O Sweet Mr Math

Dedekind cuts and decimals

2012-01-27T23:55:00.001-05:00

In my last post, I gave a definition for the real numbers intended to fill the holes in the rational numbers. Just to restate it, real numbers are an extension of the rational numbers so any set of real numbers which has any upper bound has a least upper bound.

There are two points about this definition. The first is that this is just a definition. It doesn't prove that the real numbers actually exist. Fortunately, there are techniques of constructing the real numbers which prove that they exist. My analysis textbook includes one of these techniques, known as Dedekind cuts. Interestingly, the construction by Dedekind cuts is in an appendix rather than the main text. Essentially the books says, "here's the definition, here's the proof in case you really want to know, now let's ignore that and start talking about the properties of the real numbers." The course notes I have actually go further, and say, "there's a proof in the book, but don't bother to read it." I went ahead and read it anyway.

I think I followed the Dedekind cuts proof, although I certainly couldn't reconstruct it. On first reading, I'm not convinced it gives me any insight into the real numbers not provided by the definition. I expect that if I study the subject enough, I will get to the point where I can say that this is what the basic outline of the proof looks like and that is why it is useful, but I'm nowhere near there yet.

The second thing is that in thinking informally about real numbers, I have thought of them as numbers that can be written as decimals with an infinite number of digits. The textbook has a very short section on decimal numbers, basically showing that this is valid. The last sentence of the section is, "Since we shall never use decimals, we do not enter into a detailed discussion."

Beyond the fact that going to the trouble to define decimals just to say that they won't use them is kind of funny, I think there's a useful point here. I work with decimals every day, basically every time I turn on a computer or use a calculator. As a result, It's really easy to think that "numbers" means "decimals". This flat out rejects that premise.

The slightly subtle part of this is that thinking of real numbers as decimals does not actually help understanding of the basic properties of real numbers. In fact, this thinking obscures more than it reveals. I can't help thinking that various Internet debates about real numbers are driven by the thought that real numbers are decimal numbers, and rejecting that thinking might help defuse these debates.

Least upper bounds and real numbers

2012-01-26T23:12:00.000-05:00

Last time we I outlined the proof that the square root of 2 is not a rational number. Now I will discuss why this is a problem, and define the real numbers as a solution.

Consider a set of rational numbers, for example all rational numbers less than 1. If there exists a number such that every number in the set is less than or equal to that number, that number is an upper bound for the set. For example, 2 is an upper bound of the set of rational numbers less than 1. Of course, 1 is also an upper bound. If a set has any upper bound it has lots of upper bounds.

So instead of talking about any upper bound for a set, we want to talk about the least upper bound. That is, the smallest number that is an upper bound. If we choose a smaller number, numbers in the set will be greater than the number we choose. If we choose a larger number, then there exists a number smaller than that which is still an upper bound. So the least upper bound sits right in the middle, and could be thought of as the "best" upper bound.

The least upper bound for the set of rational numbers less than 1 is obviously 1. The problem comes when we try to find a least upper bound for the set of rational numbers less than the square root of 2. Since the square root of 2 is not a rational number, it cannot be the least upper bound. Any rational number greater than the square root of 2 is an upper bound, but if you try to claim that any particular rational number is the least upper bound, it is always possible to find a smaller upper bound. The only conclusion is that there is no rational least upper bound for this set.

An informal interpretation of this result is to say, okay, the square root of 2 is not a rational number. Why don't we use the "best" rational approximation of the square root of 2 instead? Leaving aside the fact that approximations aren't really a mathematical way of thinking, the least upper bound result shows that there is no best approximation, because you can always find a better one.

Essentially, this means that the rational numbers have holes. It's true that given any two rational numbers, you can always find a rational number between them, so the rational numbers are very close together. However, there are still numbers we would like to use (like the square root of 2) which just don't exist in the set of rational numbers, and there's no rational number we can use as a substitute for these missing numbers.

This is where the real numbers come in. The real numbers are defined as an extension of the rational numbers with the property that any set of real numbers with an upper bound has a least upper bound. The least upper bound of the set of rational numbers less than the square root of 2 is the square root of 2. Therefore, the square root of 2 is a real number.

Now that the real numbers have been defined, I will post some thoughts about them next time.

Rational numbers and the square root of 2

2012-01-25T18:12:00.002-05:00

Okay, enough talk about studying math. I want to talk about math. I've been studying real analysis recently, and so far the discussion has been on the properties of real numbers.

I mentioned that the course notes started with the natural numbers and addition, and proceeded to derive multiplication and exponents. The notes also develop zero and negative numbers before getting to rational numbers. The textbook skips straight to rational numbers and I will too.

A rational number is any number written as a fraction where both the numerator and the denominator are integers. In this conceptualization, the natural number 2 is a different concept than the rational number 2/1, even though they have all the same properties and can be used interchangeably in most contexts. Additionally, 1/2 is a different number than 2/4, even though they are equivalent.

One sometimes interesting, sometimes frustrating thing about the project to start over and redefine everything from scratch is that you have to revisit lots of basic math. In the early stages, this involves things which are totally internalized. I know that 1/2 = 2/4, but what does that mean in the context of this definition?When trying to develop ideas from the ground up, I end up second guessing myself frequently. Just because I "know" something is true doesn't mean I've proved it. Sometimes I embrace the challenge of proving everything, and other times I want to say, "I know this is true. Why can't I just say so and move on?"

Anyway, once the properties of rational numbers are established, the next step is to prove that the square root of 2 is not a rational number. I've seen this proof many times, but it took multiple times before it really sank in, so I will restate it here. Assume that x is a rational number such that x² = 2. Then x can be written as a fraction, p/q, where both p and q are integers. Many different values for p and q can be chosen to give the same equivalent fraction, so pick p and q so that at most one is an even number. (If both are even, just divide both by 2 as many times as necessary until one is odd.) Then (p/q)² = 2, so p² = 2q². Therefore p² is even, so p is also even. But that means we can choose the integer m so 2m = p. Substituting back into the earlier equation, (2m)² = 2q², and by the same reasoning, q is also even. But we chose p and q so that at most one of the two numbers is even, and this contradiction means p and q do not exist, so the square root of 2 is not rational.

Now that we've shown that the square root of 2 is not a rational number, it's reasonable to ask what kind of number the square root of 2 is. I will go into the problems that the square root of 2 causes for the rational numbers and why the real numbers are the solution next time.

Doing is better than making excuses

2012-01-23T23:52:00.001-05:00

I want to develop some related thoughts to my last post, in a different (and hopefully more optimistic) context. For a long time I thought I wasn't interested in studying mathematics at the graduate level. I used two basic arguments to support this. The first is that I didn't even understand what graduate level math was talking about. Upper level math features things like groups and rings and differential geometry and I don't know what else, but even the basic vocabulary is totally foreign to me. The second is that, in my understanding, upper level math isn't even about numbers anymore, and I like numbers. I have since concluded that I was wrong, wrong, wrong.

First of all, it's not even true that all upper level math is unfamiliar to me. Looking at graduate math department course descriptions, I can find lots of courses on subjects like statistics and Fourier analysis and things that I actually do have an understanding of and interest in. In some cases, like Fourier, my understanding is a result of the engineering courses I took after I basically decided I should get a degree in engineering rather than math. I may not have known I was interested in the subject before I took the engineering courses, but with the benefit of hindsight it's clear that I should have skipped the engineering and gone straight for the math.

Second of all, I don't know where I got the idea that if I was unfamiliar with the subject, I wouldn't be interested in it. (This may be related to the being smart vs. working hard distinction from last time. The existence of topics I'm not familiar with may imply that I'm not smart, so if I was being motivated by seeming smart rather than by working hard, that was reason to stay away.) Regardless, this idea is clearly ridiculous. I studied some abstract algebra last fall. I went in not knowing what groups or rings were, and now I have at least the basics down. Getting the basics is less important to me than the fact that the whole time I was studying, I was thinking, "This is great. Why didn't I take this course a long time ago?" Lack of knowledge is not an excuse. There was always plenty of evidence that I was interested in all the math I did know, so it's not surprising that this interest extends to subjects I don't know.

Abstract algebra also dents the objection about not being about numbers. Abstract algebra is about sets, which are made up of things, which are not necessarily numbers. So a lot of abstract algebra isn't really about numbers. But one of the central ideas of abstract algebra is that different sets behave the same, so when you're talking about one set you are really talking about all similar sets. Which means it's often convenient, when talking about any arbitrary set, to find a similar set of numbers and talk about that one instead.

One interpretation is that abstract algebra isn't really about numbers. Another interpretation is that abstract algebra allows you to talk about all kinds of things which are not numbers as if they in fact are numbers. The numbers don't go away. Instead abstract algebra almost makes the numbers more powerful by making them more universal.

Last fall, speed was more important than depth in my crash course in abstract algebra. I ended up with a pretty cursory understanding of the subject. But now I'm looking forward to going back and studying it in more depth. It's clear I was making excusing for why I shouldn't be doing this before. But the evidence is pretty overwhelmingly that those excuses were wrong. Now I'm looking for opportunities to study more math instead, and I'm a lot happier as a result.

Psychology of learning math

2012-01-22T02:11:00.000-05:00

A side note on the psychology of independently studying math: for several years, I've been big on the idea that it's far better to praise people for working hard than for being smart. There's some independent evidence for this. People who have been praised for being smart eventually hit problems which are hard for them, and then they stop trying. People who have been praised for working hard view hard problems as an opportunity to demonstrate how hard they work.

Generally I feel this way about myself and math. My strength isn't that I'm smart (which is not to say I'm not). My strength is that I have an almost obsessive need to solve problems that are put in front of me. At least, that's what I tell myself most of the time.

However, I've been flipping through my real analysis text and noticing my emotional reactions to some topics that are coming up. There's basic cluelessness, such as topology, where I don't know the basic vocabulary. I look at the text and it's all jargon that I don't understand. There's a level on which I feel like I can't be interested in this because I don't even understand what it's about. On the other hand, I have confidence that I will figure it out when I get there.

Infinite series is a subject that's a little more complicated. I have a long standing feeling that I don't really understand them. However, I reviewed the chapter on infinite series in my calculus textbook recently, and I felt like I was on top of the material as far as it went. The presentation in my calculus textbook was more concerned with the question of whether a particular series would converge than with what particular value it would converge to. It presented a bunch of convergence tests, but I'm not even sure whether it demonstrated why the tests work. I can look at a series and use the tests, but I still feel unsatisfied. Maybe the issue is just that the calculus text didn't go far enough. In that case, real analysis may give me what I'm looking for.

And then there's vector calculus. I came out of multivariable calculus not understanding vector calculus at all. Last fall, I worked out enough understanding that I could correctly answer test problems, but I still feel like I have no idea what's going on. There's a section on vector calculus in my analysis textbook that fills me with dread. On a certain level, I think this is totally irrational. When I get there, I'll be well prepared, and I will take as long as I need to work out what's going on. In all likelihood, once I figure it out, I'll wonder what the big deal was.

I went through this process with trigonometric substitution. Completely failed to get it first time around in calculus, then tried it again recently and found it straightforward but tedious. At that point I sort of resented it, until I realized that it's actually useful for evaluating line integrals.

So I expect that similar patterns will play out in the future. Topics will pass from completely unknown to incomprehensible to doable but pointless or annoying to actually useful. They may not all go through every stage, or in that order, but as long as I keep working hard, the comprehensible and exciting stuff should keep growing.

This leads to what I feel is the big weakness of independent study. There's no feedback. I think I mostly sort of understand what's going on so far as I've been working on real analysis, but I'm not sure. When I try to work out a problem from the book, I can feel like I both overthinking the problem and missing the point. Am I assuming things which I haven't proved? Is my reasoning not rigorous enough? Or am I trying to unnecessarily reinvent the wheel? In a classroom, there's much more opportunity to see how others are approaching problems and to just ask the professor for clarification. I have the book and the notes, but mostly I have to figure it out on my own. I know I've missed it badly in some cases, but I hope most of time I'm doing well enough. I also hope that the farther I get, the easier it will be to determine what the important parts of the stuff I've already covered were.

Real Analysis

2012-01-20T18:33:00.002-05:00

This year I'm attempting to teach myself and relearn a ton of math, so resurrecting the blog (again) to think things through seems natural. Future posts will come as need and opportunity arise.

The blog has neither comments nor RSS, (both broke years ago) so I'm probably talking to myself. Also, math is currently in HTML, which is inadequate, but I don't want to find a better solution now.

I'm working on teaching myself Real Analysis from a set of course notes and the book Principles of Mathematical Analysis by Rudin. The course notes start by defining the natural numbers and addition, taking the Well-Ordering Principle as an axiom, and deriving the Principle of Math Induction. It goes on to recursively define multiplication and exponentiation on the natural numbers, and to prove the properties of all three operations using induction. OK, so far so good.

The notes then go on to prove the formulas for ∑_i=1→n i^p for various values of p. The first step is to derive the formula. Start with n^p+1 = ∑_i=1→n i^p+1 − ∑_i=1→n−1 i^p+1. Do a bunch of algebra and the right side turns into a linear combination of sums from 1 to n of powers ≤p. You can then recursively determine a closed form solution for the sum of i^p.

The notes assert that this is not sufficient to prove that the formula is valid for all n. In order to prove it, you must use mathematical induction. However, the notes don't say anything about why the derivation is not a valid proof (or why it could not be rearranged to become a valid proof). It feels to me like all the math in the derivation is correct, so it should be sufficient as a proof. I can't tell if the derivation is sufficient, but the proof by induction is there as practice in proof by induction. Another possibility is that there is some mathematical weakness in the derivation, so the proof by induction is necessary. If there is a weakness, it is not obvious to me, and the notes do not explain.

The book does not help at all. It starts with the rational numbers, and I don't think it has anything to say on the subject of induction.

This bumps into my general nervousness around math proofs. If asked to prove something, I'm always comfortable jumping in and playing with the math and working something out, but I'm often unsure of what starting assumptions I can use. Over and over I write something and ask, "is this actually valid? I mean, I know it is, but do I have to prove it before I can use it?" In this case, I don't know why the derivation is not a valid proof, but I can accept that. But for future proofs, I don't know if I will know what the standards are.

Update on Lecture 2

2010-09-22T22:38:00.002-04:00

Last time, I said I couldn't figure out how to justify the discussion of g(x) for fixed-point iteration. I just realized this was actually discussed in the class, but I was too lost at the time to figure it out.

Back up for a moment to consider an example of bisection. For example, compute the square root of A, where A is some non-negative number. The square root of A is a root of the function f(x)=x²−A. We are looking for the positive number r which satisfies f(r)=0. Let's be lazy and note that r must be between 1 and A (even if A<1), so we can start with 1 and A as the endpoints of the bracket and start iterating. With each iteration, one endpoint of the bracket will move, and x_i, which is the midpoint of the bracket and our best guess for r, will converge towards r. Three iterations increases the accuracy of x_i by about one decimal place, so if we need a lot of decimal places it might take a while.

There are a couple of things to think about if we want a method that converges faster than bisection. The first is that we are only moving one endpoint at a time. The second is that we know f(x) in this case, and we aren't using that information. This suggests the possibility of moving both endpoints with each iteration by using what we know about f(x). For f(x)=x²−A, if x<r then r<A/x and vice versa. Therefore x and A/x bracket r. So if we start with a bracket with endpoints a and b, we can get a new bracket with endpoints x=(a+b)/2 and A/x.

When we try this approach to find r, we find that the number of accurate digits in our approximation x_i approximately doubles with each iteration. This is way faster than basic bisection, which requires 3 iterations just to get one additional digit of accuracy. In addition, since our two endpoints are x and A/x, we only have to keep track of one number in each iteration, as opposed to the two numbers required for bisection. This leads to the iteration formula x_i+1=(x_i+A/x_i)/2=g(x_i).

Applying this idea of bisection but with moving both endpoints to cube roots (f(x)=x³−A) leads to a guess of an iterating function g(x)=(x+A/x²)/2. However, this iterating function gives performance not much better than bisection, which prompts a general discussion of fixed-point iteration and the demonstration that if 0<|g′(r)|<1, fixed-point iteration has linear convergence, but if g′(r)=0, fixed-point iteration has quadratic convergence.

Having established that fixed-point iteration gives quadratic convergence if g′(r)=0, we can ask how to determine g(x) from a given f(x), which leads to Newton's method. However, I think conceptually Newton's method makes more sense geometrically, so I would introduce it geometrically, then show that it meets the criteria of fixed-point convergence with g′(r)=0 as long as f′(r) is not also zero and f″(r) is not infinite.

Numerical Analysis Lecture 2

2010-09-19T16:32:00.002-04:00

My policy of being really lazy about writing equations and using images is likely to severely reduce the comprehensibility of this post. This post is screaming out for some graphs, but I'm going to try to do everything in plain text anyway. Depending on how badly this goes, I may work on figuring out how to include graphs in the future.

A quick review from last time: Numerical Analysis is the process of computing approximate values for mathematical expressions that are difficult or impossible to evaluate exactly. Last time we looked at approximating e^x as a Taylor series expansion. Computing individual terms can be done with a few multiplications and additions, and a finite number of terms can be summed to approximate the true value of e^x.

There are two important limitations here: first, the uncertainty, or error, of the approximation depends on the number of terms in the Taylor series sum. The uncertainty can be reduced by computing more terms, meaning there is a tradeoff between the speed of the computation and the precision of the computation. Second, because of the way that computers represent numbers, rounding errors are introduced into all computations. Depending on the problem being solved and how the solution is computed, this can greatly reduce the accuracy of the computation regardless of the number of terms used in the computation.

This time, we are looking at a different type of problem. Given a function f(x), how do we determine the roots, or the values of x such that f(x)=0? We will look at three approaches, each with different advantages and disadvantages.

Bisection

The bisection method of finding roots is really simple and reliable, at the expense of not being very fast. The idea is this: Suppose that you know that for a function f(x), a root r exists such that f(r)=0. Suppose also that you know two numbers, a and b, such that f(a) is a negative number and f(b) is a positive number. As long as f(x) is continuous, r must be between a and b. (If you don't believe me, draw a picture. The point (a,f(a)) is below the x-axis and the point (b,f(b)) is above the x-axis. f(x) can be any curved or straight line that connects the two points, as long as f(x) has exactly one value for every x and there are no jumps in f(x). f(x) must cross the x-axis between a and b, and the value of x where it crosses is r. Note that this is true whether a<b or a>b.)

This is kind of nifty, in that we basically haven't done anything at all, and we already have an estimate of r. If we guess r as (a+b)/2, since we know that r must be between a and b, the maximum error in our guess is |b−a|/2. Suppose that we need a better guess. Then we can set c to the midpoint of a and b, or (a+b)/2, and find f(c). If f(c) is negative, replace a with c. If f(c) is positive, replace b with c. (If f(c)=0, then c=r and we are done, but we should not expect that to happen.) With our new a and b, we can be certain that r is between a and b, and a and b are now closer together (half the distance, in fact), so our estimate of r is better.

We now have a way to get an accurate estimate of r to any desired accuracy. If we start with an interval between a and b that contains r, by repeating the process of finding c, the midpoint of a and b, and setting it as one of the endpoints of the interval, the interval will get progressively smaller. In fact, the interval is half as large with each iteration, since c bisects the interval. Eventually it will be small enough to be an accurate estimate of r, regardless of how large the starting interval was.

So, the good news is that bisection always works. Given that f(x) is continuous and has a root r, bisection will always converge to r. (Note that if f(x) has multiple roots, it may not converge to the root you want.) You don't even need to know what the function f(x) actually is, as long as you can find the value of f(x) for any given x.

There are two pieces of bad news. First, you need the starting points a and b. The function x² obviously has a root at 0, but you can't use bisection to find it because there is no starting value of a such that a² is a negative number. If a function has multiple roots and you are interested in one in particular, it may be difficult to find starting values of a and b that will converge to the desired root.

The second piece of bad news is that bisection is slow. The error, or uncertainty, in r decreases by 1/2 with each iteration, so if your starting error is 1, then 10 iterations are required to get 3 decimal places of accuracy. This may not seem like a big deal, but if f(x) takes a long time to compute and you need many digits of accuracy, bisection may be too slow to be useful.

Fixed-Point Iteration

I haven't figured out how to justify Fixed-Point Iteration. The problem we are trying to solve is how to find roots of the function f(x), or values r such that f(r)=0. Fixed-point iteration, however, is based around the equation g(x)=x, for the function g(x). If you have a function g(x), and a value of x = r such that g(r)=r, then r is a root of the function f(x) where f(x)=g(x)−x. And if you start with the equation f(x)=0, then it can be rewritten in a variety of ways to get an equation of the form g(x)=x, but it's hard to argue why you would want to, or given multiple choices for how to rewrite f(x)=0 as g(x)=x, why you would choose a particular form for g(x).

That said, let's go ahead and look at g(x)=x anyway. The premise behind that equation is either neat or pointless depending on your perspective. The idea is that you start with a number x, do a bunch of things to that number, and get the same number x out at the end. If you had g(x) programmed into your calculator, you could enter the value x, compute g(x), take the result of that calculation and calculate g(x) again, and keep doing that as long as you want and the number you entered originally would never change. I'm assuming that if you think that's a colossal waste of time that you've already lost interest in this blog and moved on, so if you're still reading, you're like me and think that this is pretty neat.

Let's look at an example. x² is a standard calculator function. If you type either 0 or 1 into your calculator and then start repeatedly tapping the x² button, absolutely nothing will happen. This means that 0 and 1 are roots of the function f(x)=g(x)−x=x²−x. 0 and 1 are also the fixed points of the function g(x). In general, a fixed point r of a function g(x) is a value of x such that g(r)=r.

The interesting question is what happens when g(x)≠x. If you start with a number between 0 and 1, square it, square the result, and continue indefinitely, the result will converge towards 0. If you start with a number greater than 1 and square it repeatedly, the number will grow towards infinity. This suggests that by rewriting the equation f(x)=0 in the form g(x)=x, plugging a starting value x₀ into g(x) and then plugging the result back into g(x) repeatedly may converge on the solution g(r)=r. In this case, r is a root of f(x). But then again, it may not converge. In the case of g(x)=x², repeated iteration converges on the solution g(0)=0, but will not converge on the solution g(1)=1.

So, when is this useful? This is fairly easy to visualize. Draw a graph with the diagonal line y=x. Now imagine a function g(x) crossing the line y=x. The point where it crosses is the fixed point r we are trying to find. There are several possibilities for what g(x) could look like. If g(x) is rising as x increases, then it could be rising faster than the line y=x. In that case, when x<r then g(x) is below the line y=x and when x>r then g(x) is above the line. If g(x) is rising but more slowly than y=x, g(x) will be above the line when x<r and below the line when x>r. (Again, the words may be hard to understand, but if you draw it, it will be clear.) In the case where g(x) is falling, it's a little more complicated but the important question to keep in mind is whether it is falling slowly or quickly.

Repeatedly calculating new values for x can be easily visualized on this graph by creating a cobweb graph. Start at the point (x₀,0). Draw a vertical line to the function g(x). It doesn't matter whether the line is going up or down, just that it's vertical. The point where the line intersects g(x) is the point (x₀, g(x₀)). Next, move horizontally to the line y=x. Again, it doesn't matter whether you are moving to the left or to the right. The point where the horizontal line intersects the line y=x is the point (g(x₀),g(x₀)). If we label g(x₀) as x₁, this gives a new starting point. From (x₁,x₁), move vertically to g(x), at the point (x₁,g(x₁)). Then move horizontally to the line y=x. This pair of vertical and horizontal moves can be repeated indefinitely.

Depending on what g(x) looks like, there are 4 basic paths that this cobweb can take. If g(x) is rising, then the cobweb will be a staircase. If g(x) is rising slowly, the staircase will move toward the fixed point. If g(x) is rising quickly, the staircase will move away from the fixed point. (It doesn't matter whether x₀ is greater or less than r. In either case, the cobweb will move towards r if g(x) is rising slowly and away if g(x) is rising quickly. Don't trust me. Draw it!) If g(x) is falling, then the cobweb will follow a spiral path. Like the rising case, if it is falling slowly, it will spiral towards the fixed point, and if it is falling quickly, it will spiral away.

So that's what's going visually. What's going on mathematically? g(x) rising or falling slowly or quickly sounds vague, but it can easily be made mathematically precise. Rising or falling slowly means that |g′(x)|<1, while rising or falling quickly means that |g′(x)|>1. In the rising case, this is pretty obviously true from the fact that the slope of y=x is 1 and since g(x) is either rising faster (slope greater than 1) or slower (slope less than 1) than y=x. The falling case is basically a mirror image and the comparison line is a line with slope −1 through the point (r, g(r)).

So, now we know when this is useful. If we want to find a root of f(x), we should rewrite the equation f(x)=0 to look like g(x)=x, such that |g′(x)|<1 near the root. We can then start with a guess x₀ for the root and repeatedly calculate g(x) starting with the results of the previous calculation, and x will converge to the root r. There is an important question here: how do you rewrite f(x)=0 so that |g′(x)|<1 near the root of f(x)? Newton's method, which is described below, will give one answer.

Leaving that question aside, assuming you have the function g(x), how fast is it? Starting with the guess x₀, the error is |x₀−r|. But of course we don't know r, so we can't use that to put a bound on the error. But we can estimate the error if we know g′(x). Suppose we know that 0≤g′(x)≤m<1 for some constant m for all values of x between x₀ and r. (It's fair to ask how we can know m if we don't know r, but at least in some cases we can estimate it, and this is a worst case estimate anyway. We hope that Fixed-Point Iteration will work better than this.) Then if we draw a straight line with slope m through the point (x₀,g(x₀)), this line will intersect the line y=x at some point, and r will lie between x₀ and the point of intersection. Note that if m>1, r will not be between the intersection and x₀, and if m=1 the lines will not intersect at all, so this approach is only valid if m<1. (Again, this is using lots of words to describe something easy to see with a picture.) This point of intersection gives an upper bound on the range of possible values for r, and thus an upper bound on the error |r−x₀|. The x value at the intersection point x_int=(g(x₀)−mx₀)/(1−m), so the error is |g(x₀)−x₀|/(1−m).

We can use this to estimate the change in the error in each iteration. Starting from the point (x₀,g(x₀)), we can set x₁=g(x₀) and compute the new point (x₁,g(x₁)). In the worst case, g′(x)=m for all points between x₀ and x₁. Then g(x₁)=(1+m)x₁−mx₀ and therefore the error is |mx₁−mx₀|/(1−m). In other words, the error in x₁ is at most the error in x₀ multiplied by m. As long as m is less than 1, the error will decrease with every iteration. If we are lucky, m will also decrease with every iteration, meaning the closer we get to r, the faster we get closer.

