The Architecture of Radio is an iOS and Android app visualizing the data infrastructures that enable contemporary practices of observation, communication, and navigation. In this paper, we take the app as a starting point for exploring the broader issue of how notions of data as evidence wield their influence in visualization practice — even in the context of projects that their makers explicitly position as artistic or speculative endeavors. In the process, we consider the app’s interface and its functionalities, along with the discourse around it, including reviews. The latter in particular are highly revealing of how, despite disclaimers in promotional and other related materials, the apps and their makers are held to precisely the sort of evidentiary standards that they are seeking to break away from. In looking for explanations, we relate those reactions to conventions for the selection and representation of data that producers of (all) visualizations rely on.
Data visualization: Definitions and common categories
Visualization and objectivity/truth
The Architecture of Radio as imaginative intervention
The Architecture of Radio as (perceived) truth claim
Over the past decade, commentators have expressed concerns about the ubiquity of data infrastructures and the constant connectivity it invites. Authors of work on digital media and the Internet have exposed, and have often taken issue with, the ways that new technologies persistently demand our attention (e.g., Carr, 2010; Turkle, 2011; Wu, 2016).
In media art as well, the pervasiveness of information and communication technologies is a popular topic. Recently, a good deal of work has sought to make visible how data and data infrastructures are now omnipresent, either to create awareness (whether of this ubiquity or its sociocultural, psychological, or political implications) or to critique the current state of affairs. Some creators have also appealed to the audience’s desire to withdraw from our networked society.
One such work is The Architecture of Radio (henceforth AoR), released in 2015 by the Dutch information designer Richard Vijgen. AoR is an application, available for iOS and Android devices, that provides a 360-degree visualization of what its associated Web site calls the “infosphere”. The maker uses this term to refer to “an interdependent environment [...] populated by informational entities”. 
Using GPS and processing data from several global open datasets, the app generates an image of nearby data ‘hardware’: the physical objects that connect us to the infosphere (Vijgen, 2016).  In addition, it also visualizes digital radio signals, specifying their distance from the user. In other words, the app not only identifies discrete elements of the relay infrastructure (cell towers and satellites) and physical access points (WiFi routers) that facilitate our practices of “observation, communication and navigation”, it also simulates the density and the movement through space of the signals they emit.  This way, AoR draws attention to the presence of data in our physical surroundings and lived spaces, showing where they originate and how they move around — without being visible or otherwise perceptible.
Figure 1: The Architecture of Radio app in use. Picture credit: Juuke Schoorl.
The write-ups in the iTunes and Google Play stores, where the app is available for download, appeal to users’ wish to understand the hidden information infrastructures that are currently “changing the world”. The app, these statements argue, ensures that they can be looked at, thought about, and discussed. 
A year after AoR’s initial public rollout, its visualization functionality was integrated into another app, called White Spots (2016, WS), which was developed as part of a larger multimedia project involving not only Vijgen but also the documentary filmmaker Bregtje van der Haak and the visual artist Jacqueline Hassink (both also based in the Netherlands). WS not only makes visible the information flows that surround us but, unlike AoR, invites users to escape from them by seeking out nearby places “off the grid”.  In other words, it appeals to their desire to seek (temporary) release from being perpetually ‘connected’.
In this paper, we discuss how both apps, in particular AoR, visualize the invisible. Our purpose is to demonstrate how in the process, they (inadvertently) invoke commonly held ideas: that data is drawn directly from reality, and data’s visualization offers proof for certain claims about the world. We argue that this is remarkable, as the apps are explicitly positioned as artworks and as speculations about a reality that cannot actually be charted (let alone made visible). In other words, both works are held to the very sort of evidentiary standards they seek to break away from. In a more general sense, our examples serve to illustrate how visualizations inevitably reinforce previously constituted ideas about the relations between data and the realities they provide insights into. In doing so, they show how powerful notions of data as evidence are — notwithstanding the ample criticism directed at such assumptions.
First, we provide a brief introduction to the topic of visualization. Apart from explaining what this practice entails, we discuss common distinctions made between different categories of visualization: pragmatic, scientific, artistic, and speculative. While useful for practitioners, these distinctions raise certain theoretical issues, which we touch on next. They force us, among other things, to question common assumptions about the relations between data on the one hand, and reality, objectivity, and truth on the other.
In the two sections that follow, we discuss how AoR, even as it is positioned as an imaginative intervention, perpetuates certain ideas about data as evidence — even if the makers may not have intended to do so. We conclude by relating this latter circumstance to the conventions for the selection and representation of data relied upon by producers of visualizations (including AoR), and, especially, the expectations those conventions raise in the people who use visualizations.
