Abstract

Both the theory of the public sphere and the utopian rhetoric surrounding the Internet have been a focus of scholars for some time. Given the ability of people to connect with others around the globe through the Internet, could the Internet give rise to online public spheres? If so, how would such spaces work? This article proposes that public spheres do exist on the Internet, and details how one functions. The case under study is the website Slashdot (http://slashdot.org), an online community of computer enthusiasts. The article studies the mechanisms, both normative and in code, that are vital to Slashdot's functioning, and shows how they help Slashdot function as a public sphere.

Introduction

The utopian rhetoric associated with the Internet has a long history, going back to the telegraph. Carey (1989) has labeled it "the rhetoric of the technological sublime." Primarily, this rhetoric focuses on the potential to connect previously unconnected people, so they may discuss ideas and reach new understandings, often involving democracy. The concept of the public sphere, originally detailed by Habermas (1962/1989), also concerns the coming together of peoples and the discussion of ideas, again often related to governance and democratic ideals. Given this connection, the question naturally arises: Could the Internet allow for an online public sphere or spheres? If so, how might such a sphere function?

This article details the workings of the website Slashdot, which labels itself "News for Nerds." Commentators have presented Slashdot in different lights, including as a public sphere, an online community, an online bulletin board, a computer news site, or a massive blog (a recent term, derived from "web log"). Unlike offline social movements that use the Internet for organization, however, Slashdot started online, and as the Internet and computer technology have spread, Slashdot's user base has grown and the areas it covers have expanded. Despite this growth, no one offline group or coalition claims Slashdot as its online forum. The main cultural force behind Slashdot is open source software, which permeates many of Slashdot's stories, its norms, and the code that runs it. Slashdot's open source code is used and supported by several Slashdot users. The idea behind open source is very similar to the idea of the public sphere, as will be shown.

This article analyzes Slashdot as an online public sphere. First, the concept of the public sphere is detailed. Scholars have reformulated and expanded the idea over time, and the article briefly details this history. The article then discusses how the concept of the public sphere relates to the Internet. Slashdot is described, and the mechanisms that help Slashdot function as an online public sphere are detailed. It is argued that computer code and social norms interact in a manner that makes Slashdot a public sphere.

The Public Sphere

Habermas (1962/1989) originally described the public sphere as a new opening in the social and political fabric of Europe during the 18th and 19th centuries. Formed around discourse through meetings in salons and coffee shops, as well as in newspapers and other printed forms, the public sphere gave a voice to members of the public who were previously not included in issues of governance. The public sphere was a new, budding form of democracy. Eventually the public sphere would be corrupted and co-opted in part by the commercialization of the press through advertising and entertainment.

This sphere, as described by Habermas, was strictly a "bourgeois" public sphere. Many discuss "the public sphere" without the qualifier, as a more generalized form. Although it represented a new avenue for some people, it was restricted in terms of class and gender, in that only bourgeois men could participate (Fraser, 1992). Habermas's original conception includes three important elements. One is that the public sphere was formed through discussion, often mediated. Second, it represented a new space of discussion for many who had previously been excluded. Last, ideas presented in the public sphere were considered on the basis of their merits, and not on the social standing of the speaker.

These three elements of a public sphere are important to the present analysis for the following reasons, which are detailed in greater depth below. Discussion of ideas, often in mediated form, is characteristic of Slashdot as a mediated discussion space. It is difficult to imagine the public sphere without discussion of ideas, because it was such discussion that brought the sphere into existence. This new space of discussion also allowed those who had been excluded from issues of governance to have a voice, much as Slashdot itself is a space that allows many who could not have done so easily before to discuss political (and other) issues. That ideas were evaluated on their merits flattened the political hierarchy into a more equal, democratic arena. Slashdot has a system for judging users' comments that is primarily based on the comment itself, and not the identity of the commenter, and it could be described as a space of rational-critical debate.

The "public" of the public sphere represents the idea that the people taking part in such discussions are acting not as private citizens, but in a public role by speaking in public and discussing issues of relevance to the wider public, issues of governance. They have the potential to affect public opinion and public policy. The "sphere" of the public sphere denotes that it is a space, often a mix of formal and informal institutions and organizations.

Since Habermas's original work, there have been various reformulations of and additions to the idea of the public sphere (Calhoun, 1992), and in fact the literature on the public sphere has become quite large (as noted, e.g., by Dahlgren, 2001b and Sparks, 2001). One important addition to the public sphere theory is the notion of multiple publics, rather than of one overarching public sphere (Asen & Brouwer, 2001; Dahlgren, 2001b; Fraser, 1992). The possibility for multiple public spheres is relevant here, given the large number of people who use the Internet. It is doubtful that a single public sphere could consist of millions of people and still function, since deliberation would be difficult. Allowing for multiple publics, with different interests, allows for smaller and thus workable, yet still global, public spheres through the Internet.

Fraser (1992) and many others (e.g., Palczewski, 2001; Squires, 2002) discuss counter-publics, which are identity-based groups that are in opposition to the public sphere. Even if there is one primary public sphere, in this light there are possibly many counter-publics. Garnham (1992) holds the view that there is one overarching public sphere, but notes there may be other public spheres within it, "each organized around its own political structure, media systems, and sets of norms and interests" (p. 371).

