From 1.1. Data vs. Information
…we can begin with a notion of data from empirical science, as a set of measurements extracted from the flux of the real. In themselves, such measurements are abstract, blank, meaningless. Only when organised and contextualised by an observer does this data yield information, a message or meaning. The concepts are converse, two sides of the same thing: data is the raw material of information, its substrate; information itself. In this context it is not surprising that new media art has, is the meaning derived from data in a particular context.
From 2. Indexical Data: We Feel Fine and the Dumpster
These are critiques of the automated analysis that the works use; but even if the analyses were perfect, the more fundamental representational issue remains. These works rely on a long chain of signification: (reality); blog; data harvesting; data analysis; visualisation; interface. Yet they maintain a strangely naive sense of unmediated presentation.
These works construct a notion of data — of its capacities, qualities, and significance — in the ways that they use it. Data here is first of all indexical of reality. Yet it is also found, or to put it another way, given. These works gather existing data from the network, drawing together thousands of elements that are already, unproblematically, “out there”. This reinforces the sense of collapsed indexicality; these data points have causes (authors) of their own that in some sense guarantee their connection to reality, or at least defer the question of that connection. Data’s creation — in the sense of making a measurement, framing and abstracting something from the flux of the real — is left out.
From 3. Alex Dragulescu: Abject Data
Spam is both a literal and figurative resource here: it is a cultural and a digital dataset. It embodies the failures (or perhaps the cost) of frictionless connectivity and techno-libertarian ideals.
Structure as junk is the darker alternative: that what we appreciate as order, form, and coherence is not only ubiquitous and immanent, but mundane, valueless, empty.
Taken together, Spam Plants and Spam Architecture evoke a sense of data as both structurally rich and substantially, vertiginously empty. In this figuration data is an abstract set of potentials, an array of values waiting to be mapped. A dataset feeds a process, that produces an artefact; the process doesn’t care what the dataset is, or was; whatever it was, now it’s just input: the process (the map) reconfigures the dataset completely, arbitrarily, rewrites it not by altering values but by reprogramming them, altering their potential. The process takes the data as whatever it wants (a wall, a shard, a petal, the difference between this petal and the last), irrespective of what it once was (a word, number, number of characters in a word, difference between this word and the last). Anything is anything.
From 4. Lisa Jevbratt : Data Material
Yet Jevbratt’s work is quite unlike conventional information visualisation: like Dragulescu’s work it is anti-information, in the sense of information as a formed message. Rather than transform data into information, Jevbratt transduces one form of data into another — symbolic or logical into visual.
Yet this textural quality also leads back to the inevitable choices involved in mapping data. In IIL and 1:1, one extrinsic structure dominates, to the extent that patterns in the data are literally wrapped around it. The structure is the rectilinear picture plane, a central obsession of twentieth century visual art and a given in digital media culture. (…) Jevbratt’s picture plane mapping is not based on an information visualisation rationale. It is a cultural structure, highly functional information in itself. As the artist says, it connects these works with a whole tradition, it literally frames the data and offers it up to be read in a particular way, as an abstract “picture” (rather than a graph) and also as an artwork. Of course this mapping does “impose its structure,” but that imposition only underlines the functional differences between art and data visualisation.
From 5. Borevitz and Salavon: Anti-Content and the Artist’s Squint
Data practice here is a kind of artist’s squint. This technique is used in painting and drawing as a kind of perceptual abstraction: a way of attenuating, and abstracting, visual information. Squinting blurs detail, so that recognisable objects are abstracted into visual forms: shape, tone, line. The artist’s squint overturns visual information in order to access its “raw data,” before transcribing that data onto paper or canvas. Ironically the aim here is most often realism, the accurate transcription of visual data. To see “reality”, discard information and observe data.
From 6. Data Immanence, Data Agency
Data art’s resistance to information is not unique. Underdetermination is a contemporary artistic staple; much recent visual art works to defamiliarise the cultural vernacular of images and objects, undermining their known “information” in order to make them available anew, as data.
If Digg offers a crude transcendence (top ten) approach to data excess, data art moves in the other direction, towards the many rather than the few. It turns towards immersion and sensation; it emphasises openness and intuition, rather than the extraction of value or meaning. Most of all it confronts us with immanence itself, a multiplicity of relations; with structure as potential, latent, and emergent, not given and named. This stance is in turn a kind of self-referential affirmation of the networked society.
… these artists also provide models of what might be called data agency: more than browsing and navigating — being subject to the data flows — data agents munge, analyse, map and display.context. (…) abstractions. This propagation of data agency is now well underway, supplemented by the data feed ethos of Web 2.0 culture; a growing culture of data practice is evident in communities around the net.
From 7. Data Figures and Critiques
As much as this work pursues data, it cannot escape information. The data is unreachable in itself, always inflected, at the very least, by its particular, concrete manifestation, no matter how plain. These artists seek to turn the data over to us to explore; yet it arrives already shaped, metaphorically primed, conditioned by the processes that created it, informed by the contexts and genres of its presentation.
This is not to say that data art should be somehow more pure or faithful to its datasets, only that it should embrace, and acknowledge, its impurity. Information leaks in, however slight the artist’s intervention; even (or especially) cultural defaults, like the rectangular picture plane of Jevbratt’s visualisations, shape our interpretation of the work in ways that are extraneous to the data.
A related problem is the sense of data as pre-existing or given. The prominence of networked data, and the increasing availability of data from social web services, contribute to a sense that data has an independent being and existence. Because it comes from somewhere else, typically in real time, its creation is abstracted: it is naturalised. Yet data always comes from somewhere…
This severing of data from its creation leads to two related figures. The first is a notion of data as matter or stuff.(…) The second is a sense of data as concrete and objective, rather than contingent and relational.
Agre’s proposal also addresses a third concern, which is the tendency towards data mysticism. Data here becomes a reservoir of potential, a field of the unknown and emergent. Again it seems self-sufficient, rather than part of a wider set of processes; it also slides away from discourse and critique, which are too prosaic to gain any traction.
Does data art become simply an aestheticised (and perhaps functionally impaired) form of scientific data visualisation?
Manovich suggests that one of the roles of data art is to reflect on data subjectivity; I would go further and say that data art is involved in the construction of that subjectivity. It involves a practical exploration of data’s potential uses and meanings; it literally offers us images, figures, for data itself. It pulls us away from information, from the well-formed messages that dominate our experience of digital media. By directing us instead towards data, it opens spaces for potential, for the distributed reconstruction of information. Yet in the process it inevitably encodes its own specific metadata — data about data — that can be read out through the artists’ processes, as this paper has demonstrated. This metadata must in turn inform us data subjects, if we are to move past immersion and navigation to a more critical, and active agency.