I should start off by saying that I am not a fan of the patent system as a means of establishing IP rights in software. My critique of the system is (at least) five-fold: 1) there is widespread evidence that the system of granting claims in the United States, at least, is deeply flawed (see the scholarship of Doug Lichtman, Mark Lemley and others, for empirical evidence and policy arguments related to it); 2) the process is prohibitively expensive for small firms and individuals to have a fair shake when compared to the patent-filing goliaths (witness not just the cost of prosecution of patents, but the $3 – 5 million price-tag of most patent litigation); 3) the extent to which these first two factors and others favor incumbents over newcomers — and potential innovators; 4) there are also special cases, such as the standards-setting processes for software, where these and other problems arise and are particularly acute (Lichtman’s piece on patent hold-outs is helpful here, too, and a bit counterintuitive); and 5) the patent system often works at cross-purposes with the goals of teaching and learning, in contexts including computer science, biotechnology, and so forth (we at the Berkman Center have an active research project on this topic, funded by the Revson Foundation). Others — Terry Fisher, Lawrence Lessig, Yochai Benkler (see pp. 437 – 439 of Wealth of Networks for a particularly strong form of the critique of software patents), Jamie Boyle, for instance — have made these critiques more forcefully and more convincingly than I have; this is not novel stuff. While of course I disagree on some fronts, often at the margins, with each of these scholars, I owe much to the line of thinking that they’ve blazed in all of what I do.
OK, enter the complicating factor: what if you are an entrepreneur who is devoted to creating a wonderful new generative (to use Jonathan Zittrain’s term) technology, say in the Web 2.0 space? One hard problem faces you early in the process. So, you get the part about being part of a development community, building your cool new platform, sharing it in various ways, making a market for your services, and generating a return for your investors. But what should you do about patenting? Say you, like me and others, are queasy about the current patent system (“yuck, I just don’t like them,” or “I don’t want to participate in that mess”), you don’t have much time or money, and you face an uphill battle in your crowded marketplace already. What’s a sensible, reasonably public-spirited, honest entrepreneur trying to make a living and a return for investors (if you’re lucky, or unlucky as the case may be, enough to have them) supposed to do?
Some investors I’ve heard speak or write on this topic — Joichi Ito, Brad Feld, A VC, and others — are opposed to software patents generally. As I noted above, I’m in the same camp. But most VCs do not share this view: IP rights are increasingly viewed as an asset, or as a defensive necessity, or both. Why? Well, some argue there’s a market emerging in patent rights (see Kenn Cukier’s long and provocative piece in The Economist on this score; but see the largely failed Ocean Tomo patent auction). Others believe that patents are necessary to create freedom of action in most fields of software.
This conundrum is real, and I’ve learned a great deal from finding myself caught in it. As with many people who teach, I have “outside activities” beyond my work at HLS and Berkman. Outside activities are good because they help pay the mortgage in the Boston area. They are also good because you learn things about the real world and keep you from becoming an ivory-tower-bound caricature of an academic; my teaching and research are plainly informed by outside work in the technology field. My outside activities have been primarily as an investor in RSS-related technologies and helping real entrepreneurs found companies in this space. Outside activities are a pain in the neck because I, anyway, constantly feel a tension between my academic identity (teacher and researcher of interesting Internet law topics) and my outside-activities identity (investor and participant in the marketplace). Much of this tension can be dealth with through disclosure — more on that in a moment. Nowhere is that tension more acute than in this patent space.
So, here’s the hard question: if you are an entrepreneur or investor who dislikes the patent system (either you want it scrapped altogether for software or you want it reformed), what do you do? Do you sink the money and time into participating, one-off, against the big guys and try to patent what you’ve done; do you sit it out and take your chances; do you license from someone the protective cover of some patents; or do you try to find some other solution? I do not know the answer, but I’m genuinely trying to puzzle it out. There probably is no right answer for every entrepreneur and every part of the software space (think of it as a variant of Terry Fisher’s disaggregation argument).