Dahlgren (2001a) describes publics as "issue publics," where, similar to Garnham (1992), publics are organized around various issues. For instance, Squires (2002) details several types of counter-publics, focusing on African-Americans. These publics are organized around the issue of identity (race or ethnicity). Anderson (2003) studied the Islamic public sphere, in which identity is based on religion.

Erimbayer and Sheller (1998) detail a schema for approaching different types of publics, and take account of three different dimensions of a public sphere. They do not study one particular sphere, but instead allow for the possibility of many different kinds of spheres. Given the work of these scholars and others, using a multiple public sphere framework is appropriate.

Other scholars have studied specific issue publics on the Internet, such as right-to-die advocates (McDorman, 2001), and various non-governmental organizations (Mater, 2001). Others have investigated the relationship between the Internet and different concepts of the public sphere, and have raised important questions about its functionality (Dahlberg, 2001; Palczewski, 2001; Papacharissi, 2002; Poster, 1999). Overall, their findings are mixed. One common refrain is that the Internet has the potential to form a public sphere or to assist the public sphere, depending on the scholar's approach, but that this is no guarantee that such an outcome will occur.

Papacharissi (2002) identifies three contradictory issues for the Internet and the public sphere. One is that although the Internet allows for great amounts of information storage, access and literacy are likely to be unequal. Second, although people around the world can communicate with each other far more easily with Internet technologies than with previous technologies, there may be audience fragmenting. This would be similar to strict issue publics, or Sunstein's (2001) "daily me" newspaper. The third issue, essentially, is that any online public spheres will face the problems of Habermas's bourgeois public sphere, and become corrupted by commercialism.

Dahlberg (2001) finds that "the Internet is facilitating discourse that replicates the basic structure of rational-critical debate and that in various ways approximates the requirements of the public sphere." Yet, he feels that this discourse falls short of the requirements of the public sphere.

The work of these scholars suggests that there is a clear connection between the Internet and the public sphere. However, the Internet is a network of networks, and over its history has continually changed in terms of technology, both underlying and on the user end. To say that "the Internet" is like a public sphere may be an intuitive statement, but it must be empirically grounded. Currently the Internet is a combination of chat, Usenet, the World Wide Web, email, and other protocols. How the Internet or spaces on the Internet actually function as a public sphere remains to be seen.

Dahlberg (2001) has proposed six criteria that an online space must meet in order to be considered a public sphere. They are: autonomy from state and economic power; exchange and critique of criticizable moral-practical validity claims; reflexivity; ideal role-taking; sincerity; and discursive inclusion and equality. However, Dahlberg focuses more on an overarching, single public sphere and less on one sphere within a multiple public sphere framework. In as much as the present study examines an issue public or counterpublic (in the context of multiple publics), an alternate set of criteria is proposed. Dahlberg does, however, point out that, even with the difficulty of generalizability from specific cases, further research such as case studies are invaluable. By not just characterizing Slashdot as a public sphere, but by further analyzing the mechanisms that make it so, this study hopes to contribute to the conversation about the Internet and the public sphere.

The Public Sphere Online

Based on the research detailed above, we can come to a useful definition of a public sphere that may be applied to the Internet, and that may allow us to identify a space for analysis. Drawing from Habermas's original conception (as noted above), we have the four following criteria:

1. Public spheres are spaces of discourse, often mediated.
2. Public spheres often allow for new, previously excluded, discussants.
3. Issues discussed are often political in nature.
4. Ideas are judged by their merit, not by the standing of the speaker.

Given that there are multiple public spheres, there are potentially multiple choices available for study on the Internet. Slashdot meets these four criteria, as will be shown. Note that the four criteria do not make allowances for a radically different form of public merely because the public is online. An online public sphere is still a form of public sphere, online or not, and so must meet basic public sphere criteria.

In a multiple sphere framework, people can be divided along identity-based or interest-based lines. Common high-level delimiters are race, ethnicity, class, gender, language, nationality, and religion.¹ These are, in Dahlgren's (2001) terminology, "issue publics," public spheres formed around issues of interest, much as Habermas's bourgeois public sphere was an issue public formed around issues of interest to the bourgeois in Europe at the time, primarily trade and politics. Within this public sphere framework, Slashdot can be viewed as an issue public where the issue is computer-related concerns.

With regard to the first criterion, Slashdot is a website dedicated to the discussion of computer news and related issues. The majority of the site consists of user comments, and users' replies to those comments. Often considered a discussion, it is clearly mediated since it takes place over the Internet. Any online public sphere currently must be mediated by definition.² Face-to-face communication, however, was an important part of the original conceptualization of the public sphere (Thompson, 1995).

With regard to the second criterion, Slashdot was originally a space where previously disparate computer users could come together in large numbers and discuss issues they felt were relevant. Slashdot has been in existence since 1997 and can no longer be considered new, but its members were previously excluded from discussions of governance and other issues.

