The bottom half of the internet

An edited version of this article first appeared in Village magazine.

Not too quietly, but without a great deal of fuss, NPR (National Public Radio), the American public radio network, killed commenting on its website in mid-August.

In a statement announcing the change on 17 August, the network explained that an analysis had revealed that only a fraction of one percent of their monthly 25 to 35 million audience was making use of the comment facility on their website. Over a three month period, only 2,600 people had left a comment –– 0.003% of the 79.8 million NPR.org users who visited the site during that period.

Trolling bait
Trolling bait

On 23 August, after eight years, the boards went silent, joining may other North American news sites who have decided that reader comments just aren’t worth the bother.

That isn’t to say that NPR no longer talks to –– and listens to –– it’s audience. The network runs a gamut of social media accounts, including more than 30 Facebook pages and 50 Twitter accounts, as well as accounts on Snapchat, Instagram, and Tumblr. The primary Facebook account reaches over five million readers. Individual journalists also run their own social media accounts.

*

In Ireland, different media outlets had taken a variety of approaches to commenting, from free-for-all to no comment capability at all. Differing defamation regimes also affect how commenting is managed. The courts are still working out the niceties of when a reader comment is published by website, and when the website becomes liable for defamatory content, so different publishers have taken different routes to ensure their comment sections stay within the bounds of the law. The need to police harassment and trolling by site users also influences those editorial decisions.

The Independent website requires commenters to register, either using Facebook or Google Plus (both of which have real-name policies), or by registering using an email and password combination. In theory, such measures can reduce harassment and aggressive posting, but the ease with which throw-away email addresses can be created reduce ts effectiveness. Sister publication the Herald doesn’t host commenting on its stories at all, while the Irish Daily Star (through its online vehicle Buzz.ie) simply requires users to provide their name and email before posting comments on their stories. Both the Star and the Independent require –– labour-intensive –– pre-moderation, so commenters have to wait and see if their insights and observations are deemed worthy before being shared with the world. Pre-moderation has the advantage of reducing harassment and improving the tone of the comments sections, but it does mean that the publication becomes liable for any defamatory statement that slips through.

The Examiner requires registered users to log in (using any of their Facebook, LinkedIn, Twitter, or Google Plus identities, or by registering an email and password combination), but doesn’t offer an option for readers to comment on stories in most cases. Registering does however allow the Examiner to collect data on what stories readers are interested in, and to tailor news alerts on phone and tablet apps. Like the Examiner, the RTE website offers users the opportunity to register, but this is in order to access and keep track of programmes on the RTE Player across different devices. No commenting facility on their news or entertainment offerings is available. RTE does ask for a lot more data than most sites during registration (full name, gender, date of birth, and location data).

The Irish Times, meanwhile, limits commenting to paid subscribers. The Sunday Times, through its Times of Ireland daily product, is behind a hard paywall, which also has the effect of restricting access to comments. This raises the intriguing question of what might happen when trolls who have paid their subscription fees cross the bounds of acceptable speech. Is moderation made more difficult when it could lead to a potential loss of revenue?

In summary, although most Irish traditional news outlets offer some form of reader participation, there are various requirements before comments can be left on sites. The particular software solutions managing comment sections can also act as an impediment to commenting (the Examiner site in particular was extremely slow in loading comments) and pre-moderation can lead to delays which reduce the likelihood of participation.

With news outlets born on the web, commenting seems much more vigorous. While The Journal, for example, requires a sign-in with Facebook or Twitter, comments are not as a rule pre-moderated (although on some controversial stories, commenting is disabled). Joe.ie/Her.ie do not offer a comment facility. Broadsheet, arguably the site most dependent on comment, is mostly unmoderated, although software may occasionally stop a comment until it is vetted by a staff member, if for example it is from a previously unknown account, and presumably if it contains any of a list of blacklisted phrases. Broadsheet articles are often little more than a handful of sentences, linking to a tweet or facebook entry, and the real reason readers stay is to read and join in commenting. That audience, in turn, can expect an occasional longer article, including some longform journalism running to several thousand words.

