The video of my Princeton President's Lecture, "Rip, Mix, Burn, Sue: Technology, Politics, and the Fight to Control Digital Media" is now online. The lecture, which lasts about an hour, is a layperson's introduction to the technology/copyright wars. I gave it on October 12. The first six minutes of the video consists entirely of introductions, which can safely be skipped.

UPDATE (Nov. 22): This entry has been updated to point to a new page for the lecture, which includes a Creative Commons license for the lecture video, and will eventually link to the lecture in more formats.

Princeton offers a great set of lecture videos on the net.

Malcolm Gladwell has an interesting piece in the Nov. 22 New Yorker, reflecting on the discovery that Frozen, a Broadway play, included language lifted from an earlier Gladwell article.

Equally interesting is the reaction of Dorothy Lewis, a New York psychologist who was the subject of Gladwell's earlier article. One of the characters in Frozen is very similar to Lewis, to the point that Lewis's friends and colleagues see the character as being Lewis. Lewis may or may not have a legal remedy, but she feels violated, especially since the character commits an indiscretion that Lewis herself did not.

And yet the public outcry over Lavery's copying, such as it is, points to the copying of wording, rather than any of the other copying that occurred. Why? Gladwell's exploration of this question is worth reading.

Gladwell correctly separates questions of copyright law from those of plagiarism.
Perhaps this is what distinguishes his view from the conventional wisdom that he describes and then ultimately rejects.

Little by little, the legal rules of copyright are infecting our understanding of plagiarism. Kids, who used to be taught that certain kinds of copying are wrong, or at least disapproved by adults, are now taught about copyright instead. Last year a third-grader told me that it's wrong to copy a friend's homework, because the homework is copyrighted.

Adults make the same mistake too. High-profile plagiarism accusations usually point to textual similarities, as if the underlying sin could have been blotted out by a bit of rewording.

This is a shame. Our plagiarism norms exist for a good reason. Intellectual life is about more than copyright.

[link credit: Joe Gratz]

I just finished reading Brian McWilliams' new book Spam Kings. It's an entertaining read that offers an interesting, nontechnical peek at some of the personalities behind the spam wars.

The book's central figure is Davis Hawke, an amoral character responsible for most of the spam promoting male anatomical enhancement products. (The only thing that these products have reliably expanded is Hawke's cash flow.) Hawke turned to spamming after the failure of his first career as a neo-Nazi leader, and he displays all of the class and moral judgment that one would expect given this background. His (financial) success as a spammer is due to his gift for writing persuasively sleazy ad copy, his persistence, and his willingness to do anything to make a buck. One almost wonders whether such a character could really exist outside of detective fiction; but apparently he does.

The anti-spammers, while somewhat more conventional in personality (but then again, who isn't) also make interesting character studies. Some are system administrators whose jobs are complicated by spam, but many are like Susan Gunn (a.k.a. Shiksaa), an ordinary netizen who transformed her anger at spammers into a crusade to unmask and frustrate them.

When these personalities collide, the result is a kind of online trench warfare, involving technology, old-fashioned detective work, lawsuits, and occasional threats and intimidation. The spammers are far from unified, double-crossing each other like Elmore Leonard villains. The anti-spammers are at most a loosely organized coalition. Both sides are plagued by leaks and defections. The anti-spammers occasionally break the law; the spammers routinely hide in the quasi-legal murk.

When things heat up, a veteran anti-spammer asks an insightful question: Are we fighting against the spammers, or fighting against spam? Sometimes it does seem that the whole exercise has turned into an elaborate capture-the-flag game between the two communities. Yet despite the occasional excess, one ends up admiring the determination of the anti-spammers, and happy about the results they have achieved. Even as spam multiplies, the life of a spammer is not a happy or comfortable one.

(O'Reilly, the book's publisher, sent me a free copy of the book. Thanks! Hint to other publishers: I'm more likely to read and review your books if you do this too.)

One of the underreported stories from last week's election was the effect of long waiting lines at polling places. Many polling places in Ohio, for example, had lines of three hours or more. Though many voters waited, determined to cast their votes, quite a few must have been driven away. Not everybody has three hours to spend at the polling place.