In terms of the third criterion, many issues discussed on Slashdot are political, although the majority are not and some are consumer-oriented. Slashdot stories often involve law, such as the digital millennium copyright act (DMCA); legal cases involving individuals in the U.S., Europe, and Australia; legal battles between companies as well as between countries and companies; patents (primarily in the U.S.); and there have been interviews with lawyers and even a U.S. Congressman.³ These discussions are always related to computer issues. Slashdot has made different sections (like a newspaper), and one section is "your rights online" (YRO), which deals with legal issues. Slashdot only creates sections for topics that consistently have a large number of stories. One important topic for Slashdot is open source software, which is explained below. Open source software, briefly, is software where the source code is available to everyone. It is "free as in speech," and many coders see it as a form of democracy.

The fourth criterion is important, as many spaces and communicative forms on the Internet fail to meet it. There are two components to this criterion, the first being that ideas (comments in the discussion) must be judged, but also that the comments must be judged on their own merits. Slashdot fulfills the first part of this criterion and, in theory, fulfills the second part, although in practice it only comes very close (close enough, in fact), as this study will show. Baoill (2000), while describing Slashdot as a public sphere, fails to mention Slashdot's system for judging comments (called moderation). However, moderation of comments is a vital part of Slashdot (Lampe & Resnick, 2004). The criterion of judgment was important in Habermas's (1962/1989) original description of the public sphere, and it is still important to public sphere scholars today (i.e., Ferree, Gamson, Gerhards, & Rucht, 2002), Without its moderation system, Slashdot would not be a public sphere.

This section has laid a basic groundwork for approaching an online public sphere, and has argued that Slashdot meets the proposed four criteria for an online public sphere. In the following section, a brief overview of Slashdot is presented, detailing some non-obvious but important details. The rest of the study describes the mechanisms that allow the site to function as an online public sphere. By studying these mechanisms, this study hopes to show not only how one form of an online public sphere works, but also suggest how other online public spheres may be formed. Suggesting theoretical criteria is one step, but the next is to move towards actual practice.

Slashdot

News for Nerds, Stuff that Matters

In early 1997, Rob Malda, then at Hope College in Holland, Michigan, started a website called "Chips & Dips," using his college account (chromatic, Aker, & Krieger, 2002). That fall, he and his friend Jeff Bates registered the domain name "slashdot.org" and moved the site to a dedicated machine. Over time they recruited some of their friends to help with programming new features and adding more automation to the site. Slashdot is still run by Malda and Bates, but they have expanded and currently OSTG (Open Source Technology Group, formerly Open Source Development Network), a large computer company, owns Slashdot.

The first-time visitor to Slashdot will notice that the site seems like a news site, with a column of stories running down the middle, a variety of other items along the sides, and an advertisement along the top of the page or on the side. Stories have headlines, synopses, and a small icon that relates to the topic. However, if a visitor clicks on a link to read more about a story, one of the main differences about Slashdot becomes apparent. The story page is not written by a reporter or staff editor; in fact, it is not written in a traditional offline manner at all. It is instead constructed, over time, by Slashdot users who post ideas and responses about the story and about each other's posts, similar to threaded discussion forums such as BBS and Usenet newsgroups. The main part of the initiating story is usually at another website, perhaps a news site, a press release, or an individual's site. Slashdot provides a synopsis and a link to the story, leaving Slashdot as primarily an online discussion space. Occasionally the story content is on Slashdot, such as with some reviews or interviews, but this is usually not the case. The point is that Slashdot is formed by online discourse.

Stories cover a range of issues, although since they are submitted by readers and then selected from these submissions by the people who work at Slashdot, stories almost always relate to computers and computer culture in some way. Topics include movie and book reviews, legal issues, computer companies, operating systems, programming languages, and interviews, to name just a few. It is, as Slashdot describes it, "news for nerds." One important topic discussed on Slashdot is open source software.

Open Source

Open source software is software where the source code is available for anyone to download and, importantly, anyone is allowed to change the code in their copy to their liking. The majority of consumer software in the 1980s and 1990s was not open source; consumers purchased a compiled product that was ready to run, and were not allowed access to the source code.

Open source can be explained in terms of automobiles. People who own a car are allowed to open up the hood and to tinker with the engine, repairing or tweaking it as they see fit. If a car owner does not have the skills to repair an engine, then they can take the car to any car shop they want, the dealer, or have a friend work on it. Open source is the same way. Users are allowed to change the software they have, or can have someone who can code fix bugs or add features. Closed source software is like having a car where you can't open the hood, and where you have to take the car to the dealer to get it repaired.

Open source software is usually "free" in two ways. The saying among open source advocates is that it is "free as in speech, not as in beer," although most open source software is gratis (free, no charge) as well as libre (free, no restrictions). There is, in fact, somewhat of a division among people advocating this approach to software, between calling it "free software" and "open source software." Free software advocates stress the ideological aspects of free access to the source code, while open source advocates stress the benefits to businesses. Both sides fundamentally agree about the underlying issues, and the difference is one of nuance, for the most part. Either way, access to the source code means that if something is broken, you can fix it, and if you want a new feature, you can add it. You can play with the code.⁴

Open source is the primary focus of Slashdot, although topics range far and wide, and as long as they are of interest to Slashdot, stories can be accepted. The culture of openness, however, permeates the site, so not only is the code that runs Slashdot open source (and it is used by a fair number of web sites), but the ability for users to post content and judge other users' comments is also a reflection of the openness of the site.

