This should be before chapter 3, but who would read it? It's worth explicitly going through our goals, so you can easily evaluate other proposals against the same standard.
It's a chestnut by now, but ``Perfection is achieved not when there is nothing left to add, but when there is nothing left to take away.'', attributed to a wide variety of sources.
No solution to any communication issue that works in the real world will ever be simple in execution, and there's no way we can totally avoid ambiguity that the courts will have to clean up every so often over time. But we must strive for a conceptual framework that is simple in its heart, that is based on some simple concept.
Complexity in implementation could be tolerated when these issues were more or less separated. Unfortunately, when large complex systems that were not designed to interact with one another start interacting anyhow, the result is vastly more complex. A common metaphor here is that the complexities tend to behave like multiplication, and I would consider that essentially accurate. Those with good mathematical intuition will recognize that I am claiming that it will be utterly impossible for our legal system to continue with its current policies; the number of systems interacting is high and they are all combining.
We all sense this. We all feel the system is flying out of control and that nobody has a grasp on it. The major content production groups (RIAA, MPAA) react by getting more and harsher laws passed to try to return some semblance of control to the situation, but that only makes the landscape even more complex, as the laws are broad and ill-conceived, and thus open to abuse. The common man watches a bewildering, incomprehensible miasma of new laws pop up and tell him that any number of previous acceptable activities or devices are now illegal (http://www.freedom-to-tinker.com/archives/cat_fritzs_hit_list.html). The domain of activities one can safely engage in without a lawyer steadily shrinks, which is especially economically damaging when you're trying to sell to the general public as they can become afraid of the complexity and simply not buy. The complexity of the copyright system is rapidly exceeding the abilities of any one individual, no matter how highly trained, to keep track of it. Among the other already-stated reasons, this is partially why I do not spend too much time getting into the specifics of the current system; I simply don't have time to absorb all of its complexities and still get my Computer Science degree, or do my non-lawyer job.
Odds are any set of solutions attempting to address areas individually will fail this criteria, unless the solutions truly can strictly partition which parts take effect under which circumstances so that they do not interact. I can't prove this because it's really hard to prove a negative, but I'm willing to bet no such set of solutions exists such that there will never (or close to never) be any ambiguity which part of the solution applies to a given case, simply by virtue of the fact that the flexibility of the Internet does not seem to allow the drawing of clear lines between the capabilities.
We must drastically lower the number of special cases, preferably even the number of classes of distinct kinds of communication from its present high count. If we do not, the system will (continue to) self-destruct.
It is only reasonable to assume that we've only begun to see the types of interactions that will be occurring and the legal actions that will result.
Any proposed solution must not just work for yesterday and today, but work reasonably well into tomorrow. This characteristic is required for much the same reason as simplicity; as the interactions between the various legal domains occur, the guiding principles of the solution need to provide some measure of guidance. No matter how pretty a theory somebody may spin, any real-world legal solution will have its share of special cases and exceptions. The guiding principles themselves, as explained by the theory, must still provide guidance if they are to be of any use, even if society deliberately chooses to do something else. If the theory itself is contradictory on some issue or combination of issues before the real world is even considered, then the theory is of little use, and may even be a hindrance.
And not just tomorrow as in ``next year'', but hopefully for decades to come. Companies need to know that their business models will still be legal (or still be illegal) ten years from now with some level of confidence. The new technologies that are developed need to not affect the viability of the solution too much.
In order to accomplish this goal, it is clear that a solution tied too intimately to technology is doomed to fail this measure. Technologies are very ephemeral. This tells us that for better or for worse, we must create guiding principles that do not really relate to the intricate engineering details of what a technology is doing. We must take a higher level view of what is going on, and somehow manage to not get bogged down in details.
This will be a great challenge to correctly perform that balancing act, but if we can manage to find a solution that has long-term viability (a big if!), it also means that the precise details of the technology disputes of the future won't matter to the judge so much. Once the dispute's relationship to the guiding principles are ascertained, the technology will have to fade into near irrelevance. This is good news, because it means our judges and lawyers won't necessarily need degrees in engineering, just enough training to understand the principles of engineering (although even this lower standard is curiously absent in current lawyers).
It should be clear that the path we are currently on is the very antithesis of robustness. Once you start using technology to move content in a way not possible in the early 1980's, you are in trouble. There are simple issues on the Internet that remain largely unresolved, and are largely ignored because they provide the foundation of the Internet as we know it.
Another example of this, in addition to ones described elsewhere: Do you have the right to access content on public servers, or can I selectively deny certain corporations from accessing the information? For instance, do an Internet search for ``eBay vs. Bidder's Edge''; eBay tried to deny Bidder's Edge access to their servers, which Bidder's Edge was using to download huge quantities of information to provide services to other people. This case ``resolved'' the issue only in the narrowest sense, and using the rather dangerous doctrine of trespass, which really doesn't make much sense in this context.
