December 20, 2010, last update

If we’d had the same rules millions of years ago as we do now, we’d still be living in trees. Some bright wit would have locked down the concept of spending the night on the ground and set up a toll booth.

Ideas as property are based on a fundamental fallacy, and as with all fallacies, acting on them does not work. The flaw is that ideas in the broad sense, encompassing concepts, inventions, and creativity, do not share the main characteristic of property. They are not decreased by use. It doesn’t matter if the whole world sings the same song. That doesn’t change it and it still provides each individual with the same enjoyment. Arguably, it provides more enjoyment because sharing ideas and feelings is more fun than having them alone.

Property, on the other hand, is a way of distributing limited resources that can be used up. Food, clothes, land, or phones are all things that can only be shared a little bit, if at all, without becoming useless to the sharer.

Ideas don’t need to be apportioned among too many users any more than sunlight does. In fact, the only way to force them into that mold is to artificially limit their availability, just as the only way to make people pay for sunlight would be to set up a space shield and extract a ransom. Setting up artificial barriers and waylaying people trying to get past them offends against an intuitive sense of justice. It breeds resentment followed by workarounds. Which is what’s happened with the counterproductive attempt to lock down creations, whether artistic or medical or technical.

Confusion arises because there’s also a sense that creators have a right to benefit from their good ideas. The sense is justified, just as anybody has a right to be paid for work useful to others. That’s different from pretending an idea can be transformed into disappearing ink if too many people look at it. Paying the creator doesn’t magically change a limitless resource into something else.

Thus, ideas are not property and creators have a right to be paid proportionally to the benefit they bring. With those two concepts in mind the rational way to align creativity and its benefits becomes clearer. Don’t make futile attempts to limit the spread of ideas. Try to see how widespread they are. Don’t try to extract a toll. Try to make sure the creator gets paid.

Census methods are available to count how widespread something is. There are many complications associated with counting the results of creativity, and I’ll get to a few of those in a moment, but for now let’s stay with the general idea. A census of usage can tally the distribution of a given creation. The creators are then paid based on that tally. The money to pay them comes from a tax on the physical goods needed to use their creations, in other words from a tax on the paper, storage media, phones, screens, pills, or other substrates that carry the benefit to the user.

I need to discuss a terminology issue parenthetically. Since I’m insisting that the products of creativity aren’t property, I can’t use the convenient term “intellectual property” to describe the whole class of patentable, copyrightable, and trademarkable things. I’ve used “creations” instead, even though it’s a clumsy-sounding term. There are also of course differences among those three subgroups, some of them necessary, some of them mere historical accidents. For instance, having a different standard for patentable objects as opposed to copyrightable expressions is necessary. Having a different term of protection — close to 100 years at this point for copyrights, twenty years for most patents — seems arbitrary. Most of what I’m discussing applies to new creations generally, rather than either patent or copyright specifically. Trademarks are a small subset where rights extend for as long as the mark is used. That seems sensible, and I don’t delve into trademark-specific issues.

A census method with subsequent payout is superficially analogous to the market system used now in that sales are a rough tally and the price the market will bear determines payment. However, markets can only handle property. Like the proverbial hammer to whom everything is a nail, markets have handled creativity as if it was property. When the nature of creativity causes it to escape the inappropriate container the market isn’t able to use an appropriate non-market-based approach. Instead it keeps attempting the useless job of trying to bottle the equivalent of sunlight. That by itself is a big waste of everyone’s time, energy, and money.

But there are other, bigger problems. The category error has generated injustices. Since creativity can’t be bottled, who gets paid and for what is rather arbitrary. That leads to the usual result: the powerful get paid, the others not so much. Those powerful people are very rarely the creative people themselves. The examples are legion, but to take just one instance, Charles Goodyear invented the vulcanisation of rubber (without which it’s about as useful as chewing gum) but died poor. Imagine carrying out the industrial revolution without rubber, and yet it wasn’t Goodyear who saw much benefit from his work. The inventor, programmer, or artist cheated of their work by those with more money is such a common occurrence it’s a stereotype.