In the case that g(x) is a decreasing function, the logic is a little more complicated since the cobweb path is more complicated, but the same basic conclusion holds. For any function g(x), if |g′(x)|≤m<1 for all x near r and including x₀, the maximum error in the estimate of r decreases by a factor of m with each iteration. Obviously, if we want to use fixed-point iteration to find a root of f(x), we should try to rewrite the equation f(x)=0 in the form g(x)=x so that g′(x) is as close to zero as possible around to the root. Again, Newton's method, described below, can be thought of as an application of fixed-point iteration that generates g(x) such that g′(x) is near zero when x is close to r.

We can now compare fixed-point iteration with bisection. For any function f(x) with a root r, bisection will converge to a root given two starting points a and b such that f(a)<0 and f(b)>0. In comparison, fixed-point iteration only requires one starting point, x₀, but only converges if the equation f(x)=0 can be rewritten in the form g(x)=x, with |g′(x)|<1 near r. The maximum error in the estimate of r is exactly known in the case of bisection, and the maximum error decreases by a factor of 1/2 with every iteration. For fixed-point iteration, the maximum error is only known if m≥|g′(x)| for x near r is known. The error decreases by a factor of m with each iteration, which means that fixed-point iteration can be potentially much faster or much slower (or not work at all) than bisection depending on g(x).

Newton's method

The thing about bisection is that it is dumb. Given the two starting points, our next estimate is always the midpoint of the two points. If we know something about the shape of the function, we should be able to make smarter guesses about the value of r than just taking the midpoint. Newton's method is one way of making smarter guesses.

Like with fixed-point iteration, Newton's method starts with a single point, x₀, but Newton's method works directly with f(x). Also, while fixed-point iteration can be computed with g(x), but g′(x) is necessary to compute the error in the estimate, Newton's method requires computing f′(x).

Start with x₀, and find the point (x₀,f(x₀)). Next, find f′(x₀) and use it to draw a line through the point (x₀,f(x₀)) which is tangent to f(x). The point where this line crosses the x-axis is the estimate of the root r, and this process can be repeated to get better estimates of r. The new point x₁ has the value x₀−f(x₀)/f′(x₀). Conceptually what this process does is start at the point (x₀,f(x₀)) and then follows along the curve f(x) towards the point (r,f(r)) (where f(r)=0). It does this by approximating f(x) as a straight line, which has the advantage of being fast, but the disadvantage that for some functions f(x), this is not a very good approximation.

If we define g(x)=x−f(x)/f′(x), then we can see that Newton's method is an example of fixed-point iteration. The error and rate of convergence of fixed-point iteration are dependent on g′(x), which is equal to 1−((f′(x))²−f(x)f″(x))/(f′(x))². When x is close to r, f(x) is close to 0, so g′(x) is close to 0 if f′(x) is not close to zero and f″(x) is not approaching ∞. In fixed-point iteration, the error in x₁ is equal or less than the error in x₀ multiplied by g′(x) for some x between x₀ and r. Since g′(x) is close to 0 when x is close to r (as long as f′(x) is not close to zero and f″(x) does not approach ∞), this implies that Newton's method converges extremely quickly.

If we define e₁ as the error in x₁ (=|r−x₁|) and e₀ as the error in x₀, then for fixed point iteration, e₁/e₀≈g′(r) when x₀ is near r. In the case of Newton's method, g′(r)=0, which implies that e₁/e₀≈0. This is great, but not very specific. To be a bit more specific about what ≈0 means, we can look at e₁/(e₀)². The Taylor series approximation of f(x) tells us that f(r)=f(x₀)+(r−x₀)f′(x₀)+(r−x₀)²f″(c)/2, where c is some point between x₀ and r. Since f(r)=0, we can rewrite the Taylor series approximation as x₀−f(x₀)/f′(x₀)−r=(r−x₀)²f″(c)/2f′(x₀). However, x₀−f(x₀)/f′(x₀)=x₁, |r−x₁|=e₁, and |r−x₀|=e₀. Therefore e₁=(e₀)²|f″(c)/2f′(x₀)|.

This means that for bisection and for fixed-point iteration in general, e₁=ke₀ for some k. As long as k is less than 1, this will converge to r steadily, but relatively slowly. (Convergence of this form is called linear convergence.) However, for fixed-point iteration when g′(r)=0, including Newton's method when f′(r)≠0 and f″(r)≠∞, e₁=k(e₀)² for some k. This means that when e is small, it gets much smaller with each iteration, and x converges towards r much faster than the other methods. (Convergence of this form is known as quadratic convergence.)

Newton's method requires knowing f′(x), which means that it requires more information about f(x) than bisection and fixed-point iteration, which only require f(x). Like fixed-point iteration, it only requires a single starting value for x. Also like fixed-point iteration, it can fail. While fixed-point iteration fails if |g′(r)|≥1, Newton's method fails if f′(r)=0 or f″(r)=∞. Like fixed-point iteration, it can be difficult to estimate the error. (The maximum error is dependent on the maximum f″(x) between r and x₀.) However, when it converges, Newton's method converges quadratically, meaning that many fewer iterations are required to reach the desired accuracy than the linear convergence of bisection or fixed-point iteration.

Additional thoughts on Taylor series approximations

2010-09-10T20:52:00.004-04:00

The Taylor series approximation of e^x converges rapidly for small positive values of x, but gives incorrect results for negative values, because computing the series requires subtracting large numbers from each other to get a small number, and on a computer this will introduce rounding errors.

The solution is to rewrite e^x as 1/e^−x when x<0. This can be computed without rounding errors reducing the accuracy of the final result. This solution is straightforward, but it was not obviously necessary in the original computation. The lesson, again, is to watch out for rounding errors.

The Taylor series approximation of e^x is accurate for larger positive values of x, but the number of terms that must be computed in order for the approximation to converge increases as x increases. This is a problem, because each term of the Taylor series has xⁿ in the numerator. If the number of required terms gets too large, xⁿ will become too large for the computer to represent and the computation will fail. My calculator gives e⁷⁹ as 2.038281066651×10³⁴. The Taylor series approximation returns infinity, because it requires more than 163 terms and 79¹⁶³ is greater than the computer can represent.

Once again, the solution is to rewrite the original function. x can be broken up into an integer part and a fractional part, and then e^x=e^{integer part}×e^{fractional part}. e^{fractional part} can be calculated as a Taylor series expansion. If e is stored as a constant, then e^{integer part} can be directly multiplied out, and this can be computed when a strict Taylor series approximation of e^x fails.

Taylor series can also be used to approximate sin(x) and cos(x). Like with e^x, the series require fewer terms to converge near 0, so the identities sin(x)=sin(x+2πn) and cos(x)=cos(x+2πn) for all positive and negative integer values of n should be used so the Taylor series is only computed for −π<x<π.

Finally, Taylor series can be used to approximate ln(x). The catch here is that e^x, sin(x), and cos(x) are all computed based the difference between x and 0, while ln(x) must be computed based on the difference between x and 1. This makes the math a little messier but isn't otherwise a big deal. However, the Taylor series for ln(x) does not converge for x>2. Since ln(x) = −ln(1/x), to compute ln(x) for x≥2, compute −ln(1/x) instead.

Numerical Analysis Lecture 1

2010-09-09T18:47:00.005-04:00

I have been considering resurrecting this blog to talk about my current interest, which is math. This fall I'm taking a course in Numerical Analysis, and I'm finding that in order to make sense of the class, I have to rethink through the lectures after each class. As long as I'm doing this thinking, I figured I should post it. Hopefully posting about math in a generic blogger blog won't turn into too much of a headache.

I'm going to go ahead and start with Lecture 1, which was last week's class. I likely don't know what I'm talking about. If anyone sees this and wants to leave a comment, please do.

The big picture on Numerical Analysis is that math gives us all kinds of tools for understanding the world. However, it's not always easy to go from a mathematical formula to a number, and often, we care about the number rather than the formula. Numerical Analysis is the techniques used to approximate the values of formulas and turn them into numbers.

Certain mathematical operations are easy. Addition, subtraction, multiplication, and division are all easy, and you can do a lot with them, such as evaluating polynomials. If f(x) is a polynomial, you can find the value of f(x) for any x just by a bunch of multiplications and additions. But there are lots of functions that are not polynomials, and these are hard to evaluate. For example, f(x)=e^x, f(x)=sin(x), and f(x)=∫_−∞^z (e^−t²/2)/(2π)^1/2 dt are common functions with many applications which cannot be directly evaluated at more than a few points.

Numerical analysis shows how to approximate the values of these functions. For example, we can look at how to approximate f(x)=e^x for values of x other than 0. Math tells us that lots of functions, including the exponential function, can be written as a Taylor series. The general form of a Taylor series is f(x) = f(a) + f′(a)(x−a)/1! + f″(a)(x−a)²/2! + f⁽³⁾(a)(x−a)³/3! + … . Written compactly, this is ∑_n=0→∞ f⁽ⁿ⁾(a)(x−a)ⁿ/n! .

The good news is that for f(x)=e^x, f′(x)=e^x=f(x), and therefore f⁽ⁿ⁾(x)=e^x for all integers n≥0. In combination with the fact that a⁰=1 for all a≠0, this gives the Taylor series for e^x=∑_n=0→∞ xⁿ/n! = 1 + x + x²/2 + … . This means that a Taylor series turns functions into polynomials, and since polynomials can be evaluated with addition, subtraction, multiplication, and division, we can evaluate functions other than polynomials, as long as they can be expressed as a Taylor series.

The bad news is that this is an infinite polynomial. We have turned e^x from a function that can't be evaluated at all into one that can be evaluated, but requires an infinite number of calculations. Fortunately, math tells us that if you take a finite number of terms of the Taylor series, that sum will approximate the true value of the function, and the approximation gets as close as you want as long as you take enough terms. Math tells us to figure out how many is "enough", so you can write a little computer program that does the calculations and spits out the value of e^x for any value of x. This value is an approximation, but it can be a good enough approximation for whatever purpose you're using it for.

Let's do a quick demonstration of this. The table below shows a sequence of approximations of f(1)=e¹=e. For each value of n, the table shows the sum of the first n terms of the Taylor series expansion of e^x, evaluated at x=1. For the record, my calculator says that e=2.71828182846.

n	sum
0	1.00000000000
1	2.00000000000
2	2.50000000000
3	2.66666666667
4	2.70833333333
5	2.71666666667
6	2.71805555556
7	2.71825396825
8	2.71827876984
9	2.71828152557
10	2.71828180115
11	2.71828182620
12	2.71828182829
13	2.71828182845
14	2.71828182846

The Taylor series reaches the same level of accuracy as my calculator in 14 steps, which is relatively quick. According to math, this calculation works for any value of x, although it may require more steps depending on the value of x to get the desired accuracy. In the real world, however, if x gets too large, then e^x can be bigger than the computer can represent. For example, if x=10²⁰, the resulting sequence of finite sums is:

n	sum
0	1
1	1e+20
2	5e+39
3	1.6666666667e+59
4	4.1666666667e+78
5	8.3333333333e+97
6	1.3888888889e+117
7	1.9841269841e+136
8	2.4801587302e+155
9	2.7557319224e+174
10	2.7557319224e+193
11	2.5052108385e+212
12	2.0876756988e+231
13	1.6059043837e+250
14	1.1470745598e+269
15	7.6471637318e+287
16	Inf

Actually, much smaller values of x will return a value of ∞ on most computers. e⁷¹⁰ is larger that the computer can represent, although it takes many more steps to get there.

There's a much bigger problem with using computers to calculate e^x using a Taylor series expansion. Using the Taylor series expansion, calculating e⁻²⁰ requires 87 steps, but the result eventually settles on 4.1736374994×10⁻⁹. This looks reasonable, but my calculator says the answer is 2.06115362244×10⁻⁹. We can see the problem by looking at the individual sums in the expansion.

n	sum
0	1
1	-19
2	181
3	-1152.3333333
4	5514.3333333
5	-21152.333333
6	67736.555556
7	-186231.69841
8	448688.93651
9	-962245.80776
10	1859623.6808
11	-3271048.1166
12	5280071.5457
13	-7875497.1655
14	10918172.422
15	-14140053.695
16	17182728.951
17	-19667603.573
18	21277210.343
19	-21822593.779
20	21277210.343
21	-19770222.154
22	17545625.57
23	-14902937.669
24	12137531.697
25	-9494843.7955
etc.

The problem is that the final result is a little tiny number, but early steps in the sum involve adding and subtracting really big numbers from each other. According to math, the Taylor series gives the correct result. On a computer, however, numbers are slightly rounded off. When a bunch of big numbers are added and subtracted from each other to get a small result, the rounding errors can make the final answer totally wrong.

This is kind of terrifying. You can write a program that math says will give the right answer, to the accuracy that you need it, but the computer program could introduce rounding errors that make the final answer totally wrong. By looking at the intermediate steps in this case, we could guess that there might be a problem with the final answer, but if you are handed a computer program that claims to evaluate e^x, or any other function, it's possible that the program will be mathematically correct but give incorrect answers for some inputs. These errors may only be detectable after the fact. The appropriate response is probably justifiable paranoia.

2005-07-25T21:25:00.000-04:00

The Electronic Frontier Foundation is having a Blog-a-thon to celebrate its 15th anniversary. They are asking for stories about "click moments," the moment of first taking action in support of freedom in the digital world. This is my contribution to the EFF15 Blog-a-thon.

Like many people, I'm sure, I've taken many small steps towards online activism. But my click moment is obvious. I started this blog in January 2003. The direct impetus for starting the blog was that I was quitting my job to follow my fantasies. My fantasies, which I had been kicking around for at least six months, were based on music on the Internet. Specifically, my contention was that it was impossible to stop music from being distributed on the Internet. Rather than trying to fight it, I thought musicians should try to take advantage of that fact.

But that wasn't my real click moment. That came a few months later. By that time, my blog had mostly stalled. Then I saw a flyer for a symposium on copyright being run by the Harvard Journal of Law & Technology. Initially, I thought it sounded kind of interesting, but I'm not sure I thought much more than that. But before too long, I decided I had to go.

The decision to go was really the critical moment for me. I went. I came home and blogged about it. My blog instantly turned into a copyright blog. I attended legislative hearings. I started meeting other bloggers. And I went to more conferences. My friends started threatening interventions to save me from the copyfight lifestyle.

Copyright issues in the digital age are overwhelming. There's so much going on all the time that I eventually fell behind. I tried to play catchup, but catching up was even harder than staying on top of things in the first place. I fell out of the blogging habit. Since then, my blog posting has been somewhat hit and miss. I still post about the most important developments, but I haven't been posting regularly, despite attempts to get restarted.

But the important thing is that I still believe. The EFF is one of many groups fighting the good fight. I'm still a participant in that fight. I have faith in freedom on the Internet. I believe that action makes a difference, and has made a difference. I'm not sure that this is a fight that will ever be truly won, but I am sure that we'll keep fighting, and I'm proud to stand with the EFF on its 15th anniversary.

2005-06-29T21:37:00.000-04:00

It's apparent from Uzbradistan's comment on my last post that my writing about Sony and MGM v. Grokster hasn't been that clear. So now that I've read the opinion on Grokster (which was the real point of my recent posts), I'll attempt to explain what the rulings in Sony and MGM v. Grokster mean.

I'll start with copyright infringement generally. Copyright law is a combination of the written law as passed by Congress and of various rules that have come about through court decisions. The written statute defines copyright infringement, but court decisions have created additional forms of infringement.

The copyright statute grants a set of exclusive rights to copyright holders, limited by various exceptions also defined by the statute. If you write a book and I start publishing copies of it without your permission, I am directly infringing on your copyright, as defined by the statute. If, on the other hand, I write a review of your book and quote your book in my review, I am not infringing on your copyright even if I don't have your permission. Again, that's defined by the statute.

Court rulings have created two forms of indirect infringement. These forms of infringement are not defined by the statute. They also do not require violating the exclusive rights of the copyright holder. Contributory infringement occurs when one person "materially contributes" to the direct infringement committed by another. An example of a contributory infringer is a concert promoter. If the performer infringes on copyright by performing songs without the permission of the composer, the promoter contributes to the infringement by creating the audience for the performer. Without the audience, there would be no performance and no infringement, but the promoter does not directly infringe on the copyright of the composer. However, the promoter is still liable for contributory infringement.

Vicarious infringement occurs when someone has the ability to control potentially infringing behavior and receives benefits from the infringement. For example, someone who sells infringing copies of a CD at a flea market directly infringes on copyright. The owners of the flea market vicariously infringe because they are paid by the seller and because they can refuse to allow the seller to operate at the flea market. Like contributory infringement, the flea market owner does not directly infringe on copyright by selling the infringing items. The owner benefits from the infringing action, but does not participate.

Contributory infringement and vicarious infringement are not defined by the copyright statute. They have been defined through court rulings, with the consequence that they are not as clearly and distinctly defined as they might be if defined by statute. Many lawsuits alleging indirect infringement argue both possibilities, leaving the court to determine which applies, if either.

That brings us to Sony v. Universal. Sony invented the Betamax VCR. Universal Studios sued Sony under a theory of indirect infringement. The VCR does not infringe on copyright itself, but purchasers of VCRs can use them to infringe. The Supreme Court ruled that Sony was not liable for copyright infringement. VCRs can be used for infringing purposes, such as copying a copyrighted movie. They can also be used for noninfringing purposes, such as recording a TV program to watch at another time. Since VCRs can be used for noninfringing purposes, Sony was not liable for indirect copyright infringement.

The key holding of Sony is that the creator of a new technology is not liable for copyright infringement if the technology is "capable of commercially significant noninfringing uses." However, the precise meaning of that phrase is unclear. It could potentially include anything from "nonfringing uses are theoretically possible" to "a majority of actual use is noninfringing." Recent court rulings have widely varied in their interpretation of that phrase.

MGM v. Grokster, the first of these cases to reach the Supreme Court, gave the Court an opportunity to define "commercially significant noninfringing use." Alternatively, the Court could have rejected the Sony standard and created an entirely new standard for contributory infringement. Despite its unanimous opinion, the Court failed to clarify this issue. The opinion of the Court did not address this question at all, instead defining a third form of indirect infringement. Justice Ginsburg and Justice Breyer each wrote a concurrence to the opinion. The concurrences battle on the definition of commercially significant noninfringing use.

Ginsburg's concurrence sets a high standard for substantial noninfringing use. While it's not clear how much noninfringing use is enough to avoid contributory infringement, the noninfringing use of Grokster's file sharing software clearly is not enough. She indicates that the proportion of actual noninfringing use to the total use determines whether it is commercially significant, but she does not state what proportion would be sufficient. She would hold Grokster liable for contributory infringement.

Justice Breyer argues the opposite result in his concurrence. He states that a mere theoretical possibility of noninfringing use is not sufficient to avoid liability. However, the evidence provided for noninfringing uses of Grokster's software demonstrates a capability for noninfringing use which is sufficient to avoid liability for contributory copyright infringement.

Three justices signed Ginsburg's concurrence and three signed Breyer's. Three signed neither. This gives a three to three ruling, which has basically no legal force and gives no indication of how the Court might rule in the future. The question of commercially significant noninfringing use remains unanswered. Despite this, the Court ruled unanimously against Grokster. The ruling avoids the question of noninfringing use and creates a new form of indirect liability instead.

The ruling states that Grokster may be held liable for inducing copyright infringement. Inducement can be explained through an analogy. Suppose I sell crowbars, advertising them as "Matt's Crowbars: Good for Construction." If someone buys one of my crowbars and uses it to break into a house, I can't really be held responsible for that. Suppose that instead I advertise them as "Matt's Crowbars: Good for Breaking and Entering." If someone uses one of my crowbars to break into a house, it makes sense to hold me indirectly responsible. By suggesting an illegal action and providing a tool to do it, I am causing, or inducing, that crime to occur.

The question from the Sony decision is how much noninfringing use qualifies as a commercially significant noninfringing use. That question is still unanswered. The question from MGM v. Grokster is what is sufficient to cause inducement. The Court found many inducing actions by Grokster and StreamCast, meaning that how much is enough is totally unclear. The inducement listed in the ruling includes:

Advertising to users of Napster
A name derived from Napster
Advertising showing infringing uses
Technical support to users who are committing infringement
A business model based on use of the software, given that most use is infringing use
Failure to attempt to block infringing use, in combination with the other evidence of inducement

Finding inducement is easy when there's lots of evidence for it. The question is when there's only limited evidence. For example, suppose a manufacturer advertises a use for a product which it believes is noninfringing. Apple advertisements for the iPod and iTunes use the phrase "Rip. Mix. Burn." Apple clearly believes that although it is encouraging copying, the copying is noninfringing. What if the advertised use is ruled to be an infringement? Suppose a court rules that copying songs to an iPod is an infringement. Would Apple be vulnerable to an inducement claim? At this stage, there are no definite answers. Further lawsuits may clarify things. But they may just create more of a muddle.

2005-06-28T20:56:00.000-04:00

Yesterday I posted my initial thoughts while reading the Supreme Court opinion of MGM v. Grokster. Today I will post any additional thoughts on the two concurrences on the ruling. Here's a quick review of the case and the Supreme Court decision, to be followed by my comments. Grokster (and StreamCast, its co-defendant) offer products which can be used to infringe on copyright. They were sued by MGM and other copyright holders for contributory and vicarious infringement on copyright. Grokster was held to not be infringing by the District Court and by the Appeals Court on the basis of Sony v. Universal, which held Sony not liable for copyright infringement for producing VCRs because the VCR is capable of substantial noninfringing uses. The Supreme Court overturned the Appeals Court ruling on the basis that Grokster's intent was to enable copyright infringement, making them liable for inducing copyright infringement.

Ginsburg's concurrence briefly acknowledges the discussion of inducing infringement in the opinion and then goes on to discuss contributory infringement. She focuses on "substantial noninfringing use" as applied in Sony. That ruling held that recording broadcast television programs to watch at another time constituted fair use, and therefore substantial noninfringing use.

The opinion did not address the question of contributory infringement, finding inducing infringement sufficient to create liability. While it raised the question of what type of behavior is sufficient to induce infringement, it avoided the question of what makes a noninfringing use substantial. Ginsburg would raise the bar for substantial noninfringing use. She dismisses the evidence presented in favor of noninfringing use in the case and indicates that the proportion of noninfringing files distributed over the network to the number of infringing files distributed should determine whether the noninfringing use is substantial.

She doesn't set any numbers, but it is clear that the noninfringing 10-25% of files available on Grokster is not sufficient. This is contrasted to Sony, in which 10% of taping was authorized by the broadcaster. The distinction here may be less clear than she intended, although it is true that Sony held that most of the taping of the other 90% of broadcasts constituted fair use, while she denies the possibility of any fair use in copying files over filesharing networks.

Breyer's concurrence is a response to Ginsburg. He states that because of the ruling on inducement, consideration of contributory infringement is not necessary. But since Ginsburg brought it up, he will discuss it as well.

I had the impression that Sony clearly stated that recording a program to watch at another time, even without the permission of the broadcaster, constituted fair use. Furthermore, recording programs to build a personal library was at least potentially fair use. Breyer asserts that Sony indicated that the first use might be fair use and the second is clearly not. As in Souter's opinion, Breyer's concurrence makes an interpretation of Sony that disagrees with my previous understanding of the ruling. Again, I can't judge at this time whether my previous understanding was inaccurate or whether the current ruling is seeking to change the meaning of the prior ruling without acknowledging that it is doing so. (Breyer later states that Sony did ultimately conclude that recording programs to watch at another time is in fact fair use. He initially minimizes that conclusion to strengthen the argument that a relatively low percentage of authorized use qualifies as a substantial use.)

Like Souter in the opinion of the Court and Ginsburg in her concurrence, he turns to patent law to resolve this question of copyright law. Ginsburg argued that capability for substantial noninfringing use was not sufficient. Substantial noninfringing use also had to occur in practice. Breyer quotes both previous rulings in patent cases and Sony to conclude that capability is sufficient.

Breyer then turns to the same statistics I questioned in Ginsburg's ruling. It's always exciting to find that a Supreme Court Justice shares the same thoughts you did on a case. Of course, he goes a step further, pointing both to the anecdotal evidence and the statistical evidence of authorized copying in Sony, rebutting Ginsburg's dismissal of the anecdotal evidence and her interpretation of the statistical evidence in the current case.

Breyer's discussion of contributory infringement is explicitly not directed toward the current case. He's protecting the freedom to create new technologies which may have both infringing and noninfringing uses. "But Sony's standard seeks to protect not the Groksters of the world..., but the development of technology more generally." I think that's awesome.

Dude. Breyer just gets better and better. He rhetorically asks if Sony should be modified or interpreted more strictly. "In particular: (1) Has Sony (as I interpret it) worked to protect new technology? (2) If so, would modification or strict interpretation significantly weaken that protection? (3) If so, would new or necessary copyright-related benefits outweigh any such weakening?" He answers, "Sony's rule is clear," "Sony's rule is strongly technology protecting," "Sony's rule is forward looking," and "Sony's rule is mindful of the limitations facing judges where matters of technology are concerned." That's just beautiful.

It's probably obvious that I think Breyer engaged in a glorious smackdown of Ginsburg. The fact that I'm inclined to agree with him probably has biased my judgment. Ultimately, my judgment isn't the important one. Three justices signed Ginsburg's concurrence and three signed Breyer's. That leaves three who signed neither.

Grokster and StreamCast have been nailed by inducement, but they are far from the only sources of file sharing technology. Some sources may have been more careful to not promote copyright infringement. Whether they can be found liable for contributory infringement depends on the three justices who signed neither concurrence. File sharing will return to the Supreme Court, and the next case will determine more about the relationship between technological advances and copyright than this one did.

2005-06-27T22:51:00.000-04:00

Today was a very big day for copyright law and the Internet. The Supreme Court, in a unanimous decision, ruled that manufacturers of technologies which can be used to infringe on copyright are potentially liable for any infringing use of that technology by end users. The ruling in MGM v. Grokster is available as a PDF (thanks to SCOTUSblog for the link). My notes on the decision follow.

The ruling states that "one who distributes a device with the object of promoting its use to infringe copyright, as shown by clear expression or other affirmative steps taken to foster infringement, is liable for the resulting acts of infringement by third parties." In retrospect, this is not a surprising outcome. It was clear from the oral arguments that the Court still basically supported Sony but that they were uncomfortable with letting Grokster escape liability. By introducing the intent of the distributor, they can find Grokster liable without creating liability for any device which can potentially be used to infringe copyright.

The ruling calls Napster "notorious." Is that fair? It strikes me as hyperbolic, but it may well be accurate. There's a bit of a guilt by association thing going on. Napster is obviously evil, in the Court's view, and anything tied to Napster is also evil. StreamCast worked to get users of Napster to adopt their software, and Grokster gets in trouble because its name is derived from Napster.

