I first hit affordances in Don Norman’s book The Design of Everyday Things. Here’s some key points:
The term affordance refers to the relationship between a physical object and a person . . . . An affordance is a relationship between the properties of an object and the capabilities of the agent that determine just how the object could possibly be used. A chair affords (“is for”) support and, therefore, affords sitting.
The presence of an affordance is jointly determined by the qualtiies of the object and the abilities of the agent that is interacting….
But affordance is not a property. An affordance is a relationship. Whether an affordance exists depends upon the properties of both the object and the agent. (11)
The door handle on a car door is an affordance. It looks like something you can stick your hand into, and poof! you feel a latch and the door opens.
My mother had all sorts of trouble with a television cabinet because she couldn’t get the doors to open. She didn’t know that she had to press on the door to trip a latch to make it open. Pressing on a door to make it open towards you is counter-intuitive. And there was no affordance to suggest pressing. So she was prying the doors open with a wooden ruler. Why not?
Today I was reading The Edge, looking at responses to John Brockman’s annual question–“What scientific term or concept ought to be more widely known?: Daniel C. Dennett, the philosopher, had chosen “Affordances.” Dennett traces the idea back to psychologist James J. Gibson, who was considering how any organism adapts to its environment. Here’s Dennett:
The basic idea is that the perceptual systems of any organism are designed to “pick up” the information that is relevant to its survival and ignore the rest. The relevant information is about opportunities “afforded” by the furnishings of the world: holes afford hiding in, cups afford drinking out of, trees afford climbing (if you’re a child or a monkey or a bear, but not a lion or a rabbit), and so forth.
I have been reading A History of Technology, a collection of essays on what we know about the development of technology from paleolithic times to the present–er, well, until the late 1950s when the five-volume set was completed. Current technology is pretty keen on the emergence of the airplane, and space technology, semiconductors, and the like–that’s still science history fiction. But for the deep past, the essays chronicle what we can surmise from the careful inspection of artifacts, eventually supplemented with pictures and later, text. In one essay, H.S. Harrison develops his idea of technology as “mutation” of natural things into more useful forms. A stick is just a found object. A sharpened stick is technology. The technique of sharpening mutates the stick into a poker or spear point.
What’s not so clear in this idea is what leads to someone realizing how to sharpen a stick, or how to flake a stone to create a blade or how to exploit an electric current to create Christmas tree lights. One might consider, then, the idea that the natural world consists of affordances of all sorts–but it is a matter of having a “prepared mind” (as Pasteur called it: “in the fields of observation chance favors only the prepared mind”). We might then think of a prepared mind feeling the wind of observation blow through it, and finding at some opportune moment, and affordance, a possible thing to reach to, to use in a way suggested by its properties. We might say such an event is rather an epiphany than, say, an “innovation.” But we might also say that if we are looking for something new, then one way finding something new might come about is that we recognize an affordance.
Not every detail of an object is an affordance–or rather, an affordance connects a concern for survival (or doing something involved with survival, like, say climbing a sheer rock face for the fun of it) with something in the natural world that offers an advantage for whatever living being (human, tick) that has the concern. Any number of features on a sheer rock face may be the subject of discovery, but only those that offer an advantage to the interest of the rock climber at the moment are affordances. When a rock climber realizes that a tiny crack can be an affordance as readily as can a tiny ledge, we have an insight that, when used, might be called an innovation. But someone scanning a cliff face for discoverable features, without the slightest clue that someone might wish to climb that face, is unlikely to see any features of the cliff as affordances for a climb. Reports of the cliff’s features may well be useless to someone intent on climbing, unless by chance the reports happen to include ledges and cracks in some detail, and those bits are easy to get to.
On a flight once I sat next to a senior engineer from a major electronics firm. We got to talking about innovation and university research. He said (I paraphrase), “You know, you could give me for free all the inventions that universities produce and it would be meaningless. I don’t have time to paw through all that to try to find something useful. What I need is something useful that is directly related to the problems I am working on.” Or, one might say, he was interested in finding affordances for the work he was doing, not looking through reports compiled by folks who generally had no idea what he was doing. His purpose was to climb an engineering task, if you will, not to paw through hundreds of reports of inventions. He wanted a tailor, not to go through the racks at Value Village.