The ironic result of “intellectual property” laws is that their stated purpose of rewarding innovation is an almost accidental outcome. The actual history shows that they were established as a tool for governments to control content. Distributors were the enforcers in return for a time-limited monopoly providing guaranteed profits. The government control aspect has been beaten back, but the philosophical tools needed to see the illegitimacy of guaranteed profit haven’t been widespread enough yet to correct the other half of the injustice. The rewards for innovation go to uninnovative third parties, and as the system is pushed ever further away from its official goal it breeds mainly cynicism, not creativity. Markets, by trying to pretend ideas are property, create a situation in which all that matters is who “owns” the idea, not who created it. That perverts the system of rewards and takes them away from the people with a right to them.

Once “intellectual property” is recognized as a matter of rights rather than markets, the institution which should handle it is clear. Administering a system whose primary purpose is enforcing rights is necessarily a government function. Granting patents and copyrights is already done by a government office because it must be transparent and accountable, without any considerations besides correctly logging who had which idea. Distributing royalties is equally a government function because it requires an objective census untainted by any motive except accuracy, and transparent, accountable payments based on that. No other entity has (or should have) the enforcement power and accountability required.

Variants of the idea of royalty distribution based on a census have cropped up repeatedly in recent years (e.g. 1, 2, 3) because it’s an obvious way to fairly distribute the rewards for useful creations. The idea is applicable at any stage of technology, but it is easiest to apply in a wired world. Headers in computer files can include attribution lists of any degree of complexity, similar to software version control systems, and they’re also much easier to census than physically tallying actual products. (This just in, as they say: Google is experimenting with tags that trace sources.) However, a wired system is also easier to game, and it should go without saying that stringent safeguards and punishments against fraud have to be in place. Physical sampling has an essential place, I expect, as one of several parallel tallying methods. Used together, they would provide one type of safeguard against fraud.

One difference between a centrally administered census system and market-based distribution of royalties is that a census does not pretend to have a perfect one-to-one correspondence between usage and payment. Markets do have that goal, but their distribution of payments is on the whole wildly unrelated to how much a given work is used. Anyone who finds the current system good enough in terms of apportioning royalties could not fault a census-based system for imprecision because it would be far more precise than the markets for this purpose.

Now that the internet has made it easier for everyone to publish and broadcast, a fundamental problem with assigning credit for creativity is becoming increasingly evident. Creations never happen in isolation. Every inventor, author, artist, and scientist stands on the shoulders of others. The only way to assign credit is to draw more or less arbitrary lines that delimit enough difference from previous work to merit separate acknowledgement. The cutoff for a US copyright, for instance, is more than 10% difference, although how that 10% is quantified is rather subjective. In practice, it seems to mean “some easily noticeable and material difference.”

The imprecision is inevitable — there’s no way to quantify a work of art, for instance — and it’s difficult to see any way to avoid arbitrary and subjective demarcations. That implies that the bar should not be set too low because the smaller the contribution, the murkier the distinction between significance and insignificance.

Making the determination of originality is complicated by the need to be fair. Now and in the past the “solution” to the complexity of creativity has been to give up and simply assign a given work to the applicant(s) for a patent or copyright. In this day of remixes, a more calibrated system is essential. As I mentioned when discussing legislation in the fifth chapter, I think that methods used in software version control can point the way. Software, and legislation for that matter, are just special cases of works with multiple contributors. Sometimes those tools are called content management systems, but that’s an unspecific term covering everything from tracking minor changes in a word processor, to blogging software, and to education course management which may not have the necessary contribution tracking. Something like Plone is perhaps the closest thing now available.

The software version control or Plone-type systems I’m aware of (I can’t say “familiar with”) are used in the open source community by volunteers. In that situation, most participants are honest to begin with. Plus the community is rather small, skilled, and generally aware of the facts, all of which acts to prevent cheating. A similar system adapted to patents and copyrights, where money from royalties might be in play, would need a much stricter system to prevent cheating. I’m not sufficiently close to the field to have an idea how that could be applied, but it’s a problem that must be solved in a fair system of creator’s rights.

