other threads (collected entries on recurrent themes)
Monday, August 4
I have lost the reference, but sometime within the last 72 hours, I read a quote by an official of the Defense Advanced Research Projects Agency (DARPA), the agency that helped launch the Internet and recently got into trouble for creating a "futures market" in terrorism. This official bemoaned the stupidity of his laptop, which doesn't know what he wants it to do; he called for much more public investment in artificial intelligence (AI).
I have an interesting colleague in computer science, Ben Shneiderman, who strongly criticizes AI research. His argument is not that the machines will take over the world and make us do their will. Rather, he argues that AI tends to make machines less useful, because they become unpredictable. When, for example, Microsoft Word tries to anticipate my desires by suddenly numbering or bulleting my paragraphs, that can be convenient—but it can also be a big nuisance. Shneiderman argues that computers are best understood as tools; and a good tool is easy to understand and highly predictable. It lets us do what we want. All the revolutionary computer technologies have been very tool-like, with no AI features. (Think of email, word processing, and spreadsheets.) Meanwhile, untold billions of dollars have been poured into AI, with very modest practical payoffs.
Tuesday, August 5
The latest technological phenomenon to get the attention of the New York Times is "mobbing." An announcement spreads around blogs, listservs, and bulletin boards: everyone is supposed to show up at a particular time and place to do some particular, but random, thing, like asking a Macy's sales clerk for a "love rug" or shouting "Yes, Yes!" Thanks to the viral nature of the Internet, the idea spreads and people actually show up.
Are smart mobs "The
Next Social Revolution?" as Howard Rheingold is arguing? They
certainly fit the current ideal for social organizations: completely decentralized,
Tuesday, August 26
If you compare a newspaper website to a conventional newspaper page, I think the results are a little surprising. We're used to seeing the Internet as a great expansion of possibilities, compared to print. But news websites only display about 15 words on each line, plus advertising and navigation bars. That means that a reader must essentially scroll down one vertical column of text at a time. A traditional sheet of newsprint, by contrast, is very wide and can contain an elaborate array of stories (some linked together), diagrams, and photographs. The reader can spread out a newspaper, scan it quickly, and select what to read and in what order.
As a result, news sites are perhaps more like broadcast programs than they are like conventional newspapers. A broadcaster can only transmit one stream of content at a time. There is always a danger that listeners will switch channels if they don't like what they see and/or hear. Thus broadcasters feel pressure to cater to as large an audience as possible with each of their programs. In contrast, a traditional newspaper is a diverse bundle of material, which readers can navigate and read selectively. The more diversity of content, the better, at least to a point. One would think that Internet sources would be more interactive and diverse than newspapers, not less so. But I think that the width of our current screens may actually make websites more like broadcast channels. They have to emphasize a few headline stories and try to keep their visitors from "clicking" away to other sites.
Of course, there are other differences between newspapers and news websites. (To name just a few: the lack of any final edition on websites; visitors' ability to search current and archived editions; and the prevalence of links to sites beyond the newspaper's control.) Still, the difference in width deserves mention.
Monday, Sept. 8
A well-known experiment, run by Iowa Electronic Markets, allows traders to place bets on the outcome of political elections, including the current California governor's race. According to a paper by Joyce Berg and others, the Iowa Political Market has outperformed polls in predicting 9 out of 15 elections. Its average error in predicting election results is about 1.5%, compared to about 2% for an average poll. In some past elections, the Market avoided major errors that marred all the major national surveys, whereas it has never made a gross mistake itself. The apparently uncanny ability of the Iowa Electronic Market to predict the future was one of the reasons that the Defense Department recently floated the grisly idea of a futures market in terrorism.
I'm struggling to understand the theoretical explanation for this phenomenon. I realize that markets efficiently aggegrate the knowledge of investors (who must try to make honest predictions, since their money is on the line). But where do the investors in a political futures market get their knowledge? They cannot simply ask themselves how they intend to vote. As Berg et al. note, traders are "not a representative sample of likely voters; they are overwhelmingly male, well-educated, high income, and young" (p. 2). Some are not even US residents. Thus their own choices in the real election, assuming they vote at all, will be very different from those of the American people. Yet they seem to be able to predict the actual result more accurately than a random-digit telephone poll.
One clue is that a relatively small number of "marginal traders" drive the market; they make many more trades than other people and are less prone to sticking with an unlikely bet out of loyalty. I would guess that these "marginal traders" are political junkies: people who have no sentimental attachment to any of the candidates but love to prognosticate about elections. We can assume that they have seen all the polls—but that still doesn't explain how they outperform surveys on average. Could it be that they instinctively recognize a consistent error in polling, and adjust accordingly? For example, maybe polls tend to pick the real winner but predict a larger margin of victory than actually occurs. (Races tend to "tighten" right at the end.) Or maybe polls tend to make inflated predictions for the Democrats' share of the vote, because they count too many low-income people as "likely voters." It's also possible that the marginal traders rely on one or two polls that are better than the average. (Then we would find that the market outperformed polls in general, but was no more accurate than the best of the polls.)
