You Can’t Stop a Frack Truck with Thumbnails: YouTube’s interface and the “individually wrapped” viewing public
by Leah Shafer
On October 23, 2014 a group of protesters opposed to the storage of highly pressurized methane, propane, and butane in the abandoned salt mines beneath Seneca Lake in the Finger Lakes region of NY linked hands and blocked the trucks of the Texas-based company Crestwood Midstream from entering their lakeside worksite. The act of peaceful resistance was organized by a diverse group of local citizens who had formed a coalition called “We Are Seneca Lake” to oppose gas storage in the historically unstable salt caverns. The We Are Seneca Lake movement is partly composed of a pre-existing network of citizens who’d been involved in New York’s successful anti-Fracking movement, so there was an organizational infrastructure in place to support the arrestees and to plan future actions. This organizational structure has yielded a group that employs both time tested old school grassroots techniques and the use of digital tools for activist activities. The group created a robust website (http://www.wearesenecalake.com/) that offers links to relevant environmental information and showcases photos of the nearly daily stream of committed activist citizens who are ready to get arrested in the name of protecting Seneca Lake’s supply of drinking water and the rural character of an area known for its organic farms, thriving wine industry, and agri-tourist attractions. The website also links to We Are Seneca Lake’s YouTube channel. As the We Are Seneca Lake movement has gained momentum, more and more self-produced videos have been posted on the YouTube channel where they remain, basically un-watched. These videos have not spread.
I teach at a college situated right on Seneca Lake. Several faculty and students are deeply involved in the We are Seneca Lake movement, so it seemed to me that the media produced by the group would be an excellent choice for inclusion in my Intro to Media and Society syllabus for our unit on amateur video.
When I pulled up the video “We Are Seneca Lake Teacher Arrestee” so we could screen it and talk about the different ways that the video frames the protesters and the ways that its amateur video aesthetics do and do not reflect the aesthetics of other amateur video we’d studied, several things caught my students’ attention. First, a student raised her hand and asked why the video had so few hits, given the robust activities of the activist community and the real threat to our local water systems, The second question I got was from a student who asked if we could watch the Stephen Colbert video after we watched the local activist video. So, what I thought was going to be a discussion about We Are Seneca Lake’s activist video aesthetics quickly became a discussion of the YouTube exhibition interface and the ways that targeted marketing shapes our experience of watching video on YouTube: especially video with activist content.
YouTube’s recommended video thumbnails appear to promote exhibition experiences with hyperlocal, affectively alluring, personal specificity. As there are roughly 20 recommended videos on each watch page, we might read the thumbnails as the metonymic appearance of the viewer’s hands within the interface. The sense of personal connection is obviously part of YouTube’s brand identity: it’s not called OurTube or GoogleTube. But, the figurative invocation of the personal, which is clearly meant to form an affective connection between the user and the tool is only a fractional part of the site’s actual organizational architecture. According to reelso.com, a site that provides advice on how to get your video selected as a recommended video, suggests that only one or two out of the 20 or so recommended videos are actually related to the user’s viewing history. Most of the videos are placed there by YouTube’s algorithm that predicts which videos viewers will watch based on the length of time other viewers have spent watching them: because the longer folks watch a video the more advertising revenue is generated by that view.
So, the larger question here is: what happens to activist content when it is framed by an exhibition interface organized by advertising architectures that present themselves as archives of personalized recommendations. My experience in the classroom is anecdotal, but it brings to the foreground a constellation of issues raised by the YouTube interface, and I will spend the rest of this brief post teasing out what I see as the preliminary implications of these issues, all of which circle around spreadability versus unspreadability, but which apply those terms less to the actual content of the videos and more to the way that algorithmic marketing shapes the YouTube interface in a way that I am now thinking of as the unspreadable, or, individually wrapped exhibition experience of interacting with the YouTube interface.
The phrase “individually wrapped” originates in some research I was doing for an unrelated project on Amercan television advertising of the 20th Century. While working on that project, I came across the patent filed by Arnold Nawrocki in 1956 for an “apparatus for producing individually wrapped cheese slices” – this apparatus was meant to counteract the stickiness of processed cheese slices by providing a method for wrapping a “slicelike slab of cheese in a transparent, pliant wrapper.”
When Sam Ford, Henry Jenkins and Joshua Green write about spreadability in Spreadable Media, they note that the idea “originated in relation to ‘stickiness’” (3) which is a term that they trace to Malcolm Gladwell’s The Tipping Point which “uses ‘stickiness’ to describe the aspects of media texts which engender deep audience engagement and might motivate them to share what they learned with others.” (4) In the case of my students’ responses to the We Are Seneca Lake watchpage, they were sharing information about their own media consumption habits and engagement, but they were not paying attention to the activist content and, due to the lures of the interface’s apparent personalizations, they were primarily experiencing distraction, rather than attention. If you will forgive the logic of the pun here, instead of being rapt, the interface made the students wrapped: like Nowrocki’s transformation of the slab into the slicelike, the packaging of the activist video made it less sticky.
