Free Speech, Quality Control, and Flame Wars

Three moderators of academic online discussions bring us tales from the trenches. Whether it’s computational chemistry, the history of the book, or women’s studies, the technology and the users can both prove difficult.

By Patrick Leary, Jan K. Labanowski, Joan Korenman

Sticking to the Topic on SHARP-L and VICTORIA
Don’t let your posters get off on tangents, warns a list owner.
By Patrick Leary

Mixing the Media on CCL.NET
Blogs, Web-based discussion, and e-mail produce a great hybrid, explains one moderator.
By Jan K. Labanowski

To Moderate or Not to Moderate on WMST-L
There’s a fine line to be walked in list owning, says the founder of this women’s studies forum.
By Joan Korenman

Sticking to the Topic on SHARP-L and VICTORIA

By Patrick Leary

I manage two scholarly discussion lists. SHARP-L, whose name comes from the Society for the History of Authorship, Reading, and Publishing, focuses on book history, and VICTORIA is devoted to the culture and society of nineteenth-century Britain. The early 1990s, when I began both lists, was an exciting time to explore the potential of the Internet for scholarly purposes. The Listserv software was comparatively primitive then, which meant that a list owner’s chores could be absurdly labor-intensive at times, but it was deeply satisfying to help build these vibrant scholarly communities.

One lesson I learned along the way is that the ecology of an active online community is surprisingly fragile. It can go wildly out of whack, and even self-destruct, in a very short time. I’ve seen any number of lists decline and die over the years. Some merely waste away through attrition and neglect until no one posts to them anymore; others become mere notice boards for calls for papers and the like, without any real interaction among subscribers. A few flame out spectacularly, bursting with so many nasty, off-topic messages that all the “lurkers” unsubscribe, leaving the disputants to fight among themselves until even they grow weary of it. Charting a middle course means achieving a high “signal-to-noise” ratio, in which the list’s content is useful enough to make it worth people’s time to stay involved.

Almost all members of both lists I manage have been thoughtful and courteous. Even so, I have put in a certain amount of time mediating disputes privately, trying to cool things down before they got out of hand. Each party to an argument wants to have the last word, of course, and on an email list, these tit-for-tat exchanges quickly get tedious. By now, we’ve almost grown accustomed to the sheer rhetorical violence of much disagreement on online bulletin boards and blogs, and something of the same can now and then find its way onto even the most mild-mannered scholarly list. It’s a curious feature of the online world that people who would never descend to insults in a face-to-face argument, or even in a private exchange of e-mail, will sometimes get carried away in an online forum and commit the grossest incivilities if they’re not checked. Fortunately, people now realize,, in a way that many didn’t ten years ago, how terribly destructive careless or ill-considered online behavior can be to their professional careers. This accumulation of experience with e-mail and Listservs has helped make the task of the list owner much easier than it used to be.

Ideally, subscribers will regard a scholarly list as a collegial discursive community rather than simply a distribution mechanism for professionally useful information. A few simple rules can help to achieve this goal. For content, the most important rule is topicality. As mundane as it sounds, keeping discussion within the list’s stated topic area is the single most important job the list owner does; it’s also the least well understood and generates the most conflict with individual subscribers. On VICTORIA, for example, our cardinal rule is that every message posted to the list must have some specific pertinence to the study of nineteenth-century Britain, counting that period as the “long” nineteenth century, from about 1780 to the end of the Great War, and, of course, including all parts of the empire.

One might think that those capacious bounds would leave plenty of room for discussion, and usually they do, but there’s nevertheless a constant tendency for discussion threads to spin off into unrelated matters about which all academics everywhere are keen to express opinions. This centrifugal tendency isn’t obvious to the individual subscriber, who doesn’t see the harm in dashing off a tart opinion in a single e-mail, whether it’s about politics, Google, the failings of students, tenure, or some event in the news. But all it takes is one off-topic message to generate a dozen others. Once such threads get started, a kind of Gresham’s law comes into play in which irrelevant postings drive out topical ones. Every list devoted to one subject has to figure out the issue of topicality right at the beginning—to decide, for example, if the list is to be a forum in which historians of books can talk about anything they like, or a forum in which anyone who likes can talk about the history of books. I decided on the latter option early, and I think it’s worked out pretty well.

The most valuable function of lists like SHARP-L and VICTORIA is their role in breaking through the various kinds of isolation—hierarchical, geographic, cultural, disciplinary—that are endemic to scholarly life. Both lists cover interdisciplinary fields and bring together people from many walks of life who have widely differing perspectives: literary scholars, historians, novelists, publishing professionals, students, librarians, archivists, and others, including a good many nonacademics who take a serious interest in the subjects of the lists. The structured but informal atmosphere of the lists makes possible connections that would be difficult, if not unthinkable, at academic conferences, where keynote speakers and first-year grad students, for example, don’t often strike up conversations about their common research interests.