Benoît Godin is out with a new article on innovation–this time dealing with what he calls the marginalization of research in studies of technology change. The important terms become things like “coupling” and “transfer” and “innovation.” The claims in the literature, starting in the 1960s, was that research alone was not enough–discovery without commercial use was a waste of effort. The implication of such studies is that research in science exists for the benefit of commercial enterprise. One direction then is that commercial products represent an efficient way of delivering the benefits of research discovery to the public. Another direction, though, is that the potential for commercial products creates a speculative market that rewards some capital investors at the expense of others–a version of horse track betting, in which a few win at the expense of others plus whatever it is the house takes for putting on the event.
We might, then, consider three different purposes (and then one more) that create opportunities to discover affordances, and from an affordance develop variations, and from the variations select those that best suit one’s purpose. First, there are research affordances. One is seeking to discover, and finds a blip in the data, an observation that isn’t anticipated. Is it an error? Or is it something important? If someone recognizes it as important, then that blip may become the subject of new research–that in itself is an affordance–“Look, here’s something to study!” Or the something may be a marker of some sort–“Use this marker to determine the state of the system!” That, too, is an affordance–other research may now use this marker. Research affordances indicate the productivity of research in both advancing frontiers and providing the tools to map out those new frontiers.
A second affordance is that arising from livelihood. The research affordance is part of this class, as well, to the extent that a researcher seeks things that contribute to the livelihood of indefinitely conducting research, rather like a whaling captain that has never killed enough whales. But a livelihood affordance may be much broader than research, and comes about because an individual is trying to do something, and in the trying discovers something better or something other–and finds some advantage in this new thing. Adopting the new thing, like a rock climber figuring out how to use cracks as well as ledges, and then developing a tool (such as a fist or a cam) to exploit the cracks. The new thing becomes part of one’s livelihood. There’s no need for a commercial product version–each person can make their own fist, or craft their own cam. If a company can supply better fists and cams, that’s interesting but any product must compete with what a person can do for their self. If that prevents a “market” from forming, then there’s not much value added by such a market. If a company uses patents to prevent individuals from making their own cams, then we have a company attacking its customers using intellectual property. This is a primary problem for universities that take patent positions and then refuse to grant licenses for local making and using of the universities’ inventions.
A third affordance, then, arises from the effort to create commercial products. This effort is not simply one of engineering, but includes seeking things that have a payoff better than their costs. There’s not much room for altruism in corporate thinking, nor is there much room for following one’s vision and values wherever these may lead. Once shareholders are involved, especially shareholders fixated on the best return for their investment (which may be only in paying for tradeable shares and represent no direct investment in a company–just investment in obtaining shares intended to be sold later for a profit) get involved, then the affordances that a company seeks are those that product the most profitable products–or the most valuable products, since a product can be valuable for beating back a competitor, even if the product itself does not make much money. The affordances of finding the most valuable products differ from the affordances of research, in which a technique may be expensive (and so “cost” more than it “earns”) but important for further research; and differ from the broader affordances of livelihood, in which the use of the affordance may not be considered as a matter of “profits” or “return on investment” but rather taken for fun, or on whimsy, or for the particular advantage at hand, even if in the long run the affordance causes health problems, say. We are not, despite economists’ insistence, particularly rational agents.
We might then look at university research and policy. How does a university’s IP policies and practices affect research affordances? Many universities intend to license IP exclusively for commercial development. In doing so, they have immediately called into question what position they will take with regard to research uses–the scope of those uses will depend on what rights are negotiated (if ever) in an exclusive license. Some universities “reserve rights” for noncommercial research and educational use–but those reservations are often specific to the licensing university or restricted to universities, and almost never reserve rights for research uses at companies. And even if a university reserves rights in an exclusive license, there’s still the matter of granting rights. Most universities don’t bother to do this–no public license authorizing all research uses–research on (how a thing works), research with (using a thing for further research), and evaluation of (validating the published claims regarding the thing).