There’s also a psychological issue that should be taken into account when apportioning royalties. One desirable effect of an equitable system of rewards should be that more people feel motivated to act on their creativity and to contribute. However, it doesn’t seem practically possible, at least to me, to assign rights to and then pay for every single dot and comma that somebody might add to a body of work. There has to be some level below which contributions are just part of the background, as it were. But people react badly when somebody else gets a better reward for what they see as equal or less work. It offends an innate sense of justice that goes right back to chimpanzees and earlier.

The point of equitable rewards for creativity is to facilitate its expression. It would be counterproductive to implement a system that’s felt as even more unfair than the one we have now. The system needs to take the psychological factors into account if that goal is to be met. I would guess that a clear demarcation — difficult as that is — between the amount of contribution that receives royalties and the amount that doesn’t would mitigate negative reactions. When work is obviously original, that’s not a hard call to make. When it’s incremental, then perhaps adding a time factor would help. To be eligible for royalties, the amount of work involved should be equivalent to some sizable part of a work week. In other words, if it’s equivalent to a half-time job, it’s significant. If it’s something one dabbles in occasionally, then official recognition is probably misplaced.

My sense is that commitment of contributors is not a smoothly increasing variable. A minority contributes a lot, then there’s a more or less sparsely populated intermediate gap, and then the majority whose aggregate contribution may be huge but where any individual adds only a little. If that sense is correct, research should be able to identify that gap. If the demarcation line for receiving royalties runs through the gap, it will align with the intuition that only larger contributions deserve specific rewards.

The next difficulty is to identify the best group to make the determination that new material has been contributed, and how much. The people with the clearest concept of the work are others working on the same or similar projects, but they are also interested parties. If there are royalties to be distributed, they may be motivated more by a desire to keep the pool of recipients small than by an honest appraisal of work. Critics or specialists in the field who are unconnected with the project seem like good possibilites. Public comment should always be facilitated because expertise often crops up in unexpected places. Once the outside expertise and input on the originality of a given piece of work has been obtained, then the experts at the patent or copyright office would evaluate all the input on its merits and make a determination, subject to appeal.

The current process is supposed to work more or less like that, too, but for various reasons it’s veered off course. Patent officials are evaluated on how much paperwork clears their desks, not how well it stands up to litigation. So it’s become customary in the US to grant patents for practically anything and to assume that if there’s a problem, it’ll come out in the courts. Furthermore, possibly related to the desk-clearing standard, patents are granted apparently without anything approaching due diligence on prior art. The assumption throughout seems to be that litigation is a solution rather than a problem. In reality, it’s laziness and dereliction of duty to expect the courts to clean up messes that should never have happened in the first place. Patent and copyright officials who are doing their jobs will make a thorough and good faith effort to determine that a piece of work is indeed new, and how new. Litigation has to be a rare occurence when a mistake has been made. If it’s not rare enough, it’s indicative that the responsible bureaucrats need to be fired.

So far, creator’s rights have been discussed in the context of payment for work. The other aspect is control over the work. Obviously, minor contributors (less than 50%?) wouldn’t have a controlling interest in any case, but what of major ones? How much control is appropriate?

When discussing money, I stressed that real free markets and monopolies are incompatible. That is no less true if the monopolist is an artist or inventor. The creator has a right to be paid if someone is benefitting from their work, but that doesn’t give them monopoly “rights.” An unfair advantage is a privilege, not a right. Creators cannot prevent others from manufacturing their invention, playing their song, or publishing their books. The creators do have the right to be paid in proportion to how much their work is used and how critical it is to people’s lives. The government would disburse the funds based on pay scales that have been worked out for different classes of products. (I would say that ringtones, for instance, should have lower royalty rates than headache cures.) Royalties received for similar classes of products should be much more consistent under that system than they are now. The pay scales themselves would necessarily be a somewhat arbitrary because they’re set more or less by consensus. (The current system also sets royalties that way, but the only factor taken into account is the bargaining power of the creator.) Consistence should help avoid wide disparities in reward for equivalent contributions.

Requiring creators to license their work is currently called “compulsory licensing,” which makes it sound like a bad thing. “Compulsory” anything meets with reflexive resistance. But all it does is take away the ability to impose a monopoly. That reduces short term gain for a few people, whose objections are no different from those of anyone losing privileges. They’re not valid if the goal is equitability.