Open source is similar to the concept of the public sphere, in that part of the idea behind open source is access to power through access to code, mirroring the public sphere ideal in which the public sphere is seen as access to power through access to discussion. With open source, anyone can change code and affect their copy of the software (if they can program). With the public sphere, many are allowed to take part in discussions and (hopefully) affect political decisions or outcomes. Open source is often seen as a speech issue, closely related to basic human rights and democracy, much as the public sphere ideal is as well. Further, open source is often seen as a way to avoid the power and control of corporate interests, while counterpublics, as part of the public sphere concept, are seen as spaces outside of the control of larger powers.

Demographics

Another perhaps less obvious aspect of Slashdot that is relevant to this analysis is user demographics. Habermas's original, bourgeois public sphere was primarily restricted to men of certain classes, and was European. Slashdot, as it turns out, although open in theory, is restricted in similar ways. This suggests that rather than Slashdot being the online public sphere, it is an issue public, a counterpublic, or a small part of the online public sphere.

The publicly available statistics for Slashdot are from the OSDN (2002) press kit, and as such represent all of the OSDN (recently renamed OSTG; see above) sites. However, they are most likely very close to the actual numbers for Slashdot.

The majority of requests for OSDN web site pages come from the United States of America, with 23% coming from other countries. This must be contextualized, however. The OSDN sites (and Slashdot) are in English, and relate to the computer industry, which is particularly strong in the U.S.. The U.S. has a high percentage of Internet connectivity compared to many other countries. With this in mind, it is impressive that nearly one-quarter of hits come from non-U.S. countries.

Of this 23%, however, 63% are from Europe and 10% are from Canada.⁵ The rest of the world does not fare as well. Only 2.3% of all page views are from Asia. The southern hemisphere fares particularly badly: the Southern Pacific region accounts for 2.3%, South America 1%, Central America 0.2%, and Africa 0.2%. Language and connectivity are likely factors that limit use, as may be the lack of large computer industries. There are also several non-U.S. Slashdot-like sites; for instance, OSDN runs a Japanese Slashdot, and there are sites in Russia, Switzerland, Romania, Germany,⁶ and a site in Castilian Spanish.⁷

The gender gap is exceedingly large for OSDN sites. It is unclear how OSDN determined the statistics for gender, but they report that 94% of OSDN site viewers are male. Although there is absolutely no restriction in theory with regards to the gender of computer users or Web surfers, those in the computer industry are predominantly male, and the Slashdot culture may be self-reinforcing in this regard. This is also interesting in that some numbers have shown that there are actually slightly more women online than men in the U.S. (Muehlbauer, 2000). The UCLA Internet study (UCLA Center for Communication Policy, 2003) found nearly equal numbers of men and women online, albeit slightly more men, for 2000, 2001, and 2002.

Slashdot itself receives an enormous number of page views every month. According to the Slashdot press kit, the number of page views is 43 million, although according to Slashdot's Bates more recent estimate, is it 55 million (personal communication, March 2002). This number has increased greatly over time: in June of 2000, Slashdot served 30 million pages a month (sm300).⁸ Managing the online conversations of these thousands of visitors across millions of page views is challenging; the mechanisms that make it possible are detailed next.

Mechanisms

Slashdot's Moderation System

Moderation and its counterpart, meta-moderation, are in many ways what make Slashdot work. Habermas' coffee shops can only fit so many people, but an online bulletin board does not face the same restrictions. With so many posts to Slashdot every day, there needs to be a way to filter them. Moderation is a mechanism by which users judge comments: is a comment insightful, off-topic and to be avoided, or perhaps funny? Users can then view comments ranked by score, filtering out the ones that were moderated down (not useful, or unrelated to the story, or in other words, not an appropriate contribution to the discourse).

At first, when Slashdot was small, the "Slashdot guys" moderated posts themselves, but as the user base expanded, the job of moderating became larger and more difficult for a small group to manage. Eventually, moderating powers were given to others. Today there is a group of permanent moderators, and every user who is logged in using his or her account has the chance to moderate. As described by Malda:

Each day we grew, adding more and more users, and increasing the number of comments submitted. As this happened, many users discovered new and annoying ways to abuse the system. The authors had but one option: Delete annoying comments. But as the system grew, we knew that we would never be able to keep up. We were outnumbered. (cm520)

That users abuse the system is in itself interesting, because if Slashdot were more of a closed system then this might not happen. Moderation allows for an open user base and anonymous posting because it helps weed out the "annoying comments." Malda continues,

[s]o, I picked people to help. Just a few. 25 or so at the end. They were given the simple ability to add or subtract points to comments. The primary function of these brave souls was to weed out spam and First Post and flame bait. Plus, when they found smart stuff, to bring it out. (cm520)

Malda's construction of the moderation function is to mark both unwanted, unconstructive posts (spam, first posts, flame bait), and "smart stuff." Calling these early moderators "brave souls" acknowledges that this is not an easy job, both because there is a large amount of material to moderate, and because sometimes deciding how to moderate a comment is difficult. This system worked for a while, but as Slashdot grew, it needed even more moderators.