Data visualization: Definitions and common categories
Contemporary understandings of data or information visualization broadly align with Lev Manovich’s (2010) definition of those terms as “a mapping between discrete data and a visual representation”.  Manovich assumes here that in practices of visualization, a translation of some sort takes place: select non-visual properties of the data — features or relations expressed either numerically or as text — morph into some kind of image.
The contemporary forms of visualization that Manovich focuses on are strongly indebted to the field of graphic design as it developed in the 1990s. But their visual language — one that ultimately consists of “points, lines, rectangles and other graphic primitives” — has remained essentially the same since the nineteenth century. 
The visualization practice Manovich has in mind is sometimes called ‘pragmatic’ visualization. Robert Kosara (2007), arguing for a critical theory of visualization, uses this term to designate technical applications of visualization techniques to analyze data. “The goal of pragmatic visualization”, he claims, “is to explore, analyze, or present information in a way that allows the user to thoroughly understand the data”. 
Kosara and other authors define this category in opposition to ‘artistic’ visualization . Unlike the pragmatic kind, this category is deliberately interpretive and expressive: it seeks to voice a viewpoint or concern rather than present a universal truth (Kim and DiSalvo, 2010; Dörk, et al., 2013).
Some argue that artistic and pragmatic visualizations differ because the artistic representation of data is not easily legible — and is in fact not meant to be. Artistic visualizations, these authors propose, have appeal precisely because in their use of data they render them somewhat enigmatic (Kosara, 2007; Kim and DiSalvo, 2010). The data can acquire a ‘sublime’ quality and elicit an emotional or intellectual response (Kosara, 2007). In this respect, they also differ from ‘scientific’ visualizations, designed to serve the purposes of scholarly research.
Another often-heard term is ‘speculative’ visualization (e.g., Kim and DiSalvo, 2010). This category partially overlaps with the artistic kind, but is identified on the basis of the work it does rather than the intent behind its production. ‘Speculative’ visualizations imaginatively engage with, and provoke thought about, often hidden but socially relevant issues or problems, envisioning, for instance, their possible futures or solutions.
Visualization and objectivity/truth
The distinctions between ‘artistic’ and ‘pragmatic’, and especially between ‘artistic’ and ‘scientific’ visualization, are fundamentally problematic. One reason for this is that science, and specifically practices of scientific image-making, are inextricably intertwined with artistic visualization — as much today as they have been historically.
In their seminal work Objectivity, Lorraine Daston and Peter Galison (2007) point to the key role played by certain forms of artistic sensitivity at several junctures in the history of atlas-making. Now, they observe towards the end of their book, this sort of sensitivity is perhaps even more important than in the past. They argue that “presentational images have begun to circulate at the blurred edge of science and art” — not only in their outlook, but also in terms of where they are displayed. 
Chiara Ambrosio (2015) takes this historical observation a step further, arguing that scientific visualization has always been “imbued with aesthetic commitments”.  Among the reasons for this is the involvement of visual artists in the process of representation. In her paper, she demonstrates how for centuries, artists have decisively helped shape what counted as ‘accurate’ scientific images.
In addition, we need to challenge the more fundamental assumptions that inspire distinctions between ‘artistic’ and ‘pragmatic’, or ‘artistic’ and ‘scientific’. For instance, we need to scrutinize the common association of scientific visualization with objectivity as an epistemic value — to examine, in other words, the view that objectivity is a reliable route to establishing ‘truth’.
Daston and Galison (2007) point out that over time, a wide range of criteria has determined whether a practice could be considered ‘objective’. As the notion of objectivity emerged in the mid-nineteenth century, it was associated with emotional detachment, but also with the use of automatic procedures to register data and with practices of quantification — and, by implication, the belief that there is a reality ‘out there’ with an existence that can be ascertained independent of a human observer.
Figure 2: Illustration from the book Nature through microscope & camera (1909), by Richard Kerr. Source: Public domain review, at https://publicdomainreview.org/collections/nature-through-microscope-and-camera-1909/.
Today, interpretations of the term vary according to field. Historians, for instance, tend to use objectivity as a rough synonym for impartiality or disinterestedness. Philosophers, among others, regard it to mean “standing in an immediate relation to a nonhuman reality”, but also as a position “formed by [...] critical discussion [...] among a plurality of individuals”.  A common denominator of notions of objectivity at different points in time, however, is an attitude of non-interventionism on the scholar’s or maker’s part (Daston and Galison, 2007).