These are hypotheses backed with no evidence. But if one of them turns out to be true, then we don't need a market to improve on surveys. We just need to make the same adjustment to poll results that the marginal traders (a.k.a., the political junkies) are making. Likewise, we would not benefit from a futures market in terrorism, but we should strive to understand how the best informed and least sentimental observers of terrorism make their predictions.
Tuesday, Sept. 9
I was in New York City today, at the offices of the AOL Time Warner Foundation, discussing evaluation with people who work in the field of "youth media and technology." This means people who help adolescents to produce videos, websites, radio broadcasts, and magazines for the benefit of their communities. I won't attempt to summarize the other participants' views (or even list who was present), because they weren't warned that they might be "blogged."
Speaking for myself: I think it is appropriate for people who are running relatively low-budget programs to assess themselves by making internal comparisons. For example, they can compare their own performance in 2004 with that in 2003, and if they are making progress, they can declare success. Or they can compare one of their own programs with another.
Such evaluations will not answer questions like: Is work with youth media effective? Or, What are its outcomes? To answer these questions, you have to compare one or more youth media projects with something else (such as athletics or conventional arts programs) or with no intervention at all. I would call this approach "research," as opposed to mere "assessment." Research is crucial in any field that wants to expand, find new funders, and gain acceptance in schools. But it is also difficult, expensive, and a diversion from the day-to-day goals of a service organization. That's because solid research requires random assignment of adolescents to the program or to an alternative, or at least elaborate statistical controls that mimic random assignment. And elaborate statistical analysis requires lots of expensive data-collection. In short, the field of youth media and technology would benefit from research; but for each practitioner, the costs and obstacles of doing the research are too great.
Thursday, Sept. 18
Right now, Hurricane Isabel is howling around us and most work has ceased. The University has taken its server down, blessedly cutting off my email. Yesterday afternoon, when the skies were still clear, I met with Marty Kearns of Green Media Toolshed, who is full of fascinating ideas about how the Internet and other distributed technologies (including billboards and buttons) can be used for political activism. Meanwhile, I was reading reviews of Bruce Bimber and Richard Davis' new book, Campaigning Online: The Internet in U.S. Elections. Apparently, they argue that the Internet is effective for mobilizing strongly committed partisans, but it does not increase net participation in politics and elections. This is consistent with CIRCLE research on young people, and also with my predictions in a 2002 essay on the Internet and politics.
Marty Kearns makes me optimistic about the political power of digital technologies and their value for progressive organizations. But I also worry about the chief barrier to participation. It's not the digital divide, or technological literacy, or the power of major media companies to constrain the ways that the Internet is used. It's rather the lack of motivation to participate politically—the lack of identity as citizens—among many marginalized people. In the past, people developed that kind of identity and motivation by enrolling in disciplined organizations with strong cultures: unions, political parties, religious denominations. I'm not convinced that we've found replacements for such organizations in the digital age.
Wednesday, October 1
I buy the argument in Cass Sunstein's book, Republic.com. Sunstein predicts that the Internet allows people to choose news and opinion that already interests them, while filtering out any views and facts that they find uncomfortable. As a result, the population splits into small communities of like-minded people who reinforce their shared views. Another result is a widening gap between those who have a lot of interest in public issues and those who are not interested. Motivated citizens benefit from all the news and opinion online. Unmotivated ones can ignore the broader world in a way that was more difficult back when they relied on TV for entertainment and the newspaper for want ads and crossword puzzles. Whether they liked it or not, they used to see news on television and on the front page of the newspaper.
Sunstein's book was mostly based on his theory of democracy and some experimental evidence about deliberation in narrow groups. His empirical evidence about the Internet was relatively weak. Thus many reviewers criticized him and offered anecdotes about the Web as a place for diverse public deliberations. Even Sunstein seemed to back off his own claims in the face of these criticisms. Yet I never thought he was proved wrong.
If Sunstein were right, then those who start off uninterested in politics would be less informed and therefore less likely to participate once they gain Internet access. Now a scholar named Markus Prior shows that Internet access correlates with a lower probability of voting among people who start with a low interest in the news. (In other words, these people are more likely to vote if they don't have net access.) The article is entitled "Liberated Viewers, Polarized Voters: The Implications of Increased Media Choice for Democratic Politics," and it's in the Good Society.