At first, Slashdot tried picking 400 more permanent moderators in the same way as the first 25 extra moderators. This did not work out, however, as Malda details: "Immediately several dozen of these new moderators had their access revoked for being abusive, but they settled down" (cm520). Malda knew that there needed to be more than 25 moderators, but needed a way to curb would-be abusers in a larger pool of moderators. Foreseeing further growth in the number of users and thus moderators, Malda noted, "I knew that someday I would have even less control since I intended to eventually give access to even more people." (cm520)

Moderation is relevant to the public sphere because, in the discussion that creates the public sphere, comments are judged solely on the merit of the comment. For the most part, this is how Slashdot comments are judged, although moderators can see the commenter's account name. A comment's score may also be affected by other factors (discussed below).

Over time, moderation evolved to the point where "any regular Slashdot reader is probably eligible to become a moderator" (cm520). Malda appeared very aware of issues of control and access. The control had to be moved away from the managerial end and towards the user end by creating the moderation system. But, there also needed to be a balance, as Malda realized. "I learned that I needed to limit the power of each person…" (cm520). Otherwise, as he noted, there were those who would abuse the system. In a democratic public sphere, all are considered equal, and people's words are judged not by who they are but solely by what is said. However, there is still room for abuse of Slashdot's moderation system. For instance, a user could favorably rate his friends' comments if and when he gets moderator access. To combat this potential abuse, there is metamoderation.

Metamoderation

"Metamoderation began as an experiment to see if the primary moderation system was actually working. Now, it helps us to remove bad moderators from the [moderator] eligibility pool" (mm300). Metamoderation is the moderating of the moderation done to comments, thus the "meta."⁹ Whereas moderation judges comments, metamoderation judges the moderation done to comments. Instead of rating the usefulness of a comment, metamoderation rates the fairness and accuracy of that judgment.

Metamoderation is not as statusful as full moderator access, since everyone with an account that is not in the newest 10% of accounts can metamoderate. One has to be logged in from a qualifying account, so as to identify oneself as part of the community. If so, a user can metamoderate once per day.

According to the metamoderation statistics, 92-93% of moderations are judged as fair (cm605). There is probably some error on the metamoderators' part; however, given that metamoderation is similar to inter-coder reliability, a 92-93% reliability measure is excellent, even though the metamoderation statistic is not as analytically robust.

Scoring

When posts are moderated, they can receive a score ranging from -1 to 5, although any given moderator can only adjust a post by one point. Posts by different people can start at different scores, however. Posters who choose to post anonymously (either by not having a login or by using the "anonymous coward" option if they have a user account) have their posts start at 0. Users who are logged in initially start at 1, but depending on a user's record this can go down to 0 or up to 2. Thus identifying oneself and having a good track record help give one's post a better score.

With regards to gaining moderator access, five considerations are built into the system (cm520). The user must be logged in, which is done so that moderation can be judged (meta-moderation); users who abuse the moderation system by doing a bad job may be refused moderation access in the future. Moderators must also be regular Slashdot readers: they must read some stories, but not be obsessive about it. As Malda puts it, "no obsessive compulsive reloaders, and nobody who just happened to read an article this week" (cm520). The system is designed not to give moderator access to new Slashdot accounts, in part because abusers might create large numbers of new accounts to try to get access, and also because users need to be part of the community and to understand how moderation works in order to apply it themselves. Also, users must be "willing to serve," in that there is an option on the preferences page which allows people to indicate whether or not they want to moderate. The option is on by default.

In Habermas' conceptualization of the public sphere, comments are judged by their merit, not by the identity or status of the speaker. The same is true within the Slashdot system, although people with good karma (discussed below) can start off with a slightly higher score on their comments. Slashdot's anonymous posting option reflects the public sphere ideal, although not perfectly. Posting anonymously is seen as cowardly (thus the label on such posts, "anonymous coward"), and the score of such posts is lower than for someone using their own account. It does nonetheless allow the comment to be judged solely on its own merits, since the identity of the poster is unknown. This is not to say that it is easy to determine who a user is if that user posts from their account; however, it is possible to become familiar with other users and their comments solely through their account names, and a few people use their real names.

Karma

Karma on Slashdot attempts to be a reflection of a user's actions, and is comprised of three things. If your comments are moderated up or down, you will gain or lose a point of karma (until 2002, Slashdot reported users' karma by numerical score). Thus if you post a useless comment and it is moderated down, you will lose karma (some people don't care about this, but others do). If you submit a story and it is accepted, you gain a point. Metamoderation can also affect karma, so poor moderation can cause one to lose karma. However, a user's karma is mostly a reflection of moderators' judgments on that user's comments.

