So there you have it; a whirlwind tour of nearly every major intellectual property and communication issue of the day, based on a coherent communication model. I've shown you all of these details so you can see how each issue, with the explicit exception of software patents, can all be expressed in the terms of the same basic communication model, with the complete context model as described in section 20. Now, for the final major point of this essay: To show how these issues are all deeply inter-related, and why for consistency's sake, we are limited in what solutions to the problems we can truly choose.
The best way to do this is to determine what degrees of freedom we have in determining our ethics consistently. Degrees of freedom is a term I find myself borrowing from mathematics because there is no true pure-English equivalent.
Put (very) loosely, the degrees of freedom describe how many fully indepedent directions something may move in, for a very abstract definition of ``direction''. For instance, while we live in a three-dimensional world, for many practical purposes we only have two degrees of freedom: We can move North/South, or East/West, or a little of both at the same time, but we are not free to move Up/Down. (We do have a little freedom that way, but compared to N/S and E/W, it's quite small; you can go ten miles north far more easily then ten miles straight down or up.)
The key word in that definition, as I've bolded, is independent. We can move North without moving East. We can also if we choose move in both directions at the same time (``northeast''); that's OK too. Generally, we can not and do not move up or down without also moving north, south, east, or west to some degree, so it is not an independent direction.
Similarly, the ``Dow Jones Industrial Average'', over the course of a single day, may either move Up or Down (or not at all), but there is no ``sideways''. The Dow Jones Industrial Average only has one degree of freedom. The ``Entire Stock Market'', on the other hand, is a complicated beast, with all kinds of stock prices independently moving around. Thus, the ``Entire Stock Market'' actually has one degree of freedom per stock.
(This is simplified from a full discussion of ``degrees of freedom'', of course.)
To refresh your memory, one of the earliest parts of this essay observerved (section ):
Rather then taking the time to truly map the domain of discourse and look at all of the issues in a coherent way, laws (most especially judicial decisions...) exploited the independence of the media types, and each individual segment got its own laws. ... The Internet in a period of just a few years has taken each of the bubbles that we saw in the previous section and rapidly expanded each of them until they all touch, overlap, and envelop each other. ... the fundamental problem with the current legal system is that the foundational assumption that the legal domains are independent is no longer valid.To phrase this another way, the old legal system had many degrees of freedom. A law concerning book copying would not affect the laws concerning audio tape copying. The practical realization by the computer industry that all of these things are just various sets of numbers, and they can all be expressed simply as data significantly reduces the degrees of freedom we have in creating our ethical system, because suddenly, laws affecting how audio content is copied does affect how books are copied. Consider how laws like the DMCA, which it is probably fair to say was only intended to protect video, audio, and software by the authors of the bill, has also been found to affect things like printer ink refills (http://news.com.com/2100-1023-979791.html), or protecting prices charged for retail goods by Wal-Mart (http://www.counterpane.com/crypto-gram-0212.html#4, do a find for ``DMCA'' or ``Wal-Mart'') (retracted, but voluntarily by Wal-Mart after scrutiny, not in court). It is no longer plausible to try to create significantly different systems for different types of communication, unless you can define a particular type carefully enough that it is not easy to ride the boundary condition and cheat the system. Again, there's nothing theoretical about any of this, it's happening all around us. The point of all of this theory is to give us words to communicate with and models to think with; what the theory describes is already here.
Thus, the foundational question we must answer about ethics is how many degrees of freedom are there, and what are they? I submit to you that there are only three.
Let's examine those lines in more detail:
By this I mean things internal to the communcation itself. Is the content truthful? Is it slanderous? Is it covered one of the commonly-accepted exemptions to free speech, such directly threatening someone?
I mention this because up to this point we have not examined the content of the communication itself very much. I think this is because there's no need to do so. Except for some terminology issues (the slander vs. libel distinction isn't really useful anymore), we know what to do with fraud. We know what to do with threats. Nothing is changed by having new media to make threats in, or spread lies in, or make fraudulent claims in. Thus, I am comfortable invoking existing ethics to cover ``internal issues'', and explicitly discarding them as an issue for this essay.
Many people try to muddy the issue to justify more strict laws but there is little need or call for new harsher laws. When it boils down to it, there's nothing you can say on the Internet that can't already be said through any number of conventional channels, and even the reach isn't different enough to justify any extensive new laws; more people have louder voices, but we have already extended these doctrines to distinguish between people based on their varying reach.
Thus, even though this is theoretically a degree of freedom, we have already decided as a society what our position on these internal issues are, and there's no need to reconsider it.
This is a ``degree of freedom'' because how we handle this does not directly impact the other two degrees; for instance, fraudulent messages are conceptually removed from discourse entirely so there is no ethical way for the fraudulent message to be sent to anyone anyhow.
Message integrity is essentially a binary issue; either we defend message integrity as described in my Message Integrity chapter (chp. 9), or we do not. There is little or no middle ground because any crack in the integrity can be exploited to do nearly anything, thanks to the patch hole (sec. 9.5). While it is not strictly speaking impossible to imagine an exemption that can not be so exploited, one can still question why it's worth the risk to free speech.
