Update: See the Rational Patent Exchange’s (RPX) Defensive Patent Aggregation Service.
I’ve blogged a lot lately about all the various patent battles in the smartphone and other spaces, leading to wasted money on litigation, patent acquisition, licensing, and so on–billions and billions of dollars.1 We have Google spending $12.5 billion on Motorola Mobility in part to obtain 17,000+ patents,2 to defend itself from patent threats by competitors like Apple and Microsoft. Patents are being “weaponized,” forcing even non-aggressive companies like Google to acquire patent shields.
This is good for Google–though a waste of its precious capital; think how many jobs and innovative research could those billions could have gone towards–but not everyone has the money or clout to acquire such patent arsenals for defensive purposes. Thus, the larger corporations are protected from competition from smaller players, leading to barriers to entry and oligopolies.3
I’ve often wondered if there could be a way for some kind of “open access” patent pool to be formed for smaller companies and little guys–something solely defensive. I discussed something along these lines in my 2007 Mises blog post Open Source Patents, which mentioned the Open Invention Network (OIN) (Wiki entry), a type of patent pooling arrangement which seeks to defend the Linux System from patents. It does this by first purchasing patents, and then offering them to anyone, on a royalty-free basis, so long as the licensee “agrees not to assert its patents against Linux.” If the pool grows big enough–and it’s funded by “a diverse group of companies including IBM, Novel, Philips, Red Hat, Sony and NEC”–then many companies fearing liability from one of the patents owned by OIN may enter into this arrangement for defensive purposes. OIN does “not anticipate that there will be any royalty streams.” Its purpose is simply to force companies not to sue “programmers, independent software vendors, distributors and businesses” who use Linux.4
Nice, creative idea. But one problem is that if this idea takes off, the size of the OIN pool snowballs and pretty soon there are no companies left who can afford to assert their patents against Linux, because the pool has so many potential patents that can be asserted against this company. What it means is millions of dollars have been spent: to pay salaries of OIN employees; to purchase patents in the pool; to acquire the patents in the first place; and by third parties who acquired their own patents that they now agree not to assert. So we have tens of millions of dollars of investment, and thousands of companies holding patents that they have all agreed not to use against each other.
In other words, it’s similar to the situation that would exist in the absence of IP–without all the effort and waste to get a bunch of flaccid scraps of paper.
In any case, given the existence of patents and their increasing use to squash competition and erect barriers to entry and to establish de facto oligopolies, given the rise of companies like Intellectual Ventures and other “patent trolls” who acquire vast arsenals of patents to use aggressively and offensively against innocent companies, some kind of “defensive” patent pooling might make sense. This would be akin to what Google is doing, except that Google’s defensive shield is only for its benefit. A patent pooling arrangement along the lines of the Linux pool noted above could be of value to more people.
One could envision a more general purpose patent pooling arrangement: either devoted to a given sector of technology, or open to all forms of technology. For example, let’s imagine I form the Patent Defense League, a non-profit corporation that allows any individual or company to join, so long as they abide by certain rules. These could include: (1) the company can never sue a fellow member of the PDL for patent infringement; and (2) the company has a contractual obligation to temporarily assign the patent to any other PDL member who needs it for defensive reasons. In this way, the PDL would effectively create a huge arsenal of patents any member could use defensively. The PDL could also accept patents assigned by its members, or could acquire some third party patents using member dues.
In return for giving up the right to use patents aggressively (against PDL members) and making them available for defensive use by all fellow members of the PDL, if a member is ever sued for patent infringement by someone outside the PDL, it would have access to the (hopefully huge and growing) PDL patent pool, to countersue the patent aggressor. And of course the PDL could have litigation insurance and arrangements with law firms ready to help small companies in patent countersuits. Lots of companies, I believe, would happily give up their right to sue a large (and hopefully growing) number of member firms in exchange for immunity from suits from those members and access to a huge defensive patent portfolio (an informal web poll I did a few years back indicated many people would give up their right to assert patent rights in exchange for patent immunity–see Patent Rights Web Poll).