We can relate the profound debt that discourse on data visualization owes to notions of objectivity to common understandings of ‘data’ as empirical traces, unaffected by manipulation or interpretation. In this respect, ‘data’ is often taken to differ from ‘information’, or data already structured or shaped. 
What this view ignores is that data are always the result of processes of selection, according to parameters that are never neutral (Gitelman and Jackson, 2013; Drucker, 2014). As Johanna Drucker (2014) observes, data are really capta: ‘taken’, rather than ‘given’. By implication, they are “constructed as an interpretation of the phenomenal world”, not inherent to it. 
Moreover, understandings of data along these lines fail to acknowledge the interpretive ‘work’ that all visualizations do and that, according to Drucker, is “encoded in their form”.  Yet as Helen Kennedy, et al. (2016) argue, such work tends to be shielded from view, as visualization conventions make data seem factual and transparent.
To explore how deeply engrained these conventions are, we now turn to The Architecture of Radio.
The Architecture of Radio as imaginative intervention
On the AoR project Web site, the app is introduced as a “data visualization”.  Roughly speaking, it generates visualizations of two kinds. The first consists of image elements that are largely based on data about the external world.
Once the user has installed the app on her tablet or phone, the software generates a location image of nearby data points. It shows the positions of relay devices such as cell towers (the tall spikes in Figure 3a) and satellites (the squares) as well as physical access points such as WiFi routers (the short triangles). In addition, it visualizes the distance between those hardware devices and the viewer, at the time of its use.
Figure 3: Screenshots of The Architecture of Radio in operation on iPhone and iPad. Picture credit: authors and Juuke Schoorl.
In both cases, the image elements generated are profoundly data-based. The app, after all, uses (publicly accessible) information to determine the locations of relay devices and access points. Moreover, it relies on GPS data to calculate the distance between those points and the user in real time.
A second series of image elements, however, are conceived in the manner of a simulation. For instance, the app also makes visible the digital radio signals emitted by cell towers and WiFi routers. In the image generated, these signals show up as clouds of parallel lines (somewhat resembling contour lines on topographical maps), composed in turn of smaller geometrical shapes that move in space (see Figure 3).
In addition, AoR produces a low rumbling, combined with crackling sounds and static noises that change quality as the user pans her device. The app’s maker has suggested that this soundscape was loosely inspired by noises picked up while listening to a software-defined radio (SDR): a receiver that allows the user to tune into signals from cell towers, satellites, and so on.  Even so, the result here is an impression — not an actual sonification (a translation into sound) of the emitted signals.
The app’s promotional materials place particular emphasis on this last quality. As mentioned, its associated texts speak of a ‘visualization’. They nonetheless undermine the perception that the image generated is realistic, let alone reliable. Instead, they highlight that it is an imaginative intervention.
For instance, the app store blurbs for the application insist that AoR is “not a measurement tool”. As the waves used for cell phone and WiFi communication are outside the spectrum of visible light, they point out, any representation of them is necessarily an interpretation. Therefore, the app can provide only “an impression of the infosphere, a way of seeing it”. It cannot be a source of accurate information about a particular state of affairs. 
The same tendency also emerges in interviews with the maker, who casts his project as a mere “theoretical simulation” (Vijgen in Campbell-Dollaghan, 2015) produced with artistic intent.  By framing the app in this way, he not only denies its images evidentiary status, but presumably forestalls appeals to objectivity among potential users.
Arguably, such statements by Vijgen align his practice with two of the categories mentioned above: ‘speculative’ and ‘artistic’ visualization. AoR is speculative, in that it exposes an issue of public concern: the ubiquity of all kinds of data infrastructures — and specifically, how dependent those infrastructures make us on “a global ecosystem of digital signals.”  In doing so, it basically ‘fills in’, or speculates on, what we do not know — both what we cannot see, and what we cannot hear.
But AoR’s visualization practice is also ‘artistic’. First, because of its intent. Conceived as a piece of data art, it has been exhibited in several site-specific versions, in design museums, centres for (new media) art, and elsewhere (see Figure 4).  As fitting an ‘artistic’ project, it elicits as much an affective response as reflection on the issues addressed.
Figure 4: Site-specific (360-degree) version of The Architecture of Radio, exhibited at STRP Biënnale 2017, Eindhoven, the Netherlands. Picture credit: Richard Vijgen.
As an artistic project, AoR shares data art’s concern with data as both subject and material. Mitchell Whitelaw (2008) has related the emergence of such work to the rationalization of networked culture in our post-industrial world, which involves “a creative grappling with the nature of our now ubiquitous data systems”. Data art, he writes, “draws data out, makes it explicit, literally provides it with an image. It also probes data’s constitution, potential, and significance”. 
