Popular Science’s recent decision to do away with reader comments on its website unleashed a hurricane of—ha!—commentary. The storm broke not on the PopSci website, however, but all over the Internet, in news outlets The Guardian, The Washington Post, The Hindu, the BBC, on KPCC’s Airtalk radio show and NPR’s To The Point, and on a passel of influential blogs, including the Columbia Journalism Review, Slate, GigaOM, and BoingBoing. A broad range of interesting conversations emerged about how to manage reader participation online. If PopSci’s intent was to have a substantive talk, one free of illiterate political slurs, get-rich-quick schemes, supplements for the cultivation of gargantuan genitals, and yelling at the author and/or fellow commenters, their strategy seems to be working, so far.

When the magazine’s editor in chief, Jacob Ward, appeared on To the Point to discuss the closing of the website to reader comments, he made a glancing but vitally important observation: “Comments are a vestige of the pre–social media world.” This is true, and it’s largely because the Internet from its infancy grew into a two-way communications system: It’s like a telephone rather than just a speaker system. The implications of this founding principle reverberate through every part of our lives on the Web. User participation was bred into its very bones.

Consider Usenet, the earliest forum service. The oldest Usenet post archived at Google is dated May 11, 1981. According to Steven D. Krause, a professor of English at Eastern Michigan University, by the mid-nineties there were around eight thousand Usenet newsgroups, with some seventy thousand articles and comments posted every day. And that’s not even counting private e-mail lists. Though there was as yet no World Wide Web to speak of, the online world of the late eighties and early nineties was a thrumming hive of activity, and very much like today’s web experience in its character and conventions. As an example of Usenet’s early vitality, in 1992, shortly after he’d joined the staff of The Simpsons, writer and producer Bill Oakley found a five-inch-thick set of dot-matrix printouts from the Usenet board alt.tv.simpsons in the show’s writers’ room. Newsgroups like these predated the wide adoption of not only the Web but also e-mail, LiveJournal, eBay, Facebook, and Google. The early newsgroups established an ironclad understanding and expectation of the Internet as a space for talking as well as listening.

Public discussions online, whether in a Usenet newsgroup, mailing lists, the comments sections of publications, or on Reddit or Metafilter, have a way of taking on a life of their own. They can generate heat, like a craps table, with interest and participation building on itself until it all flames out. They can bloom and die like flowers, or the dynamic of the group can morph from one incarnation to the next, as funny, clever, or well-informed participants join, form alliances and enmities, leave, and return. (I met my husband on a Prodigy literature list in the early nineties, so I experienced this firsthand.) In my experience, a good online group can only survive if there is a dedicated moderator to keep things civil and “on topic” and to boot out the trolls and weed out the spam. That’s been true from the start; none of this is new to the Internet’s old hands.

But for this reader, at least, the PopSci website’s commentariat had long seemed bizarrely and consistently vitriolic and misinformed, even by the lax standards of ordinary Internet commenting—a point that went largely unremarked in the flurry of articles following PopSci’s decision to close comments off completely. Here’s an example, from a recent article about the increase in aircraft turbulence allegedly arising from global warming, that is typical of the PopSci commentators in its grace of expression, charm, and good sense.

An informal review indicates that every recent PopSci article related to global warming has been accompanied by the similarly constituted ravings of climate-change deniers, who may well have scared off everybody else. At least, it’s easy to imagine that sane readers would hesitate to make a comment in such an environment.

PopSci is in the business of producing and publishing responsible science journalism. At this late date, is it really possible that anyone still believes climate-change denial to be “the other side” of an intelligent debate?

It appears the PopSci editorial team doesn’t think so. Online-content editor Suzanne LaBarre wrote, “A politically motivated, decades-long war on expertise has eroded the popular consensus on a wide variety of scientifically validated topics.” She also cited a study from the University of Wisconsin-Madison, “The ‘Nasty Effect:’ Online Incivility and Risk Perceptions of Emerging Technologies,” which sought to measure the effects of uncivil Internet commentary on readers, concluding that rude comments have some measurable effect on the interpretation of news stories. The study raises a number of interesting questions, among them the possibility that more-informed readers are at lower risk of being unduly influenced by online yelling. This seems instinctively true, a scientific expression of Alexander Pope’s stirring lines:

A little learning is a dangerous thing; Drink deep, or taste not the Pierian spring: There shallow draughts intoxicate the brain, And drinking largely sobers us again.

As coauthor of the UW Madison study Dietram Scheufele told Mother Jones earlier this year, trying to gather information online can be like “reading [a] news article in the middle of the town square, with people screaming in my ear what I should believe about it.”

There’s another possible reason for the constant stream of fire-breathing antiscience reader comments at PopSci, one that has received far too little attention in the current debate: These comments might have been bought and paid for by antiscience astroturfing organizations (interest groups that pay people to write online comments under cover of fictitious personae). Astonishing as this practice is, such groups have been busted regularly, by the press and by activists, since at least the early aughts.

Perhaps my imagination fails me, but the only reason I can think of for making the climate-change-denying comment quoted above—whether it was paid-for propaganda or the real noodlings of a blithering idiot—is to discredit the editorial content above it. This commenter is not trying to strike up a conversation with the author of the article or spark a meaningful debate with other readers. Is it reasonable to expect PopSci to pay to host and publish these remarks alongside its actual journalism, thereby creating an impression, however faint, of equivalency?