Consider how this would work. In the typical “oligopoly” situation noted above, Apple (say) may now be afraid to sue the Android smartphone platform, for fear that Google can mine its own patent portfolio to find something to countersue Apple for. If the PDL idea worked, Apple would likewise be afraid to sue any small company who is a member of the PDL for fear that somewhere in the tens or hundreds of thousands of patents in the PDL patent pool, there would be something that could harm Apple.
Now, there are several aspects of such an approach I am not sure about. Presumably the PDL would charge membership fees. Perhaps the fees would be discount for every patent the member can contribute, so that those contributing more have to pay less in terms of money. (I’m envisioning $50k or so for smaller companies, $100k for larger ones.)
But what incentive do members have to keep filing and acquiring additional patents, instead of halting patent acquisition effort and expenses and “free riding” on the other patents in the pool? If all the members stopped filing for patents, eventually the patent pool would shrink and be of less value. So incentives would have to be built into the structure of the pool so that member companies still file for and acquire patent applications. Presumably the right to sue non-PDL members would provide some incentive to keep acquiring patents.
Also, if members with 1 patent or 10,000 patents could join, why not companies with no patents? As long as they pay a suitable membership fee, let them in and have access to the patents for defensive purposes.
One problem is that the PDL’s pool would be useless against patent trolls. However, patent trolls usually just want money. So they just serve as a kind of tax. But competitors often want to get an injunction to shut down the product lines of their competitors. Having to pay a “tax” to a troll is usually less of an existential threat to a company than is the patent injunction threat from a competitor. So if the PDL deters this kind of patent injunction threat, that is a huge benefit.
Even if these challenges do not prove insurmountable, it’s possible the state would torpedo this as some kind of antitrust violation, or some kind of patent “policy” grounds. Sounds absurd, but then the state does absurd things. On the one had, use of patents seems to be contrary to the spirit of antitrust law. Indeed, as I’ve noted in the past, there is said to be a “tension” between the federal antitrust law (which penalizes the formation of “monopoly power”) and patent law5 (which grants monopolies).6 And for this reason there has been talk of possible antitrust scrutiny of the Apple-Microsoft-RIM-et al. consortium’s $4.5 billion purchase of Nortel’s patents–the consortium beat out Google’s $3 billion bid, thus depriving Google of a defensive patent shield from patent suits against its Android smartphone platform by these and other competitors.
But one could imagine the state clamping down on the PDL scheme. After all, it’s unfair to let companies have too big of a defense against the patent threat. That would thwart the very purpose of the patent system, heavens to betsy! Or the FTC could jump in and claim that this pooling is anticompetitive, even though the purpose is obviously to permit competition to thrive, to block the anticompetitive effect of aggressive patent lawsuits. Who knows what the schizo feds would do.7
Now, as noted above, even if this scheme worked, it would be a huge waste, but maybe a necessary one, given the patent system. Still, it could reduce the barriers to entry and antitcompetitive threats posed by patents. Maybe if there were millions of patents held by the PDL and other patent pools and by larger companies, but no one was using them for fear of retaliation, people would wake up and say, hey, why don’t we just get rid of this hundreds-billion-dollar deadweight loss on the economy? If we are not suing each other, why not just do that for free?
Update 3: See my post Taiwan’s Defensive “Patent Bank”
Update: In the comments someone sent me a link to this interesting talk on “A Defensive Patent License Proposal” by law professors Jason Schultz and Jennifer Urban of the Samuelson Law, Technology & Public Policy Clinic at UC Berkeley Law, which gives some good background on some of the issues discussed in this post, and in fact has a somewhat similar proposal to the one sketched here.
Update: After further considering the DPL proposal, I sent an email to Schultz and Urban and had a bit of correspondence. An edited version of my comments on the DPL is below:
I’m a patent attorney in Houston, and general counsel for a high-tech startup. I have also long argued for the abolition of patent and copyright on libertarian/property rights grounds (my lectures and publications opposing IP are collected at http://www.stephankinsella.
com/publications/#IPor http://c4sif.org/resources/, including my monograph Against Intellectual Property).