The profit motive, at least as expressed on page 8 of the ruling, strikes me as problematic. While the software is being used, advertisements are displayed to the user. Obviously the companies want to encourage users to use the software more, so as to increase the advertising revenue, but there seems a logical leap from that statement to the conclusion that therefore the principal intended use was to infringe on copyright. Sony's income was from the sale of the VCR, not its use, but it's undoubtedly true that some VCR sales occurred for the purpose of infringing on copyright. Selling a device with the knowledge that it may be used for infringing purposes doesn't strike me as inherently different than advertising while software is being used, possibly for infringing purposes.

The business model itself isn't the problem. Encouraging users to engage in infringing uses, where those uses support the business model, may be a problem, but the ruling already established that the companies were encouraging infringing uses. I'm not sure how an advertisement based business model in itself suggests that the companies were dependent on infringement to succeed.

The ruling implies that the proportion of infringing use to noninfringing use is relevant to the question of liability. Is this expected use or actual use? Suppose I develop a technology with the expectation that it will be largely used for copyright infringement, but in fact the primary use turns out to be noninfringing. Am I liable for any infringing use which does occur? Either outcome feels problematic.

The ruling argues for indirect liability, stating that "[w]hen a widely shared service or product is used to commit infringement, it may be impossible to enforce rights in the protected work effectively against all direct infringers, the only practical alternative being to go against the distributor of the copying device for secondary liability on a theory of contributory or vicarious infringement." I agree that it's the American way to sue someone when you have been wronged. I also agree that an easy target for the lawsuit will always be preferred. I'm not sure I agree that in the absence of a desireable target, it's fair to create one by expanding the scope of liability.

One of the significant questions about this case was whether it would modify the Sony Betamax decision. Sony held that the manufacturer is not liable for a product which can be used to infringe copyright if the product is "capable of commercially significant noninfringing uses." MGM argued that noninfringing uses of Grokster were not commercially significant. This ruling recasts the Sony ruling as dependent on "presuming or imputing intent to cause infringement solely from the design or distribution of a product capable of substantial lawful use." It's not the noninfringing uses themselves, it's whether it was intended to be used for infringing purposes. The court sidesteps MGM's argument.

At the same time, I can't judge the validity of the Court's conclusion without reviewing the Sony decision. Sony is typically presented as hinging on capability of noninfringing uses. It's possible that this is a misrepresentation of the original decision, and the decision in fact is based on the intent for noninfringing use. It's also possible that the current ruling is changing the meaning of Sony while simultaneously denying that it is doing so. This merits further research and discussion.

I tend to resist lumping together copyright, patent, and other rights in information under the umbrella of "intellectual property." While copyright and patents both give rights to the use of information, the laws governing them are completely independent and referring to them as a single unit blurs their distinctions. The Supreme Court feels no such need to keep them separate. In supporting the claim that encouraging copyright infringement creates liability for copyright infringement, the ruling cites a list of cases of patent infringement. In this instance, at least, the Court does not distinguish between copyright and patents. But then the ruling states that the Sony ruling was also derived from patent law, so perhaps this is nothing new.

I generally trust the Supreme Court to have a sufficient understanding of new technology to be able to rule reasonably on the technology. This trust is reinforced by the proper use of the terminology and jargon of the technology. So the statement that "StreamCast beamed onto the computer screens of users . . . ads . . ." (page 20) makes me nervous. What exactly does "beamed" mean? It's not as bad as an egregious misuse of technical terminology, but the failure to use more specific language reduces my confidence that the Court knows what it's talking about. This is reinforced by the ruling's discussion of OpenNap, which strikes me as fuzzy. On the other hand, I don't know enough about OpenNap to evaluate the ruling's use of the term.

This concludes my notes on the first reading of the opinion in this case. There are two concurrences, but it is late and they will have to wait until tomorrow.

2005-05-25T19:40:00.000-04:00

I just have a quick observation following the record breaking opening of Revenge of the Sith. The original maximum term of copyright under United States law was 28 years. Coincidentally, Star Wars was originally released 28 years ago. The significance of this should be obvious to anyone who thinks the Special Edition versions of the movies were not improvements on the original theatrical versions of the movies.

George Lucas considers the Special Edition versions of the original Star Wars trilogy to be the official versions of the movies. Consequently, the original versions of the movies have never been released on DVD. Lucas also will not allow theatrical reshowings of the original movies. Since the duration of copyright has been extended to 95 years, it's safe to assume that almost no one old enough to have seen Star Wars in its original release will live long enough to have the opportunity to see the original movie in a movie theater or to view the movie on a legal, commercially released DVD.

Under the original terms of copyright, Star Wars would enter the public domain next year, giving fans the ability to watch the version they want. Under current law, relying on illegal bootlegs or accepting the changes Lucas has made are the only alternatives.

I'm mentioning this less as a call to action and more as an observation. Copyright law impacts fans as well as creators. This week, it's Star Wars fans' turn to be caught by copyright law.

2005-01-30T17:05:00.000-05:00

I've recently been playing a video game that's shockingly amoral. The striking thing is that it's not a game that's likely to get the attention of the Parents Television Council. Katamari Damacy is exceedingly weird, but it isn't violent. However, it strikes me as amoral in a way that no other videogame I've played has.

My goal here isn't to judge either the game or people who play it. I find Katamari Damacy addictive and can't blame others for enjoying it. My goal is just to explore my reaction to the game. I'm not certain I can adequately express my thoughts on the game, but I'm going to try.

The premise of Katamari Damacy is absurd. The King of All Cosmos has accidentally destroyed all the stars in the sky, and it is up to you to replace them. You do this by rolling a ball, called a katamari, around various environments. The ball sticks to everything it touches. As it accumulates stuff, the ball becomes larger and becomes able to pick up larger objects. Most levels have the objective of growing the katamari to a certain size within a time limit. If you succeed, the katamari becomes a star and you proceed on to the next level.

The game has an amazing range of scale. At the beginning of the game, the katamari is very small and you roll it around inside a house, picking up objects like stamps and thumbtacks. By the end of the game, the katamari becomes large enough to pick up buildings and even entire islands. Along the way, you pass through a range of intermediate sizes, including levels in which the goal is to pick up individual people.

I don't have a good answer to the question of why I care about the objects I pick up in the game, but I do. Part of the answer is that the game has realistic environments. On levels that are set indoors, I feel like I am rolling around in someone's house. As I pick up objects in the house, I identify it as someone else's stuff which I'm just taking.

But realism isn't the only reason why I respond the way I do. Burnout 3: Takedown, another game I've been playing recently, is the type of game that parents groups like to complain about. Burnout 3 is a car racing game in which one of the explicit goals is to crash and destroy your competitors' cars.

In spite of that, it doesn't make sense to me to speak of Burnout having any morality, good or bad. The scope of the game doesn't allow for morality. While Burnout has a very detailed environment, the only reason you exist in the game is to get to the finish line and to destroy other cars. Likewise, your competitors have no function other than to try to defeat you. Even innocent traffic only exists to be an obstacle, and one to be used to your advantage if you can.

Speaking of the game in moral terms implies a depth to the game that doesn't exist. Of course driving in real life as if you were playing Burnout 3 would be immoral, but Burnout 3 is just a game. Playing the game is neither moral nor immoral. And destroying a car in the game isn't a moral act. It's just playing the game.

The same logic should apply to Katamari Damacy. But that's not how I respond emotionally to the game. I respond to the objects in the game as real objects belonging to characters in the game. Even more, I respond to the characters as people. They flee from the katamari and scream when they are caught, and I can't blame them. Nonetheless, I gather them up, placing my goal of creating a star ahead of the the property rights or personal freedom of others.

My concern about the morality of Katamari Damacy hasn't stopped me from playing the game. But I do think about it. As videogames have developed, their detail, realism, and emotional depth have increased. Videogames can be more complex than they may appear to casual observers. And that complexity is worth studying. I'm not sure what my response to Katamari Damacy means, but I'd like to find out.

2005-01-24T21:32:00.000-05:00

Last month I posted about Jason Kottke and Jeopardy! The central idea of that post is that individuals who post material on the Internet are at risk of legal action from corporations under copyright law. However, the power of legal action isn't limited to corporations. The DMCA amplifies the power of individuals to take legal action against other individuals for what they post online, as demonstrated by a recent conflict between two LiveJournal users.

I am including links to the individuals and posts involved to demonstrate that this is a concrete example but not to get involved in this particular case. Things have recently settled down after being extremely heated and I do not want to cause things to flare up again. If you are tempted to post comments on the journals of the people involved, please reconsider.

The DMCA (Digital Millenium Copyright Act) is at the heart of this conflict. The DMCA was passed in 1998 and made several changes to copyright law. The relevant portion here is Section 512(c), the "takedown" section.

Section 512(c) makes Internet Service Providers responsible for the actions of their users. If a copyright holder reports to an ISP that one of its users has posted material which infringes on that copyright, the ISP must "act expeditiously" to take down the material. The ISP must also "take reasonable steps" to notify the user. The user then has the right to counter notify the ISP and state that the material is in fact not infringing. The ISP must then report the counter notification to the complainer and restore the material in "not less than 10, nor more than 14, business days" unless the complainer states that it is taking legal action against the user.

LiveJournal is a community oriented blogging tool (and a service provider under the DMCA). One of the features of LiveJournal is that posts can be either public or "locked," so that only LiveJournal users specified by the poster can read the post. Access is controlled on a post by post basis and can be changed after the fact.

On December 2nd, the LiveJournal user Ginmar posted publically about an article in The New York Times. Her original post provoked a strong response, both in comments on her post and in new posts by various other users. Ginmar responded in other posts in her journal as well and then locked the original post.

Another LiveJournal user, Cheshyre, responded by reposting Ginmar's original post, along with her commentary, in her own journal on December 4th.

On December 30th, Ginmar complained to LJ Abuse, LiveJournal's group for handling complaints about users, about Cheshyre reprinting her post. Ginmar may not have thought she was taking legal action when she complained to LiveJournal. In a comment thread in another user's journal, she asserted that Cheshyre was the first to use the DMCA. Nonetheless, LiveJournal responded by informing Cheshyre that they had received a DMCA takedown notification from Ginmar, and that if she did not remove the post by January 3rd, her account would be deactivated.

On January 2nd, Cheshyre edited the post to remove Ginmar's post and declared her intention to file a DMCA counter notification. According to LiveJournal's page on the DMCA, if Cheshyre did file a counter notification, they would inform Ginmar. If Ginmar failed to state that she is taking further legal action, the content would be replaced "after 14 business days."

As of January 24th, that's where things stand. Cheshyre's post is still edited. There has been limited public discussion of further action by either Ginmar or Cheshyre. Ginmar has locked some further public posts that she had made, making the record frustratingly incomplete. It has been about 14 business days since Cheshyre stated she would counter notify. Whether her post will be restored remains to be seen.

I think both Ginmar and Cheshyre would agree that things have gotten out of hand, although neither seems willing to back down. What I am struck by is the role of the DMCA and LiveJournal's implementation of the DMCA in the dispute.

The DMCA has gotten a large corporation involved in what is fundamentally a conflict between two individuals. Furthermore, it has pushed both of them into a legal escalation to resolve the situation. It is not clear that Ginmar thought of her original complaint to LJ Abuse as a legal action, but because of the DMCA, her next step in this conflict would be to contact a lawyer.

Hopefully, you will read this with alarm, for two reasons. Fundamentally, it's stupid for this to have become a legal dispute, but there they are. But also, like Jason Kottke and Jeopardy!, this could happen to you. Blowing up personal conflicts into legal actions is inherent in the DMCA. Copyright law isn't just some abstract thing that affects book publishers. If you post any material on the Internet, it has the potential to affect you, and you don't want to be on the receiving end.

2004-12-11T20:44:00.000-05:00

I realize this is a month late, but I want to talk about the election. I haven't really discussed my political views on this blog, but most readers probably already know that I am a liberal Democrat. The outcome of the election was dismaying and I've been considering how to respond.

There are continuing accusations of failures of the electoral process on Election Day. I think it is important to follow them even though I don't have any expectation that they will change the outcome of the election. Unlike Florida in 2000, Ohio was not stolen this year. I think pursuing a recount in Ohio in the hope that the outcome will change is a lost cause.

However, I think disenfranchisement was a real problem, both in Ohio and around the nation. Stories of excessively long waits to vote and attempts to prevent people from voting, and doubts about whether votes were correctly recorded, are far too common. These are fixable problems, and it's important that they are pursued and corrected and that the voting process is watched so they do not continue to occur.

Voting is a right. Like all rights, the theoretical right isn't the important part. The important part is the practical exercise of the right. Ineligibility to vote, long waits to vote, and lack of guarantee that the vote was correctly recorded and counted all deny the right to vote. Loss of the right to vote destroys the legitimacy of the government, and maintaining faith in the legitimacy of the government is in everyone's interest, regardless of political interests.

But that's the future. Today, I'm more interested in the present. And the fact of the present is that George W. Bush has been elected President. From a liberal perspective, this is a crisis. The next four years will bring decreased protection for the environment, civil rights, reproductive rights, and workers' rights, just to name a small handful off the top of my head. In the near future, the Federal government will not serve the interests of individuals.

This is a frightening prospect, and one I feel compelled to respond to. If the government will not serve the roles which it should fulfill, organizations outside the government must do what they can to fill in. And we, as individuals, must ensure that they succeed.

I feel strongly that I should do something, anything, to try to combat the harm the government will do, through neglect or malice, during the next Presidential term. In some way, this feels like an opportunity to me, even if it is an opportunity I would rather have not had. After my unfocused efforts to choose a career direction for myself, this comes as a call to action. There is a need to step up where the government will not. My lack of other commitments makes me the right person to do it.

Unfortunately, I am facing an excess of choices. I wish it was an exaggeration to say that I oppose everything George W. Bush is doing as President, but it is at most a small one. Every day, it seems, there is something new to respond to. Many groups come to mind as worth working for, in any capacity. Anything I could do, with any of them, would be better than doing nothing. But that leaves so many choices that the apparent call to action is just as unfocused as I was before.

I discussed the way I feel about needing to choose something to work on and to work on it with my cousin over Thanksgiving. She pointed out that it's much more likely that I will sit around and moan about how I can't decide rather than actually taking action. I'd complain about it being unfair if it wasn't accurate. I was surprised by her ability to express the things I struggle with better than I can, but perhaps I shouldn't have been. The inability to choose a course of action and take it runs in the family, and she was speaking from experience.

So here I am then, with a desperate need to take action, but no conception of how to do that. I don't want to end up spending all my time talking about how I should be doing something while I sit around not actually doing anything. I fear that is the most likely outcome, but I intend to fight it. I'm just not sure how.

2004-12-03T19:40:00.000-05:00

It can be easy to think of copyright issues as abstract problems which don't really directly affect you. However, if you have a website, there's always the possibility that you will find yourself on the wrong side of copyright law. Jason Kottke's recent experiences demonstrate the risk you face posting online and how copyright law is fundamentally unbalanced.

On the Tuesday, November 30th episode of Jeopardy!, Ken Jennings failed to win after winning the previous 74 episodes. This was a truly exceptional streak and he acquired many followers along the way. Episodes of Jeopardy! are taped well in advance of when they air, and information about the loss leaked well in advance of the broadcast date. Jason Kottke announced the date of Jennings' defeat on his blog on November 26th.

On November 28th, Jason posted the audio of the end of the episode and a transcript. Shortly thereafter, the copyright hammer came down. A lawyer for Sony, the corporation which produces Jeopardy!, contacted Jason and demanded that he remove the audio and transcript. He promptly did, but not before the post had gotten the attention of The Washington Post. Their article, which includes a transcript that appears to be taken from Jason's blog, was published the morning of November 30th, before the episode was broadcast.

Jason has not stated exactly what Sony's lawyers have said to him, but it's not hard to guess. Sony is likely claiming that Jeopardy! is a copyrighted broadcast and by publishing the audio clip and transcript, Jason was violating their copyright. They probably threatened to sue for damages if Jason did not remove the clip and transcript. In the face of Sony's lawyers, Jason has given up, although he states that he is still resolving the legal issues. More dishearteningly, he appears to be considering giving up his blog entirely. As he says, "As an individual weblogger with relatively limited financial and legal resources, I worry about whether I can continue to post things (legal or not) that may upset large companies and result in lawsuits that they can afford and I cannot."

The issue here isn't whether Sony is legally correct in forcing Jason to remove the materials from his site. Sony undoubtedly holds the copyright for the show, but from a non-lawyer's perspective, Jason appears to have a slam dunk fair use defense. The issue is that Jason can't afford to challenge Sony. Fair use or not, the cost of fighting a lawsuit from Sony is beyond what Jason or any individual posting things on the web can afford to pay.

It may be easy to dismiss this. After all, he published an actual recording of the show. And besides, it's happening to someone else. But that would be missing the point. The point is that the threat alone of legal action is so powerful that it can cause people to consider stop posting on the Internet at all. I guess some people might see this as opportunity. But most people should see it as a cause for alarm. Do you ever post things on the Internet? Do you ever include song lyrics? Or photographs? It's not just paranoid to worry that the copyright hammer could fall on you next.

Bonus irony: the article on The Washington Post website includes a "Permission to Republish" link. For kicks, I filled out the form, stating I wanted to republish the article on a non-profit Internet site for one year. They want $400. If I took them up on it, would Sony come after me for posting the transcript? If I don't pay for it and post it, will both companies come after me? If I post the transcript from the Post but not the rest of the article, will the Post come after me, or will I only have to worry about Sony? I'm curious about the answers, but I'm not posting anything. I'm too scared to find out.

2004-12-01T22:01:00.000-05:00

It's been a while since I've posted anything here. (You may have noticed.) As I get back into the swing of things, I'd like to go over events of the past several months and tell you where I plan to take things from here.

When I left off six months ago, I was thinking about what kind of career I would like to have. At the time, I was unemployed and considering various options which I mostly only touched on briefly here. Based on my writing here, it's not a surprise that I'm interested in law. Some other ideas I've been considering are journalism, teaching, and music conducting.

Meanwhile, I have been working through a temp agency to cover my bills. As luck would have it, my most recent assignment was as an administrative assistant to a professor at a law school. My conclusion after some close up observation is that law school probably isn't for me. However, I was working for a criminal law professor, and I knew going in that my interests are elsewhere. I could have told you that I don't want to be anywhere near a courtroom before I worked on mock trials.

I haven't completely rejected the idea of law school, but at this point I'm a lot more skeptical. Much as I have an interest in some areas of law, I don't think law school's the way for me.

Temp work is hardly the only thing that's occupied my time recently. I just finished a vocal directing gig during the recent production of Iolanthe with The MIT Gilbert & Sullivan Players. Working with my co-vocal director Jess and the other members of the cast and crew was a great pleasure and I'm very proud of the end result. We had two weekends of excellent performances, and it would have been smart of me to write it up before it happened rather than after to encourage my readers to come see the show.

That points out one of the downsides of my vocal director role. I had very little free time outside of the show and blogging was a logistical impossibility. There's another downside which I think is more significant. I can have a tendency to be something of a loner, and conducting greatly exacerbated that. I spent a few hours a day in front of the cast five days a week. Whenever I wasn't actually conducting, I found myself wanting to avoid other people entirely.

I generally like to be more social than that and I'm not happy about this apparent side effect of conducting. As a result, I don't plan to be a conductor again soon. I think it's an unfortunate result, since I did enjoy the experience itself and I was extremely pleased with the performances the cast delivered. This experience also does not bode well for the possibility of making a career of conducting or of teaching which has the similar position of being directly in front of and controlling a large group of people.

Like with my experience in the law office, I may be overgeneralizing and the people avoiding response might not normal for me in that position. Nonetheless, I'm still concerned.

What comes next? Generally, I don't have any real answers. I intend to continue doing temp work while looking into more long term career plans. And I intend to start blogging regularly again. Blogging at the rate I did last year isn't sustainable. I had trouble sustaining it while I was unemployed, and that's not really an option I can afford at this point.

My current plan is to experiment with an op-ed column format. I plan to write posts of about 750 words twice a week. The goal is to write on a regular basis, but to not set my target at an unattainable level. If this format doesn't work, I will make changes until I find something that does.

This post is my first post in my new format. I will continue to write about copyright issues and will also cover other thoughts about my career options and other topics as they come up. Hopefully this post will be the first of many.

2004-05-19T16:48:00.000-04:00

I'd like to follow up on last week's hearing on the DMCRA. Here are some links to other comments on the hearing.

Derek Slater is doubtful about how the hearing went, but he's heartened by Fred von Lohmann's comments. Derek also wonders why no one tried to argue that the DMCA doesn't really reduce infringement and why their side quoted Ed Felten, but our side didn't.

Meanwhile, Ed Felten points out that Valenti distorted the meaning of his quote. Felten was responding to a question about the possibility of allowing fair use rights within a DRM scheme, and Valenti turned it around to make it appear that Felten was saying that fair use rights would destroy DRM. In other words, Felten was saying that we don't know if it's possible to have a technological solution that grants fair use rights without destroying DRM, but Valenti made it appear that he was saying that it's definitely not possible.

Seth Finkelstein has his notes about the hearing at Infothought. He refers to the discussion of the DMCA rulemaking process, saying "I definitely thought I had something to say!" I thought his point of view was well represented by various panelists, including Miriam Nisbet of the American Library Association and Gigi Sohn of Public Knowledge.

Frank Field was taking notes at the ILaw conference last week, which discussed law and the Internet. Lawrence Lessig, speaking at ILaw after the DMCRA hearing, expressed frustration about the fact that the scope of fair use was a major topic of discussion during the hearing. His comments on the DMCRA appear in Frank's notes on FurdLog about Lessig's talk at ILaw.

Frank also had some comments on the written testimony from the hearing. Like Derek, he's interested in Valenti's use of the Felten quote.

FurdLog also has some links to comments by the EFF. Frank also includes a letter from his Representative in response to a comment Frank made about the bill. It isn't clear that Rep. Capuano is on board, but it does demonstrate the importance of contacting Congress. If you make enough noise about this bill, we can get it passed.

Bag and Baggage includes links and quotes from the Professional Photographers of America and the Cato Institute. Both groups oppose the bill. The PPA is calling for its members to contact Congress to oppose the bill. There's a real fight ahead and we need to take action if we're going to win.

Brian Flemming describes his experience contacting his Representative. (from Copyfight)

JD's New Media Musings has the American Library Association's comments on the hearing. Miriam Nisbet, Legislative Council for the ALA, was one of the witnesses at the hearing. JD also points to coverage of the hearing at PC World and Fox News.

TeleRead points to a CNET article about the hearing and takes umbrage at remarks by Jack Valenti and Rep. Butch Otter. Otter, responding to a comment by Lawrence Lessig about how copyright infringement isn't theft, stated, "Theft is theft and property is property." Lessig even acknowledged at the hearing that he was being overly lawyerly in making the distinction. Of course, Lessig is correct. Copyright infringement, though a crime, is not theft.

At Copyfight, Donna points out that one of the benefits of the DMCRA would be to legalize research on electronic voting machines.

After thinking about the hearing some more and reviewing my notes, I have a few more comments. First, it's really hard to take good, comprehensive notes at a talk like this. The best I can do is paraphrase and hope I'm not misrepresenting anything too badly or leaving too many important things out. My notes suffice as rough notes for now, but I'm looking forward to the official transcript.

My comments about the argumentative nature of the hearing look a bit overstated in retrospect. Now that I've been repeatedly quoted on other blogs because of my comments on the questioning resembling a trial, I'd like to step back from that a bit. It is certainly true that some of the questioning was like that. However, it's also true that some Representatives were truly interested in learning about the subject based on their questions. It wasn't as one-sided as I originally made it out to be.

A recurring line of questioning was on the technological possibility of providing for fair use rights while still having DRM protection to prevent commercial copying. Answers on this varied all over the place, from Valenti quoting Felten to argue that the answer is no to Robert Moore's emphatic yes. Given how often that question was re-asked, it felt like the Representatives were determined to keep asking until they got the answer they wanted. I hope they recognize that there was far from a consensus among the witnesses.

It's also important to note that the question was frequently asked as, "Is there a technological way to allow a single copy of a protected work to be made?" as if granting the right to a single copy would substitute for the full range of fair use rights. As Lessig said, the issue isn't what fair use rights should be. Fair use rights are defined by copyright law. The issue is whether it should be legal to use technology to take away fair use rights. My opinion is that it should not be legal. I hope Congress comes to the same judgement.

2004-05-12T17:39:00.000-04:00

Today's subcommittee hearing on the DMCRA has concluded. I was taking notes during the hearing which I am including below. As usual, I'm making no promises as to the comprehensibility of my notes. Also, while I tried to correctly identify the speakers, I may not have always gotten them correct, particularly with the names of the Representatives. I have a few comments, mostly on procedural issues rather than substantive issues.

By now, opinions on copyright law are pretty well set for most of the speakers. It's not a surprise that Lawrence Lessig and Gigi Sohn favor the law while Jack Valenti and Cary Sherman oppose it. Most of the arguments presented today on both sides of the debate are not new. It begins to feel like both sides are set in their arguments and are determined to win by shouting out their position until the other side gives in. I found the most interesting testimony to be by Allan Swift and Robert Moore.

Allan Smith is a former Representative who was a member of the Subcommittee on Commerce, Trade, and Consumer Protection. He spoke in favor of the DMCRA as an individual who has been copying music for over 50 years. He makes mix CDs and gives them as gifts. He thinks there is a creative component to creating a mix CD that should be recognized. Some of the committee members seemed dismayed by his opinions and some strongly disagreed. I hope they will consider what he had to say.

Robert Moore is the CEO of 321 Studios. 321 Studios produced software to allow DVD owners to make backup copies of the DVDs. He is currently in a legal battle to be able to continue to produce the software, which the movie studios want banned under the DMCA. I actually found the most interesting part of his comments to be his responses to questions posed by one of the committee members on the use of encryption on his own software.

321 Studios does use encryption on their software, but the stated purpose is not to restrict copying of the software itself. Moore states that it is so they can track the users of the software so that if a user did try to use the software for commercial copyright infringement on the DVDs, they would be able to identify the user. Various parties claimed to be on both sides of the copyright issue during the testimony, but Robert Moore seemed unique in the way that he both relied on copyright but also sold a product with the purpose of defeating copy protection.

One thing I was surprised by was the tone of the questioning. I expected more information gathering, but the hearing sometimes had the feel of a court trial, with the Representative either asking a series of leading questions to build a case or essentially cross examining the witness in an attempt to discredit their arguments. I found myself wondering how useful that actually is. It's clear that Rep. Boucher supports the bill while Rep. Otter opposes it. How useful is it for them to ask questions when they're deliberately asking questions that they already know the answers to?

True, there were occasional surprises in the answers they got, but I would have liked to see more questions from more undecided Representatives. Perhaps this just shows that I'm inexperienced with hearings and they always go like this. Regardless, it wasn't quite what I expected.