Similarly, universities generally do not reserve rights for broader livelihood practice of the inventions they acquire. I’ve called this area the “make-use commons.” In lab medicine, disease assays fall into this category. Any capable lab medicine operation can implement whatever assay is described in the literature. Self-implementing might be a bother, but in turn self-implementing might lead to more specific tests, and more accurate tests, than might be had with a commercial version of the assay, which may make compromises for the sake of economics, or ease of use, and the like. Self-implementation may also spark variations, allowing labs to teach other labs, or to compete with other labs–their choice. Open source software works in this way as well. One can purchase open source software as a product, but one may also write one’s own code adapted from the code one’s purchased. And one can share that new code–and even charge for it, so long as the recipient is not restricted in what the recipient can do with the code.
Universities generally focus on the third version of affordance–finding a way to make a profitable product to sell in a “market.” One of the efforts of Research Enterprise has been to show how limited this third affordance is, not only for IP profit-making but also to innovation assisted by research generally and more directly to the productive work of research itself. If an invention is promoted as an affordance for making a profitable product, those that would have the most interest are those looking for profits, for value in return for their investment. What might give them an advantage over others? What might prevent others from gaining any bit of value when a product proves to be popular? That is, how might one maximize the return by controlling network externalities in which others benefit with no profit to the owner of the property, the product?
This sort of affordance focuses on competition, on development of commercial product, of profitability, on control of a market. Here, to sweeten the offer, a university might work to exclude others–from commercial research, from make-use livelihood–in order to reserve as many “rights” as possible to offer to a commercial or speculative partner in return for payments and a commitment to do things that will produce more payments in the future. If the partner aims to profit, then the university aims to share in that profit. (For some time, one rule of thumb for university patent licenses was to go after 25% of the after tax profit.) The university then seeks out those potential partners willing to commit to aiming for profit above all else. University exclusive patent licenses are typically drafted to prevent a commercial or speculative partner from doing anything else with licensed rights–no dedication to a standard, no settling infringement by offering royalty-free licenses, no cashless cross-licensing. None of that–or, if an exclusive licensee does such things, it owes the university licensor the cash equivalent of the value of such things. A university exclusive patent license, then, by its very nature, is primarily about the money, not about the outcome–such as the practical application anticipated by Bayh-Dole. Or, another way, the money (as far as the license is concerned) is the most important thing. Failure to pay terminates the deal. Failure to product a product with public benefits on reasonable terms–“oh, well, that’s okay, let’s try something else tomorrow.”
You might see, then, two possible conditions on which someone finds a commercial or speculative affordance. In the first condition, the affordance must compete with research uses (which might produce better variations faster than one can produce a commercial product based on the initial finding), and the affordance must compete as well with livelihood uses in which people (and companies) make their own versions and get on with their activities, without seeking a license nor waiting for a commercial product version. In the face of such competition, a commercial product version has to be better than what someone can do for themselves and has to be at least as good as the better variations in circulation, so that it is easier for someone to upgrade from their own practice by buying a product than by figuring out what’s better than their current version. In this first condition, then, the product has to exceed not just the absence of practice, but the development of new practice and the “progress” of these new practices.
There’s a line of thinking that says that if a university allows research affordances and livelihood affordances, then its inventions will have less value for commercial and speculative purposes. That is, to increase the value of a university’s patents, university patent managers should wherever possible withhold licenses for local practice of an invention in favor of providing a future commercial or speculative partner with the broadest possible rights, for which the university then expects to receive greater payment. The added value of that payment reflects the exclusion of research and livelihood practices and limits the affordances that might be found in such practices, so that these affordances do not arise, or if they do, they will be blocked by the patent position licensed to the commercial or speculative partner. In this way, a university patent broker might argue that the patent now has greater value and will be more attractive to a commercial or speculative partner. This, then, indeed might be considered a primary goal of being a university patent broker–to “maximize” the value of each patent, or (to moralize it a bit), to “maximize the return to the public for its investment in university research.”
