Nicholas Carr’s When links turn inward discusses Mark Coddington’s recent study, Building Frames Link by Link: The Linking Practices of Blogs and News Sites. Carr’s post is a good read, but in the following I refer directly to the study. Coddington analyzes the links in political articles drawn from three sample sets:
- Six big mainstream news websites (ABC, CNN, New York Times, Time, Washington Post, Wall Street Journal)
- Six political bloggers at the above websites (Political Punch, Political Ticker, Caucus, Swampland, Plum Line, Washington Wire)
- Six independent political bloggers (Americablog, Crooks & Liars, Doug Ross @ Journal, Hot Air, Michelle Malkin, TPMMuckraker)
Comparing articles on three major 2010 political events, Coddington found a striking polarization between mainstream news and independent bloggers (with bloggers on mainstream websites holding a middle position):
- Mainstream news frequently include no links at all: 36% of stories, versus 8% for independents.
- When they do link, mainstream news love linking to their own website: 91% (!) of all links, versus 18% for independents.
- When they do link elsewhere, mainstream news still prefer other mainstream media: 71% of external links, versus 54% for independents.
Internet cranks may feel vindicated by these figures. Just as they always claimed, mainstream news systematically insulate themselves – and their readers – from outside voices. If you want multiple perspectives on a subject, especially ones that aren’t already top-ranked on search engines, mainstream media are your worst choice. Coddington’s own interpretation is more restrained but still quite forceful:
Opinionated voices and partisan sources have little place in this body; it is a space dominated by supposedly dispassionate, factual information provided by established, authoritative, and largely professional media sources. Far from helping to form a conflict frame, the realm of news organizations’ links is almost anti-conflict, devoid of nearly all of the perspectival distinctiveness and boldness that characterizes the Web’s discourse, but that might be perceived as a threat to the norm of journalistic objectivity.
“Supposedly” dispassionate is a key term which Coddington sadly fails to examine. All journalistic organizations have a political bias that influences their reporting. Rampant self-linking reinforces that bias, hindering rather than helping “journalistic objectivity.” There’s a baser motivation as well: commercial operations need to capture either subscribers or advertising impressions to survive. These goals are served by keeping visitors on your own website as much as possible, and away from your competition.
However, when the intention is not to entrap readers in your ideological or commercial bubble, linking to your own work can be beneficial and even necessary. Consider this example from Hadas Shema’s report on self-citing science bloggers:
Melissa Terras, who has a blog and over 2,000 Twitter followers noticed that each time she blogged and tweeted about one of her past referee articles (they are freely available in her university repository) the download numbers went up. She decided to try and blog about three articles from the same project, but not blog about a forth. The result: the promoted articles received at least 11 times more downloads than the un-promoted one.
Clearly, there was a strong demand for the promoted articles that had gone unsatisfied because potential readers were simply unaware of them. This matches my own experience. Some of the best bloggers I follow frequently repost links to older articles, and that’s a good thing as I would otherwise have missed them! Such self-promotion may seem tacky but it allows people to find lesser-known material in a sea of content – without relying on Google’s search algorithms or the accidental goodwill of others.
2012-07-31: On 24 February 2012, MG Siegler ranted about an Apple story that he broke. His article was linked by other publications which reported on the story – except for WSJ. Siegler:
They routinely do not properly credit sources of information on the tech side of things. They seem to have a mentality where nothing is really news until they break it and/or confirm it. Another anecdote to support Coddington’s findings.