Karma is clearly important to some, as it plays a role in moderation.¹⁰ Inevitably, some people take karma too seriously. Some users are called "karma whores" because they employ explicit tactics to raise their karma. One such tactic is to post funny comments to a story very early on. This works because funny comments can be moderated up, and because it is easier to get moderated before there are too many comments attached to a story. As an indication of Slashdot's culture and a reflection of how seriously some take karma, Malda makes the following comment:

Karma is used to remove risky users from the moderator pool, and to assign a bonus point to users who have contributed positively to Slashdot in the past…. Karma fluctuates dramatically as users post, moderate, and meta-moderate. Don't let it bother you. It's just a number in the database. (cm720)

Nonetheless, because of some users taking karma too seriously, in July of 2002 Slashdot stopped reporting users' karma numerically, and switched to six text labels (terrible, bad, neutral, positive, good, and excellent) (cm703, cm706).

Moderation

Most users want to view the better comments, and this would not be possible without moderation. Moderating well requires users to visit the site often, be familiar with Internet culture, and make value judgments in keeping with the values of the Slashdot community.

Users who have temporary moderator access receive five moderator points, which are good for a few days or until used up, whichever comes first. Story pages appear to moderators as they normally do except for an option to moderate every comment included below that comment. Postings may be left as is, or may receive a variety of moderations that can score them up one point or down one point. The +1 moderations are "Insightful," "Informative," "Funny," and "Underrated." The -1 moderations are "Off Topic," "Flamebait," "Troll," "Redundant," and "Overrated."

Some of these classifications, such as funny, are self explanatory, while others are Internet culture terms, such as flamebait and troll.¹¹ Some overlap at times (for example, insightful, informative, and funny can all overlap). As in the concept of the public sphere, insight and information are valued above other types of comments, although humor is also appreciated on Slashdot.

Moderating can be a challenge. Although giving users only five points with which to moderate is meant to balance out power, it has the effect of requiring moderators to decide which comments out of several dozen they want to spend their five points on. It is easier to moderate the newest stories, since they usually have fewer comments than older stories. Moderation also favors viewing a story's comments by the newest first, so as to see posts that other moderators have not yet judged. This is in contrast to the way that some Slashdot users view postings when not moderating, which is by highest score first.

It is in a user's best interest to moderate well when given moderator access, since it makes it easier for everyone to read better comments. Metamoderation hopefully weeds out those who do a poor job of moderating. Thus users with an interest in reading interesting and insightful comments should understand the importance of moderation on Slashdot; it is part of what makes Slashdot what it is.

Conclusions and Future Research

Scholars have made it clear that Habermas' original conception of the public sphere was incomplete. Their reconceptualizations have proven useful in that they acknowledge the complexities of the public sphere idea, and provide a useful starting point from which to examine online public spheres. Viewed from this broader perspective, Slashdot is a form of online public sphere, specifically a type of counterpublic or issue public, depending on the lens used.

Slashdot meets the four criteria laid out for an online public sphere: (1) it is a space of discourse; (2) it opens a space for a wide range of computer enthusiasts to come together and discuss many issues, (3) including many political ones; and (4) ideas are judged by their merit. These criteria have mostly been applied to offline publics, but online publics should not have a form that is radically different from their real-world counterparts.

Recent work has taken new approaches to public sphere theory, and some work has expanded the theory so that it encompasses the Internet and online forms. This article has argued that it is necessary to move beyond theory and study specific online publics, with attention paid to the mechanisms that allow them to function. The analysis should not be taken to imply that web-based bulletin boards such as Slashdot are the only form of public sphere on the Internet; other online publics may form on the Usenet, or through email lists, for example. However, Slashdot is an interesting case, not only because it is a public that is purely online, but because the site is open about how it works, both in terms of its social norms and its computer code. All of Slashdot's content is available through the web, and there are detailed explanations as to how it functions. This makes the mechanisms that make it work more transparent.

Slashdot's primary mechanism, moderation, supports the concept of the public sphere in that it allows comments (ideas) to be judged on their own merit. The other mechanisms, metamoderation, scoring of comments, and user's karma, are all a part of the moderation system. At the same time, moderation and preferences also support the open culture of the site by allowing users some control to filter content. The democratic aspects of moderation recall the public sphere's focus on participatory democracy.

It is nonetheless troubling that a site like Slashdot that is open in terms of its technology and its norms is still relatively closed in some ways, e.g., to women and non-U.S. users. As Palczewski (2001) reminds us, the Internet "creates the appearance of being open to all publics, yet [it] presents economic barriers that make it difficult for the ideal to be realized" (p. 181). Viewing Slashdot as an issue public makes the demographics more understandable, albeit not entirely unproblematic.

One avenue of future research is related to the ongoing debate about the potential of the Internet to create global connectedness (Carey, 1989) at one extreme, or separation and isolation—a cyberbalkans (Van Alstyne & Brynjolfsson, 1997) or a "daily me" (Sunstein, 2001)—at the other. The website Slashdot and its imitators around the world represent an interesting case in relation to these issues. Slashdot is a computer news portal of sorts, and links to stories from places such as the U.S., Asia, Europe, and Australia. Based in the U.S., it is in English. However, there are several imitators, some of which even use the same webserver code as Slashdot, for example, Barrapunto in Spain (barra = bar or slash, punto = point or dot), Slashdot Japan, and Symlink in Switzerland (using Swiss-German). These sites all try to be a form of "Slashdot," striving to be similar in terms of their embrace of (global) computer culture. Yet all come from culturally different geopolitical and linguistic settings. Measuring their overlap in terms of stories covered could provide a preliminary indication of where Slashdot lies between the utopian "oneness" and dystopian "atomized" extremes.