When the students were presented with the YouTube interface, they looked across the surface of the page to the thumbnails of recommendations. They looked with what Jinying Li, in her essay, “"From Superflat Windows to Facebook Walls," calls “an animated shopping gaze.” (203) Li suggests that the animated shopping gaze reflects the “heightened experience of consumerism” of a visual field dominated by walls and windows. (214) As Li points out, monetized personalization creates a visual field in which our gaze is “databased and computerized:” a visual field “in which our identities are nothing more than a list of products that the computer, or the website, decides we’d like to purchase.” (216) The YouTube interface forms an exhibition space that draws our gaze to its synecdochal representation of our person: the siting of our hands and our algorithmically generated preferences alongside the primary video.
YouTube’s interface appears to suggest that it offers a limitless archive of related materials with the added function of personalization, and that appears to reduce the effects of what John Ellis calls “choice fatigue.” But the appearance of transparency and actual transparency are quite different. YouTube closely guards and makes invisible the systems and algorithms that determine which videos are added to the recommended thumbnails. As Lauren Berliner has pointed out in her essay “Shooting for Profit: The Monetary Logic of the YouTube Home Movie,” “the invisibility of these systems helps to naturalize the appearance of YouTube as a democratic platform driven by users’ tastes and interests.” (293) But, of course, the architecture of the site is neither “natural” nor “democratic.” YouTube’s architecture uses personalization as a way to control user attention, and as a way to generate data that can be monetized.
YouTube’s superflat, monetized design is not precisely an invention of the digital age. As Jeremy Groskopf point out in his study of “Silent Era Precursors of Online Advertising Techniques,” in 1916 Frank C. Thomas patented the design for a “collection of light diffusers flanking a movie screen that would allow for the stacking of advertisements along the screen in basically the same way that YouTube’s thumbnails of texts that are ‘recommended for you’ appear on the right side of the YouTube interface.” (86)
Of particular concern to early advertisers was the creation of an easily viewable system that added to the theatergoing experience without annoyingly distracting viewers from the primary cinema content and which offered a way to distinguish not only between the primary content and the advertisements, but between the advertisements, themselves. On YouTube, this two-pronged directive (to focus on primary content and to distinguish between different ads) is accomplished through scale and with visible metadata, though the advertisers’ imperative to not annoy, distract, or mislead the viewer is no longer a primary (or even secondary) concern. In fact, the interface’s display of metadata and its emphasis on personalization appear to be designed precisely for distraction and misdirection: two things that are not plainly useful to the site’s tertiary function as an exhibition space for activist video.
The interface appears to offer transparency – it is very indexical about its personalization, and it offers phrases like “recommended for you” in order to forge an affective connection between the viewer and the data being produced by their immaterial labor. But it is precisely the customization, the “individual wrapping,” of the YouTube interface that marks its slicelike-ness. The deck is stacked. Where it appears to offer unlimited avenues for viewer choice, the YouTube interface is, instead, ordered by what Daniel Chamberlain, in his essay “Television Interfaces,” calls “vectors of customization and control” (85)
When YouTube was established in 2005, they identified themselves as a personal online archive with their tagline, “Your Digital Video Repository.” To be certain: the existence of an easily accessible archive with free storage and a professionally designed interface is of great use to activist organizations. For We Are Seneca Lake, for example, it offers a “free” space for the archiving of their videos. Since the advent of Web 2.0, activists have used consumer-oriented social media tools to mobilize viewing publics and to archive their labor. Further, as Ethan Zuckerman’s Cute Cat Theory of Internet Activism posits, “Internet tools designed to let ordinary consumers publish non-political content are often useful for activists because they are difficult for governments to censor without censoring innocuous content.” Videographic activism that employs consumer tools for its spread, for example, has the ability to reach larger audiences and does not have to rely on untested technology for the spread of its media. So, the members of We Are Seneca Lake who are working tirelessly to protect the Finger Lakes region from becoming a fracked gas transportation and storage hub for the entire Northeast, use YouTube as a repository for videos that chronicle their movement.
YouTube switched its tagline from “Your Digital Video Repository” to “Broadcast Yourself” the year it was purchased by Google. The switch from emphasis on being an archive to being an exhibition space is reflected in the changes to its interface and its emphasis on recommended videos. Even as the site has become more invested in the promotion of revenue-generating personalization features, it has provided a space for the circulation and distribution of activist video content. And, the availability of that space facilitates activist labor in the digital age. But, the consumer-oriented, individually wrapped, exhibition interface with its ingratiating and coercive targeted personalizations affects the viewer experience of activist content by fragmenting attention and by offering the suggestion of curatorial choice where that choice is limited at best. It models a type of viewing that anticipates pliant consumers rather than activists. If we look to the protests of the We are Seneca Lake activists as a model of vital activism, we see the way that the YouTube interface’s promotion of ad-driven thumbnails drives attention and intention away from activist-driven linked hands.
Leah Shafer is an Associate Professor in the Media and Society Program at Hobart and William Smith Colleges where she teaches courses that explore the culture and history of television, film, advertising, and the Internet. Her criticism appears in journals including FLOW: A Critical Forum on Television and Media Culture, Afterimage, and Film Criticism as well as The Journal of Interactive Technology and Pedagogy, Teaching Media Quarterly, and Cinema Journal Teaching Dossier. A scholar/artist, she was recently awarded a research residency with the experimental media art collaborative Signal Culture, and her experimental documentary Declaration of Sentiments Wesleyan Chapel was included of the Iterations as Habitats exhibition of the Finger Lakes Environmental Film Festival.