So while this is theoretically a degree of freedom, if we are to maintain our commitment as a society to the principles of Free Speech for all, we must choose to defend message integrity, and thus this degree of freedom is already chosen for us by our pre-existing ethical values.
And thus we come to the crux of this essay. What restrictions are the senders allowed to lay on the receivers with regard to what the receivers can do with their messages? I see
However, even in the domain of the multiple acceptable answers, not all answers are possible simulatenously. Some choices we make will constrict our freedom to make other choices, and we will either need to decide which is more valuable to us, or work out some way to cleanly delimit the boundaries between the domains where different choices take effect. As I take you through at least a partial analysis of the issues, the most important things to watch for are the fundamental conflicts between different choices made for how the sender can restrict the receiver.
On the one hand it seems rather odd that it took this long to get to this result. On the other hand, I don't think anyone who has been tracking these issues can deny that there's a lot of confusion and fuzzy thinking out there. One could probably spend two or three times more verbiage then I have just clearing away the fuzz, and until the questions are clarified, it's impossible to give good answers to any question. I believe that you can not just jump into this chapter and truly understand what I'm trying to say without the context and terminology provided by the previous chapters.
I'd like to take a moment and make concrete the points initially made abstractly in chapter 2. Our problem lies in the fact that we have chosen wildly inconsistent answers to these questions, both because of accidents of technology being perceived as fundamental ethical principles and because certain issues like Privacy have not been carefully analysed as in this essay to show how connected they are with other communication issues. Consider the following matrix of Issues and summaries of what seem to be our current system's answers to the preceding questions:
| Issue | When can it be received? | What can the receiver do? | Can those interact? | Can the message be forwarded?
|
| Music (Radio) | Freely broadcast with no contract (no money from receiver) | tape it for later listening (though the industry would like to eliminate this) | can only make ``personal'' use (businesses can't just turn on a radio without paying) (codified in law since there's no contract) | Only Fair Use |
| Music CD, Movies DVD | receiver must pay for content + media | somewhat up in the air, but space shifting and format shifting seem to be OK | some CDs are trying to use DRM-like techniques to prevent certain uses like format-shifting | somewhat up in the air; can you send your grandmother an MP3 from a CD you own legally? (Does whether your grandmother lives with you impact that?) |
| TV | same as Radio | same as Radio | same as Radio | same as Radio |
| Movies (theatre) | controlled by physical access | controlled by physical access, banning recording equipment on the premises. | pay once, view once (but one can make the case that one is renting a seat for a period of time) | never in physical possession of the movie, so nothing is possible |
| Privacy-sensitive Information | information must be sent out to perform certain transactions | nearly anything they want. (Some very limited exceptions like HIPAA restrictions) | no, sender of PSI is typically allowed no control over the PSI after it is sent | anything the PSI receiver pleases |
| Software | heavy contract restrictions in the general case | based on contract restrictions, seems to be no limit to what those contracts can contain (UCITA), though they've never been tested | based on contract restrictions with no apparent limit | based on contract restrictions with no apparent limit |
| Actual Speech | controlled by physical proximity; admission charges or physical barriers may bar access | may be recorded by listener, copyright for recording belongs to the recorder, not the speaker (speaker may have other copyright) | none I know of | none I know of |
| Books | controlled by physical access mechanism | unrestricted (historically impossible to restrict book users) | no restrictions technically possible, so historically no-one ever thought of it | traditionally, since books can't be copied they can be given to anyone and nobody cares. There are some interesting issues if a society wants to have public libraries. |
| Website text | in general, may be as restricted as software is (usually freely available) | technically, messages can usually be archived, but it's an open question if this is generally acceptable (and also an open question what it means if it's not) | theoretically as rich as software contracts, usually not used | theoretically as rich as software contracts, usually not used |
There's a lot of summarizing going on in that table; for instance the ``Privacy Sensitive Information'' answers were specifically written with the common case of addresses, phone numbers, credit history, etc. in mind, not the more esoteric privacy sensitive information that still falls under my defintion. ``Software'' discusses the general case, even though not all software lays all the restrictions on the receiver. This isn't even complete, either, but I think this is enough to show the point.
The key here is to note the incredible discontinuity between the current answers to those questions, even though it's all the same thing, just numbers. Even very similar things like ``Website text'' and ``Books'' have wildly differing answers; as they merge (``E-books'') it is impossible to simultaneously reconcile the differences while also keeping everything the same as it was before; one of those has got to go.
In fact, that's a general problem with effectively every technology here; you can not simultsneously preserve all of these historical accidents in the Internet era.
As technology continues its march, what we need is a uniform set of answers that we can apply across all of these ``content types'', because the very concept of ``content type'' is becoming meaningless. Or we at least need something that can be applied ``nearly uniformly''; deviations from the uniform for specific circumstances will certainly be necessary and desirable, but we should start from a firm foundation.