Ultimately, open source software, Slashdot's main theme, may play a key role in making Slashdot an online public sphere. To the extent that Slashdot's norms and codes reflect the ideas of open source, the community's norms and code can be said to reflect ideals of the public sphere. Trying to create other public sphere-like spaces on the Internet is a laudable goal, but it may prove difficult unless the cultural norms of the space in question contain public sphere-like elements to begin with.

Acknowledgments

I would like to thank the following people for their contributions to this article: the anonymous reviewers, the JCMC editorial staff, Cliff Lampe of the University of Michigan for his help with Slashdot, and Susan Douglas of the University of Michigan and Margaret Young of Albion College for help with early drafts.

Notes

1. Some of these identifiers are used by Huntington (1996) in his work on civilizations.
2. This will become a point of contention when the Internet easily supports real-time video and many people can use such technology, thus replicating face-to-face discussion.
3. Virginia Representative Rick Boucher.
4. The specific terms are laid out in the license for the software. One of the most common open source licenses is the GNU General Public License (GPL). See http://www.fsf.org/licenses/licenses.html for more information.
5. Canada is bilingual, although many French speakers also know English.
6. These sites are both in the aforementioned languages and domains, so for instance the Russian site is in Russian and in the .ru domain (http://slashzone.ru/).
7. See http://slashcode.com/sites.pl for a list of all sites that use Slashdot's code, most of which are in the U.S. and have little to do with computer culture.
8. APA style guidelines do not include a style that is appropriate for referencing Slashdot FAQ Web pages. Thus I am using a modified format that falls under general APA guidelines. The in-text reference, sm150, indicates the FAQ (sm, "slash meta" for the Meta FAQ) and the in-page link anchor (sm150) that points to the exact FAQ. Thus "sm150" indicates http://slashdot.org/faq/slashmeta.shtml#sm150 and points directly to the correct section of the correct Web page. An additional problem for APA style is that Slashdot pages are dynamic and often undated, making it difficult to identify an exact and unchanging source.
9. Slashdot refers to it as M2.
10. As of February 27th, 2002, I had a karma of 27, which is below the maximum of 50 but above 25, which gives me a bonus when posting. Under the new system, this is reported as "excellent." Most regular users have this level of karma.
11. Flamebaiting is trying to cause a flame war, which is an overly heated and unthinking series of rants among the participants. Trolling is similar since it is a form of baiting, but instead of trying to cause a flame war a troll might make a one-sided statement about something, "trolling" for people who don't share that opinion and trying to irritate them into a response.

References

Anderson, J. (2003). New media, new publics: Reconfiguring the public sphere of Islam. Social Research, 70 (3), 887-906.

Asen, R., & Brouwer, D. (Eds.). (2001). Counterpublics and the State. New York: State University of New York.

Baoill, A. (2000). Slashdot and the public sphere. First Monday, 5 (9). Retrieved May 1, 2003 from http://firstmonday.org/issues/issue5_9/baoill/index.html

Boal, M. (2000, November 23). Rob Malda: King of the new new media. Rolling Stone.

Calhoun, C. (Ed.). (1992). Habermas and the Public Sphere. Cambridge: MIT Press.

Carey, J. W. (1989). Communication as Culture: Essays on Media and Society. London: Unwin Hyman.

chromatic, Aker, B., & Krieger, D. (2002). Running Weblogs with Slash. Sebastopol, CA: O'Reilly.

Dahlberg, L. (2001). Computer-mediated communication and the public sphere: A critical analysis. Journal of Computer-Mediated Communication, 7 (1). Retrieved May 1, 2003 from http://jcmc.indiana.edu/vol7/issue1/dahlbergold.html

Dahlgren, P. (2001a). Media and the transformation of democracy. In B. Axford & R. Huggins (Eds.), New Media and Politics (pp. 64-88). London: Sage.

Dahlgren, P. (2001b). The public sphere and the net: Structure, space, and communication. In W. L. Bennett & R. M. Entman (Eds.), Mediated Politics: Communications in the Future of Democracy (pp. 33-55). Cambridge: Cambridge University Press.

Emirbayer, M., & Sheller, M. (1998). Publics in history. Theory and Society, 27 (6), 727- 779.

Ferree, M., Gamson, W., Gerhards, J., & Rucht, D. (2002). Four models of the public sphere in modern democracies. Theory and Society, 31 (3), 289-324.

Fraser, N. (1992). Rethinking the public sphere: A contribution to the critique of actually existing democracy. In C. Calhoun (Ed.), Habermas and the Public Sphere (pp. 109-142). Cambridge, MA: MIT Press.

Garnham, N. (1992). The media and the public sphere. In C. Calhoun (Ed.), Habermas and the Public Sphere (pp. 359-376). Cambridge, MA: MIT Press.

Habermas, J. (1962/1989). The Structural Transformation of the Public Sphere: An Inquiry into a Category of Bourgeois Society. Cambridge, MA: MIT Press.