However, there is one sense in which creators would have more control over their work in a fair system than they do now. The entertainment industry has something called “moral rights,” which refers to how a creation can be used in other contexts. Consider, for instance, Shakespeare’s character, Lady Macbeth. There’s a famous scene in which she sees the blood of murder on her hands, and nothing can wash it out. Shakespeare did not retain any moral rights. They hadn’t been invented yet in the early 1600s. So a soap company could have made a zippy ad showing a relieved Lady Macbeth after one handwashing with their EverClean soap. The only thing saving Shakespeare’s legacy is that Lady Macbeth is too old and forgotten to be worth selling.

Moral rights clearly belong to creators. Their creations belong to them in ways that money can never buy. They have rights in them that money can never buy. Creators, therefore, have veto power over what they see as inappropriate alteration of their work. Like the fact of their authorship, moral rights are also permanent and inalienable. At least for expressions, but not tools, i.e. for copyrights but not patents, that control over usage is permanent. Creators who felt strongly enough could include instructions about moral rights in their wills. Those rights, however, cannot turn into a back door to permanent copyright. The scope of moral rights needs to be limited to relatively broad classes of usage. Creators have the right not to have their work perverted, but they don’t have the right to prevent legitimate use of it. Making moral rights explicit and enforcing them is probably more important in an age of remixes than ever before. It is only fair that if someone wants to use a character or another expression in a way that’s odious to the creator, then they’re under an obligation to come up with their own concept and not copy someone else’s.

There’s a point that may need stressing in a broader sense. The rights of a creator are inalienable because they’re matters of justice, not property. They’re not something that can be bought or sold or passed on to someone else. The creator, and no one else, has the right to moral control over their work, and the right to royalty payments at a level and for a period of time stipulated by law.

That would mean some changes in business practices. For instance, corporations or any other institutions would not be able to take rights to inventions made by workers. The corporation’s benefit comes from the profit of being first to market with a new product. The royalties go to the individual inventor(s) no matter whose payroll they’re on. The corporation recoups the costs of research from profits, not from taking the rewards for creativity that isn’t theirs.

In the case of celebrities, the valuable “property” may not be a work, strictly speaking, except in the sense that it’s the carefully constructed public persona of the celebrity him- or herself. Private citizens already have the right to control their personal images and data, as discussed under privacy in the Rights chapter. The creativity of entertainment and sports personalities is packaging a specific public face, but the mere fact of being a public person doesn’t make them lose all rights. In the private aspects of their lives, they have the same rights to privacy as any citizen. In the public aspects, they have the same moral rights to their personas that other creators have to their work. That would make paparazzi jobs obsolete and complicate the lives of gossip columnists, but those aren’t sufficient reasons to deprive people of their rights.

The more equitable rules of copyrights, patents, and trademarks envisioned here would obviously require big changes across a range of industries. The main effect would be to render armies of middlemen superfluous. As always when the application of just principles damages entire business models, the simple fact that profits vanish is not a sufficient reason to reject fairness. I’ll repeat myself by saying that slavery was once a profitable business model and that didn’t make it right. Nor does implementing justice mean increased poverty. On the contrary, every single time, there is more wealth when there is more justice. The illusion of lost wealth comes from the few people who lose the ability to milk the system.

Furthermore, insofar as the middlemen provide a real service, there’s no reason why producers or publishers or agents would necessarily all disappear. Expertise in packaging, distribution, and sales is not the same as making an invention or an artwork. Artists, especially, are stereotypically poor at those jobs. There’s a need for some middlemen. The difference is that they would have to be paid from actual added value and not simply because they’re gatekeepers who can extract a toll.

Moving on from creators to the creations themselves, one current large source of problems is the definition of what can be patented. In the US, things went along reasonably well for a while when the definition was narrowly focused on the sort of original work an amateur would recognize as an invention. New technology, however, introduced gray areas, and the unfamiliarity of judges and lawyers with technical issues made that standard hard for them to apply sensibly. The desire to give US business a boost also worked to promote the granting of patents, although that motivation doesn’t appear in the dense legalese of court arguments.