This may sound hopelessly Utopian but I think that is only because we have become used to the gelatinous blob that passes for current law. The reality is we will eventually have something like what I am proposing; the question is whether we get there in fits and starts, only after years of painful lawmaking and judicial oversight, and with lots of useless cruft still in place, or if we get there relatively smoothly because we had the foresight to see change. Lest you think I personally am hopelessly Utopian, yes, I'm betting on the former, but I hope this essay in some small way can help us move towards the latter.
I'd like to look at each of these four sub-questions and examine the issues more closely now; like I said, I don't necessarily have answers but I think I can ask some good questions with the foundation I've laid down.
Up to this point we've only talked about the relationship between individuals, because even large-scale one-to-many or many-to-many communication can be modeled as a series of one-to-one communications. As it comes time to choose as a society what restrictions we will allow senders to lay on receivers, it's appropriate to start analysing social effects.
Social effects of course arise from the combination of lots and lot of individual interactions. Some things only matter when lots of people do them; for instance see ``tragedy of the commons'' (Google it if you don't know precisely what that is). Some effects only become obvious when you think about lots of people doing them. Since we have freedom on the individual level, we need to decide what effects as a society we want to encourage or discourage. After all, ``intellectual property'' law is supposed to be derived from the idea of managing social effects via legislating individuals: (constitution quote here)
We have to consider social effects because the ethics at an individual level can't guide us; there are too many possible legitimate answers.
Generally, message reception can be handled as a contract issue, with optional explicit media-based exceptions for compulsory licensing. In the future, and even to a large degree in the present, the licensing model will be dissociated from the content type, and I think this is overall a good thing as it increases the freedom of the market to set prices. It will be important for governments and informed consumers to make sure no one source of information gets too powerful and starts trying to exert asymmetric control over the receiver or the sender (for instance, one can make a good case the music industry exerts too much control on the music authors because of their collective market dominance and oligarchial actions), but I do not think those are truly new ethical issues.
Easily the thorniest question we face as we decide our answer to the reception issue is ``What is the role of a public library in the Internet era?'' The strictest possible contractual restrictions, which are ``only one person shall use this content for some finite period of time'', backed up by well-implemented Digital Restrictions Management techniques, are simply antithetical to the existance of public libraries. If strong digital restrictions management is to be elevated to the status of law, as the DMCA strongly hints at if it does not actually already accomplish, it will eventually be the death of libraries.
In the end, we must really choose either public libraries, or strong DRM. It's impossible to have both with the current conception of a ``public library'' because we'd have to poke holes in the DRM to allow libraries special access, holes that would end up ``leaking content'' out of the DRM system. (For instance, photocopying resources is still provided by most libraries; the digital equivalent of photocopying, which is ``printing'', would take text resources out of the DRM system. Speaking as a computer scientist, I have no confidence in the long-term viability of techniques such as ``watermarking'' that try to keep even such printed resources ``in'' the DRM system; once it's out, it's out.)
One solution is restrict libraries to what are technically allowed by DRM. In the current environment, that may well be equivalent to eliminating libraries, since there is little or no (perceived) motivation for a sender to allow one's works to be lent out by a library when, from the sender's point of view, the lendee could instead be purchasing them. There are some arguments on whether libraries still provide a net gain to the sender in the end, but truthfully, most of them sound strained even to me, and if this is the only reason we have libraries, then it should be left up to the sender to decide whether they wish to take that risk. Without some sort of law protecting libraries as the First Sale doctrine does now, I can not see the majority of future senders willingly allowing libraries to lend their messages. This also interacts with my question of whether it is even reasonable to imagine taking a message away, which I'll explore in a bit in connection with ``How can a sender restrict the receiver?''.
Complete elimination of libraries is another option. Since libraries are build upon the First Sale doctrine which I've previously explained as being dead in the current era (see 7.3.6), if no other library-protecting legislation is passed this is the ``default fate'' for libraries.
I think we need to preserve libraries for lots of good ethical reasons that fall outside of pure ``communication'' ethics. I think they are a net good for society. I think the value of libraries to our society far, far outweighs the value of DRM. But there is nothing in these communication ethics or the legal mechanisms being developed that will support the libraries. Our current path will lead to the casual elimination of libraries by the large copyright interests. If we want to stop this, we will need to come to a concious decision as a society.
Librarians, if you're not close to retirement you really ought to be pounding the pavement on this point a bit more; your jobs are in imminent danger over the next couple of decades!
In the past, once the message (in the form of an expression) was delivered to the receiver, there was no way to control how this message was used by the receiver. Now there are ways to control the message, even after delivery, via the Digital Restrictions Management technology discussed so often before.
For example, nearly all DVDs require the viewer to sit through the FBI warning at the beginning for the full period of time the DVD producer decides. Many DVDs actually require the viewer to watch previews before the main movie could be viewed.
Can the receiver skip commercials, like with TiVo? Can they rewind or fast forward at all? If they schedule a movie for 5pm does that mean they have to be there or miss out?