The free-for-all spirit in the Broadsheet comment section is strongly reminiscent of the culture of early bulletin boards, dating back to the earliest days of the internet, before Tim Berners-Lee came up with the idea of the web.

*
Commenting has a chequered history on the internet, although of course commenting predates the internet. It simply wasn’t always called commenting. Letters to the Editor are as old as newspapers themselves. Proprietors realised very early in the history of periodical publication that people would buy a copy of the paper simply because their name appeared there. And what better way to encourage that impulse – and exploit that vanity – than to offer readers a forum where they could comment on the news events of the day.

To use the modern parlance, Letters to the Editor were heavily pre-moderated and curated, in addition to being sub-edited for length, readability, and spelling errors. Despite that, over time, legends grew among those subeditors charged with maintaining the Letters to the Editor pages about their correspondents. There were the old familiars, who ran off at least one letter a day, often stereotyped as a retired colonel in the Home counties fulminating on the pages of the Times about young people these days, or the intrusion of modern conveniences – always referred to as contraptions – into their rural idyll.

There were the hardy annuals (“Sir, I just heard the first cuckoo of the Spring. Is this a record?”). And there was the Green Ink Brigade – letters written by the clearly deluded, ranging from conspiracy theorists fulminating about alien lizard overlords and illuminati to believers in oddly garbled religious fundamentalisms who expected the apocalypse to begin by the weekend. For some unexplained reason, these seemed to correlate heavily with the use of green ink biros.

And of course there were the undeclared interests, the member of a local lobby group who neglected to mention that fact, or the brother-in-law of a county councillor, or the TD’s cousin, who regularly wrote to defend their relative, but always neglected to mention the familial connection.

Despite these shortcomings however, the letters to the editor pages provided an often lively and usually interesting feedback on stories and how they affected communities, and indeed helped foster a sense of community among the readership. So it was understandable that newspapers wanted to capture the same dynamic online.

*

In the early days of the internet, there was no world wide web, and there were no newspapers online. But there were discussion groups, first with emails carbon copied to multiple recipients, then with mailing lists managed through central subscription software listservs, and eventually, through user networks anyone could subscribe to. The Users Network (Usenet) was built on earlier electronic bulletin boards, and became the first widely-distributed, chaotic, user–regulated social network. Godwin’s Law was born there, and spam, and much of the received wisdom about how to deal with online nuisances (filter, blocks, mutes, and Do Not Feed The Trolls) were first formulated as cultural artifacts on usenet, back when the number of users online numbered in the hundreds and thousands, not millions and billions.

Usenet was divided into a hierarchy of subgroups (science, arts, recreation etc) and using subgroups within subgroups, any topic could be catered to (eg rec.arts.movies.star-wars). Users could even crosspost between the newsgroups if a topic was relevant to more than one community. “Who Would Win A Fight Between the Borg and Death Star” would cross post to rec.arts.movies.star-wars and rec.arts.movies.star-trek, for example. Of course, the person who asked such a cross-posted question was often a troll, more interested in starting a flame war between the rival fandoms than getting a straight answer. More seriously, discussions in the science groups were often disrupted by crossposts from religious groups, in particular from creationists who took the presence of biologists online as a personal affront to their religious beliefs. (Trolling had nothing to do with Scandinavian ice giants, rather, trolling was the act of fishing for comments, baiting users with inflammatory statements, analogous to how one might fish by trolling bait along a riverbank). Paradoxically, one of the most consistently readable groups on usenet, talk.origins, exists because of this trolling tendency, having been designed specifically to attract and debate creationists, thus allowing the scientists to get on with their work in (relative) peace.

A regular feature of newsgroups was a debate initiated by a news article. In the earliest days, users would type out newspaper reports manually, which they would then post, igniting a discussion. Later, when the world wide web came along and newspapers decided they needed an online presence, and began uploading their content for free (a move many in the industry now lament) this was no longer necessary. Posters could simply cut and paste the article, or post a link to the relevant web page, perhaps accompanied by a relevant quote. Again, the point of the exercise was to initiate debate among the online community in the newsgroup.