Other kinds of boundaries are routinely crossed as well. I recall one particularly lively discussion about how to inspire one’s students to grapple with Darwin’s ideas; the thread’s many participants included the head of the history of science program at Cambridge and a junior professor of English at East Texas Baptist State University, both of whom brought valuable but very different experiences to bear on the subject. On SHARP-L, it’s not at all unusual for members to discuss some aspect of an important new book with its author, who also subscribes to the list. And, of course, for every discussion “thread” on both lists, innumerable other ones begin in public but continue by private e-mail. Over the years, we’ve seen many scholarly collaborations, and even the occasional marriage, begin with a fortuitous on-list exchange.

Typically, a scholar looking for sources or trying to figure out whether certain questions are valuable will post a research query. Online colleagues can be enormously helpful in working through and framing a project at this early stage; the list also helps people to connect with others who are working along similar lines. Dozens of scholarly monographs, and even a few novels, acknowledge the wonderful generosity of the people on these lists. The usefulness of such exchanges extends beyond the particular discussions that give rise to them. Every message ever sent to VICTORIA or SHARP-L is accessible in online archives and searchable in different ways. These archives of tens of thousands of messages constitute extraordinary research resources in themselves. They also make for a fascinating record of many of the fields’ preoccupations over the past fifteen years.

We welcome new members to both lists every week, and the constant influx helps to keep our exchanges—even about some much-discussed topics—fresh and interesting. Many other members have subscribed for five or ten years or more; some who joined as first-year grad students are now tenured professors who recommend the lists to their own students. Our base of long-term members helps instill a sense of collective responsibility for the health and well-being of both lists as communities. The sheer thoughtfulness and generosity of the lists’ members have been a continual inspiration to me over the years. I’m proud of both VICTORIA and SHARPL, and as list owner I feel fiercely protective of them, but I don’t think of them as “my” lists—they’re our lists. They belong to the people who belong to them, and whose willingness to share ideas and information with the rest of us continues to make them such welcoming, stimulating, and collegial spaces.

Patrick Leary is a historian and publisher of VictorianResearch.org, a collection of resources for teachers, students, and researchers. Comment on this article.. Back to top 

Mixing the Media on CCL.NET

By Jan K. Labanowski

Ihave managed CCL.NET, an Internet-based public forum on the use of computers in chemistry for almost sixteen years.1It is a privilege and a great satisfaction to provide a useful service for fellow professionals. But professional need is not the only reason that CCL is operating after so many years. Subscribers stay with us because CCL is truly a virtual community. They value the informality of the medium, the quality of materials contributed by participants, and our Web-based archives of forum discussions. Maintaining the quality of our community’s discourse is not always easy, however, because the manager of an Internet forum cannot be too imposing or overly restrictive. One needs to master the tools that can control traffic on and access to the forum, especially if it is public. In online debates, a list manager must not take sides, even if he or she can offer a solid opinion about an issue. I sometimes think the job is half art, half magic.

Some Web-based forums and blogs rely on Web portals, where articles are submitted by way of a Web form, or USENET-style bulletin boards, called news groups, that are accessed with special programs called news readers. Web portals and USENET discussion groups have no defined readership. Users can peek at messages at irregular intervals, post something, and never return.2 Unlike these kinds of groups, e-mail forums have a defined subscribers’ circle whose members receive all posted messages. As a result, such forums tend to have a more stable readership and devoted participants. There is another important difference, however. With a Web-based forum, you can quickly remove an inappropriate message after it is posted and limit the number of people who read it. With the e-mail-based systems, you cannot recall what is sent.

For CCL, I chose a hybrid approach containing elements of blog, Web, and e-mail. I will also be installing a feature that alerts subscribers when new content is added to the site. You do not have to subscribe to CCL to post to the list, and you can review CCL discussions in the Web archives without subscribing. People can also submit messages to CCL by sending them to chemistry@ccl.net or by entering them into a Web-based submission form without subscribing.

In 1991, when CCL began, the forum was unprotected and unmoderated; all messages were sent to a common e-mail address. The messages were archived and made available for downloading. Once the Internet evolved into a commercial medium accessible by people from outside academia, this model could not be sustained. Unsolicited commercial advertising (spam) started to appear on the forum, and we had to switch to a fully moderated mode.

This mode has its drawbacks. Moderation is incredibly time-consuming for the moderator, slows down the pace of discussion, and can expose a moderator to legal liability. When you approve a message for distribution, not only the original author of the message is responsible for its content, but you, the moderator who approved it for distribution, are responsible as well. Imagine approving a message that contains trade or military secrets or causes “loss of revenue” for a company whose product had been bashed on your list. It can happen.

