Digital Interfacing

Action and Perception through Technology

Daniel Black

• Publisher: Routledge
• Year: 2018
• ISBN:
  9780429757204
• Edition: 1
• Topic: Informatik
• Subtopic: Human-Computer-Interaktion

This book takes the interface – or rather to interface, a process rather than a discrete object or location – as a concept emblematic of our contemporary embodied relationship with technological artefacts. The fundamental question addressed by this book is: How can we understand what it means to perceive or act upon the world as a body–artefact assemblage? Black works to clarify the role of artefacts of all kinds in human perception and action, then considers the ways in which new digital technologies can expand and transform this capacity to change our mode of engagement with our environment. Throughout, the discussion is grounded in specific technologies – some already familiar and some still in development (e.g. new virtual reality and brain–machine interface technologies, natural user interfaces, etc.). In order to develop a detailed, generalizable theory of how we interface with technology, Black assembles an analytical toolkit from a number of different disciplines, including media theory, ethology, clinical psychology, cultural theory, philosophy, science and technology studies, cultural history, aesthetics and neuroscience.

1
The Myth of the Myth of Transparency

If there has been an age of cinema, an age of radio or an age of television, then our current media era might be described as the age of the interface (Yonck, 2010). But the way in which the term is used is often hazy and contradictory. While definitions of older media technologies might not be beyond contestation, when asked what television is, for example, most people would confidently point to a specific physical artefact in answer; but where should we point when seeking to answer the question of what an interface is? Over the past two decades, work has already been done to place the human–computer interface within a wider media context, but this work can obscure the ways in which human–computer interfaces do not fit into existing frameworks for discussing media technologies.

When trying to answer the question of what an interface is, perhaps it is best to start by stating what an interface is not. Most importantly, and contrary to most existing accounts, the primary purpose of an interface is not representation. Certain kinds of interfaces might utilise forms of representation, but these representations are not themselves interfaces. In addition, we can also at least problematise the idea that interfaces are media. If it is appropriate at all to refer to an interface as a medium, then it must be understood that the term ‘medium’ in this context means something different from what it does when applied to other kinds of media technology: if a medium is generally understood to be a technology that mediates between the human and something ‘beyond’ that technology, an interface must be understood to mediate between the human and the technology itself.

In this first chapter, I want to spend some time looking at how the term ‘interface’ has come to be a part of our everyday understanding of human–technology relations, both reflecting and inflecting common-sense conceptions of the role of digital technologies in our lives, as well as informing the values and priorities that are applied to the development of new devices. This understanding has also had an important influence on scholarly investigations of digital technologies: while the values and priorities of technology makers have frequently been critiqued by scholars, these critiques nevertheless tend to share with the objects of their criticism a set of basic assumptions about what interfaces are and what they mean.

Debates over what an interface is—or should be—provide a good starting point when trying to characterise a dominant conception of our relationships with technological objects: cutting-edge research into new kinds of personal digital device is most often associated with new kinds of interface (e.g. direct brain interfaces; see Chapter 4); the marketing of current digital devices commonly focuses on interface features rather than the (largely invisible and incomprehensible) hardware; and ‘the interface’ has become the object of study presented as equivalent to media forms such as television or film in new media scholarship. Because it is, by definition, connected to our physical interactions with digital devices, discussing the interface will allow me to tease out a set of widespread assumptions concerning how we use technology, and the opportunities for interaction technology can provide—and show their failings. Having done so, I can then draw attention to important dimensions of technology use that an emphasis on the interface has obscured before turning to a more detailed and sustained investigation of these dimensions in Chapter 2.

The Evolution of the Word

The word ‘interface’ is a relatively new one—the first example of its use given by the Oxford English Dictionary comes from 1882—but today it tends to be used as if its history is a century shorter. This is not surprising given that it only began to enter common, everyday usage in the mid-1980s. It has entered the popular lexicon via a very specific route, a route that has importantly shaped common understandings of what an interface is and how it works.

Branden Hookway has already given a detailed prehistory of the term ‘interface’ prior to the spread of its current popular usage in the 1980s (Hookway, 2014, Chapter 2). Originating in Victorian research on fluid dynamics, the word’s nineteenth-century meaning of ‘a face of separation, plane or curved, between two contiguous portions of the same substance’ was appropriate for specialised technical, rather than everyday, usage; however, it later became caught up in the post-WWII infiltration of everyday life by specialised technical concepts brought about by the rise of cybernetics and the spread of the digital computer. In a post-cybernetic moment that often saw the world as composed of systems in interaction with one another, it was necessary to talk about the regions where those systems came into contact, and those regions were, literally, interfaces. As new knowledges and technologies allowed new kinds of interactions between systems to be created artificially, the noun ‘interface’ even gave rise to the ‘space-age verb’¹ ‘to interface’ (to which I will return later).