One other fact to be noted: large companies in the ICT space — IBM, Microsoft, HP, Sun, whomever — have massive-scale patenting operations. Some are newer to the patenting game but getting geared up, like Google. These companies have patent portfolios in the thousands or tens of thousands, with claims many times that many. IBM alone has over 31,000 patents and applies for thousands more each year. In the Web 2.0 space, Apple has begun to publish a series of applications. Google has applied for patents related to embedding ads in RSS feeds. No doubt the Intellectual Ventures team has thought about working in this space, or will soon. These entities are far from alone — patent applications related to RSS and other Web 2.0 technologies are coming out all the time. (Smaller companies, like Technorati, have applications pending as well.) As I’ve written before, there’s a difference between obtaining IP rights and enforcing them, so it’s not certain that this emerging thicket of patents will preclude innovation. Apple may never sue anyone at all for infringement of the many claims that may well be granted to it. But should an entrepreneur run that risk?
Here’s one idea. What about working with other small-scale entrepreneurs to do what the big guys do? What if you were to hold your nose and apply for patents that protect your work; share your patents with other little guys (and gals!); agree to reasonable cross-licensing terms for other entrepreneurs; and create a dual-licensing regime to allow highly favorable (no- or low-cost) licensing terms for .edus and .orgs (to solve problem number 5, above)? This is the idea that Jim Moore has been championing in the Web 2.0 space. He’s got a long and thoughtful post about it here. You may not agree with the strong form of his argument, but it’s very provocative. A company that we’ve both been working with, Newsilike Media Group, has applied for a series of Web 2.0 patents, which will soon begin to be published (as applications, to be clear, not as issued patents). Jim’s idea, which I support as an important experiment in this space, is to try to create freedom of action for start-ups and others in the Web 2.0 space by blazing a trail. While Jim and I disagree on some aspects of this matter — we have a genuine, long-standing, always-spirited conversation going on this score from which I’ve learned and continue to learn enormously — I have to say I welcome his efforts to bring innovative thinking to the space.
A key component of this strategy is to try to innovate in the area of no- or low-cost licensing for non-profits and educational institutions. Patent Commons, Science Commons, and others have been working on similar or related ideas. We should all be eager to learn more about this line of thinking. I think it’s brilliant. I very much hope that entrepreneurs will one day have an easy way to subscribe to a license — just like a Creative Commons license for copyrighted works — to give away some or all of the IP rights that you’ve obtained. Maybe it’s limited to certain classes of users (the .org and .edu idea), or fields of use, or the like, but allowing you to stand behind your beliefs while competing on reasonable terms in the marketplace. I’d love to see a way where the little guy doesn’t have to unilaterally disarm him or herself, but can do so in a way that lets him or herself sleep at night.
I am wondering what Yochai would say about this idea as a practical matter. In Wealth of Networks, he writes: “Even if the patent owner has a very open licensing policy — say, licensing the patent non-exclusively to anyone without discrimination for $10,000 — most free software developers will not be able to play.” He continues, “If working on a problem requires a patent license, and if any new development must not only write new source code, but also avoid replicating a broad scope patent or else pay a large fee, then the conditions for free software are thoroughly undermined.” (p. 438). So, if you buy this argument but you also live in a future in which large companies get broad claims via their Web 2.0 patents, what, absent reform that anti-patent people would support, should the entrepreneur do? To me, it’s a hugely vexing problem that requires innovative thinking, challenging the “yucky” feeling many of us have about software patents and honing in on short-, medium- and long-term solutions.
(One further, personal note: I’ve long had on my disclosures page my Personal Patent Profits Pledge: if I make money from any patent activity that is not consonant with my beliefs as to what is good public policy, I pledge to donate those profits to the Berkman Center or a similarly-situated institution that is working on the study and appropriate reform of the patent regime.)
There is much more to be said and learned on this topic. I welcome debate and critique; I think that the cyberlaw community, myself included, has been so focused on copyright that we haven’t delved as deeply into patent (and trademark and trade secret, perhaps, too) as we might. I look forward to participating actively in a pro-innovation, good-for-the-entrepreneur, good for society-at-large outcome. Is there a form of software patenting that can help drive innovation, not frustrate it?
* * *
Addendum, with thanks to Phil for a pointer, here is a bit of a terrific short essay from Paul Graham’s “Are Software Patents Evil?“, which covers much of the same ground:
“Do patents encourage or discourage innovation?
“This is a very hard question to answer in the general case. People write whole books on the topic. One of my main hobbies is the history of technology, and even though I’ve studied the subject for years, it would take me several weeks of research to be able to say whether patents have in general been a net win.
“One thing I can say is that 99.9% of the people who express opinions on the subject do it not based on such research, but out of a kind of religious conviction. At least, that’s the polite way of putting it; the colloquial version involves speech coming out of organs not designed for that purpose.