To implement such licensing practice, by holding as many broad rights as possible and then conveying these rights to a licensee, universities have to bend the rules. Their exclusive licenses slip into virtual assignments (for copyright, an exclusive license is defined in the copyright statute as one form of conveying ownership in a copyright; for patents, whether a license is exclusive or is an assignment follows court analysis of whether all substantial rights in the patent have been conveyed, including all of making, using, and selling, and the right to enforce the patent–most university template exclusive licenses are in fact assignments).
It’s not that assignment in and of itself is such a bad thing–one might argue that if universities assigned more patents outright, things would be better all around. The university would get even more money–payments would reflect the relief from being exposed to university payments, audits, fussing over diligence, review of sublicenses, and the like. And the assignee would be free to exploit the patent in whatever way it wanted. And the university would not get dragged into court cases or have to bother with keeping the patent asset on its books. In fact, if Bayh-Dole were to have meaningful procedures, it would give a university owner of a subject invention three years from the date of reporting the invention to the federal government to get the invention assigned to a commercial concern, giving preference to small businesses. That would cause universities to unload patents as quickly as they could and not try to acquire the ones they didn’t think they could move.
Bayh-Dole, however, goes the opposite direction and restricts how universities and other nonprofits can assign subject inventions–only to patent management organizations “organizations that have as a primary function the management of inventions,” unless a federal agency approves, and even then with all the requirements of the standard patent rights clause, including those that apply to nonprofits (sharing royalties with inventors, limiting expenses deducted from income streams, using profits only for scientific research or education). Universities, then, insist that they don’t assign inventions when they grant exclusive licenses–even when those exclusive licenses have the effect of assignment–in order to appear to comply with Bayh-Dole. But they bend the rules to follow this particular line of thinking about affordances.
In following this path of thinking, however, university patent brokers also end up in a strange place–they have greatly limited the possible commercial and speculative partners and they have suppressed the value they might have realized from making rights available early and often to the research and practice community rather than delaying access and making local development and use uncertain if not perilous, subject to later enforcement by the university’s licensing partner–one chosen for its willingness to focus on cash over other purposes. Only 1 in 200 inventions at the University of California end up as commercial products. That’s 0.5%. The rest, whether licensed or not licensed, mostly are withheld from public use, waiting that commercial or speculative partner that never showed up, or who showed up but never produced.
This sort of commercial or speculative affordance represented by a patent is distinctly different from that of discovering things that might have a public benefit. Doing things for “public benefit” is itself a difficult concept and has its own problems. What is a “public” and when does it “benefit”? If one part of the public benefits and another part suffers some loss, which part is more important? If the part of the public that benefits is companies, and investors in companies, why is that such a bad thing, since investors are part of the public, too? And if the public’s only benefit is that it isn’t exposed to lawsuits for stuff that it doesn’t care to use anyway, what possible benefit is that, beyond a happy ideology about access?
We might, however, consider “public benefit” from a different direction, without setting up a monolithic view of the public nor reducing the public benefit to service to corporate investors. We might think, instead, of how IP may be used to make affordances available for research and livelihood activities. The owner of an ordinary patent has no obligation to do such things–even if making a patent available for such use might create tremendous value. The owner of an ordinary patent can sit on it, like Smaug the worm in his mountain cave, and do nothing with it but blow smoke.
We might ask whether such practice, however, is within the scope of what “public benefit” ought to mean. Yes, in some invisible hand way everything might be said to work out for the most optimal economy. If that’s true, then our definition of public benefit doesn’t matter, but then neither does enforcement of contracts or laws or the use of intellectual property. Let’s not go there right now, interesting as that line might be. Instead, if we agree that sitting on a patent and refusing to allow anyone to practice the claimed invention for twenty years is not the thing that we expect when we allow the federal government to fund research projects hosted by universities, then we have to come up with something better for “public benefit”–something that involves advantages that we might view as on behalf of the public, to their benefit, even if early on some segments of the public gain and others lose. Any new vaccine causes the companies a loss that sold remedies for the target malady. Any new device that obsolesces old devices causes investors in the old devices to lose.