But all of these questions first presume that these restrictions are acceptable and we merely need to quibble about which ones are OK. We seem to have managed just fine without them until the present time. We're mostly still managing just fine without them, in the publishing and music industries. Why should we suddenly start trying to restrict people, just because we ``can''? Is that a net value to society?
What about a model where once a receiver experiences your message, they have every right to experience it again?
If you could tape every moment of your life for your own personal viewing, would you like to be told that you can't legally re-experience moments in your life over again because they are copyrighted by someone else?
What if I told you that you were already taping your life through the wonders of ``memory''? What's the real ethical difference between ``remembering'' a song you experienced before, and listening to it again? Especially since some of us have photographic memories and can literally re-read a book, rendering most of the obvious differences moot?
Does it even make sense to allow the sender these sorts of restrictions?
Clearly, I'm trying to bias you in favor of ``no''. I admit that. I know there are counterarguments. I just want you to consider all sides of the issue fairly. Again, I issue the ``this is not just theoretical'' disclaimer; the technology to tape every moment of your life for your personal use already exists, and will almost certainly be available to the public in five or ten years, barring legal challenges. Moreover, there's no theoretical reason why we won't eventually be able to produce devices to tap into the human memory directly, though that is even further off. Are we going to extend ``copyright''-like concepts all the way into the brain? (Are we willing to extend the ``right to free speech'' into a ``right to free thought''?)
As our technology continues to improve, and our theoretical ability to control the receiver continues to improve, it's worth carefully examining the restrictions we are considering for the definition of ``experiencing'' the message they imply. Half of analyzing a Digital Restriction Management system is determining how it defines ``experience'' for the types of messages it is trying to restrict.
I don't support requiring senders to help receivers archive their works; there's nothing fundamentally wrong with the idea of pay-per-view as such. But what does it mean if we restrict pay-per-view customers from recording the pay-per-view event for their own archives? In the days before it such restrictions were possible, such as in the mid-1990s, one could say that when a customer purchased a pay-per-view event they were paying for delivery. If we restrict what people can do with pay-per-view content, does that mean that the customer is now paying for a viewing, rather then delivery? Does that mean that the customer has legal grounds to demand a refund if they do not actually view the event as planned? In the 1990s, we would have considered that a specious argument because the goods were delivered as contracted. Now, if the customer doesn't view the work, it seems they paid money for a viewing and got nothing in return. Can the pay-per-view provider just tell the customer that it's too bad?
To maintain the principle of symmetry, restrictions on what a receiver can do imply a corresponding responsibility for the sender. DRM that does not account for this is unethical. All DRM systems I know of are strictly in terms of giving the senders control with no mechanism for the receiver to hold the sender to their responsibilities.
Further, the contracts from the providers that boil down to ``The customer shall give us money, and we may or may not deign to provide them certain services'' need to be stopped as well. They are probably already unenforcable even under current law, so this is no great loss. The more tightly the reins are pulled on the receiver, the higher quality service the receiver should be able to demand.
I think it's worth taking a moment to differentiate between ``natural restrictions'' and ``unnatural restrictions''. Natural restrictions flow from the technology itself, and are often removed by later technology. It is not inherently easy to record television. It was never the responsibility of the television industry to make it easy for the home user to record. However, once the VCR came along, the industry had no right to restrict their use, either. A natural restriction was thus removed. There is nobody who intends (in the legal sense of ``intent'') to have these restrictions, they just naturally exist.
Unnatural restrictions are deliberate limitations to the technology, intentionally engineered into the final product (again in the legal sense of ``intent''). There may be a little bit of fuzziness here but it's usually pretty clear whether a technology has been deliberately hobbled or not, especially as our ability to sling numbers around improves.
There is obviously a big ethical difference between these two. That nobody could copy compact disks cheaply when they first came out was to be expected (``burnable'' CD's are actually quite sophisticated technology and they were not trivial to develop, even though we use them casually now), and the various industries did not have an ethical mandate to wait to use CDs until people could so copy them. The inability to archive or make direct fair use of the digital data was a natural restriction of the technology.
DRM is an intentional technology, creating intentional restrictions. Even when they merely mirror older sets of restrictions, there is a significant ethical difference.
The entire question of what it means to ``share'' a message can be boiled down to ``how do we define a receiver?'' This is the other half of analyzing a Digital Restrictions Management system.
In this essay, I've very carefully defined almost every component of the communication model. The receiver is left undefined because it's actually very hard to tell who is the ethical receiver (or, for the computer folk in the audience, the logical receiver). It's usually pretty easy to point to a physical receiver but that is often not the same thing.
Most media purchased from commercial retail outlets are essentially sold to households, not people. I may have bought a DVD yesterday, but my wife or children are free to watch it, even if I am not around. So while I may have physically acquired the DVD, it is clear my entire household has received it in the ethical sense.
This obviously doesn't bring us any closer to a rigorous answer of what a ``receiver'' is, for what is my ``household''? Is it just the legal adults legally registered as living there? Who on Earth can enforce that sort of thing, anyhow? If I have a non-family-member housesit for me, is it a crime for them to watch one of my DVDs? The more you try to narrow down who the ``receiver'' is, the more very common exceptions you find.