I am a First Amendment fanatic, but I say no. Freedom of speech does not give anyone the right to spray-paint graffiti on your house.

On KPCC’s Airtalk, PopSci associate editor Dan Nosowitz explained: “People really underestimate how difficult a job it is to manage comments sections. Typically, you would need to hire at least one full-time comment moderator to go through and check everything. We run about twelve articles a day, but all of our archive is still out there as well, and open for comments.”

In other words, the only way to be sure that you publish the comments you think readers will want to see and get rid of the ones they won’t, is to hire a real, live human editor. The process can’t be automated completely. At a high-traffic site, the numbers are staggering. Andrew Beaujon of Poynter Online told KPCC listeners: “The Huffington Post deletes around 75 percent of its incoming comments, out of [around nine million] comments per month, and has a staff of around forty full-time comment moderators.” That is a lot of Viagra ads! And while most sites have only a minute fraction of that volume of spam to deal with, it’s a commensurately thorny problem. PopSci’s small staff of writers and editors has repeatedly stated that they can’t afford a dedicated comment moderator.

What should publishers spend their resources on? Wading through spam, astroturf, and insults—or creating more and better articles?

As Nosowitz told me: “We don’t think anyone else should necessarily do what we’ve done. But we do think others should look at their own situation, their own resources, audience, subject matter, and community, and decide if their comments sections actually help their sites. Because if it’s not adding to the experience of reading journalism, why bother?”


“Spam comments are an absolute nightmare,” Choire Sicha, cofounder of The Awl, told me in an e-mail. (The Awl is known to host an excellent and valuable group of commenters.) “Unfortunately, there is literally no service that effectively works against spam comments. We’ve spent a huge amount of time that we don’t have dealing with this. So far, the best services [can eradicate] most spam—but they also mark genuine comments as spam.”

Finally, there’s the delicate matter of removing racist, sexist, or otherwise offensive comments. Editors must determine what constitutes an offensive comment that must go, versus one that might be a little raw but contains a message still worthy of inclusion. It’s a weird, kind of occult skill, like dowsing. Sicha can’t explain exactly how it works. “We’ve never had a comment-removal policy. If I happen to see something I hate or get a lot of e-mail about something, we’ll happily nuke it. Or, if we’re busy, we’ll ignore it.” While it’s expensive and a headache for smaller businesses to handle their comments, he added, there are also “well-managed communities of commenters out there at some independent publications—Offbeat Bride [for example] does this very well.”


Like any other journalist on the Internet, I’ve fielded brickbats from commenters. I very much agree with Mathew Ingram of GigaOM when he says that writers have an obligation to engage with their readership, and I enjoy a spirited rhetorical duel with readers who disagree with me. It’s a big part of the reason I love my job. But that obligation doesn’t extend to offering time and attention in response to plain insults, factual errors, or baldly unsupported allegations. If a writer works hard to make sure his story is accurate—by conducting research, interviewing multiple sources, and so on—and a reader yells, in response to that good-faith effort, “Your stupid!” that doesn’t count as engagement. There’s no need to respond to such stuff, or even to publish or take any notice of it.

Nosowitz says that PopSci’s editors are “totally in favor of substantive disagreements”; he points out that there are still a lot of ways for readers to contact the authors and editors, even though comments have been disabled. His own author page offers links to his e-mail address and Twitter accounts; he and other PopSci authors are active on Twitter. He also told me that he’d like “to see authors’ Twitter handles/e-mails on each article, and eventually perhaps to surface good comments”—to use, in other words, the old-fashioned Letters to the Editor method that has served newspapers so well for so long, whereby the publication cherry-picks particularly good and interesting comments to bring forward.

Well-moderated comments sections can be a pleasure to read, whether they play it fast and loose, like they do at The Awl, or maintain stricter policies, as Andrew Sullivan does at The Dish, where readers submit content that may or may not be chosen for publication by the site’s editors. At The Atlantic, Ta-Nehisi Coates invests a huge amount of time managing the comments section of his blog in order to ensure relevance and civility—as much time as he does writing the content, he told Bob Garfield on On the Media. (Earlier this year, two of Coates’s best, most trusted commenters were invited to help moderate the blog’s comments.) Refreshingly, Coates makes no apologies for the hands-on approach to managing the commentariat. He told Bob Garfield, “It’s like hosting a dinner party. If you were in my house and insulted one of my guests, I would ask you to leave.”

There’s much to be gained by publications with the substantial resources required to manage reader comments, but we mustn’t fool ourselves that this task is anything but complex and subtle.


I can’t help giving Choire Sicha the last word, not only because he is an expert who deals with these problems every day, but because they are so incontrovertibly les mots justes:

Comments… how shall I say this? Comments on articles are a compromise. They’re technologically and structurally an insult to the commenter, who deserves more, and they’re not so great for the publication either. Basically commenter communities should be allowed to exist in their own place and/or time at a publication. Having them exist at the margins, as an after-thought to articles, is senseless. But then, shuffling these communities off-site—eliminating comments entirely and delegating that conversation to third parties like Facebook or Twitter—seems rude as well. Except for PopSci. Their commenters SUCKED.