Any good solution must be complete, and cover the entire domain of issues well. Solving ``The Napster Problem'', in isolation from anything else, is not a solution at all; any decision made in isolation in that case will only create more loopholes and special cases that others may find ways to use to their advantage. For example, there was once a bill before Congress that would specifically allowed people to do what MP3.com wanted to do with its ``beaming service''11... are they ready to guarantee the same rights for all other media as well? Is that law ready for the case when some company allows people to scan their VCR tapes somehow, beam confirmation to the company's servers that they own a legal copy of the movie, and allow people to download the digital video version? If not, why not? Movies are just like songs in most of the important respects; the only reason it's not already happening is that movies are large. (This also serves as a good example of a non-robust solution as well; what little virtue the bill had when it was proposed will only diminish over time.)
I emphasize that is only a particular example. There are tens or hundreds that could plausibly generated. The solution must be complete; it can't just deal with music, but ignore movies. It can't deal with movies and ignore computer games. It can't deal with books, but ignore websites. Far too many little solutions to little problems have been proposed, but since each interact in unpredictable ways, they can never solve more problems then they generate.
This requirement basically knocks the court systems out of the running. Courts can only properly rule on cases that are before them, and they basically lack the ability to create coherent systems of law. Congress is much better at that, so we must expect a legislative solution. It should also be clear that the legislative branch in the US is currently writing laws that are no more coherent then the court system. If it isn't clear to you yet, contrast a law like the DMCA with the general framework proposed in this essay and you'll see what I mean. In fact, Congress is several steps behind the courts, as the courts have had to deal with actual conflicts and resolve them somehow, while Congress was able to just ignore the problem, or view the issues through only one side's point of view.
Short-term viability may seem redundant to robustness, but it is not. This is a largely economic concern; not usually associated with ethics but I prefer ethics that are usable in the real world to useless abstract principles, so it should be considered. Companies need to know whether or not a business plan is viable before they open themselves up to the potential damages that Napster faced... before settling with the music companies, Napster was looking at hundreds of billions in claimed damages. But this is not the real reason that short-term viability is necessary.
It is common knowledge that the technology industry moves much faster then the legal system does. This is wrong. In reality, the legal system merely moves much less often, and deliberates those moves more carefully, then the technology industry which has a more continuous flow, with constant little changes.
The legal system may move rarely, but when it does move, the moves are massive! The Digital Millennium Copyright Act changed a whole lot of things when it was passed, and did it very quickly. Overnight, a large number of previously viable activities became illegal, and a much smaller number of previously questionable legal activities became well-determined. For instance, nobody knew how much legal protection was gained by putting access restrictions on something; the DMCA cleared that up, regrettably in favor of giving any restriction whatsoever full legal protection. Also, a large number of other activities that were previously not illegal in and of themselves, such as bypassing access restrictions, became illegal. The tendency of the legal system, especially the legislative branch, to produce huge changes is the basic reason it is necessary to deliberate so carefully in the first place.
If a solution makes sense in the short-term, this basically means that there aren't any technologies that can suddenly pop up that create contradictions in the solution. This means that while the legal system may take a great deal of time to arrive at a conclusion, if the solution makes sense to most people, the government should act predictably. Predictability is generally good in a government.
If ethical and applied-ethical (legal) solutions won't even work next month, then of course it's simply a waste of our time!
For our final criterion, I'm going to mention something that may seem silly, but it's still a necessity. The solution needs to be useful; it can't be so weak that it has no real applications. This is perhaps the biggest problem with the current guiding principles of the legal system; they are so weak now as to be nearly useless.
Intellectual Property law's basic guiding principle is to promote innovation and thereby good to society, at least in theory. It's a good principle as far as it goes, but it has the unfortunate problem that there are multiple good ways to do that. Both sides of a debate claim they are supported by this principle, and generally both sides are right in their own way. The principle doesn't have any usefulness. This means that the court cases are being decided less on the merits of the case and more on the particular biases of who the judge thinks is innovating. For instance, in the first round of the Napster vs. RIAA case, in my opinion Judge Patel essentially ruled for Big Money and against the Internet Punk Thieves... which may be a valid ruling (assuming that Napster also represents Internet Punk Thieves, which is also another discussion) but was not truly informed on the basis of promoting innovation, but rather on the basis of preserving the interests of the status quo. (Note I also would have ruled against Napster, just for entirely different reasons, so this is not sour grapes about a ruling by one of said Punk Thieves.)
So if I get done spinning a theory but that theory never says anything is right, wrong, or up to the society to just decide, it's still completely useless.
I think it's pretty obvious that current practice fails on every single one of these counts. If you find this topic interesting and have read other people's essays on these topics, I think you'll also see they fail most of these criteria, especially because there are so many essays on isolated topics. It's great that there are people writing about these problems, and many of them have been of great help in putting this together, but we need to start creating coherent structures that might be useful across many of these issues if we are to do anything.