That’s eventually landed us where we are now. Patents are given for mere and obvious ideas, such as one-click shopping. Patents are granted on life although the life in question (parts of a DNA molecule) has merely been read, not created. There have been some actual DNA inventions, such as bacterial artificial chromosomes, BACs, but those do not constitute the vast majority of patents granted for DNA sequences. In the pharmaceutical industry, when the patent on one blockbuster drug is close to running out, it’s provided in a trivially different form, such as a weekly dose instead of a daily dose. Somehow, patents are granted for something which is nothing but a dosage change. User interfaces get patented, as if in a former time somebody could have locked down the concept of using dials to control machinery, forcing any competitors to use more cumbersome systems, such as inserting pegs into holes.

The situation with trivial and proliferating patents is complicated by the fact that copyrights are both easier to obtain and last much longer. In the software industry, for instance, that’s created an incentive to seek copyrights rather than patents even though nobody reads software like a book. People use it, like a machine. Yet the legal system has let the applicants get away with the travesty of copyrighting software. That could be due to a lack of technical knowledge in the legal system, or to excessive accommodation of those with money. Either way, none of this should happen in a rational system of patents and copyrights.

Rights should be granted when the consensus among knowledgeable people agrees that there has been non-trivial original work. And insofar as there are necessary differences between patents, which are granted basically for tools, and copyrights, which are for expressions, then the creation and the type of rights assigned should be in the correct class based on the merits. The salient feature is not which office happens to be processing the application. The important point is in which category the creation actually belongs.

The criteria for what is patentable is a matter of opinion. There’s no law of nature involved, and for most of human history the concept didn’t exist. As a matter of opinion, it depends on consensus. The consensus, after a few hundred years of experience with the idea, seems to me to be coalescing around the concept of a new tool. Stripped of its legalese, that’s the core of the criterion used in US patents in the mid-1900s when the system worked better than it does now. Adding some of the legalese back in, it’s called the “machine or transformation test.” An invention is patentable if it’s a non-obvious device or method of transforming one thing into another. There are, as always, gray areas in that definition. For instance, the linked Wikipedia article points out that a washing machine is a patentable tool that cleans wet clothes using agitation, but a stick used to agitate wet clothes is not.

That example points up the fact that patentability is yet another situation in which there is no substitute for good judgment. Unusual creativity deserves recompense. Common or garden variety creativity should be appreciated and facilitated, but it’s an attribute almost everyone shares and so it requires no special recognition. Distinguishing between uncommon and common contributions is what takes judgment, and always will. Most cases don’t require exceptional ability to judge. It’s possible to discern whether a tool is new and non-obvious, even if it’s not always simple to articulate why. In the washing machine example, it seems clear to me that the difference is the obviousness of the tool. Even I could think of using a stick to work the water through the cloth more easily. On the other hand, neither I nor most people have ever been anywhere near inventing a washing machine. Another example is the weed whacker. That useful invention was patented, but also copied. The patent was not upheld when the inventor sued, because the judge felt it was an obvious idea. However, it was obvious only in a “why didn’t I think of that” way. The fact is, nobody else had thought of it, and if it was really so obvious, that mechanism for weeding should have cropped up repeatedly. Perhaps a consensus of public comment by disinterested third parties would help avoid such miscarriages of justice in patent law. All methods that prove useful in promoting good judgment should be applied to the issue in the interests of maintaining a just, effective, and, ideally, frictionless system that requires no litigation.

Fair use is a shorthand term to describe a type of usage for which royalties are not due. In practice, people tend to view it as any small scale, private usage, as well as socially important ones, such as in libraries, archives, education, or quotations that are part of other work. In the system envisioned here, the entity not paying would be different, the government rather than the individual, but the principle should be the same. And the principle should explicitly date from the time when fair use meant that people could use what they had. The current push by content owners to turn copyright into an excuse to extract rent every time anyone touches their “property” is an attempt to charge whatever the market will bear. It has no relation to a fair price or to fair use.