Friday, Oct. 10
I have just posted two new articles about the idea of a "commons." Both are defenses of a particular position, which I would summarize as follows: The Internet should be an open arena for creators to make and give away digital material. That is how the Net was born; but this commons ideal is now under serious threat from government censorship and especially from corporate control of the Internet's "architecture" and intellectual property. So far, my argument is completely indebted to work by Lawrence Lessig, James Boyle, David Bollier, and Yochai Benkler, among others. I add the view that we won't ever succeed in protecting the commons through legislation, court decisions, or clever software that circumvents corporate or state control. We need formal associations of citizens who have personal experience with the new digital media and commitment to using it for civic purposes. In "Building the E-Commons," The Good Society, vol. 11, no. 3 (2002), pp. 1-9, I discuss one such association and then move to a general argument for the "associational commons" as our ideal. In "A Movement for the Commons?" The Responsive Community, vol. 13, no. 4 (Fall 2003), pp. 28-39, I start with the legal battle over intellectual property, and again conclude that we need citizens' associations to protect and enrich the commons.
Tuesday, December 9
There's a standard version of the history of the Internet that traces it back to ARPA (the Advanced Research Projects Agency) in the 1960s. ARPA developed a way for computers to exchange information in small packets, so that two computers would not need to open a permanent and exclusive channel (such as a standard phone connection) in order to remain constantly in touch. Instead, they would send messages in small chunks that could be routed through whatever computers happened to be online until they reached their destination. ARPA was a military outfit (it soon became DARPA; the "D" stands for "Defense"), and its motive was to create a new communications network that could withstand massive disruption during wartime. The DARPA system improved, and similar processes developed separately in the academic world. Under the auspices of the National Science Foundation, these networks were brought together (starting with a process for sharing email). After a while, the NSF named this network of networks the Internet, and so we entered the current era.
This is all true and important, but it's like explaining the origins of a human being by listing all of her direct ancestors who happen to share her last name: her father, her paternal grandfather, and so on back. Lots of other ancestors have also contributed their genes and nurture, although their names are harder to retrieve. Similarly, if you look around today's World Wide Web, you'll see numerous important features that did not arise from ARPA, DARPA, or the NSF.
Two quick examples: 1) I use online library catalogues all the time, and they are a significant part of the Internet. Their genealogy begins with handwritten book catalogues (which, for all I know, are as old as ancient Alexandria), and then moves to computerized databases in major research libraries, which became accessible via modem at least 25 years ago, which then became accessible by telnet, and which are now usually searchable through a Web browser.
2) Elaborate multiplayer games like MUDs and MOOs are another part of today's net. I think their ancestors include: wargames with lead figures in the era of H.G. Wells; role-playing games after World War II; role-playing games played by correspondence, which arose roughly at the same time as computerized single-player wargames; networked computerized wargames; and finally multiplayer games on the Web.
Thursday, January 15
We released a survey today that contains a lot of data about young people--their civic and political behavior and attitudes, and specifically their reaction to the ways political campaigns are using the Internet.
Campaigns are effectively using the Internet to reach young people, and will continue to do so. But is this because young people are computer-savvy and demand Internet based campaigns? Or is it because campaigns see advantages to a cheap medium that can reach and expand their base (more cost-effectively than broadcasts and mass mailings)? Overall, our data show that young people are not particularly favorable toward new, online campaigns techniques. They favor some approaches but oppose others.
Despite a general lack of enthusiasm for many online campaign techniques, however, there are some pools of young voters who do like the new technologies. For example, those college students and college graduates who are liberal and concerned about the War in Iraq are overwhelmingly aware of blogs and favor their use in campaigns by 68%-32%. This group also likes “banner adds”and weekly email updates, which are unpopular among youth in general.
The graph shows the percentage who like each campaign technique, minus those who don't like it or call it a "turn-off." We distinguish liberal college students who are concerned about the war (blue bars) from other youth (red).
These findings put the Dean phenomenon in context. The demographic group that most likes his campaign themes is also most favorable toward electronic campaigning. It is not at all clear that blogs and Meetup events would work nearly as well for other candidates. Given that strong partisans and well-educated youth are most enthusiastic about the new technologies in politics, the Internet looks like a means to organize core voters, not a way to expand the franchise
I've been asked to write an article on how young political activists use the Internet. After an introduction about the political potential of the "commons," I turn to various types of online youth activity. One short section concerns online political organizations. Comments would be welcome. I'm planning to say:
Political, ideological, and civic organizations have formed largely or entirely online, representing virtually all ideologies, identities, and agendas. Their organizational structures also vary greatly, but compared to offline groups, they are more likely to have anonymous or pseudonymous members.