I posted The Patent Defense League and Defensive Patent Pooling today on the Mises blog (and on c4sif.org), about the possibility of forming a large general defensive patent pool to protect against patent aggression, and someone pointed me to your youtube video on your defensive patent license, which I just watched. I found it fascinating and promising, and very similar in some ways to what I’ve been groping towards. A few questions or comments, if you do not mind–First, have you written anything on this in more detail? I’d be curious to read it.Second, you talk about the “commitment to defense”. Apparently this does not mean that DPL licensees can use the patents of other licensees, defensively, in countersuits against outside patent aggressors. This is one feature of the pooling idea I have in mind that I think could make a big difference.I am general counsel of a small high tech company. Small tech companies live in fear of being sued. When tech companies file IPOs their competitors often start hitting them with suits, since the smaller company has limited resources and a limited patent arsenal to hit back with.8 If a startup company could have access to a pool of 1000, 10,000, 50,000 patents, and could pick thru them and use some of them to countersue a patent attacker, that would make make the small company as invulnerable to patent suits from big competitors as they are from their fellow large competitors.
It seems to me the threat of being sued by a fellow small competitor is limited–small competitors tend to have limited resources and can’t afford $2M each for the lawyer’s fees. So if small companies were to join the DPL, that’s fine, but it doesn’t remove much of a threat. The real threat is the big companies, and the patent trolls. The big companies won’t join the DPL and so the only defense is to be able to countersue. To do that you need a good chance of finding a patent you can use against them in a large portfolio you can search through. The pooling organization I contemplate would provide this, though I am not sure of the mechanism for this–I envision each member of what I called the Patent Defense League (PDL)–would contractually agree to somehow license or maybe temporarily assign its patents to a fellow PDL member (maybe for some specified fee–say $100k) as needed to defend from a patent countersuit.
I agree with your idea that members of your DPL have to license their whole portfolio. Otherwise, you are right, there would be gaming that would make it useless. However, this means that 1 patent gets you in, as well as 1000 patents. So why not allow 0 to count? My idea for the PDL is that anyone can join, but you pay less if (a) you are smaller; or (b) the more patents you have.
I think I’d be a bit concerned about gaming with this idea (allowing people to join by just paying a fee) because they might join only when they get sued, so they can use patents in the PDL patent pool defensively, against their opponent. However this would happen in any case, and so what? You could have a provision saying that IF you use any patent defensively within, say, 2 years of joining, then you commit to paying dues for 10 years. Something like that. Then the PDL benefits because it has more resources for staff, administration, pro bono for smaller members’ legal needs, and acquiring third party patents. I see the PDL being formed as a separate organization (corporation) as another benefit to the PDL approach–it can actually own patents in its own name–those it buys, or those donated to it by members.
I wonder also about DPL members who really only care about freedom to operate and defensive use–would they have an incentive to continue filing patent applications after they join? The earlier patents they had get them in the DPL club; so additinoal patents don’t help them there. And if they don’t want to sue people offensively, why keep acquiring patents? I suppose they still need them for defensive purposes if sued by an outsider.
I note that trolls also will never join your DPL, so this does not protect against that. But I realize you are aware of this.
You addressed the gaming issue concerning junk patents and subsidiaries or affiliates–I am curious what the mechanism is to ensure subsidiaries or affiliates or parent companies are also bound–I would think the DPL licensee cannot just bind its subsidiaries. Does it have to somehow contractually promise to make the affiliates also enter the license?
You said you have no antitrust concerns, but I wonder. It’s a wacky legal landscape. I mention this in my post.
I’m also confused about the mechanism you have in mind for “taking a license.” You said that you can take a license when you need to or want to. I gathered you have in mind A and B are already somehow DPL members though again, I am not clear on the mechanics of how this happens). And then A sees that he “needs” to license one of B’s patents, and so he just … informs B that he is doing this? This does not make sense to me. In the great majority of cases a company does not know whether it is infringing the claims of a patent. If there are 100,000 patents in your DPL pool, why should members have to affirmatively search them to decide which ones they need to “take a license” on–isn’t a covenant not to sue by everyone else sufficient? Each licensee covenants not to sue any other DPL licensee. Why does each one need to “take” a license. Seems to me it’s better to somehow signal one agrees to be bound by the license–that allows others to realize that company’s patents are now not a threat, and it then allows the company to be immune from suit from other licensees.