That's all the comments I have at the moment. Here are my notes. A transcript will eventually appear at the Committee on Energy and Commerce website. It also has written statements from all of the witnesses at the hearing.

Rep. Stearns: discussing fair use. balance between public interest in free speech and copyright holders rights in their works.

daily uses of computers challenge this balance

DMCA created civil and criminal penalties to individuals who circumvent DRM

DMCRA allows consumers to unlock DRM to make fair use

supporters argue that DMCA allows access to fair use

opponents feel copyright will be weakened

Rep Schakowsky: need to update laws because of changing technology

DMCA - important attempt to comfront new technology. DMCA swept away fair use and is abused by garage door openers and printer toner cartridge manufacturers.

DMCRA is a step we need to take

Rep. Barton - proud to be a cosponsor of the legislation

objects to spyware - trespassing

CD or DVD is mine once I leave the store. Restricted to uses that are not commercial. Illegal to buy a CD and make copies to sell them. Fair use permits personal copying. DMCA sought to provide content providers with protection.

intent of DMCRA is to permit consumers to make fair use

Rep. McCarthy

Rep. Issa - deeply interested in issue. Wants to restore fair use while protecting that which is not currently protected - reference to HDTV

unusual to have such an obvious problem.

321 Studios - circumvention prevents fair use

must be lose-lose in order to have win-win. need to craft an in-between

Rep. Green - don't want to throw out baby with bathwater

Rep Towns - this is a jobs issue rather than consumer issue

Rep Boucher - thanks for having this hearing

balance between rights of users and rights of creators

1990s - entertainment industry made appealling claim - digital is different - arrival of digital media, perfect copying - arrival of Internet as communications medium

voted for DMCA but expressed concern about overreaching nature of law and offered amendments. Now have six years since law passed. Greater tech industry now interested in this law. Support DMCRA. Large public interest involvement and supporting DMCRA.

four principle problems

1. federal offense to bypass DRM even if the purpose is innocent. Propose that bypass is legit if use is legit. Bypass for infringement still guilty of two violations.

Devices to circumvent for legit purposes be authorized. Refers to Betamax "capable of substantial noninfringing use"

notice that copy protected CDs are copy protected

exemption for research extended to research on DRM

Rep Doolittle - pulls out iPod. Take your entire CD collection or books and download them onto iPod. Manufacturers may prevent users from downloading to iPod. In 1998 didn't grasp real issues of DMCA. Now thinks they went way overboard. Fair use severely disadvantaged by DMCA. Speaking of HDTV, PVR. Use of PVR with HDTV may be limited.

Tension between copyright and technology. Never get rid of it. Line currently that favors copyright holders. DMCA would have prevented the VCR.

Lawrence Lessig - critical of copyright system but believe in role of copyright. Copyright law is broken. Hinders more than helps. Tool that dominant industry uses to protect against competition. Massively overextended.

Small changes can restore balance. DMCRA is important first step.

Speaking of circumvention. Valenti argued that fair use is not protected by law. Supreme Court has ruled that removing fair use creates a First Amendment problem

Valenti refers to his own terrorist war and some suggest that rights should be set aside during war. But free speech trumps.

Gary Shapiro - urges balance to copyright laws. In tech industry, IP is our lifeline. support fair and balanced IP laws. Radical departure from balance. Refers to Betamax. Betamax protection should be strengthened.

Jack Valenti - DMCA does not exile fair use. Letter from DVD-CCA. opposing DMCRA.

what is public interest in this bill?

unfixable flaw of 107 is that it legalizes hacking.

when encryption is broken, copying cannot be restricted. DVD purchased in Chinatown has 321 Studios statement.

Reference to Ed Felten

IP is greatest US export. creating new jobs.

Would weaken international law

Robert Holleyman - software industry also has surplus of trade. Believe that DMCA established that balanced. labeling provision - software industry voluntarily labels and thinks that's the right approach.

research exception dangerously overbroad. advancement of encryption is lifeblood of industry. DMCA has not stood in way of advancement of encryption technology.

Must not shield hacking. Congress rejected this change in 1998.

Product activation would be directly stymied by DMCRA.

Allan Swift - formerly a DJ. testifying as a private citizen. bought reel-to-reel machine in high school. Been a home recordist for 54 years. Given many friends collections of music. Never directly copied a record and never charged a penny. Respect copyright laws. Some peoople carry good idea to bad extreme. Hollywood would have smothered VCRs and DVDs.

Bought CD duplicator. Doesn't allow copying anything other than originals. Took money from my pocket.

Presumption of innocence taken away by DMCA.

Taking a hammer to consumer will not resolve this matter.

Own 3000 CDs. Profit center for business. Should be treated with respect.

Miriam Nisbet - Libraries are large consumers. Libraries provide access and preserve knowledge. Some uses do not require a permission slip. Uses include fair use, first sale, special library exemptions, the Teach act.

HR107 would permit preservation or archiving. Must be able to do this as formats become obsolete. Foreign language teachers could play DVDs purchased abroad.

Gives taxpayers full value of what they purchase.

Rulemaking under 1201d has not worked.

fear that metered use will be imposed on digital materials.

Stearns: Does a consumer have a right to make a single copy of a DVD or CD?
Lessig: Absolutely yes.
Valenti: No. Can't distinguish between one copy and many.

Stearns: Runaway Jury copy that Valenti had
Lessig: Copying beyond fair use is wrong and should be prosecuted. Giving absolute control to copyright holder is not in our interest.

Stearns: Supreme Court ruled that fair use cannot be denied. Why would you deny that right?
Valenti: Fair use is alive and well. Plead ignorant on ruling.

Stearns: National security. Export surplus.
Swift: This is not the only way to deal with problem. Takes problem out of content holders and puts it on consumers. Not the only way to respond.

Schakowsky: Public interest. How do you justify using such a large brush?
Valenti: DMCA written with involvement
Valenti: wanted a copyright royalty fee on videotapes. Libraries have CDs, DVDs, that you can take home.

Nisbet: Fear restrictions on interlibrary loan.

Schakowsky: Is there no way to allow legal purposes while preventing illegal purposes?
Shapiro: Technology can do anything. But law of unintended consequences. Cars are not restricted in speed. Conduct should be focus, not regulation.

Ferguson: Not a lawyer. You believe that someone who buys a DVD should be able to make a copy of that. How is this different from buying a bag of apples, and lose them, I can't go back a get another one for free. Don't get second car for personal use in another location.
Lessig: Very different principles apply to creative property than physical property. Different balance for creative property than for physical property. Fair use is a balance built into the law. Technology restricts fair use. Technology doesn't map onto complexity of law.
Swift has given us the map. Ordinary consumers obey the law.
Valenti: Legalize hacking for everybody. Machine can't distinguish between honest users and felons. How do you know how it will be used.

McCarthy: How did we lose respect for creative property? Teenage nephews believe creative works are not owned by the creator. Speaks of cheating, copying on a test. While can't you buy a second copy?
Swift: Granddaughter had a debate with him. Maybe we need to make a better distinction between creative property and physical property.
Lessig: is economic issue, but affects multiple sectors of economy. Xerox machine - technology is not banned. Ban legal use would be bad.

Issa: royalty on recorded material. Would you support royalty now?
Valenti: Trying to find secure environment. If you could find a secure environment I would support personal copies. Difference between digital and analog is like difference between lightning and lightning bug. Must protect property until that day.

Issa: What is fair use? When you make partial copies, do you think that's fair use?
Swift: I believe I am creating something new. Create a different theme with mix CDs. I think it's under fair use.
Issa: I disagree.

Issa: Difference between patent and copyright
Shapiro: Duration.

Gonzalez: Magnitude of potential harm? Is prosecution inadequate?
Lessig: Lots of ways to make sure people obey the law. Should you have fair use when technology takes it away. Prosecution may not be best solution. Would endorse compensation. Written several books. Had conference at Harvard. This is separate issue from this bill.
Gonzalez: Wants to focus on prevention model. Self interest in not developing model.
Lessig: AHRA gives explicit right to make copy of CD. Litigation against new business models rather than protecting creators.
Green (subbing for Valenti): Working toward celestial jukebox. Depend on DRM. Recognize that no system is perfect.

Shimkus: Tough sledding. Job loss because of 321 Studios ruling. Surprised that entertainment drives technology. Both sides need each other. Clearplay allows editing of smut and violence in movies. That's a good deal. Should we have the ability to edit?
Lessig: Absolutely. This legislation would guarantee that technology that would allow you exclude smut.
Shapiro: Technology is like a fork. You could use a fork to kill someone.
Green: This bill has nothing to do with this example. Derivative works.
Holleyman: Bill creates exception that swallows rule.
Nisbet: Support HR107.

Gonzalez: Never have secure environment. Troublesome part is where we are today and where we'll be tomorrow.
Green: The system is working.

Otter: Asking Lessig about his books.
Lessig: Third book is available free online for noncommercial use.
Otter: Is every student charged in law school or just one?
Lessig: Everyone.
Otter: Similar question to Shapiro
Shapiro:
Otter: Theft is theft and property is property.

Shadegg: Fair use is being horrendously misused. Is a single duplicate copy fair use? Are multiple copies fair use?
Lessig: Fair use is not the issue here.
Shadegg: four factors - non-profit educational purpose This bill is going after technology.
Lessig: this bill doesn't change the definition of fair use.
Shadegg: Why isn't this problem solved by the marketplace? If fair use allows single copies, while did we need the AHRA? If you want fair use, why can't you ask permission?
Lessig: why do you need to ask for permission? You don't have the right to infringe. If you are exercising fair use, it is not theft. It is wrong that technology steal that right.

Barton: Not only do we not have fair use, we have no use. Is there a way to balance tech, copyright, and fair use? Is this a debate about the number of copies or the intended use?
Lessig: The costs of adjudicating is extremely high. But that is independent.

Barton: Is it technologically possible to make a device that allows a small number of copies without allowing a large number of copies. Is fair use technologically compatible
Shapiro: AHRA enforces that.

Barton:
Green: Fair use is alive and well. No right to make individual copies even if you're not selling it.
Barton: I disagree. Is there no way to compromise?
Green: We see ourselves on the side of right
Swift: I do not agree with Mr. Green. We've provided draconian protections unlike what we do with anything else. Compares to shoplifting. Handcuffing every customer. Exaggerated, but comparable to DMCA. This is not the only way to deal with legitimate concerns of piracy.

Dupak: You would favor encryption with one copy.
Green: It is possible to build in that technology. The incentive to do it is based on preventing cracking.
Lessig: That doesn't guarantee fair use. Use a copy of a chapter of my book to criticize it. No simple tech because of the complexity of fair use.
Dupak: isn't the store responsible if your CD copier?
Swift: doesn't give an ounce of protection to the industry. It just takes four times longer to copy.

Bono: You are violating multiple copyrights when you make a mix CD. We have DRM. We have come a long way with technology. Speaking of iTunes FairPlay restrictions. This bill will undo that. Disney will replace broken DVDs, true?
Green: Yes.
Bono: people can put their creations out there freely. And is your book copyrighted?
Lessig: yes it is.
Bono: Xerox analogy is disingenuous. You value it enough to copy it but you're not willing to pay for it. You can post your list on the web.
Swift: Why would I want to do that?

Green: I have a few minds on this. Single copy is different than massive piracy. What is the process of DMCA rulemaking?
Shapiro: By most accounts that process doesn't work. Broad public access has been totally ignored. Radio put musicians out of business. Downloading is split.
Green: protect both innovation of technology and creators.
Holleyman: We have been able to deploy very simple technologies. It is not burdensome and it reduces piracy because of the DMCA.

Boucher: Copyright Office has issued the opinion that making an archival copy is fair use. Being able to excerpt small portions would be a fair use. Are you concerned to DMCA holds the potential for extinguishment of fair use?
Lessig: Absolutely.
Boucher: The creator can lock the content and protect the content.
Boucher: Fair use is an American doctrine. Is there a connection between fair use and growth?
Lessig: Absolutely. Foreign corporations do not understand the value of fair use.
Boucher: Are you concerned that the DMCA could lead to a pay-per-use society?
Nisbet: yes we are. We support the HR107 to protect free access to our patrons
Boucher: How does the rulemaking process work? My sense is that it is all but useless. Have you gotten the exemptions you have requested by any of the commercial groups.
Nisbet: We've gotten opposition, but no support. We've been disappointed, as has the Librarian of Congress.
Boucher: How different would the CES look in the event that the DMCA can be used to thwart competition (garage door openers and toner cartridges and Sony Aibo)?
Shapiro: Some of the members of our group are doing this. ReplayTV litigated out of existence because of DMCA.

Sherman: The DMCA has been characerized as anti-consumer. Disagree because creators were afraid to use technology. Now using them.

Mentions cable descramblers.

iTunes, DVDs. If HR107 were law, wouldn't have DVDs today.

Fair use rights have not been lost.

Not just facing commercial piracy. Also facing consumer copying. Sales down 31% in four years since widespread Internet copying.

Everyone relies on misconceptions of fair use.

HR107 is about getting creative content for nothing.

No discussion of marketplace. CDs and downloading services allow copying. Consumers are benefited by options.

Jaszi: Testifying for Digital Future Association. We make and use copyrighted works so we support both copyright and fair use. Support HR107.

This is a debate about freedom and fairness. DMCA lost sight of copyright norms. Overrode fair use.

Rulemaking has done nothing to bring back fair use.

HR107 will restore public respect for copyright which the DMCA has undermined.

Hollywood owes much of its success through fair use.

Will promote electronic commerce in many industries. The DMCA protects only a few industries.

Will allow lawful use. As the uses increase, the value will also increase.

Cybersecurity is important.

Rose: The entertainment software industry opposes the HR107.

Videogame development is expensive. Piracy would harm the industry. DMCA is essential to vitality and growth of videogames. Game publishers use technological measures to protect the games.

The DMCRA would make the DMCA meaningless.

DMCA rulemaking is working.

Without the DMCA, the software industry would enter a dark age.

Sohn: consumer perspective. Strongly support HR107 because it narrowly gives rights to customers. Requires labeling on copyprotected CDs.

Would restore the original intents of DMCA to protect consumer rights. Protections have been failed.

Murray: We make a living based on the protection of copyright. Garage door openers, toner cartridges, aftermarket car parts. DMCA used to block those products. Question isn't whether DMCA is broken. The question is how can it be fixed.

What should consumer expectations be in regards to labeling?

The Betamax decision didn't strangle the industry.

Moore: 321 Studios is a software company. Provided more than one million consumers a way to backup DVDs. 321 Studios is on the brink of annihilation because of DMCA.

Customers are people, not pirates. Want backups for various reasons.

Piracy is bad. But if people can use tools to make copies of CDs, why are they criminals for making copies of DVDs.

Software can only make backup copies and it typically takes an hour or more. Watermarked. Nothing on computer to upload.

321 has offered to work with movie makers. They have turned a deaf ear.

Stearns: You make a good case for your software. Let me ask about GamesX copy.
Moore: It's a virtual CD drive. We've had no challenges.
Rose: We are analyzing and keeping options open.
Stearns: Do you support restoration of sony decision.
Moore: Absolutely.

Stearns: Companies assert protection allows creativity.
Murray: concerned about anti-competitive effects when copyright wouldn't normal be in play. These systems will never be 100% secure. The status quo is technology doesn't allow fair use.

Stearns: Is it possible to allow one copy and that's it? Isn't it possible to reach a compromise?
Sherman: The problem with Mr. Moore's product is that it strips out the encryption.
Moore: That is an impossible task given the technology we have. We could reencrypt it. Second time in three years that Runaway Jury copying has happened. And we have forensic evidence we could use to find the source.

Otter: Does 321 Studios encrypt the software it makes?
Moore: Yes, to give teeth to antipiracy measures.
Otter: Would HR107 remove that protection?
Moore: I would not have a problem with that for noninfringing use.
Otter: Can I make a backup copy?
Moore: Yes, multiple times. We want control over the use of our software.
Otter: Isn't that what the studios are doing?
Moore: They have a right. But I think that ends at tools of consumption of the product.
Otter: Why encrypt with CSS?
Moore: You'd need to ask the studios.
Otter: Do you agree with Mr. Valenti's conclusions that it would affect the balance of trade?
Moore: I don't not agree.
Moore: Give the benefit of the doubt to the consumer. We expect most people to obey the law.

Issa: You know what tokenism is? You're it. You're the example of a good faith effort to make backup copies. Swift owns licenses, copies of records, and he edits records together. If you did the same thing with DVDs, would that be wrong?
Moore: Not a lawyer. Copyright is an act of distribution or act of consumption.
Issa: Only one person can use a book at a time. If you make multiple copies and watch them simultaneously aren't you violating that license.
Moore: There is no license on DVDs, unlike software.
Issa: If you could allow for only one copy per user, could you do it and would your product sell?
Moore: Provided that it's not drowning in DRM, it would continue to sell. This bill would not determine fair use.
Issa: This committee has a role to play in educating about fair use. Doesn't think Swift's behavior is fair use.

Boucher: We heard from Prof. Lessig that the DMCA can extinguish fair use.
Jaszi: yes
Sohn: yes.
Boucher: Content providers say. Not to worry. We have a process and it works. Do you agree?
Sohn: No. The Copyright Office has raised the burden of proof beyond the statute. The Office does look at the types of use. Assistant Sec of Commerce has protested the burden of proof or the types of use.
Jaszi: There is a structural problem with the rulemaking that is a function of the law. Holds out the possibility of an exemption for conduct, but not the devices or technologies. Therefore it's a false promise. Gives theoretical right, but no practical way.
Boucher: in 6 years, only four exemptions have been granted. 25 groups have been rejected. The Librarian of Congress has recommended that the law be changed.

Subcommittee adjourned.

2004-05-11T19:01:00.000-04:00

Tomorrow the House Subcommittee on Commerce, Trade, and Consumer Protection is holding a hearing on H.R. 107, The Digital Media Consumers' Rights Act of 2003. The hearing begins at 10:00 AM (Eastern Time) and will be webcast. The witness list has not yet been announced, but Lawrence Lessig has stated that he will be testifying. He has made his written testimony available as a PDF. Other written testimony and transcripts of the session will appear eventually at the hearing website.

It's probably obvious that I'm excited about this hearing but the reason why may not be as obvious for people who don't follow copyright legislation. The Digital Media Consumers' Rights Act (DMCRA) is an attempt to correct some of the imbalances of the Digital Millennium Copyright Act (DMCA).

The DMCA attempts to protect the rights of copyright holders in the face of digital technology, which allows easy, perfect copying of copyrighted materials. Copyright law grants copyright holders exclusive rights to duplicate, reuse, distribute, perform, and display copyrighted works. Copyright law also provides for damages against copyright infringers who violate one of the five exclusive rights of copyright.

Digital copying and distribution reduces the difficulty of duplicating works. It is easier to copy a PDF and email it to a friend than it is to manufacture a physical book and ship it. The DMCA was written to address the fear that it would be harder to identify infringers and collect damages against them given the ease of infringement.

In an attempt to make duplicating digital works more difficult, publishers have taken to adding copy prevention to their works. One example that many people encounter regularly without necessarily realizing it is with DVDs. Commercial DVDs are encrypted with CSS (Content Scrambling System). This isn't a big deal most of the time because DVD players are licensed to decrypt CSS. If all you use DVDs with is a DVD player, you may not even notice the difference.

If you have a DVD drive on your computer, you may be surprised at what you can't do with DVDs. You can't take screenshots of the DVD or make copies of the video or audio from the DVD. This is in obvious contrast to CDs, which allow you to freely copy the music they contain.

There's a reason for this. Computer manufacturers license the right to decrypt CSS, and one of the conditions for the license is that the software can't allow the user to copy the DVD or any part of it.

There is, however, a weakness to CSS. It has been broken. Software that decrypts CSS but that hasn't been properly licensed is available. Since it isn't licensed, it doesn't follow the terms of the licensing and it allows users to make copies of the DVD.

This is the weakness of all copy prevention systems. They will be cracked. The current copy prevention Apple uses with music downloaded from the iTunes Music Store was cracked in less than a day. Once it has been cracked, the copy protection technology is useless for preventing copying.

This is where the DMCA comes in. The DMCA makes it illegal to circumvent copyright protection systems. This is why, although it's not difficult to download unlicensed software to play DVDs, you can't purchase commercial software to play DVDs that is unlicensed. The DMCA prevents the manufacture or sale of anything that breaks copy prevention.

The DMCA exceeds its bounds in several key ways. The first is with public domain works. Public domain works have no copyright holder. Anyone may use a public domain work in any way they wish, including copying, reusing, distributing, performing, or displaying the work. However, the DMCA means that the manufacturer of a public domain work can use copy prevention technology on the work and prevent purchasers from using the work as they wish. The DMCA doesn't protect the work itself, but it does protect the copy prevention, meaning that in practice purchasers may have no more ability to use public domain works than they have with copyrighted works.

The DMCA oversteps its bounds with respect to copyrighted works as well. Copyright law provides for several limitations to the exclusive rights it grants. The two most well known limitations are the first sale right and fair use, but there are others as well. The right of first sale states that after a customer has purchased a copyrighted work, the customer can resell it freely. The copyright holder cannot restrict the sale in any way. Fair use is the use of a copyrighted work for criticism, teaching, research, or parody. The scope of fair use is not clearly defined by law, but fair use is noninfringing use of copyrighted works and it is explicitly protected by law.

The DMCA does not adequately protect the rights of purchasers of works with copy prevention. The public domain, first sale, and fair use grant rights to customers which can be taken away by copy prevention. The DMCA enforces the power of copy prevention by making it illegal to defeat the copy prevention and it doesn't provide for the protection of the customers' rights. In effect, the DMCA takes away rights explicitly granted by copyright law.

Rep. Boucher has written the DMCRA to restore the rights given under copyright law to customers that have been taken away by the DMCA. The bulk of the proposed law is focused on labeling audio CDs. In an attempt to prevent music from being copied, various manufacturers have begun experimenting with copy prevention on audio CDs. This copy prevention can prevent the CDs from being played at all on certain CD players or on computers. The DMCRA would require CDs with copy prevention to be clearly labeled.

The far larger potential impact, even though it makes up one small section of the bill, is a revision to the DMCA. The DMCRA would state that breaking copy prevention for noninfringing uses would no longer be a violation of the DMCA. This would restore the right of fair use.

This bill is expected to receive strong opposition. Major copyright holders have argued to Congress that copyright law needs to be strengthened in order to protect the rights of copyright holders and creators. Generally, they have been successful and copyright law has been repeatedly changed over the last 30 years to increase the rights of copyright holders.

I believe that copyright law requires a balance. Creativity is dependent on reuse of materials from the past and copyright law interferes with that reuse. Furthermore, I don't think copyright law should prevent customers from merely watching a movie or listening to some music, as it can today. The DMCRA would not undo copyright law. It would reduce the current scope of copyright law, restoring rights to creators and consumers that have been improperly taken away.

Please contact your Representatives and tell them that you support the DMCRA because it restores rights to customers. This is particularly important if your Representative is a member of the Subcommittee on Commerce, Trade, and Consumer Protection. The members of the subcommittee are:

John B. Shadegg, Arizona
Mary Bono, California
Darrell Issa, California
George Radanovich, California
Diana DeGette, Colorado
Jim Davis, Florida
Peter Deutsch, Florida
Cliff Stearns, Florida
C.L. "Butch" Otter, Idaho
Bobby L. Rush, Illinois
Jan Schakowsky, Illinois
John Shimkus, Illinois
Ed Whitfield, Kentucky
John D. Dingell, Michigan
Bart Stupak, Michigan
Fred Upton, Michigan
Karen McCarthy, Missouri
Lee Terry, Nebraska
Charles F. Bass, New Hampshire
Mike Ferguson, New Jersey
Edolphus Towns, New York
Sherrod Brown, Ohio
Ted Strickland, Ohio
John Sullivan, Oklahoma
Joseph R. Pitts, Pennsylvania
Joe Barton, Texas
Charles A. Gonzalez, Texas
Gene Green, Texas
Barbara Cubin, Wyoming

2004-02-09T17:23:00.000-05:00

I have a question. If someone is being a sucker, is it better or worse to knowingly be a sucker? Pepsi is giving away free downloads from Apple's iTunes Music Store. If you buy a bottle of Pepsi, you have a one in three chance of getting a free song. Despite the fact that I know it is dumb to spend $1.25 to get a 33% chance to get a 99 cent song, I've started buying Pepsi anyway. Clearly I'm being a sucker. I wonder if that's made worse or mitigated by the fact that I know I'm being a sucker.

On a more serious note, I've been wondering what Apple and Pepsi's target audience is. If their primary target is people who already use the iTunes Music Store, then this is just an attempt to sell more soda. It's a successful attempt in my case, but that doesn't really give it a higher purpose.

On the other hand, I wonder if they are targeting users of file sharing software. These are people who already consider the Internet to be a source of music, but only free music of dubious legality. Given the opportunity to download legal free music, some file sharers may try it. Then it's up to Apple to convince the downloaders that iTunes is better than file sharing.

If that is their goal, then the promotion which looks like a way to sell soda is also a tactic to encourage the use of paid music services rather than free ones. The commercial run during the Super Bowl supports this thinking. The message seems to be, "We know you're going to download music. Why not try it legally?" The commercial repeats the line that the recording industry has been suing downloaders of music, which is not correct. All the lawsuits so far have been against music sharers, not downloaders.

Regardless, if their target is file sharers then the campaign is more clever than it seems. I wonder if it will succeed.

2004-01-07T14:17:00.000-05:00

It's the time of year for retrospectives, predictions for the next year, and such. I figured that since I started this blog last January, I would put off a New Year's retrospective and and just have a first anniversary wrapup instead. I checked to see when I started the blog, and it turns out that I missed it. I started this blog on January 4, 2003, but I actually wrote the first entries on the 2nd, before I had the blog itself set up. So it's a few days late, but here's my first anniversary review and a look to the future.

Starting at the beginning really requires starting during the summer of 2002. My job wasn't going anywhere and I was trying to decide what to do instead. I hit on the idea of starting an online record label. This would neatly combine my long-established interests in technology and music and my rising interest in law. The premise was to have a network of websites to promote the bands on my label. They would include both free music downloads and music downloads for sale on both an individual and subscription basis. A relatively small number of songs would be offered free on a rotating basis with the hope of encouraging fans to return on a regular basis. All album tracks would be available individually. As an incentive for a subscription, it would include bonuses like live recordings and music that was otherwise unavailable.

I would not attempt to restrict redistribution. In fact, I would include MP3s of all the music on CDs that I sold. I figured that if they're going to be ripped anyway, I might as well try to impose some quality control. The hope was that any sales lost by redistribution would be made up in increased sales.