In doing so, it need not be based on an information-visualization rationale. In fact, he argues, the point may not even be to provide insight. Some works merely perform data’s malleability and, in the process, engage with its “susceptibility to transformation, mapping and munging”. 
Nevertheless, data art is often assumed primarily to seek to make visible something that otherwise would remain hidden.  Because AoR fashions such visibility for wireless devices and the signals they emit, it helps, Vijgen (2016) claims, to create understanding and empower users. Not only can they now ‘see’ the concealed world of digital technologies, they are ‘elevated’: no longer mere consumers, they become active participants in our contemporary information society. 
The Architecture of Radio as (perceived) truth claim
Even if AoR is positioned as an imaginative and interpretive intervention, the app’s use of data also anchors it in a reality ‘out there’. This reality, moreover, has been registered via GPS; it has been recorded automatically, or in a non-interventionist way. As we mentioned, non-interventionism is a necessary feature of — and requirement for — ‘objective’ representation.
While AoR never claims to render an objective image, it evokes associations of objectivity by relying on the measuring technologies used for many scientific images, and also by drawing on similar visual conventions. For instance, in the image it generates, the size and placement of the geometrical shapes that represent wireless devices signify their relative distance from the viewer. The user knows this: she is familiar with everyday representations of geolocation data produced by consumer navigation devices. In these devices, a high degree of accuracy of the measurements is key.
In addition, the data used are hardly rendered ‘enigmatic’. AoR is explicit not only about where its various data come from, but also about what they represent in the visualization. It is clear that the app’s data concern the location in (actual) space of devices that relay data signals and/or allow users to connect to data networks. This is all the more remarkable since the app is positioned as a work of art — for as mentioned, artistic visualizations often cloud the data they use by enigma. 
What’s more, because the user’s GPS coordinates show up as numerical information in the top left corner of the screen, the app provides a precise reference point — and ‘anchor’ in the real world — for the relative distances shown in the image. The figures, instantly recognizable as location information (produced once again by a measuring device), thus seem not only to inform but also to back up what is shown.
In accounting for the fact that the app, in this way, also enables the making of truth claims, we argue that it operates within what Whitelaw (2008) calls an “indexical paradigm” , because the datasets it relies on still serve primarily as indexes of reality.
In linguistics and visual theory, the term ‘indexicality’ is used to identify a direct, imprint-like relation between a sign (such as a word or an image) and its referent. A classic example is the footprint, which functions as an index for the foot that made it. A photographic image, or an automatically registered measurement, can also be understood as an index for the real-life values it registers. 
Viewed from this perspective, we can understand the data that AoR’s images, or data visualizations more generally, are based on to be direct (empirical) traces of a world ‘out there’. Thanks to this relation, they may legitimize the premise of what is communicated — even if promotional materials, like AoR’s, make no claims about the precision or accuracy of what is shown in relation to the reality it derives from.
In AoR’s case, however, these paratexts, crucially, make ample reference to the use of specific datasets (Kennedy, et al., 2016). Not only the AoR project Web site and the relevant write-ups on the iTunes and Google Play sites, but also the app itself explicitly mentions the sources the data were retrieved from, thus fuelling the impression that the relation with the outside world established here is indexical. 
The app’s reception — evident in appreciative and negative reviews alike — suggests that it wields considerable evidentiary power. For instance, iTunes app store reviewers both praise and denounce the piece for the correspondence, or lack thereof, between what it shows and “real radiation”, or, as another reviewer puts it, “the actual spectrum of radio frequency waves”. 
Occasionally, reviewers focus on the fact that the location data for relay devices and access points the app relies on are not real-time, but are updated only once every few months. In other words, that the app merely provides snapshots of the communication infrastructure at particular points in time.  This is not always perceived as a problem, but its being mentioned suggests that it matters — presumably, in terms of how it affects the images’ accuracy.  As it turns out, such reactions are common for other artworks using data that are not real-time but scraped at specific intervals. 
User reactions, though not studied exhaustively here, suggest that while the app’s positioning as both a data visualization and a speculative artwork leaves room for rather diverse understandings of Vijgen’s work, this interpretational openness is reduced significantly at the level of its reception.
Another example that vividly demonstrates this is the app’s fictional staging in the American television series CSI: Cyber (CBS). In the series’ final episode, special agents Mundo (James Van Der Beek) and Krumitz (Charley Koontz) are in pursuit of a man suspected of having stolen a large number of U.S. government employee files. To locate a backup hard drive connected to a network in the suspect’s home, Krumitz uses a tablet on which a version of The Architecture of Radio is installed. As the camera zooms in on the screen, we get the familiar view of nearby WiFi beams. One of those, turning red as the app locks onto it, is identified by its name as emanating from the device the men are after (see Figure 5), and this leads the agent to conclude that the hard drive is “somewhere in this room”. 