I’d like to spend some time on a topic not usually included in discussions of creativity or intellectual “property.” The government doesn’t only administer the rights involved, it’s also a major customer for the products. Under the current system where patents grant a monopoly as well as rights to royalties, that can mean the government in effect becomes a factor in creating monopolies as well as becoming one of the trapped customers. Neither of those is acceptable for an entity that must treat all citizens equally. I assume the situation would not arise in a system that requires compulsory licensing. If it nonetheless should develop, it’s clear that it has to be stopped. The government should make the explicit effort in its purchasing to buy from different companies, if their products are of similar quality. The most efficient way to do that is probably not to have preferred vendors at all, but to leave buying decisions up to individual bureaucrats subject to audit, as always, and random reviews for conflict of interest. The primary criterion, of course, is getting good value for the taxpayers’ money.

The mandate to spread the government’s custom brings up a tangential aspect of compulsory licensing. As tools grow increasingly complicated, user interfaces became more and more of an issue. People don’t generally think of interfaces as a tool of monopoly, but they can be, and they can be the most effective one of all. There’s nothing we prize more than the time and effort we have to put into learning something. If using a competitor’s product means learning a new way to do something, then it won’t be used, even if the result is better. Just look at the impossibility of getting the world to use anything but qwerty keyboards.

Inventing a new and improved interface is a real contribution and a real invention. But compulsory licensing has to explicitly include the right to use that interface wherever appropriate. The relevance to the government’s situation is that the requirement to avoid favoritism among vendors does not have to mean reduced efficiency of the workers. The desired interface could be ordered with any product, since the two must be independent of each other.

That leads to the further implication that promotion of competition requires rules that ensure interoperability. It’s essential at all levels: in simple hardware such as electrical plugs (at both the wall and appliance ends), in machines where drivers for different hardware must interface with all software, and for users where people must be able to take the interface they’re comfortable with to any application or device of their choosing. That would require an effective standards setting process whose primary mission, in an equitable society, would be the convenience of and least expense to the most users.

The requirement for interoperability does not, by itself, preclude closed and secret processes. The government itself, however, has to operate on stricter rules because it cannot be in the position of relying more on some citizens than others. The government itself must be transparent and equally accessible to all, which means the tools it uses have to share those characteristics. Unless they do, there’s a big and unacceptable loophole in the transparency requirement. There’s nothing to stop the government from using proprietary tools, but they must be open. Closed source hardware, software, or other tools has no legitimate place in government offices.

Archiving is another function where interoperability and transparency has not been an issue heretofore. Librarians preserving books for posterity might have had to worry about physical preservation, and the time and expense of transcription, but they never had to worry about losing the ability to read the books themselves. The words wouldn’t turn into a jumble of unrecognizable characters unless the language itself was lost. However, through the miracles of modern technology, we’re now in a position where the ability to read something written a mere twenty years ago depends on the continued existence of a company.

The problem is currently most evident in computer games, where it doesn’t worry most people since games are considered worthless. Some games from the early history of computing are already unusable because their programming was closed and the companies who held the secret of it have vanished. Whatever one’s opinion of the games themselves, they’re the canaries in the coal mine who are showing us the future of much larger and weightier things.

It is not right for work to be lost because of a rights holder’s desire for secrecy. Right to a patent or copyright does not include the privilege of destroying the work made when using the tool. How best to implement that limitation in practice would need to be worked out. Maybe it should be a requirement to deposit all necessary information with the government archivist whenever a product starts being used by more than some tiny percentage of the population. Maybe some other method would be more effective. There do have to be procedures in place to ensure that work isn’t lost simply because the knowledge of how to read it was once kept secret.

– + –

Getting the rewards for creativity right is probably more important in an equitable society than the kinds we currently have. In a world of peace and equitable distribution of wealth and time, people’s focus is likely to be on finding ways to amuse themselves. For many people that means having fun in the usual sense, playing sports, engaging in social life, and the like. But there’s a large minority for whom fun has a more extensive definition. It means learning and doing, as well as being amused. If the society as a whole facilitates that activity, and if it’s justly rewarded, that’ll lead to a beneficial cycle of innovations and more satisfying lives for all citizens.