Huntington, S. (1996). The Clash of Civilizations and the Remaking of World Order. New York: Simon & Schuster.

Kellner, D. (n.d.). Habermas, the public sphere, and democracy: A critical intervention. Retrieved June 24, 2003 from http://www.gseis.ucla.edu/faculty/kellner/papers/habermas.htm

Lampe, C., & Resnick, P. (2004, April). Slash(dot) and burn: Distributed moderation in a large online conversation space. Paper presented at CHI 2004, Vienna, Austria, April 24-29.

Lemos, R. (2002, June 24). Behind the Slashdot phenomenon. C|Net. Retrieved April 30, 2003 from http://news.com.com/2008-1082-938615.html

Mater, M. (2001). A structural transformation for a global public sphere? The use of new communication technologies by nongovernmental organizations and the United Nations. In R. Asen & D. Brouwer (Eds.), Counterpublics and the State (pp. 211-234). New York: State University of New York.

McDorman, T. (2001). Crafting a virtual counterpublic: Right-to-die advocates on the Internet. In R. Asen & D. Brouwer (Eds.), Counterpublics and the State (pp.187-210). New York: State University of New York.

Muehlbauer, J. (2000, August 10). Women unite, take back the site. The Industry Standard. Retrieved January 22, 2005, from http://ecom.communications.com/news_0800.htm

Net vs. norm: The Slashdot effect. (2000, February 21). Forbes. Retrieved April 29, 2003 from http://www.forbes.com/asap/2000/0221/043_print.html

OSDN. (2002). 2002 Media Kit. Retrieved April 29, 2003 from http://www.ostg.com/advertising/index.htm

Palczewski, C. (2001). Cyber-movements, new social movements, and counterpublics. In R. Asen & D. Brouwer (Eds.), Counterpublics and the State (pp. 161-186). New York: State University of New York.

Papacharissi, Z. (2002). The virtual sphere: The Internet as a public sphere. New Media & Society, 4 (1), 9-27.

Poster, M. (1999). The net as a public sphere. In D. Crowley & P. Heyer (Eds.), Communication in History: Technology, Culture, Society. London: Longman.

Priestly, M. (1999). Honest news in the Slashdot decade. First Monday, 4 (8). Retrieved January 1, 2003 from http://firstmonday.org/issues/issue4_8/priestley/index.html

Raymond, E. (1999). The Cathedral and the Bazaar: Musings on Linux and Open Source by an Accidental Revolutionary. Sebastopol, CA: O'Reilly.

Resnick, P. (2002). Beyond bowling together: SocioTechnical capital. In J. M. Carroll (Ed.), HCI in the New Millennium (pp. 247-272). Boston: Addison-Wesley.

Rheingold, H. (1993). The Virtual Community: Homesteading on the Electronic Frontier. Reading, MA: Addison-Wesley.

Slashdot. (n.d.). Slashdot: News for Nerds. Stuff that Matters. Retrieved January 11, 2005 from http://www.slashdot.org

Slashdot. (n.d.). About Slashdot. Retrieved January 11, 2005 from http://slashdot.org/about.shtml

Slashdot. (n.d.). Comments and Moderation FAQ. Retrieved January 11, 2005 from http://slashdot.org/faq/com-mod.shtml

Slashdot. (n.d.). Meta FAQ. Retrieved January 11, 2005 from http://slashdot.org/faq/slashmeta.shtml

Slashdot. (n.d.). Meta-moderation FAQ. Retrieved January 11, 2005 from http://slashdot.org/faq/metamod.shtml

Sparks, C. (2001). The Internet and the global public sphere. In W. L. Bennett & R. M. Entman (Eds.), Mediated Politics: Communications in the Future of Democracy (pp. 75- 95). Cambridge: Cambridge University Press.

Squires, C. (2002). Rethinking the black public sphere: An alternative vocabulary for multiple public spheres. Communication Theory, 12 (4), 446-468.

Sunstein, C. (2001). Republic.com. Princeton, NJ: Princeton University Press.

Thompson, J. (1995). The Media and Modernity: A Social Theory of the Media. Cambridge, UK: Polity Press.

UCLA Center for Communication Policy (2003). The UCLA Internet Report: Surveying the Digital Future: Year Three. Retrieved March 2, 2003, from http://ccp.ucla.edu/pdf/UCLA-Internet-Report-Year-Three.pdf

Van Alstyne, M., & Brynjolfsson, E. (1997). Electronic Communities: Global Village or Cyberbalkans? Retrieved April 23, 2003 from http://www.si.umich.edu/~mvanalst/papers/cyberbalkans.pdf

Wellman, B., Salaff, J., Dimitrova, D., Garton, L., Gulia, M., & Haythornthwaite, C. (1996). Computer networks as social networks: Collaborative work, telework, and virtual community. Annual Review of Sociology, 22, 213-238.

About the Author

Nathaniel Poor received his doctorate from the University of Michigan in May, 2004. He studies social, legal, historical, and technological aspects of new media, as well as the history of media technology. More information is available at http://www.umich.edu/~natpoor/.
Address: Department of Communication Studies, 2020 Frieze Building, 105 South State Street, Ann Arbor, MI 48190, USA