Figure 5: CSI: Cyber, episode “Legacy”, at the point when the hard drive is located in the room (screenshots by authors).
While the software actually does not allow for devices to be located with such precision, its imagined use here once again attests to the appeal of data as linked directly to what Whitelaw calls “the ‘real’ of its source” — even if, as here on CSI: Cyber, it still requires some effort to locate this ‘real’.
The look of AoR’s interface (see Figure 3) suggests why some reviews have designated it an ‘augmented reality’ (AR) app. Of course it isn’t, strictly speaking, because the layer it adds to reality covers it up. But it does follow the same layering logic as AR and, combined with the 360-degree viewing it imposes, this further reinforces the user’s sense of a direct connection between what she sees on screen and the world in which she moves. This link becomes even clearer if one compares AoR’s interface to that of the aforementioned White Spots (WS) app, released a year later.
Figure 6: White Spots app in operation (screenshots by authors).
As Figure 6a shows, this application uses the same ‘network scanner’ functionality as AoR. However, the user is invited here to use the app to search for places without coverage. Tapping the “get me out!” button at the bottom of the image sends her to a map functionality: a tool that allows her to navigate, with the help of GPS, to a nearby place “off the grid” (see Figure 6b). This way, connectivity is unequivocally framed as something (at least temporarily) undesirable.
As the images show, WS’ network scanner interface provides fewer details than the one in AoR. For instance, it does not include a compass, nor does it identify the user’s position in longitude/latitude coordinates. It relies on a more limited data input, showing only cell tower locations.
But in practice, the visualization here is even more overwhelming than in AoR — especially in combination with the simulation. Cell towers that are actually located farther away show up taller and more in the foreground of the picture. More data streams charge at the viewer at a higher velocity, accompanied by a more monotonous sound that contains constant, unnerving static.
Since the app, after all, appeals to the user’s yearning to disconnect (and perhaps even inspires it), this is hardly surprising — and is in fact highly functional. But arguably the navigation functionality here also serves to ground the network scanner visualization even more firmly in an external reality. Moreover, it suggests more forcefully that the information it provides can be judged on its accuracy or truthfulness, thus undermining implicit claims about the non-evidentiary status of what it offers. WS, indeed, capitalizes on the expectation of reliability raised here — even if, like AoR, it has been conceived of as (part of) an artistic project with exploratory and to some extent conjectural, rather than conclusive, intent.
As our examples show, understandings of data as drawn directly from reality, and by implication as proof for truth claims about the world, wield their influence even in visualizations explicitly positioned as speculative rather than evidentiary. The analysis of our cases suggests that data visualizations exert rather strong persuasive power, even within the context of data art. Both pieces show that artistic visualizations, much like their scientific counterpart, may still operate within an indexical paradigm. As such, they illustrate the continuing entanglement of science and art, evidence and imagination.
Explanations for such entanglement can be found, on the one hand, in the conventions for data use and representation that producers of all sorts of visualizations draw on, and, on the other, in the expectations these raise in their audiences. Below, we briefly restate their significance to our argument.
To begin with, we may conclude that the makers of visualizations, whether scientists or artists, employ some of the same conventions for data use and representation — consciously and less intentionally. In the cases of AoR and WS, one such example is the choice for data acquired with the help of GPS or other technologies for the automatic registration of real-world ‘traces’. In visualizing those data, producers also rely on shared conventions for their representation. AoR, for instance, uses well-established codes to express distance. In doing so, it elects (as many other artistic visualization projects do not) to make geolocation data seem transparent, rather than to hide their origin or meaning.
Some of these conventions, of course, are not specific to data visualization and are part of much broader representational traditions. Consider, for instance, AoR’s 360-degree image: opaque (so hardly an example of AR) but still ‘layered’ on top of the real world — currently a ubiquitous convention in software for handheld devices. Or, if we shift attention from the visual to the auditory plane: the realist impulse that lies behind the app’s sound — even if it is not actually a case of sonification. In this respect, our paper has touched upon the underappreciated role played by sound in understandings of data, as a means of capturing their density in ways that images cannot, thus adding a qualitative dimension to visualization.
Relevant also are conventions governing how such choices are subsequently framed, either as part of the representation or in the communication around it. One might think here for instance of the mentioning of data sources on Web sites, or in interviews with the apps’ makers. Texts like these often include disclaimers about the accuracy of the visualizations an app generates or the evidentiary intent of their use. Nonetheless, such references also suggest to the user that a given representation functions within an indexical paradigm.