You addressed the possible issue of the complications arising when a DPL licensee is acquired by a larger company (who is not a DPL member). I suspect this ought to just trigger the 6 month withdrawal period or something. But I fear that if there is a chance this could jinx an acquisition a lot of companies will be leery of the DPL. This is especially so now that patents are so important for defensive purposes–e.g. the $12.5B Google purchase of Motorola Mobility. On the other hand, I suppose a big parent could buy a small DPL member to use its patents defensively against its large competitors. So maybe this is not a problem.
Query: Suppose DPL member A sues outsider C for infringement. Can C simply take a license out on the patent being asserted against it, joining the DPL and causing the suit to evaporate? Maybe this reduces the offensive value of patents held by DPL licensees…
One more query regarding the issue of revoking: suppose A leaves DPL and is selling products arguably infringing some of the patents owned by DPL members. Before A left it had a license. AFter A leaves it does not–well, it might not, if the other DPL patentees “revoke” the DPL for that turncoat company. But suppose A leaves, and then rejoins DPL a year later. Now, fellow DPL licensee B sues A for patent infringement for sales of products during that year. Can B do this? Does suing retroactively revoke the license that A had during that year? Or did B have to revoke the license to A during the year A was awol? I think you need to have a clear default position and a mechanism for revocation. My view is that revocation should be presumed upon exit.
Also — you clarified that someone can use DPL with no patents. I thought I heard you say in the video that you can join the DPL with no patents, but only if you have some patent applications pending, something like that. Maybe I misheard you, in which case I am unclear how one joins. But if you meant that, what I was getting at was just a company that had no patents or even patent application or even plans to patent at all–I would think there ought to be a way for them to take out these licenses too.
One more thing occurs to me–I and others like the Mises Institute use CC-BY. However, I am not myself 100% sure CC licenses are actually enforceable. After all there is no consideration necessarily, and it’s also hard to prove that there was a license. There’s just a statement on the footer of an article (say) saying that the article “is” licensed under CC-BY. What does it mean to say it “is” licensed? When is it licensed?
Suppose A uses the article and then a year later the author removes the CC-BY notice. How does A prove that there WAS a CC-BY notice? Unlike a normal executed, written license agreement, where the parties have proof of a license, a date, names of parties–this does not exist in CC situation. I suspect that in most cases at least an estoppel would arise. But it’s murky. I hope your DPL process has more certainty about it: more of a concrete step someone takes to join: I imagine some kind of public registry on a website of current licensees, and the date they joined.
It also seems to me that to join the DPL, you do it not by “taking a license” (in part for the reasons I mentioned before: you don’t always know what patents you need a license from), but by agreeing to the DPL terms: that is, merely agreeing to the DPL terms, agreeing not to sue any other licensees, is enough to invoke the other licensees’ covenant not to sue. Of course there is a potential gaming problem here: I have no patents; I join the DPL club to get protection, but later when I have patent ideas I leave the club, and never really gave up anything. Still, this gaming issue does not seem significant.
- Google buying more patents to defend itself; Microsoft Copyrights –> Patent Dominance; Patent Cross-Licensing Creates Barriers to Entry. [↩]
- Google pays $22 million per patent to defend itself. [↩]
- Patent Cross-Licensing Creates Barriers to Entry; Intellectual Property Advocates Hate Competition; also Kevin Carson, Criminalizing Competition. [↩]
- For more on patent pooling, see In the Pool, by patent shill Gary Odom, aka “PatentHawk”. [↩]
- See my post Price Controls, Antitrust, and Patents. [↩]
- Are Patents “Monopolies”?. [↩]
- For one apparent insinuation that such patent pooling should not be permitted, see In the Pool, by patent shill Gary Odom, aka “PatentHawk”. [↩]
- See, e.g., the discussion of Optium’s and Neophotonics’ use of this tactic, in Ideas Are Free: The Case Against Intellectual Property. [↩]