Previous work

The idea of “surfacing information” regularly appears in scientific and technical literature. However, it is rarely conceptualized in the form of a dedicated term or expression such as “information surfacing” along with a definition. So far, I have found only three sources that do this, and only in technical publications. Two of these sources define information surfacing as the revealing of hidden data or patterns:

“Information surfacing: revealing previously hidden data.”McFedries, “Information at your fingertips,” 2009, p. 24.

“Latent information surfacing (finding hidden patterns and relationships that would not be perceptible otherwise).”Hai-Jew, “Static Text-Based Data Visualizations,” 2015, p. 207.

The third source defines information surfacing as a component of information architecture :

“Information surfacing is a top-down, strategic approach to design […] It’s important to differentiate information surfacing from information hierarchy. Information surfacing is more specific to the visual presentation of information on a single page or in a single instance. While it doesn’t involve categorization, per se, it does involve decisions: such as which fonts to make bolder or which elements to show on mouse rollover. It’s all about which information you want to ‘surface’.”Volnyansky, “Information Surfacing,” 2012.

Use in information science

I think a proper concept of information surfacing would be of interest to information science. Writing this post allows me to form some thoughts which I plan to develop later in a dedicated paper.

In my dissertation, I describe an experiment in personal hypertextual documentation, with a visualization program (Cosma) being used as a memory aid and for reflexive work. I speak of visualization broadly: Cosma includes data visualization in the usual sense (there is an interactive network representing interlinked documents) but there is also functionality based on presenting certain information in a specific way (for instance, showing backlinks to a document with the surrounding paragraph).

While using Cosma, I found that on several occasions I would stumble upon relevant information that I had not looked for directly, because the layout and functionality of the program created little “neighborhoods” of information: nodes that sit together in the network visualization despite not being connected to each other; two documents listed together in the backlinks section of a third one; unrelated ideas popping up when filtering on a certain keyword.

This sort of information surfacing favors serendipity, i.e., “knowing how to pay attention to something surprising and imagine a relevant interpretation”.“Savoir prêter attention à un phénomène surprenant et imaginer une interprétation pertinente”, Catellin and Loty, “Sérendipité et indisciplinarité,” 2013, p. 34.
Serendipity occurs when you make an unexpected discovery and decide to do something with it. It is not so much a matter of chance as it is about “un-sought ideas”.“Idées non cherchées”, ibid., p. 33.

So I think there is something to be done here, and not only in relation to visual epistemology, as I did in my dissertation. To be continued.