Right now it isn't a crime anyhow, because they could borrow it and watch it that way, so this can be considered as a form of ``borrowing''. But what of a DRM'ed movie, backed by the DMCA, that I paid for for only household viewing. Would it be a crime for my housesitter to watch that movie because they are fraudulently identifying themselves to the movie sender?
In fact this theory does not point toward a way to rigorously define a ``receiver''. It points to some ways how we could do that, but it does not prescribe one.
If one tries to narrow down what a receiver is, one is invariably drawn towards extremes, because the lines are fuzzy and people will abuse them. You can't define it in terms of ``household'', because that's a virtually meaningless term. Same with ``family''. One is pushed towards either a very narrow view, where only single people can be legitimate receivers, or a very broad view, where entire large, amorphous groups of people at a time are the receiver, as either the restrictions grow, or the group definition grows.
Clearly current DRM technologies are all moving towards the ``single person'' definition, where one person receives a message, and they may essentially never share it with anyone ever again, possibly not even with themselves if the message ``expires''. There's no grounds to call this inherently unethical, although the technical difficulties of restricting receivers this much while not restricting their Free Speech rights or creating an asymmetric relationship may be insurmountable. In fact I believe them to be, but this technically is an engineering issue and it is possible that with enough decades of work, a fair but restrictive DRM system could be built. If they are truly impossible then this isn't really an option, ethically, and there is also the question of the damage done while inferior systems are in place, which will certainly be for many years.
On the other hand, we are currently living in the world where the definition of ``receiver'' is very, very loose, and I'd submit for your consideration that the economy has conspicuously failed to collapse. Right now, practically speaking, I can copy a CD and give it to a friend. Nobody is seriously claiming this is destroying the music industry. (The music industry certainly talks about this on occasion but their real ire is saved for the P2P sharing systems, more about which in a moment.) Is there anything terribly wrong about sharing a movie with your parent, or a friend? Especially if one could develop a culture that still encouraged paying for it? (Anecdotally speaking, I think we live in that culture already. Certainly the purchase rate isn't 100%, but one questions who got the idea that such crime could ever be stamped out, or why these crimes justify restrictions on the rest of us.) The sheer mathematics of life and time work out such that even if giving copies of movies to our friends was as easy as pushing a button, we have better things to do with our time then worry about that.
I think the maximal value to society is obtained with this fairly loose idea of ``receiver'' that we inherited from older technologies. This includes a seperation between the idea of ``sharing with one's friends or family'' and ``sharing with the world''; we did not condone large-scale piracy in the 1990's and I think that the modern equivalent, large-scale anonymous file sharing to all comers is also unethical. But here I think we can draw a relatively sharp line between ``sharing with all comers'' and ``sharing with just people you know''. (LawMeme)
The tension here comes from the difficulty in defining ``receiver'' in different ways for different technologies. Certainly the large content providers of today are comfortable with locking down to individuals and grudgingly admitting that families can probably watch the same movie (though they'd make them pay extra if they could), but that's not necessarily the social ideal. And one must also question the costs of implementing these schemes, which are expensive in both development and maintenence costs, vs. any gains for anyone, even the actual sender, let alone society. Unfortunately either a restrictive or a free model must be chosen, and as usual it will be difficult to impossible to mix the two.
I think it can basically be boiled down to this:
Once you choose the answers, propogating them back out to the various message types and media are really relatively simple. It's the figuring out what we should answer that is that hard part, and what I'm trying to at least help with with this essay. As a society, we need to come to a consistent idea of how the sender can restrict the receiver of a message, while still maintaining symmetry of relationship. It's a subtle problem with more then one good, consistent answer, but quite a few bad, inconsistent answers.
It is probably too late to change terminology, and I'm not enough of a Don Quixote to think I can. But if I could, I would change a few words we use in this domain.
The term ``Intellectual Property'' covers nearly everything I mean by a ``message'' except patents, and it probably wouldn't be much of a stretch for most intellectual property lawyers to see privacy-sensitive information as another form of intellectual property. (It certainly already is once it's collected, becoming a normal copyrighted aggregation.)
Many people criticize the term ``intellectual property'' as inherently containing a metaphor of ``ownership'', be it of the communication itself, of the ideas (patents), of the phrasing (trademarks), or something. They believe that the physical concept of ``ownership'' does not apply in this domain, because it has no meaning. They believe that using the term ``theft'' or even worse, ``piracy'' (http://www.gnu.org/philosophy/words-to-avoid.html#Piracy) is inappropriate because there are no real analogues with the physical meanings of the term. By using these terms, the entire conversation about communication is being prejudiced in a certain direction that does not truly reflect how the things referred to as ``intellectual property'' works.