It might be benign in an odd way to argue that so long as the only folks to lose are investors–and they are used to it–that there’s no harm. But the argument is actually the other way–that it’s not okay that so long as the only folks to benefit are investors, that somehow this meets our expectations of “public benefit.” The public benefit we seek is not one that arises from making money by taking IP positions relative to the investments of other investors, but rather a public benefit that arises from the use of the underlying invention. The public benefit that investors receive by way of increased “value” of their holdings (and universities receiving royalty payments that have nothing to do with use) is distinct from the public benefit arising from use of the invention–and its variations, and its applications, and its immediate practice without delay, without having to pay, without having to wait for permission from an institution.
We might then consider whether a university’s IP practice must necessarily commit first to making sure that research affordances and livelihood affordances are authorized up front and without reservation. Such authorization is entirely consistent with the mission of the university, while withholding such authorization is not. But more so, if no one practices an invention, even when there are no restrictions based on patent rights, then making the authorizations has no particular effect on the value of a given patent with regard to making commercial product for sale. It is the effort to create commercial product that carries the significant expense, and it is there that investors, in particular, desire a monopoly. Fine. But a university should not permit that monopoly to extend to the research affordances and the livelihood affordances. And if a university does split its patent rights in this way, it can never assign rights for any commercial or speculative venture, because substantial rights have been already conveyed to the public–to make and to use the invention however they might.
Yes, this approach exposes inventions to an apparent “lesser” value. Even this exposure is often illusory. Granting upfront research and livelihood licenses forces universities to be selective with regard to the inventions they seek to acquire–or at least the ones they push into a program of exclusive licensing for commercial product development. If a commercial product cannot provide a benefit greater than local practice of the invention, then there is no public purpose in attempting to make the patent suppress the local practice in order to secure an exclusive patent deal. A university would have fewer inventions to manage as exclusive licenses for commercial or speculative investment. This in itself would be a great benefit–focusing on a few inventions allows a university to do the work necessary to secure the desired business partner for each. It’s not possible for a licensing manager to deal effectively with fifty inventions, let alone two hundred or five hundred. There were managers at the University of Washington with eight hundred inventions in their portfolios. How can anyone seek the commercial development of more than ten or twenty inventions at a time? Not possible–or, it’s not professionally appropriate to even try.
If a patent portfolio were split, however, so that research and make-use rights were made broadly available upfront, a much stronger set of relationships with regard to local practice, in the form of teaching the invention, gathering information on local variations and applications, and assisting companies that aimed to create commercial versions (such as by testing or certifying that products perform to specification–much as is the case for clinical trials), and developing standards. The financial return from such activities (if one insists on taking this view) are better, if not greater, than the financial return on an exclusive licensing program that includes suppressing the discovery and use of research and local affordances. More relationships are formed, a greater diversity of efforts are enabled, more individuals have opportunities, and the university follows practices aligned with its charter and policies dedicating its efforts to public purposes.
Call this approach “crossing the commons.” Inventions made at a university are published, and research and livelihood uses are authorized up front. If local practices develop, then the university has the opportunity to assist those practices with further research of its own, updates and improvements, new data and new tools. If local practices do not develop, then a university might think carefully about the nature of the invention and whether a commercial product version might be the only way that new practice arises. If so, then there still may be no reason to terminate local practice authorizations. It is the vitality of the commons itself that suggests the opportunities–the affordances–for commercial versions of what is locally practiced.
The first goal is use. That’s not just a mantra for Bayh-Dole patent practice, but for establishing the value of a patent that’s not ordinary, that can’t be sat on by a dragon-owner, but needs to help the underlying invention get used. Excluding all others from using an invention does not generally promote the use of the invention. Why do people have such a hard time with this idea? Yet it is at the heart not only of federal research patent policy but also buried in the policy documents at American universities.
Why might there be restrictions on patent authorizations? One reason is to make authorizations of invention use conditioned on reciprocal behavior–in company cross licensing, this becomes “we won’t sue you for X if you won’t sue us for Y.” In open innovation terms, this becomes “we grant you rights in X provided you never sue us under your Y to block the use of any feature of platform Z–that is, you grant us all a limited license in your Y.” This latter bit was one of the first rationales that universities had to take ownership of inventions–to create patent positions that faculty inventors generally couldn’t afford, to prevent (through some sort of standoff) companies from patenting improvements to university research and then blocking that research and public access to that research–essentially, a form of open innovation commons without using the current terminology.
