In lieu of an abstract, here is a brief excerpt of the content:

125 Data Matthew Fuller The question of data is a recursive one since it asks what we have to begin with, what is given, a quality made explicit in the French name for data, les données. What is given must be gotten; data must be derived, sensed, put on and in the table. This givenness of data is what makes it controversial, in that it cannot be taken for granted, but also what makes it essential to contemporary forms of computing and of culture where data is donated, abstracted, elicited, mined, shared, protected, opened, analyzed, and normalized . In short, we may say that data comes in three kinds: as data that refers to something outside of itself—encoding or representing changes outside of the computer; data as data, that is, what is handled by a computer; and data that works on data, as a program. Such a synoptic statement itself requires some metadata. To start with this triple in reverse, data is convoluted. It is not simply the content of a website, or that recorded in the tables of a database, but crucially, at a certain scale, also the material of the programs and operating systems and interfaces by which they run and are worked on (see database ). This is a legacy of the formulation of the Turing machine that conceptually stored the symbols for instructions on the same abstract tape as that of the symbols they handled , and that of von Neumann (1945), who designed an electronic version of the universal machine that stored programs as data (see Turing test). (Indeed, before computer science became the dominant term, an early contender for the name of the field was data- fi logy [Naur 1992], suggesting that human knowing, language, and discrete structures formed its basis.) What constitutes data, the programs that operate on it, and the structures within which it is placed may therefore at times be obscured, as data becomes active in different ways. A parallel way of formulating this mobility of the category of data ff ff in knowledge management has been the hierarchy of forms running from data, to information , to knowledge, in which the organization of data increases as one rises through the hierarchy and the pertaining degree of abstraction. Conversely, however, one can say that data is a dumb form of programming in that it carries, at a certain level, an instruction to a machine (display this symbol, record this value). Data then, as what is worked on, also does work. A significant example of this is fi that the question of what constitutes meaningful data has, at numerous times, had consequences for the trajectory of computing. As an example, the 1967 Lyons Electronic Of- fice was an early offi fi ce computer with hardware designed to meet the needs of a large ffi catering company, and thus it had to be able to process large amounts of information D 126 Data about numerous small transactions, occurring over a long period of time at multiple sites and driven by the kind and rate of the data to be analyzed (Caminer et al 1996). The problem to compute was, in this case, not of speed, as found in tracking the path of a plane or missile, but in understanding the requirements for thousands of cups of tea and slices of cake. In this case, the anticipation of certain kinds of data, to be handled in certain ways, acted on hardware. Thus, the convolution of data extends out, into the systems that anticipate it. Data as data is what is left when all the noise has been taken out. It may be raw, structured , a stream of variables, or of another kind, but this image of data is now its standard idealization. It is something inert that sits stored on hard drives, rather than as a magnetic charge gradually disturbing them (Kirschenbaum 2008). Data is discovered, studied , held, encrypted, sent, and received, all the while remaining stable. That’s quite an achievement, and it takes a lot of activity, or structured inactivity, a lot of stabilization, to keep it that way. At the same time, as with the Lyons machine, data is sometimes said to drive programming; as computing expands to interpolate more aspects of life, life answers back with, and as, data. As computing moves out into the world, as a governing framework for the arrangement of things, in social media, logistics, or geographic information systems, much of what it comes into contact with also becomes data, and if...

Share