In addition to selling music downloads, I would sell CDs and paraphernalia, such as t-shirts, posters, and stickers. I would also promote concerts and try cross-promotion between the bands. The goal was to think of everything as a promotion tool and expect that even if I lost money on downloads, they would increase the rest of the business.

I came up with plan without knowing how to do any of this. I figured, how hard could it be? Besides, I learn best by doing anyway. I'd figure it all out as I went. Whether that makes me optimistic or stupid is up for debate, but that was my plan. When I encountered the phrase "rolls a hoover" on RageBoy's blog, I decided it fit what I was doing. I was jumping off a cliff and trusting that I would land safely. (His original post has apparently disappeared. See the FAQ in the right column for marginally more information.)

By last January, I decided it was time to act. I left my job and started this blog. It was a deliberate attempt at self-motivation. I thought that if I had the blog I would be obligated to keep working on my plans so I would have something to update the blog with. Early posts were on technical issues of setting up the blog itself. Unfortunately, it didn't really work as a motivating tool, and by March I wasn't progressing the way I had hoped and my posting frequency dropped off.

Fortunately, in mid-March I went to a symposium on copyright at Harvard. Since this tied into what I had been thinking about for my record label, I jumped in hard and this blog has never been the same. Since then I have posted almost exclusively about the theory and practice of copyright.

My posting frequency fell in September. Since then my major posts have been on subjects that I thought were unlikely to be covered elsewhere and analysis of court rulings. I have some thoughts on why this occurred. First, it's not because I lost interest. It's fairly typical for me to be obsessed with something for a few months and then for that interest to fade, but that hasn't happened in this case. I still follow copyright issues on a daily basis and I have a long list of links that I've been intending to blog. By now it's long enough that I don't know what to do with it. Any hope of actually blogging them all individually seems unrealistic and the list just keeps growing. Maybe what I need to do is put together a single post consisting just of the list of links without any commentary at all. But if I don't comment on them, it may not be clear why I think they're interesting. I don't know.

Further proof that my interest hasn't passed is in the way that I felt compelled to blog the ruling in the appeal of Verizon v. RIAA. It's not just that I was interested enough in the ruling to read it. It's that I had to write my analysis of it immediately, even though I had intended to put it off.

I think one factor that both explains why I've been doing less blogging and why I felt compelled to blog the court ruling is that I've been looking for ways to get a job doing copyright advocacy. What I've found is that there aren't that many jobs available, and the ones that exist are in high demand. That seems even more true for people like me who don't actually have any law experience.

One solution to that would be to go to law school. I've been going back and forth on whether that's a good idea for the past several months. On the downside, law school isn't easy and I would have to take a significant number of courses that I'm only peripherally interested in. Furthermore, what would I do once I had a law degree? There are certain positions I expect I would be interested in, including policy work, appellate work, and academia. But if I ended up doing contract law or litigation, would I be happy? I honestly don't know.

On the upside, it seems like copyright law touches on nearly all aspects of law. Actually having a law degree would help me to fit all the pieces together in a way that I can't always do now. And I actually enjoy reading court rulings. My natural response when I hear of a ruling I'm interested in is to sit down and read the ruling itself. The fact that I'm interested in court rulings at all makes me unusual enough, I suspect, but the fact that I want to work through hundred page rulings myself puts me in a rare class.

I've been discussing the merits of law school with lawyers and non-lawyers. Non-lawyers have generally been more enthusiastic than lawyers and I'm not sure what to make of that. Lawyers in advocacy or policy positions have actually been the most skeptical of the value of a law degree. I'm not sure where that leaves me.

Recently I've been trying to get a position doing IP work in a law firm. I hope that a position in some sort of legal assistant role would give me an opportunity to decide if law is really for me. So far I haven't had any major breakthroughs, but I'm still working at it.

One of the reasons why I've been blogging less, then, is that I've been looking into doing this as a career. Blogging is fun and all, but it doesn't pay very well. I'm also uncertain of the impact it has. Does what I write here influence anyone with real power? It might. If I cause readers to think about these issues more and discuss them with others, there may be real effect of my blog. But it's indirect at best. So if I spend more time thinking about what else I could be doing and less time blogging, that may not be a bad thing.

I'm also not sure that a blog is really the right structure for my writing. If I make a series of posts about a particular issue, they end up scattered all over the archives. I've taken to running Google searches on my own blog, which seems really dumb but it's the only way I can find things. I've considered the merits of switching to a blogging tool that allows more categorization so it would be possible to look things up by topic. Another possibility is migrating to a wiki or some other structure that puts organization by topic ahead of organization by date. While I've been sitting around wondering what to do, it's been a disincentive to keep posting as is.

Another factor that I have to acknowledge in changing my blogging habits has been significant upheaval in my personal life. My natural tendency is to try to separate my personal life from my blogging, but sometimes it spills over. I moved at the beginning of August. Shortly thereafter, my relationship with my girlfriend fell apart. (The two events were not directly related.) Both have been disruptive. Breaking up with my girlfriend has been particularly hard. It's no exaggeration to say that I've been hurting every day since then. The fact that it's been hard for her as well is small consolation. I mention this less for sympathy than to just acknowledge the impact it's had. We may not want one part of our life to affect another, but it's not always in our control.

So that's where I am today. Here are my thoughts about this coming year. I need to get a job of some kind. Right now, I think my best possibility is a legal position, but I'm open to the possibility of unexpected changes in direction. I want to make a firm up or down decision on law school. I hope my job will lead to that decision, but I think it's important to actually make a decision one way or another regardless of what kind of job I end up with. Finally, I want to see what more I can do with this blog. Is a total redesign the way to go? I've had various ideas about a possible companion site. I need to do more experimentation with that.

As usual, my look back is far more detailed than my look forward. I wonder how much that's self-perpetuating. Anyway, here's to 2004 and year 2 of Matt Rolls a Hoover.

2003-12-20T14:23:00.000-05:00

The appeals court has ruled in favor of Verizon against the RIAA in the file sharers subpoena case. This is a huge win for Verizon, file sharers, and fans of limiting the power of copyright laws. I first heard the news from John Palfrey, but I'm sure just about everyone is covering it by now. The ruling itself is available as a PDF.

First, the background: the DMCA allows copyright holders to issue subpoenas against Internet providers to compel them to reveal the identity of users who have copyrighted works online. Note that this subpoena power exists independently of a lawsuit. Copyright holders also have the option to file a lawsuit and then subpoena the Internet provider to learn the identity of the user. The DMCA reduces the requirements for the copyright holder by not requiring the initial lawsuit and not requiring any evidence of infringement.

In 2002, the RIAA issued two subpoenas to Verizon under the DMCA to determine the identity of two Verizon customers which the RIAA asserted were distributing copyrighted music files over a file sharing network. Verizon challenged the subpoenas on both statutory and constitutional grounds. The district court ruled in the RIAA's favor and Verizon appealed the ruling. Verizon also attempted to stay the ruling during the appeals process. Verizon lost the request for the injunction and was compelled to provide the information about the customers to the RIAA.

Since then, the RIAA has issued thousands of subpoenas to Internet providers across the country and has used the information to file hundreds of copyright infringement lawsuits and threaten hundreds more with potential lawsuits. A small fraction of the subpoenas have been challenged by both the Internet providers and by their customers and a small number of the lawsuits have been challenged, but in the majority of cases the targets have settled, with typical settlements of a few thousand dollars.

This ruling brings that process to a halt by ruling that the subpoenas aren't valid.

The ruling states that Verizon made one statutory argument and two constitutional arguments opposing the subpoenas. The statutory argument is that the language of the DMCA specifies that the Internet provider provide storage for the infringing file. File sharing software, which did not exist when the DMCA was written, allows the users to store the files on their own computers rather than those of the Internet provider and uses the Internet provider only to transport the files when they are copied from one user to another. Verizon argued that since the Internet provider doesn't store the file, the DMCA does not apply.

The first constitutional argument is that the law is invalid under Article III of the Constitution, which requires a "case or controversy" before a court can issue a subpoena. Verizon argued that in the absence of a lawsuit, there is no "case or controversy" and therefore the law is unconstitutional.

The second constitutional argument is based on the First Amendment. Verizon argued that the DMCA infringes on the user's rights of speech and association.

The appeals court agreed with Verizon's statutory claim and did not come to a decision on the constitutional claims, although a footnote indicates that the court would not have accepted the Article III argument.

Section 512 of the Copyright Act, added by the DMCA, lays out protections against liability for Internet providers for the copyright infringement carried out by the users. Subsection 512(a) protects against liability for transferring copyrighted materials at the request of a user. 512(b) protects against liability for caching or temporarily storing materials. 512(b) is important for the overall performance of the Internet because Internet providers rely on caching to reduce the total data sent over the Internet, which increases the speed of the Internet. 512(c) protects against liability for storing copyrighted materials at the request of a user, provided that the Internet provider is generally unaware of the content of the material. Finally, 512(d) basically protects search engines from liability.

The important subsections in this case are the 512(a) protection against data transfer and the 512(c) protection against data storage. A significant difference between the two subsections is that 512(c) includes a take-down provision while 512(a) does not. The take-down provision requires that upon receiving a subpoena as detailed in subsection 512(h), the Internet provider must either remove or block access to the allegedly infringing material. 512(h) also requires that the provider provide the identity of the allegedly infringing user to the copyright holder.

The arguments reduce to minutiae on the exact wording of the law and whether or not subsection 512(c) applies to file sharing networks. To my mind, it's obvious that 512(c) depends on the provider providing data storage and it clearly does not in the case of file sharing networks, but the actual arguments in this case are far more nitpicky. In any event, the court decided in favor of Verizon, ruling that 512(c) does not apply to file sharing networks.

Alternatively, the RIAA held that subsection 512(h), which requires the identity of the user to be revealed, applies to all four subsections and protection from liability, so that even if file sharing is covered by 512(a), copyright holders can still issue a subpoena under 512(h) to learn the identity of the user. In response, the courts states that, "This argument borders on the silly."

Verizon argued that 512(h), because of the way that it is written, only applies to 512(c) and not to 512(a), (b), or (d), despite the fact that 512(b) and (d) also have take-down provisions which reference 512(c). The court rejected this argument for 512(b) and (d) but agreed for 512(a). The court emphasizeds that 512(b)-(d) all refer to data storage while 512(a) refers only to data transfer.

The RIAA made an argument that the legislative history of the DMCA indicates that 512(c) should apply to file sharing anyway, but the court ruled that the law is quite clear as written and if it doesn't apply to file sharing, that was because file sharing was developed after the law was passed. If Congress had anticipated file sharing, it might have written the law differently, but it didn't and that's that.

Finally, the RIAA argued that Verizon's interpretation would defeat the purpose of the DMCA, to which the court basically responded with "tough luck." It's not the court's job to make laws; it's Congress'. The court pointed to the ongoing Congressional hearings on file sharing to back up that position.

The important question is what happens next. The RIAA may appeal to the Supreme Court, but it hasn't yet stated that it will. The RIAA has stated that this ruling will force its hand and will prevent it from attempting to settle infringement claims outside of lawsuits, which implies that the RIAA intends to keep suing anyway.

My big question is on the fate of all the issued subpoenas. By now, hundreds of subpoenas have been improperly issued, leading to lawsuits and settlements. What is the status of these subpoenas? I thought that the DMCA listed penalties for knowingly filing erroneous subpoenas, but I can't track the precise language down now. Is the RIAA on the hook? They could claim it wasn't "knowingly," but it still could result in a lot of trouble for them. Does this throw the outcome of the various lawsuits and settlements in doubt? Could a target of the RIAA claim that since the subpoena was improper, the settlement was invalid? This is one of those times when I wish my knowledge of general legal process was better than it is.

This ruling only applies to the DC circuit. Will the RIAA just start issuing subpoenas through courts in another circuit? Will the RIAA and MPAA try to change copyright law to include file sharing under the take-down provisions of the DMCA? There are many different directions the RIAA could move in from here. I'm very curious about what they will do.

Another obvious point is the way in which copyright law in general struggles to keep up with technology. The Copyright Act of 1909 was written partially to account for the new technologies of the late 19th century. 60 years later, technology had changed sufficiently to require the complete overhaul of the law in the Copyright Act of 1976. Less than 15 years after that, the Audio Home Recording Act was written in response to the rise of audio tape. Less than 10 years after the AHRA, the DMCA was intended to cover digital technology. And now, five years later, court rulings on file sharing are indicating that the DMCA has failed.

It feels like we're moving toward a breaking point, where the pace of technological change outstrips the response of the law so badly that a fundamental reconceptualization of copyright will be required. I don't think we're there yet, but I hope we get there soon.

2003-11-16T18:50:00.000-05:00

The Diebold controversy has gotten to MIT. Friday's issue of The Tech reports that MIT has asked two students to remove a collection of email leaked from Diebold. (The article hasn't been posted to the website yet, but it should appear eventually.) For those not up on the story, around 15,000 internal emails were leaked from Diebold and posted on the Web. Diebold has since attempted to suppress the messages by using the DMCA to force ISPs to remove the messages. This has produced a cat-and-mouse game as the messages have propagated around the Internet, mostly at universities, followed by cease and desist letters from Diebold to the ISPs. Diebold claims that the messages are the copyrighted work of Diebold and demands that the messages be removed under the takedown provision of the DMCA.

So far, it's been a losing battle for Diebold. Although the ISPs have generally complied and removed the material from their websites, it seems to be spreading faster than the company can shut it down. Besides MIT, other schools where the messages have been posted include Swarthmore, Berkeley, and Harvard, where the messages were posted by Derek Slater of A Copyfighter's Musings. Harvard received a take-down request from Diebold and Derek is considering how to proceed in the face of Harvard's two strikes policy on copyrighted works.

It's fair to ask why these messages are getting so much attention. Diebold manufactures computer-based voting machines. Some observers have questioned the reliability and security of the machines and the email messages appear to confirm many of the problems the machines have. I don't want to get into the specific issues of the voting machines here. Instead I want to call attention to the conflict between newsworthy publication and copyright.

The DMCA takedown provision states that if an ISP provides Web space for a user and the user posts a copyrighted work on the Web, the ISP is not liable for copyright infringement if the ISP removes the material in response to a request from the copyright holder. (It's not clear to me that the ISP is necessarily liable if it refuses to remove the material. Anyone have more information on this?) The law is ripe for abuse even with regards to traditional copyrighted works like books, movies, and songs, but situations like the Diebold emails make this abuse more clear. Under modern copyright law, everything written is copyrighted from the moment it is written down. That extends even to internal corporate email which the corporation never had any intention of publication. I expect that corporate email would be considered a work for hire of the corporation and so the corporation would hold the copyright rather than the employee who actually wrote it. There might be a line of defense in the claim that the copyright is held by the employee rather than the company unless there's an explicit statement to the contrary in the company's employment contract or use policy, but I wouldn't want to have to rely on it.

I really have to ask if it makes sense that internal documents be protected by copyright. I don't question that in general companies should be able to get legal redress if internal documents are made publically available or available to competitors, but this really strikes me as more a trade secret issue than copyright issue. Under current law it's both, which allows companies to use the DMCA to try to suppress distribution of the documents. I have to admit that if as a consequence of this copyright supplants trade secret law, that wouldn't be an entirely bad thing. Trade secret law operates at the state level and I can see advantages to standard federal laws like we have with copyright. The current double protection seems redundant.

Returning to this specific case, C. Scott Ananian and David Meyer posted the Diebold emails to their web pages at MIT, provoking a take-down request from Diebold. Both students removed the emails. The Tech reports that Meyer is satisfied that the documents have been sufficiently widely distributed and will not be contesting the order, but Ananian has filed a counter notification asking to be permitted to repost the messages.

The DMCA does not require the party issuing the takedown request to offer proof of copyright ownership. To counter the potential for abuse this creates, the law allows the individual who posted the material to file a counter notice if the material is not infringing. In that case the ISP must allow the material to be published unless the copyright holder takes direct legal action against the individual. The Tech states that Ananian believes that he is entitled to publish the messages under the fair use provision of copyright law and so he will make a counter notification allowing him to repost the messages.

I'm very interested in how far this will go. If Diebold proceeds to take direct action against Ananian, the first question will be whether his fair use defense holds up. The law provides additional pitfalls on both sides, because there are damages associated with knowingly issuing either a false takedown notice or a false counter notice. If Ananian loses on the copyright issues, he may also be liable for the false counter notice. If he wins, Diebold may be liable for the false takedown notice. If Diebold is liable, that may have the side effect of reining in the notices the music industry and other copyright holders have been firing off recently.

The current issue of The Tech also has an editorial on LAMP that I figued I should mention in passing. Like the Diebold article, it is not yet online. (LAMP is a system developed at MIT to make music available over the cable system. The developers had licensed the right to distribute the music from Loudeye and were distributing the music in analog form to avoid copyright laws pertaining to digital distribution. Shortly after the system was launched, the RIAA objected and it has been shut down while the creators negotiate on the legality of the service.)

The other thing that I have to mention from this issue is the front page review of Ruddigore performed by The MIT Gilbert & Sullivan Players. All of the works of Gilbert and Sullivan have entered the public domain which has given the group the ability to pick and choose from the various versions of the show that exist. The choices made were evidently very successful on the basis of the rave review the production received. (Full disclosure: I was a member of the cast of this production.)

2003-11-10T17:34:00.000-05:00

I recently had a frustrating conversation about copyright with a good friend of mine. Fred and I have similar attitudes on a wide range of political and philosophical issues, so it was distressing to realize that he generally favors long copyright terms and strict control over derivative rights. Many people have similar attitudes to his on copyright, so those of us who want weaker copyright laws have to get better at presenting our arguments if we want to see real change.

The conversation started on the question of whether downloading music is right or wrong, but it quickly spread to a discussion of copyright in general. It covered a lot of ground and I'd like to repeat some of the key topics of discussion here.

The first thing that came up is that Fred is focused on morality while I'm focused on legality. At this stage I feel that I generally understand what the law is and what I would like for it to be. I have less concrete answers to the moral issues of copyright. As I see it, behaviors shouldn't necessarily be illegal just because they are immoral. Some uses of copyrighted works should not be illegal regardless of whether or not they are immoral because the harm done by criminalizing them is worse than the harm of the actions themselves. In some cases, I have remained basically undecided on the morality of certain actions because I think they shouldn't be illegal regardless. This has left me unprepared for certain discussions of the morality and copyright issues.

The example we worked with in our discussion was the Star Wars movies. We are both big fans of the original Star Wars trilogy, think The Empire Strikes Back is one of the best movies ever made, think the Special Edition versions released in 1997 don't add anything to the originals at best, and are very disappointed by the first two movies of the prequel trilogy. This gave us a common ground to discuss copyright issues based on the movies.

One of the major points of disagreement that we have is the duration of copyright, especially as applied to derivative works. Fred believes that the purity of the Star Wars story should be maintained until it has mythical status and other people should not be able to create Star Wars movies without George Lucas' permission. My belief is that Star Wars has been around for over 25 years and if other people want to create Star Wars stories which diverge from what Lucas has done, they should be able to do that. At my more extremist moments, I doubt that there should be any copyright protection for derivative works.

Fred responds by asking about the possibility of someone other than Lucas creating Star Wars Episode 7. We agree that any movie following up on Return of the Jedi is wholly unnecessary and if someone else made a sequel to it, not only would it likely be bad, but it might even be so bad as to damage the reputation of the original Star Wars movies. Fred finds this intolerable. I have more faith that the public would adopt knock-off Star Wars movies if they are good and reject them if they are poor. For that matter, Lucas has damaged Star Wars himself with his prequel movies. Hearing Ben Kenobi tell Luke about the Force just isn't the same now that we know about midichlorians.

Fred's response is that Lucas has the right, as the creator of Star Wars, to screw it up, but other people do not have that right. Besides, if there was no protection of derivative rights, someone could have made Star Wars 2 immediately after the first movie was made and prevented Lucas from making The Empire Strikes Back. I am less convinced that this is a real threat than Fred is. First of all, a Star Wars 2 would not have prevented Lucas from making Empire, although I think it might have made the movie more difficult. If the other movie had succeeded, it would have put pressure on Lucas to respond to it and it might have reduced the commercial success of Empire because the two movies would have competed with each other.

I think Star Wars 2 would have had a very hard time succeeding because it wouldn't have the credibility of a movie by Lucas. Furthermore, I think this possibility would be an incentive for more creation. If Lucas wanted to control the Star Wars setting, he would have to keep creating new movies. Otherwise, someone else could step in. I pointed to Don Quijote as an example of this. When Miguel de Cervantes wrote Don Quijote, copyright protection did not extend to derivative works and another author created a sequel. In response, Cervantes wrote his own sequel in which he repudiated the other author's work. What we think of as Don Quijote today is the combination of the two books Cervantes wrote, while the work of the other author has been forgotten. This did not convince Fred, who is worried that this would force authors to create sequels to what had been intended as stand-alone stories just to prevent other authors from writing their own sequels.

One issue that I did not bring up is the interaction between copyright and trademark. This might serve to reduce the confusion between versions that Fred fears. If Lucas controlled the trademark on Star Wars but not the derivative rights, then other people would be free to create movies set in the Star Wars universe and use the characters, but they would not be permitted to call their movie "Star Wars." This would reduce the incentive to create sequels in an attempt to cash in on name recognition while permitting more creative reuse of existing materials than is available today.

Speaking of creative reuse, I argued that Star Wars is itself an amalgation of existing ideas. I asked Fred whether he thought it was fair that Lucas used ideas from Joseph Campbell and Kurosawa, among others, but that new creators aren't free to reuse Lucas' work. Fred's response was essentially that Star Wars reuses abstract ideas in a concrete way and that it would be wrong to reuse the concrete ideas of the movie.

Another sticking point is in what I see as abuse of copyright. Since Lucas completed the Special Editions, he has surpressed the original versions of the movies. Fred and I have both gone as far as buying Laser Disc players and the original movies on Laser Disc because Lucas has said that he will never release the original versions on DVD. More recently, Lucas refused to allow the original version of Star Wars to be shown at a film retrospective. I think this is indefensible. One of the reasons why Star Wars had the impact it did in 1977 was the special effects, and insisting that a film retrospective show the version with 1997 special effects is immoral. Fred agrees that it is wrong, but he thinks Lucas has the right, as the creator, to control which version is shown. Like Disney deliberately releasing old movies on DVD for a few months and then pulling them from the market, I think this is wrong and an absolute abuse of copyright. Copyright exists as an incentive to the creation of new works and using it to suppress older works, as Lucas and Disney do, should not be permitted.

One point of agreement for us is copy protection. Fred thinks that once a customer purchases a copyrighted work, the customer should have free use of the work. Copy protection that interferes with this use is immoral and the customer should have the right to defeat the copy protection. This hit home from Fred when he considered watching movies on his laptop. He would like to be able to copy movies to his computer so he wouldn't have to worry about the DVD when he travels, but copy protection interferes with this. He thinks that he should have the right to override copy protection which prevents fair use.

I decided not to try to make the argument that what he is asking for is a personal use, not a fair use. To my mind, fair use applies to publishing a work that makes use of another work while all uses of copyrighted material short of publication or otherwise making them publically available are personal uses and should never be subject to the control of the copyright holder. I decided that since we agreed on the outcome, that's a distinction I didn't have to make at the time, but I think it is an important distinction that people should be generally aware of.

The truly frustrating thing in all this was that at the beginning of the discussion, I set out the differences between the utilitarian model of copyright and the artist's rights model, and Fred had no problems agreeing in principle to the utilitarian model. In practice, however, he returned to the artist's rights model again and again and didn't understand why I didn't agree with him. He made the point that personal ownership is highly valued in American culture and that I am essentially arguing against that ownership. I agree with him on that, but I believe ideas are essentially unownable, and arguing about copyright on an ownership basis is starting from false premises.

I'm painfully aware, however, that although I think they are false premises, Fred does not, and many people agree with him. Changing copyright law will require changing that thinking. In this case, at least, I was unable to do that. Changing the laws is a fight we can't win unless we get better about changing ideas for the basis of copyright.

2003-10-08T16:25:00.000-04:00

On September 29th, Derrick Ashong gave a presentation at Harvard in which he presented a new model for the music industry. He was speaking on behalf of ASAFO Productions and their Freedom Access Music (FAM) licenses. The FAM license is similar to Creative Commons licenses. It allows the music to be redistributed, provided that credit is given to the artist.

One of the major focal points of the talk is that musicians don't make money by selling recordings of their music. Musicians make money by selling things related to their music such as concert tickets and merchandise. Therefore, not only should musicians not try to control the sale of their music, but they should encourage the distribution of the music as widely as possible.

Ashong gave examples of parts of the music market that already work in this way. For example, many rap artists and reggae artists get their start by making a name for themselves in the gray market of music sales without copyright protection. They then turn the fame they have gotten into a major career. He asserts that the music industry as it currently exists is dying and it will be replaced by this sort of system generally.

My notes from the talk are below. As usual, they are not formatted particularly well.

music industry - simple principles
label is bank, distributor, own heart, life, soul
company has sole right to profit from what I do
master recordings belong to label
under FAM, no exclusive license, license for commercial or noncommercial reproduction. for the most part, you are free to do as use please
look at four artists
A is independent, selling unsigned
B is signed, but new
C is known
D is superstar

A makes money performing. Music makes selling music is directly sold. Doesn't have resources to do promotional blitz.

B makes money from live performances. Doesn't make money off sales unless goes platinum

C makes money from records, performing, merchandising

D is a marketing machine. Not in business of music. In business of brand. Sells anything you put his name on. Makes more money off merchandising than record sales.

If money not in sale of recorded music, why keep selling? 90% of bands don't make any money from records. less than 10% make profit for parent company. The 10% who do sell so much they make up for everyone else.

music industry is not in business of music. It's in the business of image. Notoriety makes everything else work. In business of fame.

FAM licensing: If I want to be successful, need to understand nature of business.
cost of production very high.
talk of home studios and Internet distribution

If in 10%, label will milk product for rest of life.

only reason to sign: promotion. label will manufacture notoreity.

reduce cost of promoting new talent by distributing among a variety of promoters
incentive to sell music and support artist if successful.

lists rappers who started on underground (illicit tracks)

questions

busting monopoly. Then again, not. System as it stands is broken. If I want to be a company, I don't want to be a record label, because no longer have monopoly on production or distribution. make my business as manager. build business around notoreity. take percentage of everything.

record label will want monopoly on brand? managers regulated. More choice of promoters

record labels decide to go after top 10%? more artists become viable with smaller sales

question about 24 Hour Party People, but also mentions value of high end studios. Sound quality not as high, but people still make money.

question about replacement for mix tape? distilling functions of record label. management company controls efforts

not trying to sell to record companies. trying to sell to artists.

will consumer buy into more fragmented market? downloading issue. good music and smart marketing always functional.