Cyberneticists saw systems all around them, but in the context of computers, interfaces might exist between computers, between different subsystems within computers, between computers and other machines, or between computers and the human beings who interacted with them. The spread of the personal computer brought the term into popular usage in the restricted sense in which it was used in computer science. In fact, it was introduced as a restricted subset of the restricted sense in which it was used in computer science: that is, while within computer science the word ‘interface’ could be used to refer to the meeting of computer hardware, computer software, and human user in any combination other than two human users, in everyday usage the term ‘interface’ has become simply an abbreviation for and thus synonymous with the human–computer interface. Technical or semi-technical references to other kinds of (usually hardware-to-hardware) interfaces can occur, but this is rare for the average computer user. In everyday discussion, the word ‘interface’ alone is understood to refer to the mode of interaction between human user and machine; for example, a computer will be described as having a ‘graphical user interface’, or a smartphone will be described as having a ‘multi-touch interface’.

Probably the most important phase of this trajectory that took the word ‘interface’ from varied technical usage, through the restricted technical usage of information technology and then into common usage in the further restricted sense of the human–computer interface began with Doug Engelbart’s famed 1968 demonstration of what would become the foundations of the modern graphical user interface, or GUI (Johnson, 1997, pp. 20–22). It was the GUI, typified by computer operating systems such as Microsoft Windows and Apple’s Macintosh, that more or less single-handedly introduced the term into the popular lexicon, and made the interface a key concept in our understanding of what a computer or other digital device is. Speaking of Ridley Scott’s famous ‘1984’ television advertisement, which introduced the Apple Macintosh to the American public during that year’s Super Bowl, Steven Johnson notes that ‘it was the first mass-media promotion that devoted as much attention to an interface as to the underlying hardware itself’ (1997, p. 50).

In 1989, John Walker, one of the founders of Autodesk, noted

the extent to which the term user interface has emerged as a marketing and, more recently, legal battleground following the introduction and growing acceptance of [graphical user interfaces]… Many people would probably fail to identify anything before a fourth-generation menu system as a ‘user interface’ at all.

(Walker, 1990, p. 443)

And in 1993, influential interface design expert Brenda Laurel commented that ‘“Interface” has become a trendy (and lucrative) concept over the last several years—a phenomenon that is largely attributable to the introduction of the Apple Macintosh’ (1993, p. 2). During this period, the GUI was presented as the magic ingredient that could transform computers from something arcane and specialised, something only available to a dedicated technical élite, into a mass-market consumer appliance ‘for the rest of us’, as the famous Apple campaign claimed.

The term ‘interface’ was turned into a buzzword (Heim, 1993, pp. 75–76) primarily through its use in marketing the GUI. The ease with which a user might operate a complex digital device is crucial to its appeal, and so advertising seeks to instil a belief that the device’s user interface is inviting— even ‘intuitive’, to repeat a common marketing expression. But this suggestion that a digital device ‘has’ a certain interface is out of keeping with the literal meaning of the term. After all, the word interface literally means the point at which the exteriors of two or more entities meet; strictly speaking, to talk about one entity independently possessing an interface makes no sense.

In fact, a fundamental transformation has taken place over the course of this word’s evolution from specialised technical term to everyday marketing jargon: the meaning of the word did not simply become progressively more narrow and restricted, but also underwent a qualitative change. An interface came to be a thing. This change is not surprising, given that it accompanied the introduction of digital computers as consumer items, and consumerism deals in things. In the late 1980s, Alan Kay, a key figure in the development of the personal computer, quipped, ‘User interface has certainly been a hot topic for discussion since the advent of the Macintosh. Everyone seems to want user interface but they are not sure whether they should order it by the yard or by the ton’ (1990, p. 191).

To explore this change, let’s consider an application of the word ‘interface’ that would have been unremarkable at the turn of the twentieth century, but is quite different from the dominant contemporary usage derived from computer engineering. The term ‘interface’ originated in the field of fluid dynamics (Hookway, 2014, pp. 59–67) to refer to the internal division created by the meeting of differing kinds of liquid, such as hot water and cold water, or still water and moving water. This interface is a kind of non-thing, a non-place. It is a two-dimensional plane defined not by what it is, but by where it is in relation to other things, and what those other things are doing. It is a place that occupies no space; it is simultaneously one entity, the other entity, both entities and neither entity—it’s often unclear if there even are two entities, or if the interface is rather internal to a single entity. If the contours of one body of liquid change, the interface alters, and alters the contours of the other body, but it is not at all clear that we should say that it is the interface that changed the contours of that other body. The interface is both a product of the contours of the two liquids and the thing that shapes the contours of the two liquids. It maintains the perfect reciprocity and complementarity of the two liquids, and exists only while they are in contact. It is purely a phenomenon of the meeting of the two liquid bodies and will vanish at the moment they disengage from one another or merge. This is an interface, but it is hardly something that you could put in a box and sell in an Apple Store.