They are correct. Certain aspects of the ``property'' idea are still useful, but as a whole, the entire ``intellectual property'' idea is a bad metaphor that inhibits understanding, rather then helping it (http://www.jerf.org/irightsSupp/concepts/metaphors.html). The few good aspects are swamped by the ways the metaphor impedes understanding. Communication is not property. It can not be meaningfully stolen. It can not be meaningfully transferred in toto to someone else, because at a bare minimum, the original owner will still remember the content of the communication. There is no meaningful analog to copying a communication in a physical pro.
As discussed in the essay pointed to by the previous footnote, a metaphor is only as useful as it correlates to the origin of the metaphor. The fatal flaw of the ``intellectual property'' metaphor is that with communication, possession is divorced from the ability to own rights on the property. I own this physical chair I am sitting in as I write this. I can do as I please with it, and you can not. If I give it to you, I do not have it any more, and I lose all rights to it. On the other hand, I can and have given you a copy of this essay, yet I believe I retain certain rights to it. This is a night-and-day difference between ``intellectual property'' and real property, much more then enough to make the metaphor useless.
``Intellectual property'' as a concept dates from a time when we did not understand how thoroughly computers would revolutionize communication and help us attain nearly 100% of the theoretical capabilities of communication. The fact is, we've outgrown this old metaphor, and we should be mature enough as a species to give up this crutch. Just as we gave up the concept of a ``horseless carriage'' and moved to ``automobile'', just as we gave up ``wireless telegraph'' and moved to ``radio'', it's time to give up ``non-physical property'' and moved to a more mature model of communication.
Note throughout this essay I've studiously avoided using the term ``intellectual property'' unless I was explicitly discussing the current modes of thinking. After all, the whole point of my essay is to clarify thinking, not muddy it up.
I'm not good at naming things, but I would propose we need to move to a message-rights based understanding of communication to completely replace the concept of intellectual property in the domain of communication. The question that truly needs to be answered is who has the rights to do what. Note the terminology involving ``rights'' is already used in practice, ``intellectual property'' is only used as an aggregate term, and as a rhetorical term. It is very rare to hear of people actually talking about licensing ``intellectual property'', though I've heard some lawyers use the phrase in public. Typically one talks about licensing the rights: ``I licensed the movie rights to the book.'' I think part of the reason can be seen in that sentence: The word ``rights'' can take an adjective like ``movie'' in a way the phrase ``intellectual property'' can not: ``I licensed the movie intellectual property from the book.'' doesn't make sense, because there is no movie intellectual property until the movie already exists.
This, incidentally, is another way the ``property'' metaphor breaks down; ``property'' can not be meaningfully traded or manipulated until it exists. Future rights to physical property can be, though, just as rights to messages that do not yet exist can be.
If the Intellectual Property metaphor's foundation looks something like this:
It isn't just the ``IP'' lawyers making the mistake of lumping many disparate concepts under one confusing umbrella term, though; by lumping so much stuff under the heading of ``intellectual property'' you have people calling for the complete abolition of ``intellectual property''. But because of their poor conceptualization of the problem, these people are calling for something that would be worse then what we have now.
Abolishing IP, be it a version of my communication-based protections or even modern IP, means abandoning everything in this essay. It means abandoning privacy entirely, because it can not be protected. It means abandoning all rights and all control to your messages, not just the messages of others. It means abandoning message integrity, because the defense for message integrity is effectively a copyright-based one. As a result, it means abandoning free speech, since integrity can not be protected and without integrity, we have no free speech. And yes, it means abandoning market rewards for new information, which may be OK in certain very limited domains like software, but will destroy other domains. Even trademarks can be important in subtle ways that we don't usually think of since we live in a society with strong trademarks.
Ironically, the IP abolitionists often hold thier views in the mistaken belief that this will increase their freedom and power, and decrease the power of corporations. They couldn't be more wrong; in the anarchaic environment that would result, it would be corporations who would be better able to exert power, more inclined to create and use DRM standards that only corporations can afford, and individuals who would be left almost entirely defenseless. Only strong IP laws even gives an individual a chance against a large corporation. If anything, IP laws need some selective strengthening, not weakening or outright destruction.
This demonstrates the importance of proper terminology to proper understanding; if we were really working in a message-rights-based model, these people could instead be fruitfully discussing what ``inalienable rights'' a receiver or a sender has to a message, which is a very important discussion for our society. Instead, these people marginalize themselves by calling for something that won't and shouldn't happen. The IP abolitionists are wrong.
Not everything that is currently called intellectual property fits into my communication model.
Patents do not fall under the domain of communication (except software patents as described earlier); physical patents can be cast in terms of communicating how to build something but they are clearly intended as protecting physical objects, not the communication itself since the entire patent application is publically and freely available to all (http://www.uspto.gov/patft/index.html). Physical patents contribute to the belief that ``intellectual property'' is a good concept because it seems like they can indeed be treated as property much more accurately then communication, because they can be truly bought and sold in a way that often deprives the original owner of all rights. Physical patents do not fit in with the other IP concepts, though, because they are not communication related. Instead, a physical patent is the right to tell someone to stop manufacturing something.