2003-10-06T13:56:00.000-04:00

Jason Goodman wrote an email disagreeing with my assertion that radio spectrum is nonrivalrous. I trust that he won't object to my posting his letter here. Here is his letter, followed by some additional comments.

I disagree that radio is inherently nonrivalrous. The FCC's band licensing rules exist because there's not enough room in the low-frequency radio spectrum for everyone to have their own TV/radio station. These rules are in place precisely to prevent a tragedy of the commons. Thus, the excludability of low-frequency radio is not an arbitrary legal impediment, but is designed to maximize the utility of a rivalrous medium.

Wireless networking is not rivalrous in practice today for two reasons. First, it is not a mature technology, and the hard limits on its use have not yet been reached. Today's high-frequency radio spectrum is a bit like the open range of the 19th-century American West: if you tried to explain the tragedy of the commons to an early Texas rancher, he'd spit on your shoes.

Second, anyone can create a wireless network because the range and power of those networks is limited by FCC regulations to reduce interference. The FCC's power limits reduce rivalry, but do not remove it: once every bar, coffeehouse, and restaurant on the block has wireless internet, interference will become a major problem. (Only 3 802.11 networks can coexist in one area!) The power limit also excludes certain types of use: there's no practical reason a company couldn't build a massive wi-fi transmitter and provide service to an entire small town or city center (ala cellular telephone service), but this use is forbidden by the FCC's power limits. Part of the reason for this limit is to allow small transmitters (home networks, for example) to exist.

Thus, FCC regulations impose legal exclusion in order to maximize the utility of a rivalrous but not physically exclusive resource. Legal action to make spectrum excludable increases the utility of the spectrum by eliminating hogging, in contradiction to your thesis in the second-to-last paragraph.

Jason

I recognize that there are theoretical limits to the density of wireless networks, but in my experience there are no practical limits. Two quick examples are that at my old apartment, my computer could detect four different networks simultaneously and I was never aware of any interference, and I was at BloggerCon yesterday and I don't know how many people were wirelessly networked simultaneously, but the answer was lots. See Dan Bricklin for photos. It may be that wireless networks today are like the 19th-century American West and they will eventually fill up, but wireless network protocols are designed to prevent that from happening. I may be naively optimistic, but I'm not convinced that the wireless spectrum will be filled.

My second point is that in no way am I claiming that wireless spectrum should be unregulated. FCC regulation controls FM radio and it controls wireless networking. The difference is in the type of control. FM broadcasts are tightly controlled. The FCC regulates who can broadcast, where they can broadcast, the broadcast power, and what they can broadcast. In comparison, the only control on wireless networks is the power level, and the same frequency range is used for a variety of devices, including portable phones and interference from microwaves. It may be that in order to maintain the wireless network frequency range as a nonrivalrous commons, the FCC may need to establish further regulations, but I hope it will never turn into FM radio style regulation.

The fact that radio spectrum depends on regulation to be nonrivalrous does distinguish it from information, which is inherently nonrivalrous. The licensed spectrum of FM radio is clearly not the only or the best way to regulate spectrum. I don't want to be too quick to assume that all ways of utilizing spectrum will necessarily result in all spectrum being filled. Analogously, current copyright laws are not the only way to regulate information and other forms of regulation may have different advantages.

2003-09-29T14:55:00.000-04:00

Brian Carver is promoting "sticky" copyright. One of the characteristics of copyright is that it is transferable. The music industry takes advantage of this to require musicians to assign copyright in their recordings to the record company. Brian argues that if copyright is intended to reward the creators, they should retain copyright in their own works and copyright should not be transferable. He goes on to consider some possible objections and his defenses.

2003-09-29T14:30:00.000-04:00

P2P United is finally up and running. P2P United is an industry association of peer-to-peer software developers, including Grokster, Morpheus, and other smaller companies. P2P United has issued a press release encouraging negotiations between all stakeholders on file sharing and copyright. (from Boing Boing)

2003-09-29T13:23:00.000-04:00

I have previously used The League of Extraordinary Gentlemen as an example of the reuse of the public domain. In a somewhat bizarre followup to that story, Fox has been sued for copyright infringement over the movie. Martin Poll and Larry Cohen claim that the movie is based on Cast of Characters, which they originally proposed to Fox in 1993. Like The League of Extraordinary Gentlemen, Cast of Characters features a team of Victorian era heroes battling Victorian villians. Fox claims that the movie is based on the comic book by Alan Moore, but Poll and Cohen state that the movie has more in common with their movie proposal than with the comic book.

Like the copyright infringement suit over the movie Underworld, I don't have enough knowledge of either the original work or the claimed infringing material to be able to judge the claims. I think it would be pretty stupid of Fox to state that they were making a movie based on one source while actually using another, but given how the movie turned out, expecting Fox to not be stupid might be giving them too much credit. In any event, this suit is probably worth following.

2003-09-29T12:40:00.000-04:00

Kerry Gonzalez has been sentenced to probation and $7000 in fines and restitution for copyright infringement. Gonzalez released an unfinished version of The Hulk on the Internet prior to the release of the movie. He pleaded guilty to copyright infringement after the file was tracked back to him. The arrest and fine has been celebrated by the movie industry.

2003-09-24T12:08:00.000-04:00

Is anyone tracking the outcomes of the lawsuits the RIAA has filed? Slashdot has a link to a story about the RIAA suing the wrong person. The RIAA accused the woman of of sharing more than 2000 songs using Kazaa. She insists that she has never shared or downloaded any songs, and beyond that, she has a Mac and Kazaa isn't available for Macintosh. The RIAA has withdrawn its suit. Anecdotal evidence on the lawsuits has been accumulating. Is anyone tracking the numbers on the outcomes of the suits? Information on the demographics of the lawsuit targets and the outcomes (number of settlements, cost of settlement, etc.) would be useful for evaluating and responding to the lawsuits.

2003-09-19T16:50:00.000-04:00

Yesterday I attended Digital Media in Cyberspace, a conference held by the Berkman Center to discuss copyright and the Internet. The conference is a follow up to the white paper "Copyright and Digital Media in a Post-Napster World," which lays out where we are today. "Five Scenarios for Digital Media in a Post-Napster World," which suggests future possibilities was distributed at the conference. To my knowledge, it is not yet online. The conference featured speakers from a variety of viewpoints discussing the merits of the five scenariios in the paper. The conference was interesting but I think ultimately disappointing.

The major disappointment was in the people not present. Because of Hurricane Isabel, some attendees were not able to make it, including some of the speakers. Most conspicuous by their absence were Cary Sherman, President of the RIAA, and Fritz Attaway, VP of Government Relations and General Council for the MPAA. If I were more conspiracy minded, I might suspect that they arranged for the hurricane deliberately to avoid coming. Many of the speakers indicated that changes are unlikely to happen without the support of the recording industry and movie industry, so holding discussions without them seems like a waste of breath. I'm very interested in their take on compulsory licensing. Are there any conditions they would want to apply to that sort of plan? Are they completely opposed to that approach? The major means of communicating they seem to employ is lawsuits, which I hope isn't the only message they have on this.

The conference presented five scenarios for moving forward. While reading the descriptions, I had the sense that although they spoke of five interested parties, creators, publishers, technology companies, ISPs, and the public, the proposal was written to favor the publishers, tech companies, and ISPs. Part of that is the language used. The proposal refers to the public as consumers, which I think minimizes the role of the public in these discussions. The public isn't just the people who will purchase the products of the creative industry (regularly referred to as "content," another word I have issues with). Theoretically, copyright serves the public. This idea is contained in both the Constitution and the Copyright Act. Copyright is a tool to increase creative productivity, because more creation is in the public's interest. I also think it's important to remember that the line between the public and the creators is a thin one. While they have different interests, people move fluidly between the two categories, which implies to me that the distinction isn't as great as it might appear.

I also had the sense that the role of creators was minimized. In these discussions, it frequently feels like there's an assumption that meeting the publishers' needs will guarantee that the creators' needs are also met. I'm going to borrow an analogy from Scott Adams and say that musicians are to the music industry as minks are to mink coats. Assuming that the publishers have the creators' needs at heart is clearly not valid. I may have been reading this into the paper. It's an attitude that I expect to see, but the paper does try to distinguish between the two groups.

Thankfully, comments by the speakers and audience members at the conference did not exhibit those biases. The interests of creators and the public were well represented. The underrepresented group appeared to be the publishers, as I already indicated.

The five models are no change, property rights, tech defense, public utilities, and compulsory licenses, which were regularly referred to as "alternative compensation." I'm not sure why the other name was used. Perhaps alternative compensation sounds less scary than compulsory licensing.

No change is, as the name suggests, continuing to do the same things that we're doing today. We could look forward to continued rampant illegal online distribution, lawsuits, and the general decline of the music industry, with the expectation that other industries, including movies and print, would follow. I think it's safe to say that no one wants this.

The property rights model assumes that the lawsuits work or attitudes otherwise change so that distributing copyrighted materials online ceases. The music industry would like this, but I don't see why that would ever actually happen.

The tech defense model assumes that actual working DRM will be developed. Again, the popular opinion is that this is very unlikely to come to pass.

I wish the public utility model had gotten more play. The idea is basically to trade increased industry control of copyrighted material for increased regulation. Public television, public radio, and the BBC were presented as possible models. Unfortunately, this idea had practically no support. Concerns about the impact of government involvement in creation pretty much killed the idea. Despite the concerns I have about that, I would have liked to see more objective discussion of the plan.

Compulsory licensing would scrap the current copyright system. Instead, taxes would be collected and distributed to publishers based on the amount of online distribution. In exchange, online distribution would be essentially unregulated. These sorts of plans hve been subject to a significant amount of debate, and the debate continued at the conference. One panelist asked who would pay five dollars a month just to make copyright issues go away (and legalize filesharing) and got substantial but not universal support. I'm still not sold on the idea. As I said at the conference, it feels like a bribe. "Please don't sue and eat us. Just take some money and go away." Beyond that, there are substantial practical problems with this sort of plan. William Fisher, who has proposed his own compulsory licensing scheme, recognizes that government opposition to this idea means that it is unlikely to come to pass without voluntary systems as test cases. He suggests that perhaps a voluntary system might ultimately be better than a government controlled system.

I can't help feeling that everyone says that they don't want the current system, but as soon as a new system is proposed everyone reflexively attacks it and knocks it down. If we can't agree to a different system, we're going to be stuck with the current system whether we like it or not. I don't know how to get away from that short of a complete industry meltdown. In the meantime, I'm reminded of the definition of insanity as continuing to do the same thing and expecting different results. The idea of success coming from either legal or technical solutions alone seems far-fetched, but people seem to think that maybe some combination of the two will. Can anyone explain to me how using two things that don't work together is supposed to create a working solution?

I guess I've gotten pretty pessimistic. How are we supposed to start doing things differently if not everyone participates in talks about it and any suggestions of changes get shot down? This is the world we are living in. It is also the world we are stuck in.

Derek Slater was also at the conference and has posted his comments. Somehow, in a room of 100 people we ended up sitting next to each other and so I finally got a chance to meet him, which was good. News reports on the conference include Information Week, AP (at the Seattle Post-Intelligencer), and Business Week. Dan Kennedy of the Boston Phoenix was there, but I don't see an article yet. I will add other commentators to this list as I find them.

2003-09-17T19:43:00.000-04:00

I've been doing some more thinking about nonrivalrous goods. This is basically me thinking out loud and I don't really know what I'm talking about, but that hasn't stopped me before.

The divisions of rivalrous and nonrivalrous goods and excludable and nonexcludable goods have a built in bias towards private property and economic determinations of value. The whole structure seems geared towards private property as the default with the other categories for cases when private property breaks down. For example, private ownership of roads would be unworkable, so roads become the property of the government and are nonexcludable.

Similarly, the whole framework favors a pure economic analysis, which isn't necessarily how things are considered in the real world. Economic theory predicts that public (nonexcludable) land will lead to a tragedy of the commons, as everyone attempts to maximize their own use of the land to take advantage of the available economic utility. The predicted consequence is that the land will be destroyed by overuse. Here in the real world, public parks are valued precisely because their utility is not maximized. Nearly everyone except Objectivists appreciate public parks even though the measurable utility of the park is near zero.

Music, and information in general, is nonrivalrous. Eugene Volokh developed the example of the well to demonstrate why music should be treated as property. In his response, Lawrence Solum argues that a well is a club good, not a toll good. That is, the well isn't purely nonrivalrous in the same way that music is. Even in a hypothetical example of a nonrivalrous well, we are influenced by our expectations of rivalry (rivalrousness?) and so the analogy to music doesn't hold.

I started thinking about other examples of purely nonrivalrous goods but I didn't come up with a very long list. Roads are clearly rivalrous as indicated by traffic jams. Water and air are rivalrous because of pollution. Even sunlight is rivalrous. If someone sets up a really huge solar panel array, whatever is on the other side will no longer have access to the sun.

The first example I came up with is gravity. Gravity affects everyone and everything, whether we want it to or not. Well, that's just great. Gravity clearly isn't property. There's also no possible way to make gravity excludable. Copyright essentially makes music excludable, so it's hard to draw any parallels between the two.

The second example that came to mind is radio spectrum. Spectrum as licensed by the FCC for radio and television is both rivalrous and excludable, but wireless computer networks are neither. Multiple networks can operate in the same area and can generally additional computers can be added to one of the networks without the networks significantly interfering with each other. Theoretically, there are maximum limits to the number of networks and computers that can be set up in one place, but within my practical experience there are no real limits. My expectation is that the more spectrum is used for point to point communication rather than broadcast communication and the more spectrum which is licensed for unrestricted use becomes available, the more natural it will be to think of spectrum as nonrivalrous.

The behavior of wireless network operators is worth studying. Wireless networks can be set up to be either public (nonexcludable) or private (excludable). A substantial number of networks are public because there's little reason to create a private network. Economics predicts that this will limit the number of networks created, because people will leech off of existing networks rather than create new ones, but the evidence is that anyone who wants access to a network creates their own.

Beyond that, it's worth considering networks which require payment for access. Businesses such as McDonald's have been creating networks and charging for access, but many have been giving free access instead. The cost of a system to charge for access is high enough that many places are deciding to give free access and assuming that the goodwill created by the network will pay for the network. It's not yet clear whether paid access or free access will win out. But one thing is clear. Charging for spectrum (making it excludable) decreases the utility of the spectrum by increasing the costs. This is obviously true in the rivalrous world of radio and television, as indicated by spectrum auctions, but it is also true for nonrivalrous use.

I'm hesitant to make too strong an analogy between wireless networks and music. The costs of a wireless network are low and the costs of music creation are much higher. However, it's clear that for nonrivalrous goods, making the good excludable increases the costs of the good. Copyright makes the cost of creating music and books higher. It's worth asking whether the benefits of copyright exceed the costs, but I'm uncertain of how to measure that.

2003-09-16T17:18:00.000-04:00

Legal Theory Blog has a fascinating analysis of arguments for copyright as property. This is part of an ongoing discussion between Lawrence Solum and Eugene Volokh which I mentioned on Sunday. Volokh had proposed the metaphor of a well as a justification for considering copyright as property. Solum reconsiders the differences between rivalrous and nonrivalrous and excludable and nonexcludable goods.

I found this valuable just for the terms common pool goods and toll goods, which I knew probably existed, but which don't get much discussion. Common pool goods are rivalrous but nonexcludable. The tragedy of the commons is a description of the predicted outcome of common pool goods. Toll goods are excludable but nonrivalrous. Copyright essentially converts information from a public good into a toll good. Now that I know what they're called, I can look for more information on toll goods. I'll start just by asking for other examples of toll goods. Anyone have any pointers?

2003-09-16T14:53:00.000-04:00

The Free Expression Policy Project has published "'The Progress of Science and Useful Arts': Why Copyright Today Threatens Intellectual Freedom" on the conflict between free expression and copyright law. The report includes good details on the Copyright Term Extension Act, the Digital Millenium Copyright Act, filesharing, and the lawsuits surrounding each. (from Sivacracy.net)

2003-09-16T13:11:00.000-04:00

Zug has transcripts of a series of calls and letters to the RIAA, Apple, record companies, and individual bands as the author attempts to pay for an illegally downloaded MP3. This is distinctly on the lighter side of the copyright debate. (from Boing Boing)

2003-09-16T12:51:00.000-04:00

Aaron Swartz proposes using cryptography to fix the problems with compulsory licensing. He states that associating cryptographic keys with the taxable products and services used in filesharing would ensure anonymity while allowing accurate counting of media use and limiting abuse of the system. I'm willing to believe that this would be a workable solution, but it fails to address the economic issues of filesharing.

2003-09-16T12:44:00.000-04:00

Verizon is appealing the ruling supporting the RIAA subpoenas under the DMCA. The original ruling was a key step in the RIAA's current campaign to sue everyone. The article reports that Sen. Brownback is planning to introduce legislation that would require a pending lawsuit before the the subpoena could be filed. In related news, The New York Times reports that SBC is continuing to refuse the subpoenas. The article reports that SBC is the only major ISP holding out, although executives at other ISPs are also unhappy about the subpoenas.

2003-09-16T12:22:00.000-04:00

The Senate has voted to roll back the FCC regulations on television and cross-media ownership. The rollback proceeds to the House and then President Bush, who has threatened to veto the legislation.

2003-09-15T18:06:00.000-04:00

A Copyfighter's Musings is thinking about the distinctions between commercial and non-commercial uses of copyrighted works. He points to a paper by Ernest Miller and Joan Feigenbaum which is definitely worth a read. It points out that the word "copy" in "copyright" is in the sense of "text" rather than "reproduction." This is a distinction that has been unimportant until now, but digital technology employs copying at its root, giving copyright holders control over copying leads to all sorts of negative consequences.

Instead, the focus of copyright should be reconstrued as public distribution. This would restore the original intention of copyright and correct current problems with fair use, first sale, and the DMCA on the Internet.

I think there's a lot of merit to this proposal. In a comment on LawMeme, I cast doubts on the ability to distinguish between public and private distribution, but it is more appealing to me than commercial vs. non-commercial uses. One thing I'm concerned about is the fact that the Internet seems designed to facilitate public distribution. I'm not wild about saying to people, "Here's a tool, but if you use it, you'll be sorry." Unfortunately, the other alternatives are compulsory licensing and giving up on copyright on the Internet entirely. Neither is very palatable.

2003-09-15T12:46:00.000-04:00

bIPlog presents a new model for the music industry. When I started this blog, my plan was to work on something very similar to what Mary is suggesting. Then I didn't make progress on it the way I intended to and got distracted by writing about copyright and it's all been downhill ever since. But I still think the idea is a good one. The way bands should make money is concerts and merchandise (counting recordings as merch.) and labels should be in the business of promoting the bands. Free (or near free) music on the Internet should be a promotional tool, not competition for the primary business of the record label.

2003-09-15T12:26:00.000-04:00

Among all the Times articles yesterday, I missed one. "File-Sharing Battle Leaves Musicians Caught in Middle" (cache) looks at the response of musicians to the lawsuits.

2003-09-14T23:00:00.000-04:00

One more thing for today. The FCC has issued a press release (in PDF format) on the new digital television rules. There's no mention of the analog hole or the broadcast flag. The rules set limits on copy protection. Pay per view and video on demand signals can block any copying, cable service must be copiable once, and broadcast television must have no copy protection for signals transmitted through the interface the rules specify. The release states that "the FCC also noted that it will address Digital Broadcast Copy Protection issues in the near future." Digital Broadcast Copy Protection is the official name for the analog hole and the broadcast flag.

2003-09-14T22:43:00.000-04:00

The New York Times has been extensively discussing copyright lately. "Suing Music Downloaders" (cache) is an editorial from Friday which states that the music industry needs to change its business model if it wants to survive. In the Week in Review, Steve Lohr writes that "Whatever Will Be Will Be Free on the Internet" (cache). He compares the record industry's battle against file sharing to whack-a-mole and mentions Jonathan Zittrain of the Berkman Center, Lawrence Lessig and Creative Commons, and William Fisher's compulsory licensing plan. The related article, "The Music Industry Reveals Its Carrots and Sticks," (cache) discusses the RIAA lawsuits and amnesty offer. In "Beyond File-Sharing, a Nation of Copiers," (cache) John Leland ties file downloads to plagiarism, counterfeits, and Enron. In tomorrow's paper (already online today. The Internet is so nifty.), "Hollywood Faces Online Piracy, but It Looks Like an Inside Job," (cache) points out that many times when movies end up online, it's the result of an internal leak. And "Crackdown May Send Music Traders Into Software Underground" (cache) points out that file sharers are likely to just switch to services that are hard to track in response to the lawsuits.

I'm actually less interested in the content of these articles than the sheer quantity of them. I guess hundreds of lawsuits will do that. But the more press coverage these issues get, the more general awareness will increase, and that can only be a good thing.

2003-09-14T21:32:00.000-04:00

For those who can't get enough of economics and copyright, Legal Theory Blog points to "Deprivatizing Copyright" by Shubha Ghosh. This is long, but it undertakes a significant rethinking of copyright so I hope I can adequately summarize it in a few sentences.

Economics has traditionally viewed copyright through public good theory. Ghosh argues that this is inadequate because it frames copyright strictly in terms of private interests and ignores the public interests in copyright. Instead, copyright should be considered through democratic governance. This recognizes the government interest in creating and maintaining culture as part of democratic society. In this view, copyright is neither a property right nor a government regulation, but a privatization of government power. Ghosh then evaluates several areas of current controversy in this framework, including model code, religious works, the DMCA, the limits on which works are copyrightable, fair use, and the conflict between the First Amendment and copyright. Ghosh generally concludes that current laws and rulings give copyright too much power, because they fail to serve the public interests behind copyright.

I'd be interested in seeing an extension of this work to include file sharing. After one reading, my grasp of the theory isn't strong enough to be able to immediately apply it to file sharing, but I wonder what it would have to say.

2003-09-14T16:16:00.000-04:00

Unintended Consequences points to a strong attack on compulsory licenses. Stan Leibowitz argues that compulsory licensing poses many economic problems that make it a poor solution to file sharing. For example, pricing is a major difficulty for ensuring market efficiency, but compulsory licensing completely disconnects the price from the demand, guaranteeing that pricing will be worse.

Of course, his solution is DRM, which I have far less support for.

2003-09-14T15:06:00.000-04:00

The question of intellectual "property" has come up again. This time around the focus is on what constitutes property in the first place. The Volokh Conspiracy defines property in terms of the right to use and the right to exclude. Legal Theory Blog responds, taking issue with Eugene Volokh's example of the well. The well (or the water from the well) isn't a public good, as Volokh asserts. Instead it is a club good, in that there exists an optimal number of users. It is therefore distinct from intellectual property, which is a public good. On the whole, Solum approves of Volokh's post.

FurdLog presents a different argument for property. Property, he says, is a way to internalize transaction costs and allows the market to function. Physical property therefore exists to reduce costs, but the entire basis for intellectual property is increasing costs. By now, I'm swimming in economic jargon. As I've summarized it, Field's argument seems circular, but I'm certain that I don't actually understand it, so I can't really evaluate its validity.

2003-09-14T14:43:00.000-04:00

Apparently TechTV ran a program called "The Music Wars" yesterday. Wish I had caught it. Fortunately, it looks like it will be rebroadcast on Monday and Tuesday. Until then, comments from various other bloggers will have to suffice. bIPlog has a description from in the audience and Bag and Baggage has a collection of quotes.

2003-09-14T14:11:00.000-04:00

Here's a piece of news that I don't want to report, but I feel obligated. BMG is planning to release a CD with copy protection later this month. A significant number of recordings have been released in Europe, but this is largely uncharted territory in the US. I can only hope that there will be a sufficient consumer backlash to deter them from continuing the experiment.

2003-09-13T13:05:00.000-04:00

I want to further develop some thoughts from yesterday. Suppose I were to make 1000 CD copies of the new Metallica album and give them away. I think most people would view that as a pretty clear case of copyright infringement, even if I wasn't charging for the copies. It's hard to look at that sort of mass copying as anything other than an attempt to replace the sales of Metallica's album. Furthermore, I would expect that if I did that on a regular basis, it would be only a matter of time before I became a target of a lawsuit from Metallica, their record company, or the RIAA. I would deserve whatever punishment the legal system doled out.

Compare that to making a single copy of the Metallica CD and giving it to a friend. Most people consider that reasonable behavior. The recording industry disagreed, but at the same time there wasn't really anything they could do about it legally. Single copies are too small scale for the recording industry to have a chance of catching it. Beyond that, it's too small for the punishment to justify the enforcement effort economically. Recording industry frustration with individual copying led to the Audio Home Recording Act of 1992, which essentially granted permission for personal copying in exchange for certain restrictions.

The AHRA has basically failed because it was too limited in scope. It regulated digital recording media, leading to the development of such marginally successful recording media as DAT, MiniDisc, and music CD-Rs (as distinct from data CD-Rs, which have become wildly successful, both for music and for general data storage). The intention was to permit personal copying within certain limits while guaranteeing payment through media taxes to the recording industry. Digital recording devices were defined in a way that excludes computers from the scope of the law. From the standpoint of the computer industry and users, that's a good thing. From the standpoint of the music industry, it's a disaster because computers are now used for unregulated wide-scale copying.

This leads us to lawsuits over file sharing. From the perspective of users, file sharing resembles individual copying, since all transactions occur on the scale of individual copies. From the perspective of the recording industry, file sharing resembles commercial copyright infringement, because a single file can be copied thousands of times, replacing record sales. The recording industry has successfully pressed Congress to view file sharing as comparable to commercial infringement, leading to the lawsuits they are now pursuing.

From the perspective of users, these lawsuits are shockingly unfair. This is evident in the statements that targets of the lawsuits are making which state that they thought file sharing was legal and don't understand why the industry is targeting individuals and in the extreme hate that average people are now directing at the RIAA.

It doesn't make sense to ask which perspective is correct. The question is how to resolve the two views. Widescale lawsuits clearly don't do that. Neither does unregulated copying. Perhaps what's needed is a modernized version of the AHRA, something which preserves the right of individual copying while establishing a payment mechanism for copyright holders.

I disagree with Ernest Miller. I don't think the choice is between supporting copyright, and therefore the RIAA lawsuits, and opposing the lawsuits, and therefore copyright. The only approach that makes sense is to preserve copyright while stopping the lawsuits. Doing that may require new ways of thinking about copyright. Hopefully everyone will come to realize that and work together on new approaches.

2003-09-13T12:16:00.000-04:00

I just thought I should point to Penny Arcade's response to the RIAA's amnesty offer.

2003-09-12T20:40:00.000-04:00

LawMeme points to a pair of articles on file sharing at Salon. Scott Matthews, a software developer, is opposed to the EFF's approach to file sharing. Jason Schultz at the EFF has written a response. Back at LawMeme, Ernest Miller adds his own criticisms of the EFF.