“Whether they encourage innovation or not, patents were at least intended to. You don’t get a patent for nothing. In return for the exclusive right to use an idea, you have to publish it, and it was largely to encourage such openness that patents were established.”
Worth reading in full.
John Palfrey 
You should also check out this:
http://www.paulgraham.com/softwarepatents.html
What should Web 2.0 entrepreneurs do about software patents?…
nice…..
Dear Professor Palfrey
I am an IT lawyer in the UK with a strong interest in software patents having worked with developers and IT companies for many years.
I can see that there is a great deal of pragmitism in your article and for US start-up software companies there is a real challenge because, whether they agree with software patents or not, they are within a system that does accept them. In such a scenario, the reality for many is that they need to file defensive patents in the US purely from the point of view of protecting the financial stability of the company by warding off possible litigation (although, even then, it costs a great deal to simply challenge a spurious claim(s) because of the ‘shades of grey’ problem of how patent claims are drafted). However, as someone who does not believe in software patents, this is as far as I would go in the US, as to consider issues such as cross-licensing, is to take a further step in accepting a fundamentally flawed IP system for software. Although cross-licensing and similar strategies might alleviate some of the negative effects of the patent system, these strategies effectively encourage the patenting of software (including the waste of human resources and money which is diverted away from software development to filing and maintaining patents and analysing third party patents) and detract from probing deeper into the very essence of patenting (monopolisation) as a mechanism for fostering innovation in the software industry. In 2005, I spent a considerable time researching (not as an academic but as a practising lawyer) whether the patent system was appropriate for software in the context of the debate over the’Software Patent Directive’ in Europe and recent case law in England. Although I am not against computer-assisted inventions (provided the software involved is not patented per se), my overall conclusion was that copyright (perhaps tweaked slightly) rather than patents is more appropriate to fostering innovation in this sector and protecting the investment made by developers. If you are interested, the article/paper can be found at http://www.groklaw.net/pdf/Swptft(final)(v2).pdf.
Kind regards
Cristian Miceli
http://www.lasporg.info
lasporg’at’googlemail.com
MARAZZI prin omnitechgroup
[…] When I researched Intellectual property in 2002 to find a suitable topic for my master’s thesis, the web was full with articles, comments and opinions on software patents. Most of them I found to lack proper significant scientific foundations and seemed to be more driven by pursuit of a more or less hidden agendas or plain religious beliefes. Given that I wanted to work on a sizeable problem for my thesis, I decided to leave software patents and decided to do some work on the European Database Directive. Revisiting the literature more than four years later, things do not seem to have changed a lot. The big voices of cyberlaw (e.g. Lessig, Benkler, just to name two authors I read last week) acknowledge the importance of the topic and the concerns they have about freedom of innovation, but only to continue with their exegesis of current issues of copyright law. John Palfrey writes in his noteworthy blog entry on Aug 3 that “the cyberlaw community, myself included, has been so focused on copyright that we haven’t delved as deeply into patent (and trademark and trade secret, perhaps, too) as we might.” This lack of academic involvement may easily explain the lacking fact base, but might also reflect the perceived practical importance of software patents as opposed to the so far more theoretical threat it had on the development of open source software. So what immediately caught my attention was John Palfrey’s claim that the VC community had recently discovered patent protection as a means of protecting their investment. This stroke me, because I hadn’t heard that before! While in other sciences, namely chemistry and biology, using patents to get seed funding for start-up companies is common practice, VCs in the software business were very reluctant to get involved in this business. This observation is mostly based on my contacts with my former colleagues from Swiss Federal Institute in Zurich, who now mostly work as patent attorneys or licensing experts in biotech and pharma, but hardly ever deal with software. I would be very interested in getting to know more about this. Is there any reliable source of facts to back this claim? The quoted article in the Economist is mostly based on qualitative evidence and non-industry specific figures on licensing revenues. How a well working patent system, also for software, could be working is well covered in Jim Moore’s idealistic essay “The software patent process can be your friend”. What is in theory a great idea (”disclose your ideas instead of keeping them secret and get rewarded with a limited state-guaranteed monopoly” that offers a “small company” protection from “the mercy of the market strength”) looks quite different in reality. Jim explains that the expensive and resource intensive system is mainly a tool of large companies and fails to offer protection to small companies, which could need the patent protection to obtain venture capital. But that is only one of the objection against the current practice on software patents. Another much stronger argument is the limited insight contained in many