A fantastic transubstantiation takes place for which visual signs and images no longer simply represent other things in the world, but become themselves tactile technologies that, when activated, cause a real event to occur correspondent with the semantic content of that sign/image. A picture of a bomb is merely a representation, whereas a button with a picture of a bomb on it that causes remote explosions is weaponized skeuomorphism. Put in more technical terms, the GUI is a visualization of a machinic network and of the outcomes that it claims to mediate; the formation of its interfaciality is an arc of translation from a set of possibilities into a visual instrument. Between the machine and what it can do and the representations of that potential are translations, and however arbitrary or integral each may be, they are necessary for our comprehension of any network we might encounter.
This is particularly true for computational machines, which can, according to software instructions, perform any calculable task we might assign to them. In the past, any one of those functions may have been performed by a single-purpose analog machine that perhaps communicated its unique functions directly by its form. Wrenches look like wrenching. Computational machines may rely on the rhetorics of the interface to do the same, or as a design strategy, the posture of the computational object and its function can evolve semiautonomously. Through that expression of function by GUI layers, the machine network can appear to be and do almost anything, and because of this (not in spite of it), the interface is essential to the machine itself. In the play of a function communicating itself, honestly or by dissimulation, the real and relative transparency or opacity of an interface ’ s distant effects are themselves disclosed, hidden, or masked. As the interface conceals and narrates, its concealments and narrations are themselves concealed and narrated. At one extreme, placebo interfaces let Users point and click all they want and yet have no real effect on any outcome other than apophenic satisfaction. Some Users manage persuasive interfaces, which deliberately train the self in response to positive and negative feedback stimulus, promising more control over life by reassigning that control to automated metrics. These micro-rhetorics of the interface are core to their social effects. Some call this “ making computers invisible, ” others call it “ interaction dissolving into behavior, ” and still others “ society without organs. ” Not so unlike my son, we are all probing our interfacial condition, trying to figure out the ultimate technical, ethical, political, and aesthetic possibilities of a world full of images of things that do (or do not do) the things they visual signify.
In the awkward, incomplete (and incompletable) semiotics of interface representability, where cause and effect is reduced to a bounded set of idiomatic and idiotic pictograms, culture becomes interfacial and vice versa. Even a maturing metaphysics of data visualization serving the ubiquity of time-based diagrams of events and patterns (and prototyping perhaps what future GUI will look like) does not provide a short-cut out of ambiguity, but it does raise the stakes. <sup>13</sup> In that they previsualize their effects in advance for us, these sorts of interface machines must be located within a historical shift in technologies of the image itself. <sup>14</sup> Taking the long view, we see that humans ’ externalized expression of visual ideas dates at least to the primordial architectures of the cave wall, and then much later, it passed through a relatively short painting-photographic-cinematic phase (lasting a few centuries, give or take) for which individual images and image sequences were produced, distributed, and appreciated as rare artifactual events. Now and for the foreseeable future, images are a specific genre of machines. How so? Like the images on paper money that appear as they do in order to support the performance of specific counterfeit-prevention technologies designed into the patterns, some images have a discrete technical capacity that is inextricable from their physical form. Consider how everyday data visualization turns the diagrammatic image into a scientific, managerial, and military instrument, or how GUIs turn icons into active, goal-directed tools mediating between human folk psychology of action and algorithms linked to screens in the User ’ s environment. Additionally GUIs don ’ t only mirror preexisting User intentions; as whole interfacial regimes (such as Windows or iOS or Bloomberg Terminal, etc.), they also train thought toward certain ways of interpreting that environment through the repetition of represented interactions. As those representations become more closely glued to the direct perception of environments (as for augmented reality, for example), their capacity to engender committed interpretations for Users will prove irresistible to various fundamentalisms (see below). The machinic image is qualified by many little sinkholes between the symbolic, the imaginary, and the real, and at a global scale of billions of Users, the interfacial image is also partially a function of sheer machinic quantity. With the comparatively instantaneous adoption of mobile devices (a Turing complete processor + camera + homing tether + telephonic voice relay), we have seen an explosion in the absolute volume of mechanical images of and in the world, dwarfing the total sum produced before the mobile phone appeared in our hands. Unlike images of the painting-photographic-cinematic era, these images do not pass into an archive only after their practical life is passed; rather, through various Apps, images are produced through the medium of the archival database itself, socialized through the archive, and assigned searchable metadata through the archive. As a consequence, the general image apparatus is slowly accumulating a comprehensive simulation of visual experience that will be of enormous value to future artificial intelligences interested in simulating its Anthropocenic origins. This may even be its most durable purpose and its true responsibility. Even today, it ’ s not difficult to see the whole Android user population (for example) as comprising individual nodes in a vast, massively distributed supercomputing sensing, seeing, tracking, and sorting platform. At the same time, visual trends like the New Aesthetic suggest the potential of an art (if that is still the right word) that is made not only by artificial vision machines generating their own autonomous aesthetic, but in time an art for such intelligences that can appreciate them uniquely and perhaps develop their own taste genres of M2M (machine-to-machine) connoisseurship. <sup>15</sup> It is also worth noting that many current applications of the machinic image (e.g., on currency, or to X-ray great works of art to verify their originality, or facial recognition to verify the identity of User) are used to authenticate the originality of either the image itself or what it represents. With Walter Benjamin in mind, we see that the image in the age of mechanical reproduction is being pushed toward the assignment of verifying the original and its aura as well.
224
Interface Layer