As the above examples demonstrate, the outcomes of production choices ultimately always depend on the expectations they raise in those who encounter them. The abovementioned ‘conventions’ can only function as such because users make certain assumptions, which they base in turn on prior encounters with other, presumably similar representations — assumptions, for instance, about the so-called ‘affordances’ of visualizations: what they allow us to claim, or to do . We mentioned that AoR draws on codes for the representation of geolocation data that are common to a wide range of navigational devices, including domestic ones. If users tend to take those as a starting point for making truth-claims (“our destination is X miles away”) or taking action based on such claims (orientation in space, driving somewhere), they may expect the same of representations that use those conventions as well. As our cases suggest, such expectations often overrule any disclaimers that are made.
Quite possibly, this phenomenon can be related to a tendency in our current datafied society toward blind trust in the objectivity of so-called Big Data (and, by extension, of any observations made on the basis of such data). Van Dijck (2014), focusing in her work on the recent impact of online platforms on social practices and institutions, has spoken in this context of an “ideology of dataism”, characterized by “a widespread belief in the objective quantification and potential tracking of all kinds of human behavior and sociality through online media technologies”.  In a broader sense, such trust in the objectivity of data also seems to inform audiences’ readings of representations that use them to artistic ends — even if, very often, artists precisely seek to question the validity and appropriateness of such faith.
Once again, this goes to show that we need to be aware of, and thus be able to critique, the strength of evidentiary power in data visualizations — regardless of the purpose or intent behind them.
About the authors
Eef Masson is an Assistant Professor in the Media Studies Department at the University of Amsterdam in the Netherlands.
E-mail: E [dot] L [dot] Masson [at] uva [dot] nl
Karin van Es is an Assistant Professor in the Media and Culture Studies Department at Utrecht University in the Netherlands.
E-mail: K [dot] F [dot] vanEs [at] uu [dot] nl
1. Architecture of Radio: A Field Guide to the Hidden World of Digital Networks project Web site, http://www.architectureofradio.com, accessed 12 July 2017.
2. The datasets used are derived, respectively, from OpenCellID (a collaborative community database that contains GPS positions of cell towers, created primarily to serve as a data source for GSM localization), Caltech’s Jet Propulsion Lab (whose key activity is building unmanned spacecraft for NASA), and mylnikov.org’s Wifi database (an API that enables users to fetch geo-location information on the basis of WiFi positions).
3. Quote taken from the Architecture of Radio project Web site.
4. “Architecture of Radio” entry on iTunes Preview, at https://itunes.apple.com/nl/app/architecture-of-radio/id1035160239?mt=8, accessed 14 July 2017; same entry on the Google Play site, at https://play.google.com/store/apps/details?id=nl.richardvijgen.architectureofradioAndroid&hl=nl, accessed 14 July 2017.
5. White Spots: A Journey to the Edge of the Internet project Web site, http://white-spots.net, accessed 12 July 2017. The project also included a television broadcast (for the Dutch broadcaster VPRO), a documentary film, a book, and a travelling exhibit.
6. Manovich, 2010, n.p. In this article, the author uses the term ‘information visualization’ (or ‘infovis’). Elsewhere, he opts instead for the term ‘data visualization’, but with reference to roughly the same sorts of data; compare, for instance, Manovich (2008). As we argue below, the often-made distinction between ‘data’ and ‘information’ is problematic.
7. Manovich, 2010, n.p.
8. Kosara, 2007, n.p.
9. Compare, for instance, Kim and DiSalvo (2010).
10. Daston and Galison, 2007, p. 385. At the same time, we might add, authors of practical guides to scientific visualization have actively promoted design principles that help extol the ‘inherent’ beauty of science. A notable example here is Felice Frankel.
11. Ambrosio, 2015, p. 118.
12. Daston and Galison, 2007, pp. 378–379.
13. See Weinberger (2011); Galloway (2012). As Mitchell Whitelaw (2008) points out, this distinction also underlies common-sense definitions, such as the Wikipedia entry for ‘information’.
14. Drucker, 2014, p. 128.
15. Drucker, 2007, n.p. Drucker makes this statement in reaction to Edward Tufte’s notion of visualization as providing “a transparent legibility that gives us unmediated access to [...] information”. See also Kennedy and Hill (2016).
16. Architecture of Radio project Web site (see note 1).
17. Richard Vijgen, e-mail message to Karin van Es, 8 January 2017; Richard Vijgen, Q&A following untitled presentation at the Imagining [urban] data visualization expert meeting, Utrecht University/Parnassos, the Netherlands, 27 February 2017.