The Internet poses scaling problems. The harm of making one copy of copyrighted work is pretty minimal. But when everyone allows anyone else to make a single copy, the overall effect is substantial. Copyright holders see the overall impact, but individuals are only making one copy at a time. Neither perspective is really wrong, but they produce radically different conclusions about the response. I'm not sure the two positions can be balanced against each other, but without a balance I just see unending conflict.

2003-09-12T15:16:00.000-04:00

Ernest Miller asks a really good question. For those of us who oppose the RIAA's lawsuits, when do we think copyright holders should be able to sue? Here are my thoughts. It seems like the penalties for an action should be dependent on two factors: the harm done by that action and the difficulty of that action. The Internet has caused the difficulty of reproducing a copyrighted work to plunge, while the putative harm is still the same. The response of Congress has been to vastly increase the penalties for copyright infringement. I'm not sure that's the right direction to go in. It's as if Congress thinks legal approaches and technical approaches are interchangeable when they're clearly not.

Rather than ask how we can maintain the status quo in the face of changing technology, I think we have to ask what systems make sense given that we have new technology. Suing 60 million people for hundreds of thousands of dollars apiece clearly does not make sense. I'm struck by the fact that the RIAA is settling with people who were sharing thousands of songs for a few thousand dollars, when under the law they could be liable for $150,000 a song. (I'll note that it's not certain that they are liable, because it hasn't yet been explicitly tested. But they could be, and the RIAA asserts that they are.) At the rate of a few dollars a song, it probably isn't worth the effort to sue file sharers, and yet that's the rate that the RIAA seems to think is appropriate. I'm inclined to agree with them. If the value of a song is around a dollar, it makes sense that the penalty for illegally copying a song be on the order of a few dollars.

If the maximum penalty for copyright infringement were, say, $5 a song, I'd be much more likely to approve of the lawsuits. The downside from the RIAA's perspective is that at that rate, it wouldn't be economical to sue infringers. I'm reminded of transaction costs problems and market failure, although in this case the "market" is operating through lawsuits. The solution to market failure is to not depend on the market. If the most the RIAA could gain through a copyright infringement suit against a file trader was $5/song, they'd be screaming for government intervention. They've gotten it too, but not in the form that we, or they, should hope for.

Increasing the penalties of copyright infringement is clearly not the way to go. What we, and they, should be looking for is a way to obviate the need for lawsuits. Selling music online at reasonable prices and terms is the free market solution. Compulsory licensing is government based solution. If there are other options, I'd like to hear them.

2003-09-12T13:44:00.000-04:00

Lisa Rein has written an article about the legality of the RIAA's subpoenas. Although the subpoenas have been filed under the DMCA, the procedure provided by the DMCA may not be legal by the existing standards of subpoenas. If that's the case, the RIAA could be in hot water over the attempt to use the subpoenas. I think it's unlikely that these subpoenas will come back to harm the RIAA, but the article lays out why the subpoenas are a problem. (from Boing Boing)

2003-09-12T13:27:00.000-04:00

The EFF has started a petition to Congress asking for reasonable copyright laws in the wake of the RIAA lawsuits. I'm sceptical of the effect this will have, but I've signed it anyway. Doing nothing definitely won't make a difference, but signing the petition may. The EFF is approaching the half way point of the goal of 10,000 signatures. Take the time to add yourself to the list. (from Boing Boing)

2003-09-11T22:46:00.000-04:00

Here's some more info on reselling songs purchased through iTunes. As the Apple Turns reports that Apple has stated that you are free to resell the file, but not the authorization. That's not much of a sale. It's sort of like selling a book, but not permission to read it. George Hotelling has reported that he succeeded at transferring a song to someone else, only by giving that person the entire account. I have trouble viewing that as much of a success.

2003-09-11T22:32:00.000-04:00

There's been plenty of excitement lately in the RIAA's campaign to sue everyone. On Friday, the RIAA announced that it was offering amnesty to file sharers. On Monday, the RIAA filed lawsuits against 261 individuals, accusing each of them of copyright infringement for making songs available for downloading on the Internet. Of the targets, the one who has gotten the most attention is a twelve year old girl who quickly settled for $2000. The recording industry is also trying to demonize file sharing as a den of pornography. CNet has a roundup of stories on the RIAA which is a good place to start. Other commentary follows.

Let's start with the amnesty offer. The RIAA declared that it would not pursue individuals who it has not already started the legal process against if they stop using file sharing software and delete all downloaded songs. Public Knowledge and the EFF issued warnings against the amnesty program, calling attention to the fact that this would only shield users from the RIAA and could actually make them more vulnerable to other lawsuits. Eric Parke has taken things a step further, suing the RIAA over the amnesty offer.

The New York Times had a trio of articles about the lawsuits. The first focuses on the details of the lawsuits. (cache) The second argues that the RIAA is "Fighting the Idea That All the Internet Is Free". (cache) The third article is on the trouble in the music industry beyond file sharing. (cache) These were followed by article on the fact that it may be children who are file sharing and the responsibility of parents to control this behavior. (cache)

The Times has an article on the pornography angle which indicates that the music industry may be overstating its case. (cache)

MacSlash raises an interesting question. Many users currently believe that file sharing services are legal. If the RIAA convinces them otherwise, will users come to believe that iTunes (or other legal music services) is also illegal? Is the difference between legal online music distribution and illegal distribution big enough to make a distinction for most people?

2003-09-11T21:33:00.000-04:00

Tim O'Reilly has some interesting comments on file sharing and ebooks. It's interesting that while he doubts the success of ebooks, saying that he thinks subscription services are a better model, the most successful music service has been based on individual sales. The subscription services have struggled. Of course, they have had different DRM policies as well, which may make a difference.

2003-09-11T17:24:00.000-04:00

You may have seen ads for the upcoming movie Underworld. It comes across as one in the current generation of Matrix ripoffs, making extensive use of black, shiny clothing and capes, guns, and kung-fu. It's the target of a copyright infringement lawsuit, but not because of the similarity to The Matrix. White Wolf, a publisher of games and books, accuses the movie of using elements from the story "Love of Monsters".

Penny Arcade has the details and commentary on the suit. White Wolf has issued two press releases on the suit. The first is an announcement of the lawsuit. The second states that White Wolf's request for a temporary restraining order has been denied, but they are proceeding to discovery for a hearing on a preliminary injunction. Penny Arcade, which has been the target of trademark infringement claims in the past, is sceptical, noting that Romeo and Juliet comes to mind before White Wolf's games or stories.

I don't know either the story or the movie well enough to pass judgement. I can't judge from the complaint to what extent the similarities are substantive or superficial, and I certainly can't predict what the judge will think. Instead, I'm wondering about how copyright should work. How similar should one thing have to be to another to be infringing. Vampires, werewolves, and Romeo and Juliet have all been around for a while, but does that mean that a combination of the three is unoriginal and undeserving of copyright? Alternatively, does it make all combinations the same, and any story combining the three a derivative of the first story to combine them?

Obviously, neither extreme is defensible. Despite the fact that stories of vampires, stories of werewolves, stories of star-crossed lovers, and stories of vampires and werewolves who are star-crossed lovers have already been written, a new story deserves copyright protection and is not automatically infringing on existing stories. I'd like a firm line, saying this is clearly different enough but that is clearly not. I think the only place that line could be drawn is at exact duplication, but I don't think that's enough. Certainly most publishers would argue that that isn't enough. I don't know.

2003-09-11T15:41:00.000-04:00

The New York Times reports that the FCC has issued new rules on digital television. (cache) Unfortunately, the article doesn't have any details on the biggest question on digital TV in copyright circles, the broadcast flag. Anyone have any more details?

2003-09-11T13:48:00.000-04:00

On this date two years ago, terrorists took control of four airplanes. Two were flown into the World Trade Center, destroying it and killing 3,000 people. One was flown into the Pentagon, doing substantial damage. The fourth was crashed, presumeably on its way to another target. I feel like I can't let this day go unremarked, but at the same time I don't know what to say.

The events of that day were so large that it seems impossible to summarize them or adequately comment on them. The attacks redefined the place of America in the world, and America is today continuing to try to redefine itself. I feel like it's not my place to pass judgement on that. I disagree with many of the things that our government has done in response, but I ask who am I to judge whether a particular response to such a devastating event is appropriate.

I think the answer, if there is one, lies in the ideals this country was founded on. The Declaration of Independence, the Constitution, and the Bill of Rights are more than the basis of the laws and government of this country. They are a philosophy of government, which places the government at the service of the people and not the other way around. The authority of the government is limited and it exists only as long as the people grant it that authority. This is a radical reconceptualization of power, taking it away from the few who have historically used power as a means of control and giving it to the many, who wield the power as both a defense and expression of freedom.

The current members of our government would do well to remember that. It is this freedom and power in the hands of individuals which has empowered the growth and development of this country, but this same power threatens many people, both outside and within this country. The ones who are threatened by it are the ones who have power, control, and authority over others. Granting power to individuals means taking power from the ones in control.

If there is anything the world learns in the aftermath of the destruction of two years ago, I hope it's the importance of freedom and individual power. The American government has not always respected that, internally or externally. The current government is pursuing control above freedom. Hopefully they will come around. If they do not, I hope individual Americans exercise their power to bring them around.

2003-09-09T19:54:00.000-04:00

The Shifted Librarian is asking whether the American Library Association's law firm has a conflict of interest. The same firm represents both the ALA and the RIAA, and there's an obvious conflict between the two sides on some issues.

2003-09-09T19:51:00.000-04:00

Lately my response to some news reports has been to just decide that "that's stupid" and move on. At calmer moments I recognize that that response is inadequate. Case in point: Microsoft is adding DRM to Office. Word, Excel, and Powerpoint files have generally been interoperable between different versions of the software, and despite the fact that Microsoft hasn't published the file formats, other software vendors have reverse engineered the formats to make reasonably compatible software.

Microsoft threatens to bring all this to an end by introducing DRM to their documents. The "good" news is that soon, you won't be able to manipulate Word or Excel files the way you expect to. The bad news is that you will need the newest versions of Office in order to open them at all. There are no upsides. Does anybody want this?

2003-09-09T19:48:00.000-04:00

Garage door openers are not protected by the DMCA. It's mind boggling to me that I should have to write that, because the DMCA is about copyright and garage door openers are intuitively, at least, outside the scope of copyright law. Nonetheless, Chamberlain sued Skylink for violating the DMCA because Skylink manufactured a garage door opener that works with Chamberlain's devices. Other people have analyzed this. I think it's just dumb, and a sign of how wrong the DMCA is.

Other commentary includes Lenz Blog, Freedom to Tinker, A Copyfighter's Musings, and LawMeme.

2003-09-09T19:44:00.000-04:00

At the wacky extremes of copyright law on the Internet, George Hotelling attempted to resell a song he purchased from the iTunes Music Store on eBay. Under the first sale doctrine, a purchaser of a copy of a copyrighted work has the right to resell that copy. This means that although the record industry may not support used CD stores, the right of the stores to resell used CDs cannot be taken from them.

This resale of a song purchased online was a deliberate test of that policy. Legally, Hotelling has the right to resell the song, but the practical meaning is unclear. The copy protection on songs sold by the iTunes Music Store is tied to the original purchaser, and while Hotelling wanted to transfer the copy protection, he likely would not have succeeded.

The sale was interrupted by eBay, who stated that the sale violated its terms of use and pulled the sale. Hotelling is now working on giving the song away.

Other coverage includes Slashdot, Bag and Baggage, CNet, and Ernie the Attorney.

2003-09-09T19:40:00.000-04:00

Universal Music is cutting the prices on its CDs (cache). The list price of their CDs will be reduced from $18.98 to $12.98. The price reduction is in response to falling music sales, which the record industry blames on illegal copying of recordings, including Internet file sharing.

I'm not sure of the impact of file sharing on CD sales, but as a music purchaser, I'm all for lower prices. The article indicates that the other record companies will probably also reduce their prices. If this leads to higher sales, as it should, this is clearly a win-win situation. Slashdot has additional comments.

2003-09-09T19:33:00.000-04:00

Wired News reports that the Appeals Court has blocked the new FCC media ownership rules from taking effect. The new rules would allow a single company to own more television stations than under existing rules and also allow companies to own both television stations and newspapers in the same market. The new rules set off a firestorm of protest when they were first proposed and both the Senate and the House have proposed legislation to overturn at least some of the new rules. This court ruling effectively adds another front to the battle. The New York Times also has coverage. (cache)

Online Journalism Review has a related interview with Michael Powell. (from JD's New Media Musings) The interview doesn't touch on the ruling itself but Powell discusses the issues behind media deregulation. Suffice it to say that while Powell is certainly intelligent, some of his comments completely miss the point.

2003-08-29T11:01:00.000-04:00

The EFF has published Jane Doe's Motion to Intervene (as a PDF) in RIAA v. Verizon. Jane Doe is the target of one of the subpoenas the RIAA has filed to learn the identities of file sharers. Doe argues that although the subpoena is against Verizon, she has a personal interest in quashing the subpoena because it is her personal information that will be revealed if the subpoena is followed. Good stuff. (from Joho the Blog)

2003-08-29T10:28:00.000-04:00

Wired has an article by Bruce Sterling which repeats all the worst exaggerated fears of the changing value of information on the Internet. It's disappointing to see this sort of thing coming from a well-known (fiction and nonfiction) technology writer.

2003-08-28T13:11:00.000-04:00

Derek Slater has returned from his vacation. He brings further discussion on the legality of file uploading and downloading with him. He and I have already pretty thoroughly discussed this, as indicated by the links he includes, so I'm not going to take up the substance of the discussion again. I'm just going to point out that the ultimate legal disposition of this issue depends on which metaphors and analogies the courts choose to accept. Until we get a definitive ruling, it's still up in the air. If there's any good to come out of the RIAA's lawsuits against individual file sharers, it will be in the courts directly addressing this question.

2003-08-27T14:32:00.000-04:00

The New York Times has a tribute to Jack Kirby (cache) that demonstrates the wheel of creativity. Jack Kirby created many classic comic book characters including the X-Men, the Incredible Hulk, the Fantastic Four, and Captain America, for starters. The article traces his influence on movies from Star Wars to The Matrix, in addition to the movies based on his characters which appeared this year. At the same time, the article notes the influences that movies had on his comics. This is creativity in action. Taking something old to make something new, so it can in turn be used by the next creation.

2003-08-27T14:13:00.000-04:00

More on Bunner: FurdLog presents his analysis of the ruling. He is concerned by the ruling's analysis of whether the injunction is content based or content free. (That should really be "content independent." The point isn't that there's no content in the enjoined speech, but that the the injunction is independent of the content of the speech.) I don't quite follow Frank's thinking here. He is focused on the fact that DeCSS is both speech and "meta-speech" because it provides access to other speech, and says that this places it in a legal blind spot, but I feel like I don't grasp what his point is.

Infothought raises a good point about reverse engineering and trade secrets. Trade secrets are explicitly not protected against reverse engineering. Does a shrinkwrap license prohibiting reverse engineering protect trade secrets? The DVD CCA seems to be arguing that it does, but at least one judge in the case disagrees. Judge Moreno states that

nowhere has it been recognized that a party wishing to protect proprietary information may employ a consumer form contract to, in effect, change the statutory definition of "improper means" under trade secret law to include reverse engineering, so that an alleged trade secret holder may bring an action even against a nonparty to that contract. Moreover, if trade secret law did allow alleged trade secret holders to redefine "improper means" to include reverse engineering, it would likely be preempted by federal patent law, which alone grants universal protection for a limited time against the right to reverse engineer.

In other words, even if it was accepted that trade secret protection could be extended by a contract, it would violate patent law. I wonder how far this would go toward invalidating reverse engineering provisions of software licenses in general.

2003-08-27T13:53:00.000-04:00

GrepLaw has an interview with Glenn Peterson. Peterson represents a target of the RIAA subpoenas who is fighting the constitutionality of the subpoena. The interview indicates Peterson's legal strategy and shows how unreasonable and potentially harmful this subpoena power is.

2003-08-26T20:18:00.000-04:00

It tends to suck when all of your applications "unexpectedly quit" at the same time. Especially when you've been collecting links all day in a file you haven't saved. But that's my problem. On to DVDCCA v. Bunner.

The ruling is online (as a PDF). CNET has a story. The EFF has a press release. Blogs with coverage include Slashdot, Lessig Blog, FurdLog, bIPlog, Dan Gillmor's eJournal, Bag and Baggage, Tech Law Advisor, Scrivener's Error, Infringing Actions, Freedom to Tinker, and The Volokh Conspiracy.

Before I get involved in the details of the ruling, I want to review the context. Commercial DVDs have a form of encryption called CSS (Content Scrambling System). CSS doesn't prevent the DVDs from being duplicated. It prevents them from being played on unlicensed players. Part of the purchase price of an DVD player is a licensing fee paid to DVDCCA for the software to decrypt CSS. A group of people in Europe reverse engineered a DVD player and wrote a program called DeCSS to decrypt (and watch) DVDs without a licensed DVD player. DeCSS was released on the Internet and promptly spread everywhere.

There have been three significant lawsuits regarding DeCSS. The first was a lawsuit in Norway against Jon Johansen, one of the people involved in the reverse engineering effort. He was acquitted, but the case is being appealed. The second was against Eric Corley for publishing the code for the program (and later links to the code) on the 2600 Magazine website on the grounds that DeCSS is a copy protection circumvention device restricted by the DMCA. Corley argued for a First Amendment defense on the grounds that he is a journalist and the code is protected speech. The court ruled against him and required him to remove the code and links to other sites that listed the code.

Like Corley, Andrew Bunner posted the code for the software on a website. In this case, the DVDCCA sued Bunner under trade secret law. The district court issued a preliminary injunction against Bunner which was overturned by the Appeals Court on First Amendment grounds. The DVDCCA appealed to the California Supreme Court, which leads us to the present ruling.

The question being addressed is whether a preliminary injunction against publication of a trade secret violates the First Amendment. The first issue is whether computer code is speech. The California Supreme Court agrees with other rulings which have held that it is speech, and therefore is entitled to First Amendment protection.

In analyzing whether trade secret law and injunctions against publishing trade secrets violates the First Amendment, the ruling is quite eager to establish information as property and cites a significant number of other rulings which draw the same conclusion. This makes me nervous. I am somewhat undecided on the merits of information as property, but the court definitely is not. I am concerned that this will lead the conclusion that the holder of a piece of information is entitled to the same rights with respect to that information as the holder would have with respect to a piece of physical property, which definitely would be going too far.

There's some definite doublespeak going on with regards to trade secrets. Trade secrets are said to prevent the fragmentation of scientific research. Unless I'm missing something, restricting access to a piece of information to a select group itself is a fragmentation of knowledge, so I'm not sure how that restriction is supposed to reduce fragmentation.

The next issue is the relative speech rights of information of "public concern" vs. "private concern." This makes me nervous a lot. What's the difference? I have to somewhat grudgingly admit that the court seems to be drawing a reasonable distinction in this case.

The court ruled that the First Amendment does not prohibit injunctions under trade secret law, but it leaves open the question of whether CSS is indeed a trade secret. It remands that question to the Appeals Court for reconsideration. Judge Moreno would go a step further and rule that the injunction is a prior restraint and the preliminary injunction should be overturned unless the DVDCCA is likely to win at trial. He then goes on to conclude that CSS cannot be considered to be secret because of its widespread availability on the Internet and therefore the preliminary injunction should be overturned.

So the next step is to return to the Appeals Court. The crazy thing is that this is still just the preliminary injunction stage. Regardless of whether the preliminary injunction is overturned, the case will return to the district court for a decision on the case, and then the whole appeals process can start again. The case started in 1999. Who knows when it will end?

2003-08-26T12:36:00.000-04:00

FurdLog has info on MIT's response to file sharing subpoenas. MIT had contested a previous subpoena because it was issued by a Washington court. The RIAA has stated that it will reissue the subpoena from a Boston court, and MIT states that it will comply with the new subpoena.

2003-08-26T12:25:00.000-04:00

bIPlog has a Technology Review article on "The Customer as Enemy". As the article says, this mentality is perverse and will lead to customers not adopting new technologies. In an ideal world, everyone wins scenarios would be developed. Instead we get everyone loses approaches.

2003-08-26T11:06:00.000-04:00

The California Supreme Court has ruled that DeCSS is a trade secret and not eligible for First Amendment protection. DeCSS is a program that cracks the encryption on DVDs, which is known as the Content Scrambling System or CSS. DeCSS was developed in Europe, in part by Jon Johansen, by reverse engineering DVD player software. Andrew Bunner posted the code for DeCSS to a website and the DVD Copy Control Association sued to force him to remove the code.

The DVDCCA asserted that DeCSS was a trade secret. Bunner's defense was that computer code is speech protected by the First Amendment. Bunner lost the initial court ruling, won at the Appeals level, and has now lost at the California Supreme Court. I'm not certain what possibility there is for appeal at the federal level. The ruling is based on the Uniform Trade Secrets Act, a state law.

I'll have more as I learn it.

2003-08-25T15:29:00.000-04:00

I'm trying to figure out how to respond to TeleRead's comments. The big issue is that I think we're talking past each other, which drives me crazy. I think the most I can do is point out some differences between his thinking and mine, and after that I have to let it go.

First, I should identify an error that I made. In my first post on TeleRead, I incorrectly stated that TeleRead is a voluntary payment scheme for online content. That is not correct, as David Rothman has since made clear to me. Payment is not voluntary, but publisher participation is, which makes the plan distinct from compulsory licensing schemes. I had intended for my more recent post to correct my previous description of the plan, but apparently that wasn't clear.

Second, David and I are focusing on entirely different problems. David's first concern is increasing access to books. TeleRead's goal is to create a mechanism to guarantee that books are available to everyone, regardless of their financial situation. My concern is that copyright holding corporations want to prevent access to their copyrighted works, except on their terms. As far as I know, book publishers aren't saying, "We wish more people were reading our books, but we just don't have a way to give them access." There are, however, plenty of companies trying to take away access to files on the Internet. They aren't really succeeding, but they're doing lots of damage as they try.

Third, I criticized the TeleRead plan for being vague. To me, it makes the plan unclear, which makes it harder to support. David presents that vagueness as flexibility.

Fourth, we have different attitudes about DRM. As I see it, there are two kinds of files. There are files which anyone can manipulate in any way, and there are files which can only be used by certain people (or certain computers) in certain ways. Any restrictions on how a file can be used are a form of DRM. Any scheme which proposes either payment or tracking for file downloads has a choice. The files can be unrestricted, in which case there's no way to guarantee accurate payment or tracking, or the files can be restricted, in which case the publisher needs to decide on the restrictions and the implementation. I see these decisions as central to any Internet distribution scheme. TeleRead glosses over them.

I could respond to the specific points that David brought up in his response, but I think that would be missing the point. Instead, I think the differences in how we choose to frame issues of online distribution outweigh the resulting differences that we have on the issues themselves.

2003-08-25T13:19:00.000-04:00

I've gotten a few responses to posts from last week which I should acknowledge. I criticized "The Escalating Copyright Wars" for not sufficiently describing nonzero-sum approaches to copyright. Peter Yu wrote to me to mention another paper he has written, "Toward a Nonzero-sum Approach to Resolving Global Intellectual Property Disputes: What We Can Learn from Mediators, Business Strategists, and International Relations Theorists", which discusses the merits of the nonzero-sum approach. He also indicated that he is working on a followup article specifically on the music industry response to file sharing.

I was fairly aggressive in my criticism of TeleRead. David Rothman has responded on the TeleRead blog, in which he interprets the phrase "rolls a hoover" in somewhat different fashion than it's been used previously. That's fine, of course. I certainly can't claim exclusive rights on the phrase. I expect to have more on the substance of his comments later.

2003-08-25T12:51:00.000-04:00

Wow. The BBC is going to put its archives online and make its television and radio programs available as free downloads. This is a radically different approach than most American television companies, which are afraid that their programs might end up online as free downloads. Maybe the American companies will learn something from the BBC. Oblomovka and Slashdot have more information. Oblomovka also has the text of the announcement and more links. A comment on Slashdot mentions the similar CBC Archives. (from Boing Boing)

2003-08-23T21:42:00.000-04:00

Peter Yu has written "The Escalating Copyright Wars", which makes the point that changing laws and changing technology have changed copyright from something that doesn't have a direct impact on most people to an open conflict between the entertainment industry and everything else. Yu states that the entertainment industry's approach to copyright issues has "become increasingly futile, disorganized, and counterproductive." He points out that "no country has ever won a war by fighting battles on all fronts" and suggests that the industry become more focused in its response.

Unfortunately, his suggested approaches aren't as innovative as could be hoped for. The first suggestion is to "bridge the copyright divide" through an education campaign. Perhaps if people understood why they have a stake in copyright, they would stop using file sharing services. Tied to that, he encourages tie-ins like password protected websites and direct competition against pirates by increasing the quality of the products. He also recommends increased enforcement, particularly internationally. Finally, he suggests cheaper legal alternatives. It may be clear from these suggestions that he is mixing international and commercial piracy with personal copying, just as the industry has done.

The other big approach he suggests is a nonzero-sum approach to copyright issues, but he is vague about what that would mean in practice, and no sooner does he suggest it than he starts talking about the difficulties of that approach. (from Legal Theory Blog)

2003-08-23T13:24:00.000-04:00

Last night I danced with the public domain at Johnny D's in Davis Square. Brave Combo played a variety of songs including such classics as "The Habañera Twist" and the "Rondo a la Turk Polka." "The Habañera" was written by Georges Bizet for the opera Carmen and the "Rondo a la Turk" is a familiar piece by Mozart. The versions we heard last night weren't quite the same as the originals, but it didn't matter because the place was jumpin'.

I suspect that most of the people at the concert last night didn't appreciate the role of the public domain in their enjoyment of the concert. That's because the public domain doesn't go out of its way to call attention to itself, or even label itself at all. But the truth is that the public domain is all around us and touches us constantly. You just need to know to be looking for it. I can't avoid pointing it out. Perhaps if more people were aware of its importance, we'd have better balanced copyright laws.

2003-08-22T14:15:00.000-04:00

Here's a story from the Washington Post (PDF, from FurdLog), nearly every detail of which is shocking. The World Intellectual Property Organization, which isn't known for taking limited positions on intellectual property, agreed to host a conference on open and collaborative business methods. This conference included a panel on open source software. The US Patent and Trademark Office opposed the meeting on the grounds that open source software is opposed to intellectual property and the WIPO supports intellectual property. And the conference has been cancelled.

It's surprising that the WIPO even planned to have this conference in the first place, but in a good way. The WIPO has a reputation for absolute copyright, and for them to be open to hearing about other models and approaches is a good sign. Unfortunately, including open source software attracted Microsoft's attention, and through them the attention of the BSA. They lobbied to shut down the conference, and got the US government to take action.

