Editors’ note: Juliette de Maeyer (@juliettedm) kicks off this month’s edition focusing on Ethnographies of Objects with a response to two questions posed to her: ‘Why is the hyperlink an interesting object of journalism?’ And ‘What’s the best way to approach this object methodologically?’ Her work on hyperlinks is a fascinating exploration of materiality, stubbornness and methods for trailing the object.
An object of journalism?
First things first: why do I even claim it is an “object”? A link is not exactly a thing that can be touched… However, a link has a material existence in the digital realm, with a beginning and an end — it is clearly defined by chunks of code, the <a> and </a> HTML tags that border it. We can define what’s a link, what’s not a link and, even, what’s almost a link: the six news sites I have extensively studied for my dissertation all contain what I call « plain text links », that is a URL effectively written in the text of the news story, but which is not a link. The idea of indicating another place on the web is there, it’s a reference to another web page or site, but it’s unclickable.
The material boundaries that define the object itself suddenly become blurry, and that’s exactly where it becomes interesting. Why did the journalists produce almost-links or anti-links? Same goes for the apparently very simple distinction between internal and external links: internal links lead to pages in the same site, the same domain, defined by its URL, whereas external links point to other sites. Alright, but what about links that lead to other sites belonging to the same owner? News site A is the online counterpart of a tabloid, and sometimes links to articles published by news site B, the online counterpart of the quality paper — all are owned by the same company, and due to convergence efforts, news sites A and B are produced in the same newsroom. Formally, that’s still an external link. The material boundaries again become interesting when they are challenged.
A link also has an unambiguous existence for the actors involved in online news making. Ask a journalist, a blogger or an editor: they know what a link is. They can recognize one, they know when they produce one. This may seem a very mundane quality, but many things that we claim to study academically don’t have such an obvious existence. Try to ask journalists about their Bourdieuan habitus… Of course, this is not to say that the Bourdieuan habitus is an invalid concept. There might even be a portion of habitus involved in the ways journalists deal with links. It’s simply a question of vantage point: studying objects—things that exist—is a bottom-up approach that allows to iteratively discover concepts, theories and issues. It’s an empirically-driven, inductive perspective: instead of saying “Hey! The issue of sourcing is an important concept in news, let’s see how journalists use links to show their sources”, the logic sound more like this: “So there’s this thing that seems quite unique to online news, it’s called a hyperlink: let’s see why journalists use it… They use it to show their sources, but also for many other reasons!”.
Approaching the study of the hyperlink methodologically
Studying a single object of journalism has a great advantage: because it is so focused, it allows the use of mixed methods. It’s a very pragmatic argument, verging on stubbornness: I’m studying the link, and just that. Sure, other online news features are fascinating, but I don’t want to know anything about the latest multimedia fad or the craze of users’ comments. This is why I can cope with doing a big data content analysis, a historical discourse analysis, and some ethnographically-inspired newsroom observations — and graduate in due time (hopefully). All these methods are extremely time-consuming: being highly selective about what I was actually going to look at was a matter of survival.
And it produced interesting results. Let’s consider, for example, the ancient debate of how outdated CMS weigh on bad or non-existent linking practices. I’ve conducted ethnographically-inspired work in two newsrooms. CMS-wise, one of the newsroom was a classic case of print-centric tools forced upon web people: journalists in charge of online news had to use the same tools as print folks. Visually, it meant that they had to write their stories in an interface that looks like a printed page, with columns and stuff. No HTML allowed, of course. Hence, no links — or more particularly, no inline links: side-column links are another story. If they wanted to add inline links in their stories, they had to circumvent the automatic workflow and log in into another system. Everything was incredibly ugly and counter-intuitive. It involved many clicks and did not exactly fit well with the pressure to publish fast. When asked about their linking practices, journalists in that newsroom complained that there were many technical barriers. They claimed that they did not produce a lot of links, because of the print-centric tools they had to use.
In the other newsroom I’ve visited, journalists worked with a spanking new CMS, a blog-like interface where everything could be dragged and dropped effortlessly. The possibility to add inline links was smoothly integrated and it could be done at any stage of the process. Journalists claimed they added links “whenever it is necessary”.
Guess which site produced more inline links? The first one, with the print-centric CMS and many alleged “technical barriers” to linking. This surprising result was only visible when looking at aggregated data over a long period of time, it wasn’t obvious when looking at a handful of articles because both sites produced rather few inline links (around 10% of articles contained at least one inline link in the second site, whereas the proportion was a bit more than 20% for the first newsroom). This is exactly why it was important to complement newsroom observation with a large-scale content analysis. Or to complement the content analysis with newsroom observation, if you prefer. Looking at a specific object allowed me to do just that: multiply the vantage points while keeping my research feasible with the time and resources I had. Nothing new, really, just good old triangulation with a pragmatic twist.
All in all, the “object” is a very useful lens. It allows a research stance focused on what’s material, but does not limit it to the study of artifacts. Discourses, representations and meaning all play an important role in my research — as much as large-scale content analysis and ethnographic inquiries. Focusing on the “object” is the only way I know of keeping it all together.
Featured image by JanneM on Flickr CC BY NC SA