suppose X is sandwiches. If I'm allowed to charge anyone who makes sandwiches, yes, my patent is worth lots of money. It's not "worthless" in that sense. It's worthless in the societal sense: allowing my monopoly provides no value to society.
Nobody thinks you can't make money from software patents. Some people think software patents do not provide value to society by encouraging innovation, which is the stated legal intention of intellectual property law.
When people call patents valueless, they're probably talking with respect to society as a whole.