The truly dismaying thing is the reasoning the Patent and Trademark Office used. They said, essentially, that the WIPO has one defined purpose and it can't be allowed to consider any thinking not in step with its declared purpose. After all, thinking is bad. Beyond that, the mischaracterization of the relationship between open source software and intellectual property that the Patent and Trademark Office presents only serves to show how much they could learn from a conference about it. The whole thing smacks of not only sticking their own heads in the sand, but trying to pull everyone else down too. (from Lessig blog)

2003-08-22T12:51:00.000-04:00

Lawrence Lessig speaks about how people don't appreciate the value of the public domain. Eliot Landrum has put together a chart that shows why. Out of the 50 best-selling classics in 2002, 10 books are in the public domain. The chart focuses on the current year that the books will enter the public domain, assuming that copyright isn't extended again. I'd like to turn it around and ask how many books would be in the public domain today under the copyright terms when the books were published.

Under the Copyright Act of 1909, the maximum duration of copyright was 56 years. This was not altered until 1962, so all books published before 1947 would have entered the public domain under the terms of copyright when they were written. Thirteen of the books on the list were published before 1947 and are still covered by copyright, meaning that instead of the one in five books we have today, nearly half of the books on the list would be in the public domain. Authors whose books would have entered the public domain include J.R.R. Tolkien, F. Scott Fitzgerald, John Steinbeck, George Orwell, Aldous Huxley, Ayn Rand, Zora Neale Hurston, Agatha Christie, and Ernest Hemingway. Given the continuing sales success of these books, I think it's fair to say that none of them would be abandoned by their publishers, and the competition that should exist for editions of these books might even increase their sales and popularity. (from The Shifted Librarian)

2003-08-22T12:08:00.000-04:00

TeleRead has posted an alert that Project Gutenberg is broke. It would be tragic to let Project Gutenberg run out of money. Go make a donation.

2003-08-22T11:54:00.000-04:00

Ed Quillen points to recent abuses of copyright and asks whether it is "Time to give up on copyright law?" I really can't go that far. The intent of copyright law is good, even if the current implementation is not. But I hope this article sends a warning the corporations which are abusing copyright law. Keep using the law the way you are, and you might find yourself without any copyright protection, and then where will you be? Somehow I doubt that message will get through. (from Tech Law Advisor)

2003-08-21T13:38:00.000-04:00

Kevin Marks has stated that he is interested in my take on mediAgora, despite the fact that the stated intent of the project doesn't match my current concerns. With that qualifier out of the way, let's take a look.

MediAgora is a proposed "new marketplace for media." The focus is on guaranteeing payment for creators. The premise is simple. When a creator makes a work available for sale (as a digital file), the creator sets the price. After that, customers pay the price to get the work. There are a few wrinkles. First, derivative works are permitted, but the creator of the original work must always be credited and paid. Second, promoters are built into the system to encourage dissemination of the created works. The role of promoters is recognized through commissions built into the payment system.

The system starts with creators. MediAgora encourages creators to give up control but to expect payment in return. Rather than trying to fight copying of digital files, mediAgora preaches acceptance of copying but trust that the copying will lead to payment. Likewise, mediAgora requires giving up control over derivative works, but in exchange it requires payments back to the creator of the original work by requiring anyone who purchases a derivative work to also purchase the original. MediAgora also allows negotiation on terms of reuse, and it encourages creators to divide their works into pieces to facilitate negotiations and reuse.

The approach to customers relies on the assumption that customers have an interest in making sure that creators are fairly paid. This assumption is at odds with both the rationale for copyright law and economic theory, which assumes that customers will always seek to maximize their own position, even at the expense of public goods. Given the demostrated lack of "rational" behavior that people exhibit in the real world, maybe this is a valid assumption. The details of how revenue for creations gets distributed to the creator and the promoters are public, so potential customers can evaluate how fair the arrangement is during their purchase decision. The system is also designed to encourage customers to also become either creators themselves or promoters.

The role of promoters is the most interesting part of the system. A fairly complex commission system, with terms set by the creators, is built in. In addition to non-linear commissions per sales, the system supports a chain of promoters for a given sale. The system is designed to remove competition by not providing an incentive for promoters to cut out upstream or downstream promoters and also not providing an incentive to customers to shop around by making the price fixed.

Taking the system as a whole, the first thing to observe is that it depends on everyone not acting selfishly for short term interests. Creators are expected to reward their promoters. Creators also are encouraged to subdivide their own works to encourage reuse of parts of the work, rather than insisting on full payment for their own works for use in any derivative works. Customers are expected to pay for what they use rather than just freely copying. Promoters are encouraged to accept downstream promoters eating into their revenues in exchange for greater overall returns.

The lack of constraints on pricing and the control granted creators creates a totally free market for digital creations. The expected result is that creativity will flourish and everyone will get paid, but the behavior of participants in the existing systems suggests that things may not work out that well. In particular, many creators want to block derivative works and this system allows them to not negotiate on price, which could make the potential price of any derivative work so high that it's an effective barrier to creativity. That may not be rational from an economic standpoint, but that's not much of an argument because the system depends on all the participants not behaving in economically rational ways.

As an aside, I want to discuss copyright and derivative works in general. I am becoming more convinced that copyright should generally not control derivative works. Little good comes of this and much harm. Sure, some authors may move from successful to wildly successful because of movie rights, but I'm not sure how that encourages creativity. I believe the idea of control of derivative works grew out of the music industry on its way to making music copyright an order of magnitude more complex than copyright in general without providing any real benefit to the creators. Why encourage that in the rest of creative work? This isn't a criticism of this plan. It's just a brief rant about copyright in general.

I'm also not a fan of the way this system encourages everyone to get into sales. I've worked in the sales department of a company. I wasn't selling myself, but I had enough contact with the people who were to conclude that I don't want to be that sort of person. The plan seems to have built into it the idea that if you're a customer, you'll naturally want to get into sales. I understand there's nothing keeping me from just being a creator or just being a customer, but the way the plan seems to call out, "C'mon, be a promoter. You know you wanna," makes me nervous.

I have one more concern, which is directly tied into why this plan won't stop the music industry's rampage. For people working inside the system, there's not much reason to try to cheat. But there's also nothing to prevent people from stepping outside the system. I suspect that anyone downloading a song off Kazaa won't care if it originated at mediAgora. On the creator's side, since acknowledging the source for derivative works will either increase the cost or decrease the creator's earnings, there's an incentive to neglect to mention the sources. I suspect that by even asking these questions I'm tagging myself an industry supporting control freak, but I'm sure these are questions that will be in the mind of any creator thinking of selling works through the system.

Finally, one last question about both TeleRead and mediAgora. These are both proposals without implementations. If these plans are really going to work the way you say they will, why aren't you doing them instead of just talking about them?

I think mediAgora is an interesting idea, but I think that in some ways it's a solution in search of a problem. It addresses compensation for creators, which I don't want to minimize, but which also doesn't have as direct an impact on Americans in general as the bad laws and bad lawsuits the music and movie industries are inflicting on us.

2003-08-21T10:28:00.000-04:00

USA Today reports on the first legal action taken by a target of the RIAA's subpoenas. The story is light on details, but it appears that the target will argue that file sharing is not copyright infringement, as I have previously argued. Daniel Ballard, the attorney for the anonymous subpoena target, offers some of his defense. The article states that "Distribution implies sending, rather than leaving something where it may be taken, he says." Assuming that this goes to trial, it will be an important legal test. (from Tech Law Advisor)

2003-08-20T12:54:00.000-04:00

Kevin Marks has posted a comment saying that I got the problem wrong in response to my post about the problem of file sharing. Kevin states that the problem is in rewarding creators for their work, and that in framing things the way I have, I'm implicitly supporting the existing recording industry. I disagree.

Ensuring fair reward to creators has been a problem essentially since the invention of movable type. Revenues for creative works have always returned to the publisher and publishers have acted to minimize the payments they have made to the actual creators. This problem is nothing new and has been unchanged by the Internet.

I'm not trying to minimize this problem. The recording and movie industries have mistreated their creative workers since their inception, and that continues today. When best selling recording artists are declaring bankruptcy, that tells me that something's wrong with the basic functioning of the industry. To the extent that the Internet allows musicians to leave the major record labels, have successful careers, and keep more of their earnings, I think that's a good thing.

That's not the problem that I find myself writing about every day, however. What I write about is the response of the the record industry to the Internet. The recording industry has been successful over the past ten years at changing the law to support their attempts to control music, regardless of the impact that has on technological innovation or the creative process. They have also wielded lawsuits effectively to destroy new potential businesses that could challenge their control. Finally, they are mounting an unprecedented attack on their own customers in an attempt to scare them in line.

Somehow I don't think the targets of the RIAA subpoenas, who are facing potential hundred thousand dollar lawsuits, are thinking, "but the important thing is whether or not the artists are getting paid." The important thing is stopping the RIAA's reign of terror.

That's why I'm focused on either changing the law to legalize file sharing or changing people's behavior to stop the lawsuits. My willingness to support changed behavior may appear to be support for the RIAA, but as open to legalization of file sharing with no compensation to the copyright holders. While I have misgivings about that approach (because I do believe that creators should be paid for their work), I think it may be more likely to succeed than behavioral changes.

Perhaps I should reframe my proposed goals to focus on the industry rather than individuals. Solutions to the file sharing problem should either take away the industry's ability to sue individuals and interfere in technological innovation and the creative process or it should take away their incentives to do so. I think that statement is nearly equivalent to my previous formulation, but the change in focus may make it more palatable.

I had originally intended to review mediAgora today, but I'm going to hold off. Kevin Marks is the principle supporter of mediAgora, and given the differences in how he and I frame the problem, I doubt that it will do well against my standards. If Kevin or others want to argue that it will in fact hold up well, I'll reconsider my position.

2003-08-20T11:03:00.000-04:00

We knew it was coming. The RIAA and MPAA have filed an appeal in the Grokster case. Grokster and Morpheus were sued by the RIAA and MPAA for copyright infringement because of their file sharing services. In April, Judge Stephen Wilson ruled that Grokster and Morpheus were not liable because the nature of the services didn't meet the conditions for either contributory or vicarious liability.

The appeal was filed Monday. The statement argues that the companies that produce the software are earning millions of dollars by providing the site for direct copyright infringement by users of the services, and they are therefore committing vicarious infringement. The original ruling concluded that the services do not have awareness of individual acts of direct infringement or the ability to stop them, so they do not qualify as vicarious infringement, even if the businesses are deliberately structured around the infringing behavior.

2003-08-20T00:56:00.000-04:00

Yesterday, I laid out my understanding of the file sharing problem. Today I'm going to discuss TeleRead's proposal and consider its likely effectiveness in solving the problem.

TeleRead proposes a "Well-Stocked National Digital Library System." Although the focus is on ebooks, the proposal would work for other types of data as well. Unfortunately, beyond that catch phrase, the proposal is surprisingly vague.

TeleRead proposes dedicated cheap generic hardware for reading ebooks. The range of functionality seems to vary from one description to the next, but functions beyond being used to download and read books in electronic format include wirelessly transmitting books from one unit to another and word processing and other general computing functions, perhaps up to the level of being a full fledged Tablet PC.

These "TeleReaders" would connect to the digital library through a generic Internet connection, whether a phone line, Ethernet, or wirelessly. The devices would allow the user to search for and download books. The system would include both a national library and local libraries, so that information of particular interest to a community would only be available to that community. When a user chooses to download a book, the download would be recorded so that the publisher could be properly paid. Likewise, any direct transfers from one device to another would also be recorded and the data collected when the device is connected to the national library.

Here's where things get fuzzy. Payment is at the core of any digital distribution scheme, but TeleRead's payment mechanisms are uncertain. TeleRead seems to be proposing both allowing publishers to charge for books (as in a bookstore) and to have a royalty system operate through the library. If I am interpreting this correctly, some books would require direct payment on download. Others would be effectively free downloads, and the payment would come out of the general library funds. I am unsure what would happen if I were to directly transfer a book requiring direct payment to you. Would it be possible? Would you be obligated to make the payment next time you connect to the library?

The source for library funding is also vague. An early version of the proposal suggested a tax on televisions, but that has been dropped. More recently the proposal has included the possibility of a monthly access charge. Other possibilities include federal funding or a major private donation. Books "for sale" could be paid for directly. Books paid for through the royalty system would require one of these external funding sources, as would the operating expenses for the system.

That sums up my understanding of the system. If there are significant errors or oversights in my description, I would appreciate a correction.

Now to measure the system against the problem. The intent here is to provide a superior system to file sharing so that users will choose to use it rather than infringingly distributing copyrighted materials. It shares many features with the various music services. The possibility of a monthly fee, possibly in combination with charges per download is familiar to the music services. The availability of materials, which is determined at the publisher's discretion, is also characteristic of the music services. Just as the music services have not made an impact on file sharing, this proposal is also unlikely to change user behavior.

There are other unanswered questions to this plan. What's the role of DRM? Given the tracking of books within the system, they can't be permitted to be distributed outside the system. How will that be controlled? What use restrictions will there be? Things I might want to do with a digital book include automated reading out loud, printing the book, and copying and pasting text (and images, if any) from the book into other documents. Will I have my fair use right to make legal but unauthorized uses of the book?

There are also privacy considerations. If personal information is associated with the individual devices, which it has to be to support charging for downloads, the system will be able to track every book an individual ever reads. This is in contrast to traditional libraries and bookstores, which offer true anonymity to readers.

It should be clear that I am skeptical of TeleRead as a solution to the file sharing problem. It could be argued that TeleRead is not intended as a solution. It is true that the original motivations for TeleRead were not related to file sharing. However, it is often promoted in the context of solutions to file sharing and is compared to other proposed solutions. I came to it thinking of it as a proposed solution and I have evaluated it in that context.

2003-08-19T13:35:00.000-04:00

Library Journal has an article on "Fair Use Under Fire". The article lays out the threat to fair use that DRM embodies and argues the need for "good DRM". I'm not sure I believe in good DRM. All DRM defines the ways in which the DRM protected material can be used. DRM authorizes the uses to which the material can be put. As the article itself states, "Fair use is an unauthorized yet lawful activity."

In the wild, there are no limits on how creative works can be used. You can, for example, read a book out loud, record the reading, chop the recording up, and rearrange the recording with other recordings into a piece of music, as John Cage has done. Not everything you can do is legal, but you can always do it. DRM has a totally different philosophy. DRM always says, "Here is this file. These are the things you are allowed to do." DRM is intrinsically about giving permission.

Fair use, as Carrie Russell recognizes, is about doing things without permission. If there were a DRM system built from the ground up to allow everything except what is restricted, I might be willing to see that as having the potential to allow fair use, but that is not how DRM is constructed today. Without that change in philosophy, DRM will never support fair use. The thought that anything not prohibited is permitted is offensive to the organizations which a clamoring for DRM, so I don't think the two sides will ever meet.

I suspect I'm not clear enough about the distinction I'm trying to make. I'll keep working at it. Any comments you have would help clarify how to say it better. (from The Shifted Librarian)

2003-08-19T12:24:00.000-04:00

Jay Currie has written an article about Canada's compulsory licensing scheme. Not that the article refers to it that way, but that's what it functionally is. Under Canadian law, private copying of audio is explicitly protected as legal behavior. Currie asserts that this extends to making files available over the Internet and copying them, but I'm not sure I trust that unless it's been tested in court. In exchange, there is a tax on blank recording media which is paid back to the music industry. This makes it functionally the same as many proposed compulsory licensing schemes in the US which propose a general tax which is redistributed to the copyright holders.

From the sound of things, this is working out pretty well in Canada. The taxes are invisible and generally accepted by the public. The music industry doesn't have significant lobbying power to try to change copyright law. And apparently online file sharing is unrestricted.

Currie is focused on the American consequences of this, and in particular, the music industry's plans to sue everyone. The music industry is going after individuals who make files available to download. The thinking is that if there are no files being shared, no files can be copied. The RIAA may stop Americans from sharing files, but it can't stop Canadians. If the files are still available, the downloads will continue and the RIAA's strategy is doomed from the start.

Canada merits a closer look. What impact has the blank media tax had? The way this article describes it, it sounds like an effective solution to the file sharing problem. (from FurdLog)

2003-08-18T14:59:00.000-04:00

About a month ago, I mentioned that TeleRead was discussing a paper about DRM. TeleRead was using the paper to question various approaches to the problem of file sharing. I concluded that the paper's challenged TeleRead's approach as well. David Rothman wrote to me to say that I was misunderstanding TeleRead, which has led me to decide to study TeleRead's proposal in more detail. These sorts of plans tend to get out of hand, so I've decided to extend this study to various other proposals, including mediAgora and various compulsory licensing schemes. I hope to get to each of these in the near future, but I think I should start by identifying the problem. Any proposed solution has to be measured against how well it solves the problem in addition to how well it achieves its own goals.

According to the recent Pew study on music downloading, 26 million American adults make copyrighted files available for downloading online. The majority of these people don't care whether doing so is illegal or not. 35 million American adults download copyrighted files. Again, the majority don't care about the legality of their behavior. While there is significant overlap between the two groups, it is not total, meaning that 46 million American adults are involved in the online distribution of copyrighted files.

These numbers horrify copyright holders, particularly the major record labels and the major movie studios. From their perspective, these 46 million people represent lost revenues and the potential future downfall of the industry. The law as it has been interpreted by the courts makes both activities illegal and the music and movie industries are determined to stop them.

This is the problem. Any potential solution must do one of two things. Either it must change the law so that this behavior is no longer illegal or it must change people's behavior so that they choose to stop uploading and downloading files. Both approaches are uphill battles. Changing the law is unlikely to happen without industry support, which in turn means that it must include a mechanism for payment for all this distribution of copyrighted works. The difficulty of changing the behavior of users is indicated by the indifference they currently show to the legality of their behavior.

It's worth mentioning the approaches that have been taken so far. Education campaigns have been unsuccessful so far, as measured by both the continuing infringing behavior and the growing indifference to the law.

Legal services have had varying degrees of success, but even the most popular of them haven't impacted the illegal use of the Internet. The iTunes Music Store may be a success as a store, but it's a failure at controlling copyright infringement.

The music industry has had some success in suing file distribution services, but not enough. New services are showing up faster than the music industry can shut them down, and they are being redesigned to be harder to shut down. Besides, they've never had any success at fixing the real problem, which is stopping individuals from copying files and violating copyright.

The music industry's plan to sue the world is too new to measure its success, but let's talk numbers. In the worst case, stopping people from sharing music by suing them could require suing all of them individually, or 26 million lawsuits. Even if the industry won, what would be the harm done to the economy through the individual damages, the strain on the legal system, and the cost of the lawsuits to the industry itself? Even if this is a solution, it can't be said to be a good solution.

So there we are. The problem is widescale copyright infringement online. Responses so far have either not had an effect on copyright infringement or will do untold damage if they succeed. I welcome discussion at this stage. If I'm not identifying the problem correctly, I won't be able to evaluate proposed solutions. If there's something I'm overlooking or getting wrong, let me know.

2003-08-18T13:56:00.000-04:00

There's a major copyright fight that I've been avoiding, mostly because it's so ugly. It's getting enough attention that I should at least acknowledge it. SCO is claiming that Linux contains code from the Unix operating system, for which SCO controls the copyright. A wide variety of allegations have been made in both directions as a result of this basic claim. SCO is requesting a licensing fee from all users of Linux, despite the fact that copyright law doesn't grant an exclusive right of use. IBM has refused, and SCO has declared that it is terminating IBM's licenses for Unix, threatening AIX, IBM's version of Unix. IBM claims in response that SCO can't terminate IBM's license, amid general claims that the history of Unix is so complex that it's not clear that SCO even owns the copyright for Unix. Things are complicated further by the fact that SCO has distributed Linux in the past. Red Hat, another Linux distributor, has also gotten involved and is threatening to sue SCO. Related to all this, SCO is claiming that the GNU General Public License, under which Linux has been made available, is inherently invalid. Trademarks, patents, and trade secrets all have their roles to play in this as well.

ZDNet has a collection of articles on SCO (thanks to FurdLog for the pointer). Like I said, it's all really ugly.

This is a clear attack on the GPL and open source concept in general. Beyond the basic claims that the GPL is invalid, there's a threat here to all open source software. The threat is that you, as a user, don't know what's in the software. If there are any copyright infringements, you may find yourself unable to continue using the software or obligated to pay to continue to use the software. This is in supposed contrast to commercial software, where the manufacturer is more trustworthy because the manufacturer faces liability for infringement. Beyond Linux, this threat applies to a wide variety of software, including for example the Apache web server, the Mozilla family of web browsers, and Mac OS X.

Copyright law doesn't grant any exclusive rights with regard to use, just to duplication, distribution, creation of derivative works, display, and performance. What that means is that if you are using software that is found to be infringing on copyright, you have no obligation to either stop using the software or to pay a licensing fee. Users have nothing to worry about. If you are a Linux distributor, such as Red Hat or IBM, you have more worries. Corporations which have distributed the software internally might also be targets for violating the right of duplication.

Something that's not clear to me is what happens if someone claims copyright on a work they don't hold copyright on. Obviously that person has all sorts of liabilities and any licenses that person issues are invalid, but what's the liability of a third party which acts according to the terms of those licenses? Suppose I claim to be the copyright holder for Lord of the Rings and "license" to you the right to make and sell copies. I'd be in a lot of legal trouble, but are you? I suppose that since copyright doesn't require knowledge of infringement to make it infringement, you are at least liable for civil damages. That would appear to apply in this case, because the GPL doesn't contain any language indemnifying the licensee for the actions of the licensor.

If Linux (or an open source project in general) is found to be infringing, the infringing code can be removed and replaced. Logic indicates that if the code is fundamentally irreplaceable because there are no other ways of achieving that functionality, then the code should not be copyrightable under the idea/expression dichotomy, but I don't know that that would hold up in court. Code has been removed and replaced in open source projects in the past, so containing infringing code should not doom the software or present any problems to users. Developers and distributors will have a strong incentive to replace the offending code as quickly as possible.

This is ugly and potentially harmful to software development in general. I use a significant amount of software which is open source or built on open source components every time I use my computer, as do many other users and the Web in general. If this attack succeeds, all of that will be threatened. So much for the promotion of science.

2003-08-17T12:46:00.000-04:00

Here are my notes from the hearing on the Piracy Deterrence and Education Act which was held a few weeks ago. Some comments at the outset: I missed a couple of key chunks of testimony, including that of Linn Skinner. From what I saw, all of the targets of the testimony seemed to be chosen specifically to say that they support the bill. No serious testimony was presented in opposition to the bill. Rep. Berman took the opportunity of this hearing to talk up ACCOPS. Rep. Boucher should be commended for the skepticism he expressed. FurdLog has previously discussed this hearing. What follows are my notes taken during the hearing, which are, as usual, rather rough.

HR2517 Piracy Deterence and Education
piracy over internet - peer to peer - alarming
suing developers hasn't worked
suing individuals expected to be a deterrent
difficult to get law enforcement to get involved
efforts made to increase efforts of law enforcement
many believe copyright enforcement not important

Berman says:
infringers can better afford litigation than holders
users of Kazaa do not fear jail time
2517: FBI warnings about dangers of infringement
criminal suit on unregistered work - registration barrier prevents protection of works
HR2752: complements
transnational issues
uploading of a single work is felonious
clear and conspicuous notice on spyware
false domain name registration information - evidence of wilfullness in copyright infringement
Janet Monroe of FBI:
activity of FBI cyber division
July 11 Chicago Tribune. computer hijacking for porn. claims tied to Peer to peer.
2 billion files transferred/month few believe consequences
few believe security or privacy at risk
peer to peer centers on music and software, also child porn
orders FBI to develop deterrence program
agree that FBI must work with private sector
FBI and RIAA in final stages of IP warning program
"Open Letter on vulnerability and illegality of peer to peer"
hope to stem tide
consequences from economic damage to espionage

Mr. Trust? photography association
pulled away

Lynn Skinner

Christiansen UniversalStudios
piracy is greatest threat to creative industry
one copy of film can be made available to millions
80% of TV and movies lose money
depends on hits, which are most pirated
tells Hulk story
10s of thousands of sources
people who see it on Internet don't pay
work print also threat
ad agency loaned it to Kerry Gonzalez
soon available everywhere
worked from watermark on work print
applaud conviction
further measures will foster legitimate distribution

break for vote

question to Christiansen: success of Hulk case, why ripe
ability to track leak source. rapid response, FBI very serious in assistance
question to Skinner and Trust: dealing with law enforcement
Skinner: not great response from law enforcement
Trust: photographers feel disenfranchised, don't register, almost impossible to do anything about stealing a photograph
question to Monroe: expect law to more. is Hulk first prerelease prosecution
Yes
Q; why don't we prosecute more?
still in embryonic stages, cyber crime #1 violation is IP. expect many more prosecutions
...
q: eliminate registration requirement. current obligation to register for criminla prosecution
must have valid copyright for prosecution, must have application for reg.
q to Christiansen: warning notice - under what authority for existing warnings
agreement in place between FBI and movie industry on warning on start of movie
q on single upload
welcome any tools - valuation automatically exceeds threshold, prosecutions rare
q on fears of civil liability
Skinner: lack of fear well founded
q to monroe: false whois data. would accurate info be useful
yes
Rep Kelly Q for Monroe: making it easier to use FBI logo
invest, edu, train
edu write letter to public on illegality to be posted and used by holders
Kelly is asking about advance leak who do they contact
local office
Kelly should there be "online SWAT team"
currently have trained agents in each field division
also regional offices for more manpower
q to Ashcroft on prosecutions on p2p (not at that time) is Hulk precedent for additional crimes

q from boucher concerned about inappropriate use, more should be done, commend FBI for Hulk. has concerns with bill sec 3 "unauthorized" consent of copyright holder or unlawful
Christiansen: infringement includes normal defenses
boucher: unauthorized means infringment
Christiansen download is an infringment
boucher is asking about fair use
Christiansen: never find fair use to upload
...

q from Kelley on ramifications of FBI actions. Editorial pages supporting IP. reexamine what IP theft really means - how does committee further the work of the FBI? Do you have the tools legally for Hulk and future?
monroe: more personnel
investigators
to Christiansen: suggestions to committee on powers to protect rights
issue of registration

q from Winters - thinks that Hulk noteworthiness is that it is noteworthy at all. Monroe: how long to find an illegally pirated movie
not very long
Q aren't you just one monolithic (after referring to Universal as Paramount)
Q why isn't there more action
have undercover operations that will be prosecuting
Q giving notice
working on it
Q no deterrent force because gov not taking seriously. How many investigations currently?
as many as we can with manpower available

expect to mark bill next week