I use the term ``physical patent'' to distinguish it from a ``software patent'', which is communication related as discussed in the Patent chapter. As software becomes more and more powerful, it may be the case that all patents will become software patents, in the sense that they all become programs running on some powerful robot. In that case, the concept of a patent will become obsoleted, being supplanted by communication-based models.
I've used corporations as an example many times in this essay, but I've not directly addressed how corporations fit into this model. This is because in theory, they are just another entity and as long as the symmetry is maintained in the communication relationship, there's no special theoretical need to address corporations. But as this chapter is addressing more practical concerns, the practical implications of our large corporations should be addressed.
You may think you've detected an anti-corporate undertone in this text, but that's really not representative of my opinions. I've used corporations as examples simply because they are interesting, they get into the news and other permenent records and thus provide citations for me, and because frequently only a large corporation can take advantage of or build new technology. You simply can't run DirecTV out of your garage.
Corporations are a tool of humanity and like any other tool can be used for benefit or harm. Many things we take for granted are difficult or impossible without at least one large company involved, such as making cars. (Automobile companies are supported by hundreds of smaller companies, but there must be one final company that assembles, tests, markets, and sells the final product; those functions really can't be broken up efficiently and that's going to be a big company.) There's nothing inherently wrong with them. But practically speaking, it must be acknoledged that they can wield significantly more power then a single person, power which can easily come into play with communication relationships.
The problem is quite simply one of man power. Today, I have roughly 16 hours of wakefulness, assuming an 8-hour sleep cycle. Every workday, a corporation receives on average slightly more then eight man-hours from each employee. That is to say, for a 10,000 person company, for my 16 hours today, that company received 80,000 man-hours of life (rounding to 8 man-hours), 5,000 times more then me.
Now suppose this company sues me, and we get into a drawn out lawsuit that occurs over the course of a year, and consumes roughly a quarter of my year meeting with lawyers, rotting in jail, preparing defense, worrying such that I can't productively do anything else, etc. (That's probably conservative on my part; it could easily completely destroy my year.) If I live a nearly-average (and conviently rounded) 75 years, that's a third of a percent of my entire life. (If you're willing to call it the entire year, that's one and a third of a percent of my life. If I died tommorow, that would be a full 4% of my life, as I'm near a rounded-by-luck 25.)
Let us suppose this lawsuit also eats three lawyers and the equivalent of one administrator year, for a total of 50 * 5 (fifty weeks, five days a week) * 8 (eight hours a day) * 4 (four people) = 8,000 thousand man-hours. Now, that may sound like a lot but it's only a tenth of one day for the 10,000 person company.
In terms of communication issues, I think that practical symmetry requires thinking about the following considerations, above and beyond normal concerns like monopolistic or oligarchic practices:
On the other hand, if I pay all of my bill but a penny, they will still come after me for the penny. I do not know how far they would push the issue and while it would be an interesting experiment I'm not willing to put my credit history on the line for this. My point is that a corporation, through the economy of scale it gets through its size advantage and automation advantage, can afford to really hold me to my end of the contract. There's an asymmetry here, because they've slipped several times.
Of course, it's actually worse then that. They're a big conpany with lots of lawyers. Technically, the contract states that while I will pay without fail the bill the company sends out for me or suffer great consequences, in return, they will try to deliver service, but if they fail, well, gee, that's too bad. They aren't responsible for anything, up to and including returning money for service not received. If they do return some money it is out of the goodness of their heart. This is one manifestation of point #1.
I honestly don't know what to do about this practically, and it obviously extends beyond merely communication issues. I'm not satisfied with any proposals I've seen so far to try to rectify this problem. But it's a real problem and it's only going to get bigger and more relevant as the armament available to corporations such as DRM improves. I do not know how to convert theory into practice here. But I think we need to consider that very carefully as a society or we will regret it as individuals in the future.
As I mentioned at the outset, I don't really think it is my place to lay out a complete legal framework, nor am I necessarily qualified to do so. As long as I was speaking only about ethics and generalities, I feel relatively safe that my opinions are meaningful, but constructing laws is a bit out of my domain. However, I do believe that the analysis of the issues can give lawmakers at least a few concrete guidelines while trying to formulate the practical ethics for the next few decades.
First, as I emphasized repeatedly, catastrophic effects occur when technology is directly addressed; contradictions are inevitable, and the legal loopholes will proliferate. Laws should not address technology directly, but be solely concerned with effects on humans. Even recent laws have been falling down on this criterion; the DMCA outlaws ``circumvention of a technological access protection measure'', instead of the more proper, more general, and more human-centric ``accessing of a message without paying the owner a reasonable fee'', or some other reasonably human-centric formulation of the real point of the law. By mentioning technology at all, one invokes a morass of issues about what constitutes circumvention, the logic behind outlawing DeCSS even when used only to view DVDs the viewer owns a fully legal copy of, free speech implications of such outlawing, etc. If one merely bans the effect of viewing a protected work without paying, with regard only to the payment and without regard to how the protections were circumvented, a much more general and yet simultaneously less objectionable law is created.