Thus, there may be reasons for a university to restrict patent permissions to address what might be considered anti-public commercial behaviors. Actions that are not necessarily illegal, but which limit access to research findings, or which prevent researchers from hunting for affordances or developing them or teaching them or helping others with what they have found. If there are deviant patent behaviors in a given market, a university patent broker is faced with a moral choice–whether to resist those behaviors or seek to participate in those same behaviors. The behaviors, after all, are legal. But are they also legal for university patent brokers (especially under Bayh-Dole and its standard patent rights clause, such as it is)? And even if legal, are they moral? Are they the right thing to do? And does making money on 1 invention out of 200 justify what happens to the other 199?
One might also restrict granting rights to sell an invention, pending a determination of whether there’s local practice. Rather than delaying local practice (by refusing to grant authorization for use upfront, at no cost, or at a nominal, non-discriminatory cost), a university might overtly delay granting a monopoly right that would interfere with local use. A university might give local users a head start in the practice of an invention, and keep commercial or speculative interests from asserting a monopoly. Yes, those interests would see what to them is a “better market,” but it is a market created by depriving people of what they could do for themselves. Consider patents on inventions that would prevent a disease (say, “a method and system of avoiding hybridized wheat gluten in one’s diet”) or for inventions that would clean up oil spills on beaches, or would build on improvements to traditional practices (farming, medicine).
By withholding authorization for these practices, one creates a market by blocking what people are prepared to do for themselves. That’s good “value” but it comes at the price of being predatory on what people do, rather than simply, plainly, directly offering them a benefit. Where the need is great, the “market” may be made valuable by forcing the satisfaction of the need to move through a “market” controlled by a patent owner. But doing so does not mean that the public has benefited, or that the benefit is at all reasonable. We might then imagine that there may be multiple potential “markets” for an invention. The first markets are not commercial, and of the commercial markets that might develop (for training, for sale of products, for sale of ancillary products needed to support new practices), many might not benefit from or require monopoly positions. We might find, indeed, that it is a rare market that requires a monopoly simply to exist at all to provide the use of an invention that cannot be practiced but by a product.
We might consider, then, whether vaccines and cures for diseases and the treatments to repair injuries should be treated like surgery techniques and not subject to patenting at all. Eliminate all method and composition of matter claims. Allow claims for hardware, for devices. In such a condition, we might then ask whether individuals acting without institutional or corporate controls might be better able to find affordances to address these areas of health better than corporate concerns might do, seeking the most profitable affordances–where the most profitable things might include monopolies to suppress local practice.
It seems weird, in its way, to live in a society that takes pride that its most valuable financial assets are maintained by financial speculation on the continuation of chronic morbidity. One might think that fortunes made in mining or manufacturing or building retail sales empires might be worthy fortunes. But why do we speculate on health? Why does a patent attorney at BIO call Jonas Salk “neither typical nor realistic” for releasing his vaccine without seeking patents and payments? Perhaps, if corporations cannot see a way to contribute to public health without monopoly profits, they shouldn’t be so involved. Maybe public health should not be such a market. Maybe public health should be a matter for research and livelihood affordances first, and perhaps private foundations and state and federal governments should do more than fund research–they should fund the development of the methods and practices that enable prevention, produce vaccines and cures, and permit the repair of injuries.
We don’t need a change in the law to move away from the creation of monopoly markets around our health–or other areas of what we might call public welfare. Universities may implement practices immediately to change how health-related patents are licensed. It doesn’t even require changes to university patent policies, because no university patent policy I have read–and I have ready scores–even authorizes exclusive licensing as a default. The reason universities don’t grant outright public licenses for research and for livelihood use are the private decisions of the people who are hired to run their licensing programs. It is, in its way, a professional ideology, and it is a tough habit to break, since like tends to hire like, B’s hire C’s, and bozonets flourish where the quality of practice is low, the standing of those practicing is low, and the power being controlled is relatively high.