When the term ‘interface’ is used in computer science to describe the point of contact between two systems, the qualities of this contact are fundamentally different. As products of human engineering, each of the meeting systems is separate and distinct, separable and readily distinguishable. If two hardware interface ports are connected to one another, this inevitably creates a new phenomenon analogous to the interface between the two liquids—one that is equally a part of both systems and neither of them—but at the same time the engineering logic of computer hardware development understands the interface as primarily a piece of hardware, a physical structure built as part of one system that will allow a connection with another. Similarly, when software interfaces with either another layer of software, or hardware or a human user outside itself, this is understood in terms of the engineering of a specialised system.

In computer science, an interface becomes a point of articulation, of translation or exchange. It becomes a point of connection across which data moves. This is quite different from the interface between two fluid bodies: in that case, the interface is precisely that plane across which there is no exchange; there is no mixing of the two fluids, nor is there any exchange or translation between them. The interface of some liquids— such as oil and water—might be characterised precisely by the fact that they cannot mix or merge; other kinds of fluids might be able to mix with one another, but if they do so the interface collapses and disappears. The borders of the interface might shift—liquid that was still, for example, might begin to move, shifting the location of the interface between still and moving water—but in that case one entity has simply become part of the other entity, which is something different from translation or exchange. Where they create an interface, the two fluids remain distinct and separate from one another, stable and unchanged in their physical composition, and yet the exterior contours and disposition of each fluid are changed, determined by its contact with the other to produce a perfect complementarity that does not originate with either body or move from one to the other. Each liquid body is changed by its relationship with the other, but there is no clear point of origin for this change and it has no directionality. Their meeting creates a new phenomenon (the interface) that brings about a convergence and reciprocity between them as long as they are in contact, but vanishes when either their contact or their distinctness ceases.

The more general engineering logic of the computer interface is shifted further by the specificity of the human–computer interface. Where other kinds of computer interfaces seek to create a bridge between different kinds of hardware or software, the human–computer interface seeks to create a bridge between the computer and a human being, something fundamentally different from hardware or software, and ultimately alien to the principles of computer engineering. This creates the initial failing of HCI (human–computer interaction) research, which is that it tends to treat the human user as simply another piece of computer engineering, whose behaviour is fixed and predictable. The human being is treated as if it is simply another piece of hardware—a ‘model human processor’ (MHP; Card, Newell and Moran, 1983; see Turner, 2009, p. 6)—to be plugged in to the larger system, and the behaviour of the user is largely treated as a product of the design of that system.

An awareness of this problem developed during the broader turn-ofthe-millennium critique of disembodied information (e.g. Hayles, 1999; Hansen, 2000; Dourish, 2004 [2001]), which took issue with a late twentieth-century discourse of ‘cyberspace’ as ‘the new home of Mind’, where ‘[o]ur identities have no bodies’, in the words of John Perry Barlow (Electronic Frontier Foundation, 1996; see introduction, this volume). Today, the designers of digital technologies are trying to move from a model of the user as a collection of inputs and outputs towards one that understands interactions with technology as embodied, affective and frequently non-rational (Turner, 2009). User experience, or UX, in particular seeks to adequately acknowledge the importance and complexity of the role played by embodied human users in the utilisation of technological artefacts, and self-consciously grounds itself in criticisms of traditional cognitivism, which have arisen from areas such as phenomenology (van Dijk, 2009) and the extended cognition thesis of Andy Clark and David Chalmers (1998; see Chapter 2).

While this shift has been important, acknowledging that embodiment is a part of our relationships with technology is not the same as understanding how embodiment is a part of those relationships, and the UX approach arguably still struggles to build a clear and consistent conceptual framework that can meaningfully inform design practices. A belief in technologies as disembodied perhaps has been more effectively discredited by changes in technology itself: mobile devices, virtual and augmented reality and wearables are self-evidently dependent on the bodies of their users (see introduction, this volume).² At the same time, however, developers of new technological devices all too often treat these new user interface technologies as necessarily and innately progressive in their engagement of users’ bodies, even as they merely reproduce an existing set of assumptions about what bodies are and what they can do. Furthermore, while the critique of disembodied technology originated in the humanities, UX’s attempts to account for the user stand in contrast to the complete absence of users in many influential theorisations of digital user interfaces from the humanities and social sciences.

Transparency and Opacity

Florian Cramer has noted the way in which ‘[h]umanities media studies have simultaneously restrained and inflated the notion of interface’ (2011, p. 120). Of the variety of possible applications of the term, Cramer argues,

media studies have historically privileged one, the human-to-software user interface, using it often ...