software patents. Jim claims that “patent documents covering software are inherently open source, in contrast to the closed world of software trade secrets—where code is buried in compilation.” But if you read a real patent application you will hardly ever find source code. A greate example are the current patent claims of IBM against Amazon over US patents 5,796,967, 5,442,771, 7,072,849, 5,446,891 and 5,319,542. While I do not dare to judge if the “person skilled in the art” will find anything inventive in these disclosures, the MSc in Computer Sciences which I am is having some difficulties. The empricial lack of reliance on patents by VCs to me was not only proof of the structural assymmetry between “the small” and “the big” in accessing the patent system, but also due to the fact that software patents were not used to protect innovation of new market entrants but to protect existing markets by putting up a thicket of defensive patents. Is this now changing? It would be interesting to get more voices especially from the VC community, what kind of protection they require for seed funding in software and what roles software patents play in their go-to-market strategy. […]
[…] As a follow-up to earlier posts on this topic: Microsoft has filed a large-scale patent application related to RSS. (A news story here and Dave Winer’s post on it here.) Microsoft’s patent application, filed on June 21, 2005, reads: […]
[…] VCs go software patents Published by wolf.richter November 13th, 2006 in Uncategorized When I researched Intellectual property in 2002 to find a suitable topic for my master’s thesis, the web was full with articles, comments and opinions on software patents. Most of them I found to lack proper significant scientific foundations and seemed to be more driven by pursuit of a more or less hidden agendas or plain religious beliefes. Given that I wanted to work on a sizeable problem for my thesis, I decided to leave software patents and decided to do some work on the European Database Directive. Revisiting the literature more than four years later, things do not seem to have changed a lot. The big voices of cyberlaw (e.g. Lessig, Benkler, just to name two authors I read last week) acknowledge the importance of the topic and the concerns they have about freedom of innovation, but only to continue with their exegesis of current issues of copyright law. John Palfrey writes in his noteworthy blog entry on Aug 3 that “the cyberlaw community, myself included, has been so focused on copyright that we haven’t delved as deeply into patent (and trademark and trade secret, perhaps, too) as we might.” This lack of academic involvement may easily explain the lacking fact base, but might also reflect the perceived practical importance of software patents as opposed to the so far more theoretical threat it had on the development of open source software. So what immediately caught my attention was John Palfrey’s claim that the VC community had recently discovered patent protection as a means of protecting their investment. This stroke me, because I hadn’t heard that before! While in other sciences, namely chemistry and biology, using patents to get seed funding for start-up companies is common practice, VCs in the software business were very reluctant to get involved in this business. This observation is mostly based on my contacts with my former colleagues from Swiss Federal Institute in Zurich, who now mostly work as patent attorneys or licensing experts in biotech and pharma, but hardly ever deal with software. I would be very interested in getting to know more about this. Is there any reliable source of facts to back this claim? The quoted article in the Economist is mostly based on qualitative evidence and non-industry specific figures on licensing revenues. How a well working patent system, also for software, could be working is well covered in Jim Moore’s idealistic essay “The software patent process can be your friend”. What is in theory a great idea (”disclose your ideas instead of keeping them secret and get rewarded with a limited state-guaranteed monopoly” that offers a “small company” protection from “the mercy of the market strength”) looks quite different in reality. Jim explains that the expensive and resource intensive system is mainly a tool of large companies and fails to offer protection to small companies, which could need the patent protection to obtain venture capital. But that is only one of the objection against the current practice on software patents. Another much stronger argument is the limited insight contained in many software patents. Jim claims that “patent documents covering software are inherently open source, in contrast to the closed world of software trade secrets—where code is buried in compilation.” But if you read a real patent application you will hardly ever find source code. A greate example are the current patent claims of IBM against Amazon over US patents 5,796,967, 5,442,771, 7,072,849, 5,446,891 and 5,319,542. While I do not dare to judge if the “person skilled in the art” will find anything inventive in these disclosures, the MSc in Computer Sciences which I am is having some difficulties. The empricial lack of reliance on patents by VCs to me was not only proof of the structural assymmetry between “the small” and “the big” in accessing the patent system, but also due to the fact that software patents were not used to protect innovation of new market entrants but to protect existing markets by putting up a thicket of defensive patents. Is this now changing? It would be interesting to get more voices especially from the VC community, what kind of protection they require for seed funding in software and what roles software patents play in their go-to-market strategy. […]
Nice…
interesting
Cool…