A story in the Boston Phoenix, by David S. Bernstein, points to significant reductions in the number of polling places in some parts of Ohio, compared to the 2000 election. According to the article, polling places were consolidated on the theory that voters would cast their votes much more quickly on the touch-screen systems that were to be used in this year's election. But then many counties put aside the touchscreens due to security concerns, and used the old punch-card system instead. The result is the same voting system as before, but with many fewer polling stations. Add in a higher than usual turnout and long lines result.

How did this affect the election results? Some data from the article:

Of Ohio’s 88 counties, 20 suffered a significant reduction — shutting at least 20 percent (or at least 30) of their precincts. Most of those counties have Republicans serving as Board of Elections director, including the four biggest: Cuyahoga, Montgomery, Summit, and Lucas.

Those 20 counties went heavily to Gore in 2000, 53 to 42 percent. The other 68 counties, which underwent little-to-no precinct consolidation, went exactly the opposite way in 2000: 53 to 42 percent to Bush.

In the 68 counties that kept their precinct count at or near 2000 levels, Kerry benefited more than Bush from the high turnout, getting 24 percent more votes than Gore did in 2000, while Bush increased his vote total by only 17 percent.

But in the 20 squeezed counties, the opposite happened. Bush increased his vote total by 22 percent, and Kerry won just 19 percent more than Gore in 2000.

This suggests that the long lines may have driven away more Kerry voters than Bush voters. But it's only a suggestion at this point, not a solid inference; and in any case the effect doesn't look big enough to call Bush's victory into question.

It would be great to see a carefully, methodologically sound study of this issue.

Continuing yesterday's discussion of the new Rob/Waldfogel filesharing study, let's look at the possible effect of authorized downloading services.

As we saw yesterday, one of the main findings of the study is that people derive lots of benefit (about $45 annually per capita for the study's sample population) from downloading songs that they don't value enough to buy. In the absence of filesharing, this would have been deadweight loss.

Suppose that CDs cost $15 and that the (marginal) cost of producing and distributing one more CD is $2. Suppose further that Alice would be willing to pay $9 for a particular CD. If the record company could somehow sell Alice the CD for a price between $2 and $9, Alice would buy it and both parties would be better off. The total welfare gain would be $7, the difference between Alice's valuation of the CD and the cost of providing it to her. But if the record companies have to charge everybody the same price for a CD, they would be foolish to lower everybody's price below $9 – that would increase their unit sales but lower their total profit. So they'll keep the price at $15 and a mutually beneficial sale to Alice won't happen – that's deadweight loss.

If it's impossible to sell Alice the CD at a price she is willing to pay, then there is economic benefit in letting Alice download the album without paying – Alice gets the music, and the record company doesn't lose anything since it wouldn't have made a sale anyway. Some value that would have been lost is instead being captured by Alice.

But if the record companies can price-discriminate, that is, if they can charge different prices to different people, then they might be able to sell Alice the CD for $8 without lowering the price they charge anybody else. If they can somehow do this, they can eliminate the deadweight loss. But they can only do this if (a) they have some way of guessing how much individual customers are willing to pay, or at least some way of segmenting the customer population into groups that are willing to pay different amounts, and (b) they can prevent people like Alice from buying the CD at $8 and then reselling it to somebody who is willing to pay more.

It's hard for record companies to price-discriminate in the traditional CD market, but maybe it's easier for them to price-discriminate by using online music services. Maybe they can cook up a pricing plan that causes people who like a song more to pay more for it. I don't think anybody really knows how well the industry could price-discriminate in such a scenario.

If, somehow, the industry could price-discriminate perfectly, so that they charged each user just exactly the maximum he was willing to pay for each song, then the deadweight loss would vanish – into the industry's pockets – and downloading would turn out to be harmful. (Remember, if our goal is to maximize total welfare, we care only that the deadweight loss is gone; we don't care who pockets it.)

But if we assume, more realistically, that the industry's price discrimination strategy will have only limited success, then we can't say whether filesharing will turn out to be harmful. We know that filesharing eliminates $45 (per capita) of deadweight loss. If price discrimination eliminated less than $45, then filesharing would still look like a good idea, considering only direct effects. We know the indirect effects (less music being produced, because of reduced industry incentives) of downloading will be negative, but we don't know if they'll be large enough to overcome the positive direct effects.