Second, once that principle is understood it becomes clear that the primary public policy question to answer is ``Shall a given effect be legal, illegal, or something in between?'' Rather then asking if a particular annotation technology is legal, ask if anything with an annotation effect is legal, or under what circumstances it may be legal. Rather then trying to define ``framing'' in the context of HTML itself, ask if the idea of re-wrapping somebody else's content in a new border is legal, and under what circumstances it is. Once a list of basic effects is created over time, through legislation and court cases, a coherent and meaningful body of law could be created that is actually fair, predictable, and workable. Even if someone creates the World Wide Web Squared, a wonderful new digital frontier as different from the WWW as the WWW is from a world with just email, that case law would still be very useful in determining what was legal and illegal in that new environment.
Prof. Touretsky's Gallery of DeCSS descramblers, previously mentioned in the Software Patents section, located at http://www.cs.cmu.edu/~dst/DeCSS/Gallery/index.html, brings up a serious of concrete questions about what constitutes a circumvention device under current DMCA law. It highlights the absurdities that are inherent in the way the DMCA tries to define a ``circumvention device''. Such a gallery is an excellent concrete example of how one can try to skirt the law by dancing around the line; there is a very smooth continuum between ``graphic'' and ``program'' shown in the gallery. This highlights the absurdities in the way the DMCA tries to define a ``circumvention device''; as soon as you try to nail down a definition of ``device'', somebody can come up with something that isn't quite a ``device'', but would allow ``circumvention'' with a reasonable amount of effort.
With my formulation of the issues, even if you do decide to ``ban DeCSS'' there's a reasonable answer to what constitutes a ``circumvention device'': DeCSS is a ``circumvention device'' the moment it is used to do something that affects a human, in this case, decoding a DVD that the law says should not be decoded. Even a copy of the DeCSS algorithm, sitting on the hard drive, one simple command away from running on a DVD, is not a circumvention device. It must be loaded into memory and actually executed in such a way as to affect a human, and only that copy is a circumvention device. Even a partial application of these principles would correct some of the outright absurdities embedded into current law. Again, this boils down to the following the effects: The question is whether a movie was illegally watched, not whether a program was executed.
In many of these cases, there is a societal judgement call in where the line is, and while I have my opinions, I certainly believe that there are a number of other valid opinions. Thus, I don't really see the point in trying to enumerate these effects and giving my opinions. What I really care about is that these principles be used to create good law, better then the inconsistent, incoherent trash currently being passed and judicated currently.
Let's take the case of the Australian television authorities considering requiring a license for streaming video as discussed in 2.3.4, and try to make a good law out of it in stages.
I've implicitly defined ``large-scale streaming'' in terms of how it is technically served out. Therefore, you've opened yourself to technical circumventions of the law. Imagine a peer-to-peer video sharing network, where no one peer ever serves out more then ten streams at a time, yet the system as a whole reaches thousands or millions of people. As is usually the case with ``technical'' circumventions, there's a good reason to do this even outside of ``getting around the law''; the bandwidth drain on any one person is much lower with such a system, and if done correctly is easier on the network as a whole as well. There's good reason to create this sort of network, technically. For a real-world example, see BitTorrent (http://bittorrent.com), which helps download large files like Linux distributions without hammering the hosting server.
The question to answer at this point is ``Who are we seeking to regulate, the broadcasters (transmitters) or content producers?'' Since for television, broadcasters and content producers are typically the same organization, or at least very closely related, this will be a new conundrum for the television regulatory agencies. The answer changes depending on exactly which regulation we're considering. Certain types of regulations, such as ``requiring X minutes of public service broadcasting a day'' are regulating the broadcasters, whereas content-based regulations are laid upon the content producers. Since we're talking the Internet, the television regulatory commision probably won't have the power to regulate the Internet common carriers, so they'd either have to give up that idea or work with the agency or agencies that do have that power. They can continue to regulate content, though, so let's try that:
Also note I've not tied this to the stream being ``live'', because that is another technical consideration that would allow people to circumvent the law by not streaming it ``live'' but offering it for download instead. On the other hand, the phrase ``in an hour'' is necessary because a small video file could collect 1,000 viewings over the course of five years; we're not worried about those.
Having brushed away a hundred years of misconceptions and layer upon layer of unclear, fuzzy concepts and terminology, and replaced them by a clear model for communication, it becomes clear that the fundamental question before us as our communications technology continues to progress is this: What restrictions are fair for a sender to lay on a receiver, and how can we back the answer to that question up by law?
The current confusion we are experiencing is not fundamental to the domain, it's caused by our attempts to be radically inconsistent in how we answered that simple question, and our inability to sustain the previous answers now that previously seperate communication domains are merging. There is no true need to be inconsistent, and the harder we try, the more the system will strain against us.
As I said above, I don't believe that there is One Perfect Solution that we must choose, I believe there is a range. But that range is really a rather narrow range, with only a handful of choices to be made, and once we have made those choices, practical application is rather simple.
![\includegraphics[width=0.65\paperwidth,keepaspectratio]{degrees.of.freedom.eps}](img35.png)