Anonymity has the advantage of allowing candor, which is especially beneficial for members of stigmatized groups (such as the only gay adolescent in a small community). It also allows people to experiment with novel identities. However, anonymity may have the disadvantage of making relationships relatively superficial and may permit behavior that is disruptive to the group itself. If members can adopt fictitious identities, then they can change their identities as soon as anyone threatens to expel or socially ostracize them.
Compared to offline groups, online ones tend to be easier to “exit” but harder to change by exercising “voice” (Hirschman), because there is no method of democratic decision-making that one can influence. Because exit is easy, groups tend not to discipline their own members by demanding contributions or particular forms of behavior in return for membership. Again compared to offline groups, online ones tend to be “thin” rather than “thick” (Bimber, p. 148). In a classic “thick” group, such as a family and ethnic group, members are committed to the survival and flourishing of the collectivity; but its purposes are changeable and subject to debate. In a “thin” group, members enter having some purpose, and view membership as instrumental to that goal. Although many online groups are “thin,” unstructured, and easy to exit, this is not true of massive, multiplayer games, whose participants invest considerable time in developing fictional characters. Often, they become highly committed to the flourishing of the game community as an end in itself. However, most games are not political.
Political or civic groups more typically allow members to visit a website, contribute money, and/or elect to receive email messages. A prominent example is MoveOn.org, a liberal organization in the United States that claims 2.3 million members as of January 2004. MoveOn was formed to oppose the impeachment of President Clinton, but now tackles issues that its members choose by voting. It has raised and spent millions of dollars to influence US policy. No information is available about the median age of MoveOn members or staff, but it has been described as an “an inter-generational grouping heavily peopled by young voters, something that most political constituencies lack” (Schechter, 2004).
here's a lively discussion of "cyberbalkanization" on the Deliberative Democracy Consortium's blog. The discussion was prompted by a New York Times article last Sunday that claimed that people use the Internet to sort themselves into small, homogeneous groups and to filter out views that don't interest or please them. I've pasted my comment below, but I recommend the full discussion:
For my money, the best theoretical account of cyberbalkanization is still a 1996 paper by Marshall van Alstyne and Erik Brynjolfsson. They predict that the Internet will help people who are so inclined to increase the range and diversity of their information and contacts. They also predict that the Internet will allow people to "filter" out unwelcome ideas or contacts and to form narrow, exclusive groups. So the technology will not determine the outcome; people's motives will. And clearly, people have various motives. Some prefer diverse ideas and serendipitous encounters; others want to shun people who are different and simply confirm their own prejudices.
I am fairly pessimistic about the cyberbalkanization problem, not because of the technology, but because of cultural trends in the US. Niche marketing has become highly sophisticated and has divided us into small groups. There's more money to be made through niche programs than by creating diverse forums for discussion. Meanwhile, people have developed consumerist attitudes towards news, looking for "news products" that are tailored to their private needs. And broad-based organizations have mostly shrunk since the 1950s. In this context, the Internet looks like a means to more balkanization. In a different context, such as contemporary Saudi Arabia, it may have a much more positive impact.
I'm writing a paper (for a conference organized by Lin Ostrom) that connects my two main preoccupations: the Internet as a commons, and youth civic development. Actually, I believe this link is very important. A "commons" is a public asset. It requires voluntary contributions, and it can be ruined by pollution or exploitation. Therefore, it depends on people who display trust, reciprocity, long time-horizons, optimism about the possibilities of voluntary collective action, and personal commitment. People have to be raised this way; they aren't born "civic" (i.e., with a deep feeling of belonging and responsibility for some common good).
Lots of evidence shows that people develop durable attitudes toward the public sphere during adolescence. They either come to see themselves as efficacious, obligated, critical members of a community, or they do not. Their identity, once formed in adolescence, is hard to shake. This theory derives from Karl Mannheim, but it has considerable recent empirical support. In the 1920s, Mannheim argued that we are forced to develop a stance toward the public world of news, issues, and governments when we first encounter these things, usually in our teens. Our stance can be one of contempt or neglect, or it can be some kind of engagement, whether critical or conservative. Most of us never have a compelling reason to reassess this stance, so it remains in place throughout adulthood. That is why generations have enduring political and social characters, formed in their early years.
Unfortunately, there is reason to suspect that young Americans are less likely to develop civic identities and values today than in the past. For instance, most of the decline in social trust since 1970 is a result of young Americans becoming highly distrustful of fellow citizens. This is bad news for any effort to develop a commons--whether a small-scale resource like a community garden or a vast social form like the Internet. There are (of course) some young people with habits and norms that are friendly to the commons, but not nearly enough.
On the bright side, we know how to develop civic identities. Adolescents need to feel that they are assets, rather than potential problems; that they matter to a group. It also helps to have direct experience with civic or public work. This is the impetus behind much service-learning. It is also what we are trying to accomplish at Maryland by helping young people to create free public goods for display on a community website.