18. Architecture of Radio entries on iTunes Preview and Google Play (see note 4).
19. Richard Vijgen, e-mail to Karin van Es, 8 January 2017. He writes: “As digital radio signals are not visual by nature, there is no ‘correct’ way to visualize them. Instead, each visual representation is an (artistic) interpretation.”
20. Architecture of Radio entries on iTunes Preview and Google Play (see note 4).
21. The app premiered as part of a site-specific art installation, first exhibited at the GLOBALE: Infosphere exhibit at ZKM Center for Art and Media in Karlsruhle, Germany (from 5 September 2015 to 31 January 2016). Since then, it has also been shown at the Currents New Media festival in Santa Fe, U.S. (10 to 26 June 2016), the Hello, Robot exhibit at the Vitra Design Museum in Weil am Rhein, Germany (11 February to 14 May 2017) and the STRP Biënnale 2017 in Eindhoven, the Netherlands (24 March to 2 April 2017), in the latter case in a 360-degree projection, ten meters in diameter (see Figure 4).
22. All quotes from Whitelaw (2008), n.p. His characterization is borrowed from Liu (2004).
24. Compare, for instance, Grugier (2016).
25. Vijgen, 2016, p. 27.
26. In spite of this, we should add, the piece also challenges some key traits of visualization as commonly defined. For instance, it doesn’t adhere to the convention that each component of the representation be based on the data used, or some of its key features (compare Manovich, 2010). Examples here are the shape, size, color, and movement of the radio waves, which are not inspired by the data. Or otherwise, in the sense that the spatial relations between the data it visualizes are not arbitrary, but highly meaningful (ibid.).
27. Whitelaw, 2008, n.p.
28. The distinction between ‘index’ and two other main types of sign (‘symbol’ and ‘icon’) was originally developed by Charles S. Peirce (1931–1960).
29. See the Architecture of Radio project Web site, the “Architecture of Radio” entries on iTunes Preview and Google Play (see notes 1 and 4), and the “About” section in the Architecture of Radio app.
30. See “Architecture of Radio” entry on iTunes Preview (note 4).
31. For instance, reviewer Sit7 in the iTunes Preview store (note 4).
32. This is evident, for instance, in the reaction of Kaufman (2015). In this piece, the fact that the information used is not real time is not seen as a major issue; however, the formulation used (“The data presented in the application is not live [...] but [...] is still remarkably robust”, n.p.) confirms that data reliability is relevant. We should add here though that overall, (online) magazine reviewers tend to take more heed of the caveats made in the paratexts.
33. Compare, for instance, Takahashi (2016). The Web site of Wind Map, the project this article discusses, does highlight that the representation is not real-time: see http://hint.fm/projects/wind/, accessed 14 July 2017.
34. CSI: Cyber, “Legacy” (season 2, episode 18), directed by Eric La Salle, written by Pam Veasey, CBS, broadcast 13 March 2016.
35. The notion of ‘affordance’ entered the study of media (originally, their design specifically) through the work of Donald Norman (2002).
36. Van Dijck, 2014, p. 198.
Chiara Ambrosio, 2015. “Objectivity and representative practices across artistic and scientific visualization,” In: Annamaria Carusi, Aud Sissel Hoel, Timothy Webmoor and Steve Woolgar (editors). Visualization in the age of computerization. New York: Routledge, pp. 118–144.
doi: https://doi.org/10.4324/9780203066973, accessed 4 September 2017.
Kelsey Campbell-Dollaghan, 2015. “If our eyes could see wireless signals, here’s how our world might look,” Gizmodo (24 August), at https://gizmodo.com/if-our-eyes-could-see-wireless-signals-heres-what-the-1726215792, accessed 14 July 2017.
Nicholas Carr, 2010. The shallows: What the Internet is doing to our brains. New York: Norton.
Lorraine Daston and Peter Galison, 2007. Objectivity. New York: Zone Books.
Marian Dörk, Patrick Feng, Christopher Collins, and Sheelagh Carpendale, 2013. “Critical InfoVis: Exploring the politics of visualization,” CHI EA ’13: CHI ’13 Extended Abstracts on Human Factors in Computing Systems, pp. 2,189–2,198.
doi: https://doi.org/10.1145/2468356.2468739, accessed 12 July 2017.
Johanna Drucker, 2014. Graphesis: Visual forms of knowledge production. Cambridge, Mass.: Harvard University Press.