A fully moderated operation should be run from within a large organization that agrees to indemnify the moderator. Or, as a moderator, you could arrange some personal liability protection by creating a corporation to house your forum. In any case, it is important to set clear rules for your forum and make them available to all participants. They will serve as guidelines for participants and as a backup for your editorial decisions.

How can you authenticate those authorized to send messages to your forum? It’s not a good idea to allow unrestricted and unverified submissions unless you have the staff and time to review every message that is posted. In these times of spam and viruses, you can expect thousands of messages a day. Filter your traffic automatically with computer-based rules and protocols. Doing so minimizes human involvement, saves time, and allows you to “blame it on the computer” rather than having to explain your editorial decisions.

Requiring users to copy a randomly generated number into a Web form (so that primitive viruses on infected computers and spam robots are helpless) is not enough. We use highly restrictive filters against spam and viruses, keyword screening, and blacklisting and greylisting of mail (see http://www.ccl.net/greylisting for a definition of greylisting). But a few inappropriate messages still penetrate these defenses. An inappropriate message can generate an avalanche of trivial exchanges among participants, flame wars, and scores of messages to the moderator. Avoiding them is very important.

You don’t want to lose your best participants, who provide quality comments and expect the same. Making sure that the identity of the author is always included with the message (no anonymous postings) improves the situation dramatically. You can require prior registration and a password for posting. Alternatively, you can request a confirmation by e-mail using the author’s “from” address. An e-mail address on the “from” line of a message can be easily faked, and no generally accepted standard yet exists for certifying the origin of messages. However, requesting confirmation that requires human action (for example, clicking on a Web link or typing something and responding) usually works. Yes, it inconveniences the participants, but the alternative is much worse. Spammers rarely risk revealing their true identities and use fake e-mail addresses, and viruses lack the sophistication to pose as humans. CCL does not require registration or subscription prior to posting and, for that reason, the moderator manually approves the first message from any given e-mail address. Once an e-mail address is “blessed,” the author can post by responding to an automatic confirmation request.

Another dilemma relates to whether or not to reveal an author’s e-mail address in a posted message. Spammers can grab addresses easily, and we try to protect our posters from them without actually removing e-mail addresses (that would prevent readers from being able to contact an author). CCL obfuscates e-mail addresses in a random way (we replace the @ sign with random characters) that makes it difficult for robots to find and extract the addresses yet allows any human being to figure it out. Remember that spy ware and viruses that collect e-mail addresses are often present on the computers of subscribers to your forum. For that reason, only the addresses of your forum (that of the list administrator and the list distribution address) should be left unobfuscated in outgoing messages.

Although changes in technology force CCL to adapt, we are conservative and try to avoid radical or frequent modifications that would require actions or changes on the part of our readership. We would rather add functionality than replace it. Stability is a critical issue of the success of the forum, and it may be the secret to our long existence and continuing growth.

Notes

1. The space constraints of this short essay allow only a superficial presentation of many problems related to maintaining public discussion forums. The CCL.NET Web site at http://www.ccl.net provides details about the procedures and rules of our forum. For more information, write to me at jkl@ccl.net. Back to text

2. Web portals, which usually come as part of a content-management system, are quite popular. USENET, among the earliest technologies used for shared professional discussions, is now losing popularity because of the cost of maintaining servers and the bandwidth it requires. Some organizations, such as Yahoo! groups and Google groups, provide free Web-based interfaces for their groups; in this respect, their approach resembles that of blogs. Back to text

Jan Labanowski is a manager of the Computational Chemistry List in Columbus, Ohio. Formerly, he was a faculty member at Ohio State University. His e-mail address is jkl@ccl.net .  Comment on this article. Back to top

To Moderate or Not to Moderate on WMST-L

By Joan Korenman

I started the Listserv WMST-L in early 1991 specifically for discussion of women’s studies teaching, research, and program administration. It now has almost five thousand subscribers from forty-six countries. I’m not sure, but it may be the largest women-focused academic list in the world.

Strictly speaking, WMST-L is not moderated: most messages go directly to the list for distribution rather than to a moderator first. I set it up to be unmoderated for several reasons. First, I like the spontaneity and rapid pace of an unmoderated list. Members don’t have to wait for someone to approve their messages: a message is sent and, seconds later, it appears. A minute or two after that, there may be replies. Second, I didn’t want to have to pass judgment on every message. What if a message had worthwhile content but was not well expressed—perhaps because it was unclear or too snide—would I return it to the writer with editorial comments? No thanks! As an English professor, I definitely did not want more writing to correct. In addition, I learn more from the list because it’s unmoderated. I’m embarrassed by the number of times I’ve thought, “What a stupid message,” only to see from the responses that the message had more to it than I had recognized. If the list were moderated, legitimate messages might be suppressed out of ignorance or prejudice.