At the end of the day, the Rob/Waldfogel study doesn't tell us whether filesharing is helpful or harmful, from the standpoint of total welfare. (Nor, to its credit, does it claim to do so.) What it does tell us is that downloading eliminates a lot of deadweight loss. And so it weighs on the scale – very lightly, to be sure – against the proposition that downloading is harmful.

There's a new study out, by Rafael Rob and Joel Waldfogel, on the effect of filesharing on music sales. The news headlines will say that the study shows that filesharing hurts CD sales (as the BBC story does); but the full results are more complicated.

The study relied on surveys of college students: a preliminary survey of a large group at four institutions, and a more detailed followup survey of undergraduates at the University of Pennsylvania. The most interesting results come from the smaller Penn survey.

It's worth noting that this kind of study is likely to overestimate the negative impact of filesharing. The survey samples only college students rather than all filesharing users – and, contrary to popular myth, students account for only a small minority of filesharing activity. As I've explained before, surveys of younger people will tend to overweight the harmful effects of filesharing. (Rob and Waldfogel are upfront about the limitations of their study.) For the sake of discussion, I'll set aside the drawbacks in the study for now, and assume it is correct and it generalizes to the population as a whole.

The heart of the study is an economic comparison, based on the survey data, between the current world, where students do a certain amount of filesharing, and a hypothetical world where filesharing is impossible. The availability of filesharing decreases consumer spending on music from $126 to $101 annually. That's $25 of lost revenue for the music industry. But it's only half of the picture.

The other half is the effect of filesharing on consumers. The average consumer is $70 better off with filesharing than without. Of that $70, $25 comes from downloading music rather than buying it, but $45 comes from getting access to music that the consumer wouldn’t have bought anyway. (In econo-speak, deadweight loss is reduced by $45. For more details, see pp. 26-27 of the report.)

This raises an interesting public policy question. The proper goal in this policy area is to maximize total welfare, and not simply to maximize industry profit. (Industry profit is good only insofar as it creates incentives for industry to do things that increase total welfare.) The direct effect of filesharing is to increase total welfare by $45 per capita (increasing consumer welfare by $70 and reducing industry welfare by $25, for a net gain of $45). If direct effects were the only effects, then the right policy would be to allow filesharing.

But of course there are indirect effects, most obviously that the drop in industry revenue reduces the incentive to create new music. The predictable result is that less music will be made. There can be little doubt that the loss of this music will reduce total welfare. The key question is how much welfare loss these indirect effects will cause. If they cause less than $45 per capita of harm, then allowing filesharing may still be a good idea. If they cause more than $45 of harm, then allowing filesharing seems to be harmful. $45 is a lot of welfare to offset – based on the study’s numbers, reducing the production of commercial music all the way to zero would reduce total welfare by only about $125 per capita.

So it's certainly not right to say that the report shows filesharing should be illegal. (Nor is it right to say that the report shows the opposite.) The report doesn't quite reach the most interesting policy question.

(Tomorrow I’ll write about the effect of authorized online music services on this analysis.)

Just in case you thought that lawsuits about pregnant chads were the worst possible election outcome, here's a story about the consequences of e-voting without a proper paper trail.

A bug in e-voting system software caused about 13% of the votes cast in Carteret County, North Carolina in last week's election to be lost irretrievably, according to a story by Kelcy Carlson at WRAL.

The state Board of Elections discovered on election night that 4,532 electronic ballots through early voting were not recorded.

"The bottom line that we have heard from the manufacturer is that these votes are not missing. They're lost," county commissioner-elect Tom Steepy said. "It's very disheartening. It really is."

Carteret County had one stop for early voting. Twelve electronic booths fed into one electronic system that was expected to hold just over 10,000 votes. In reality, it only held just over 3,000. Officials said anyone who voted after 11 a.m. on Oct. 22 through Oct. 30 did not get their ballot counted.

"The company has admitted now that it was its error and that it was a simple keystroke that should have been applied to the system perhaps several years ago and was not," said Ed Pond, of the Carteret County Board of Elections.

(See also the earlier USA Today story.)

Had these machines used a voter-verified paper ballot, the problem could have been rectified by counting the paper ballots. As it is, there is no backup to protect against software problems, so Carteret County voters will have to go to the polls again to vote in a new election.