Johanna Drucker, 2007. “Visualizing interpretation,” lecture at Miami University (Oxford, Ohio), at http://polaris.gseis.ucla.edu/drucker/VisualizingInterp_07.pdf, accessed 13 July 2017.
Alexander R. Galloway, 2012. The interface effect. Cambridge: Polity Press.
Lisa Gitelman and Virginia Jackson, 2013. “Introduction,” In: Lisa Gitelman (editor). ‘Raw data’ is an oxymoron. Cambridge, Mass.: MIT Press, pp. 1–14.
Maxence Grugier, 2016. “The digital age of data art,” TechCrunch (8 May), at https://techcrunch.com/2016/05/08/the-digital-age-of-data-art/, accessed 14 July 2017.
Michael Kaufman, 2015. “Surrounded by ‘Architecture of Radio’,” GeekDad (29 November), at https://geekdad.com/2015/11/architecture-of-radio/, accessed 14 July 2017.
Helen Kennedy and Rosemary L. Hill, 2016. “The pleasure and pain of visualizing data in times of data power,” Television & New Media (7 September).
doi: http://dx.doi.org/10.1177/1527476416667823, accessed 13 July 2017.
Helen Kennedy, Rosemary L. Hill, Giorgia Aiello, and William Allen, 2016. “The work that visualization conventions do,” Information, Communication & Society, volume 19, number 6, pp. 715–735.
doi: http://dx.doi.org/10.1080/1369118X.2016.1153126, accessed 13 July 2017.
Tanyoung Kim and Carl DiSalvo, 2010. “Speculative visualization: A new rhetoric for communicating public concerns,” paper presented at the Design Research Society conference (Montreal, 9 July), at http://www.drs2010.umontreal.ca/data/PDF/066.pdf, accessed 12 July 2017.
Robert Kosara, 2007. “Visualization criticism — The missing link between Information Visualization and Art,” IV ’07: Proceedings of the 11th International Conference Information Visualization, pp. 631–636.
doi: http://dx.doi.org/10.1109/IV.2007.130, accessed 12 July 2017.
Alan Liu, 2004. “Transcendental data: Toward a cultural history and aesthetics of the new encoded discourse,” Critical Inquiry, volume 31, number 1, pp. 49–84.
doi: http://dx.doi.org/10.1086/427302, accessed 14 July 2017.
Lev Manovich, 2010. “What is visualization?” Poetess Archive Journal, volume 2, number 1, at https://journals.tdl.org/paj/index.php/paj/article/view/19/58, accessed 12 July 2017.
Lev Manovich, 2008. “Data visualization as new abstraction and anti-sublime,” In: Byron Hawk, David M. Rieder, and Ollie Oviedo (editors). Small tech: The culture of digital tools. Minneapolis: University of Minnesota Press, pp. 3–9.
Donald Norman, 2002. The design of everyday things. New York: Basic Books.
Charles S. Peirce, 1931–1960. Collected papers. Edited by Charles Hartshorne and Paul Weiss. Cambridge, Mass.: Belknap Press of Harvard University Press.
Tess Takahashi, 2016. “Data visualizations, loops, and the taming of big data: Wind Map,” Flow (19 September), at http://www.flowjournal.org/2016/09/wind-map/, accessed 14 July 2017.
Sherry Turkle, 2011. Alone together: Why we expect more from technology and less from each other. New York: Basic Books.
José van Dijck, 2014. “Datafication, dataism and dataveillance: Big data between scientific paradigm and ideology,” Surveillance & Society, volume 12, number 2, pp. 197–208, and at https://ojs.library.queensu.ca/index.php/surveillance-and-society/article/view/datafication, accessed 4 September 2017.
Richard Vijgen, 2016. “Life in the infosphere,” Total Space, volume 50, p. 27, at https://grafisch-ontwerp.hetnieuweinstituut.nl/sites/default/files/total_space_v50_0.pdf, accessed 4 September 2017.
David Weinberger, 2011. Too big to know: Rethinking knowledge now that the facts aren’t the facts, experts are everywhere, and the smartest person in the room is the room. New York: Basic Books.
Mitchell Whitelaw, 2008. “Art against information: Case studies in data practice,” Fibreculture Journal, number 11, at http://eleven.fibreculturejournal.org/fcj-067-art-against-information-case-studies-in-data-practice, accessed 13 July 2017.
Tim Wu, 2016. The attention merchants: The epic scramble to get inside our heads. New York: Knopf.
Received 16 July 2017; revised 20 August 2017; accepted 31 August 2017.
This paper is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Visualizing connectivity: Data as evidence in The Architecture of Radio
by Eef Masson and Karin van Es.
First Monday, Volume 22, Number 10 - 2 October 2017