I do, however, make some exceptions to the policy on moderation. Many women-related lists attract
more than their share of kooks—people whose idea of fun is to upset others and create havoc. To keep WMST-L unmoderated without leaving it vulnerable to such people, I arrange for the messages of all new subscribers, and a few others who find it difficult to follow the list’s guidelines, to come to me for approval. Thus, many inappropriate messages never make it to the list.

Occasionally, a person will object that I am violating his or her right to free speech by not approving a particular message. I disagree. Everyone is free to post on other lists or in a blog. Just as people have no guarantee that their letters to the editor at a given newspaper will be printed, so, too, they’re not guaranteed that their messages will appear on WMST-L.

WMST-L’s size and diversity, coupled with its substantial mail volume, led me early on to formulate what is probably the list’s most controversial policy: WMST-L does not permit discussion of gender-related political and social problems unless the discussion relates directly to the list’s stated focus on teaching or research. WMST-L welcomes conversations about teaching issues such as abortion, rape, and sexual harassment. Over the years, however, I’ve found that discussing the issues themselves is likely to wreak havoc on the list for several reasons. First, opening the list to discussion of the world’s many gender-related societal issues would vastly increase WMSTL’s already-heavy mail volume. Moreover, people often have very strong views about such issues and are sometimes hostile to those who voice opposing beliefs. In addition, forums where gender-related societal issues are discussed tend to attract troublemakers and people determined to “educate” misguided feminists. The resulting increase in mail volume and a generally hostile atmosphere would likely drive many participants to leave WMST-L, and the list would no longer serve as a useful professional resource. I do not want to see that happen.

It’s often difficult to draw a clear line between what’s appropriate and what’s not and to decide when a discussion in progress has ceased to be appropriate. As I’ve been writing these remarks, a heated discussion has been taking place on WMST-L. At the start, it focused on the value of teaching feminist theory, with one participant asking, “How will teaching feminist theory prevent insane misogynists from murdering schoolgirls?” Certainly, this question falls within WMST-L’s focus on teaching. Gradually, however, the discussion shifted to male violence and the situation of women in Muslim countries. Some rash statements were made, tempers flared, and message writers began exchanging accusations of racism and ethnocentrism. A few minutes ago, I sent a message calling a halt to the thread. One subscriber immediately wrote back to ask that such decisions be made by the entire Listserv, “in line with feminist politics.”

This request points to a challenge I face as the person in charge of a women’s studies Listserv. Women’s studies has tended to oppose notions of hierarchy, the privileging of some voices over others. Authority is suspect: most feminist groups operate by consensus. Thus, many people expect WMST-L to be governed by consensus; they are dismayed when I insist on adhering to the policies I deem best, charging me with being “unfeminist.”

I’ve given a lot of thought to this accusation. As a feminist, I value other women’s views. Finally, though, my experience leads me to conclude that consensus is not an effective way to run a large, active Listserv like WMST-L. Consensus works well in small groups, not in a group of five thousand. It requires people to take time to work things out, not race through their e-mail. It operates best when people share similar backgrounds and assumptions, whereas WMST-L’s subscribers come from a wide variety of academic disciplines and from forty-six countries, as varied as the United States, Korea, Finland, Egypt, South Africa, India, and Brazil.

Recently, WMST-L has encountered a new problem, one that may affect women’s studies Listservs more than most: overzealous or inappropriate spam filters that have fits whenever they encounter words such as “sexuality” or “pornography.” Some WMST-L subscribers have been automatically removed from the list because their e-mail systems keep rejecting mail sent to them from WMST-L. The rejection notices contain explanations such as, “message rejected because of unacceptable content.” One system rejected a call for papers on the topic “gender and performance” because “the message contains the [words] ‘sexual’ and ‘performance’ in close proximity.”

Ten years ago, when the Internet was an overwhelmingly male preserve, it was hard to use search engines to look for information about women or girls; it was assumed that the people who made such inquiries were looking for pornography, and the results reflected that assumption. Now it’s not the search engines but the spam filters that deem women’s studies subject matter to be salacious. Plus ça change, plus c’est la même chose.

Joan Korenman is professor emerita of English at the University of Maryland, Baltimore County, where she taught from 1969 to 2005. While at UMBC, she also directed the women’s studies program and founded and directed the Center for Women and Information Technology. She continues to manage WMST-L and maintain the center’s Web site, http://www.umbc.edu/cwit/.

Comment on Article..

 Printer-Friendly Page