The big copyright owners have gotten pretty sophisticated about monitoring P2P applications to gather evidence for lawsuits. But now P2P traffic seems to be shifting to the BitTorrent system, which works differently from other P2P systems. This will affect the copyright owners' monitoring strategy in some interesting ways.

Most P2P systems allow users to choose between being only a receiver of files, and being both a receiver and a provider. (The default is typically to do both.) BitTorrent, by contrast, tries to enforce a kind of reciprocity between its participants, so that you have to provide parts of a file in order to receive other parts. If this works as designed, there is no way to be a passive, receive-only participant in the BitTorrent system. And that fact has important implications.

Copyright owners typically monitor a standard P2P system by joining the system as a receiver. That allows them to see who is providing which files, without requiring them to provide copyrighted files to anybody. If BitTorrent's reciprocity scheme works, then the copyright owners will have to participate in the P2P distribution of their own copyrighted files in order to monitor who is providing those files. That's a step they might not be willing to take.

On the other hand, BitTorrent reciprocity will require that ordinary users who want to receive files will have to contribute to the distribution of those files. They won't have the option of being silent receivers, as they can be on standard P2P systems. Every user will be exposed to detection by the big copyright owners.

This will foil one of the more subtle features of the copyright owners' P2P strategy. Their decision to monitor and sue the providers of files, but to leave receivers alone, created an incentive for P2P users to shift to receive-only mode. The idea was that if enough users did this, a larger number of receivers would contend for the services of a smaller number of providers, reducing the overall effectiveness of the P2P system. This plan has a certain elegance, but there's not much evidence it has worked.

These changes are all interesting, but they will happen only so long as the BitTorrent reciprocity mechanism works – only so long as BitTorrent users are actually forced to provide parts of files in order to receive what they want. People who are threatened by these changes – a category that would seem to include the big copyright owners as well as many users of BitTorrent – will have a growing incentive to defeat the reciprocity system so that they can passively monitor BitTorrent, or passively receive files.

My guess is that somebody will figure out how to be a passive BitTorrent user, albeit with lower performance than active providers get. The knowledge of how to do this may change BitTorrent in important ways. If I were a big copyright owner, I might even consider developing releasing a BitTorrent-leech application.

(For more thoughts on copyright owners and BitTorrent, see Cuong Nguyen's post on CopyFutures.)

Tuesday’s election created lots of data about voting patterns in places that used different voting technologies. Various people have done exploratory data analysis, to see how jurisdictions that used e-voting might differ from those that did not. See, for example, the analysis cited in Joe Hall’s entry over at evoting-experts.com.

As a commenter on Joe's entry ("Jon") notes, voting technology is not the only difference between Florida counties that might account for the observed differences. Counties that used e-voting tend to be larger, more densely populated, and more Democratic-leaning than those that don’t. Perhaps these differences explain the data.

To answer questions like these would require more sophisticated data analysis, probably performed by a person who does such analyses for a living. Such a person could control for differences in voter demographics, for instance, to see whether there is an e-voting effect separate from the kinds of differences cited above. Such a person could also tell us how big the remaining effect is, and whether it is statistically significant.

It would be great if some hotshot social science data analyst would agree to do such a study. I’m sure that the folks out there who have data would be willing to furnish it, and to suggest theories to test.

It’s also worth thinking about what a particular finding would tell us. It’s one thing to find an anomaly in the data; but it’s another thing to explain what could have caused it. If you can point to an anomaly, but you don’t have a plausible story about how a rational election-stealing strategy would have caused that anomaly, then you don’t have strong proof of fraud, no matter how much evidence of the anomaly’s existence you can present.

If real anomalies exist, I think it’s more likely that they’ll turn out to be caused by errors or technology failures than by e-voting fraud. Either way, a careful study of the data might be able to teach us a lot about how well various voting technologies work in practice.

The Motion Picture Association of America plans to file copyright infringement lawsuits against about 230 individuals today, according to a New York Times story by Laura M. Holson.

Rumor has it that studio heads had long wanted to do this but former MPAA chief Jack Valenti had refused to go along with it. Now that Valenti has been replaced by Dan Glickman, it's not surprising to see the suits starting.

Lately the movie industry has had more of a scorched-earth attitude toward the copyright wars than the music industry has. If the movie people start filing lots of infringement suits, perhaps the music people will back off a bit and file new suits.