As the world wide web grew in popularity (and people increasing called it simply “the web”) , websites sprang up seeking to mimic the features of usenet, and capture its audience.
Boards.IE, to pick one example, began as essentially a duplicate of the usenet experience. It even carried usenet groups, which were free to all since usenet, built in the original collegiate spirit of the internet, wasn’t privately owned. (In fact, it isn’t owned by anyone at all, having no central server or administrator, leading to one of the first internet cat memes: Fluffy owns Usenet). Politics.IE was an even more specialised example, limiting itself to Irish political debates. Reddit to this day has the look and feel of a clunky usenet client transported to a web page, as did such notable sites as Slugger O’Toole for a long time, before blogs and WordPress templates became the new paradigm for discussion sites on the web.

What made websites like Boards.IE and Politics.IE work was the communities they built, regular readers who initiated and contributed to discussions every day. And with every refresh as a contributor checked for new responses, a fresh page hit was recorded, which translated into advertising revenue.

Facebook is only the most successful of several social websites which were set up to capture those audiences. Like Bebo, Friendster and others before it, Facebook (and other social media sites from Twitter to Snapchat) work on the premise that there was no need to create subgroups for users. Ultimately, it doesn’t really matter what people post and talk about, so long as they do what comes naturally to human beings – building communities. The most successful social networks are those which enable frictionless community formation. Those communities, once they become large enough, generate network effects. You’re on Facebook because everyone you know is there, and not on MySpace. People don’t need to follow news groups, they just have to follow each other. Communities are an emergent property of the network.

News stories are sometimes the connective tissue holding these communities together. But so, for that matter, are radio broadcasts, TV shows, movies, comic books, podcasts, the latest Netflix download, and of course, cat memes. Anything that people can share, appreciate, and critique.

And so, it should come as no surprise that once NPR looked at the sums, they discovered far fewer than one percent f their readers and listeners are commenting on their website any more. Audience is still listening and reading, but they’re talking about it on Usenet, or Twitter, or LinkedIn, or Tumblr, or even Snapchat.

*

There’s a media myth that comment sections attract readers online, because people want to talk. It dates back to the days when independent blogging was at its peak, and newspapers were borrowing whatever emerged organically. Specialist bloggers linked to each other, and created a new kind of news community, who both created, commented on, and read each others posts. But while many blogs still survive, they are no longer primarily where debates take place. That happens on social websites. Blogs are the triggers to start a discussion somewhere else. Their comment sections mostly lie empty. So, noticeably, do the comment sections for a great many newspaper stories.

The network effect means its easier to join one or two sites like Facebook, and discuss what you read everywhere else on the web by posting links your friends and colleagues can see there, rather than registering to post comments on dozens of different sites, each requiring a different combination of login verifications and passwords, each with different audiences of strangers.

The process has come full circle. Now, instead of posting a story link and then discussing it on usenet, audiences do the same on Facebook, or Twitter, or the network of your choice. Even Google Plus, unloved even by its creators, has its fans.

Newspapers aren’t losing audiences to social networks, because they never really owned those audiences in the first place. They assumed that readers were interested in their specific stories and titles, when readers were usually more interested in community, and newspapers have for the most part been very poor at managing communities. There’s still a place for Letters to the Editor, curated and considered comments and input from readers. Huffington Post is based on the principle of turning Letters to the Editor into articles. Other publications disguise it better, but many of their online hot takes and opinion pieces work on the same principle of (often unpaid) user-generated content, commenting on what has already been reported elsewhere.

For original news outlets however, the task of editors are journalists remains the same as it ever was. Produce compelling content people will talk about, point to, and read. Good luck.

Send to Kindle
This entry was posted in Other Media and tagged , , , , , , , , , , , , , , , . Bookmark the permalink.