Code

Computer source code is written in a par ticular language, which consists of syntax and semantics. A language’s level is defined fi by how closely tied it is to the computer’s architecture and operation. Some are compiled, others interpreted, and not all languages are lists of instructions or imperatives, for example, functional languages such as Scheme. The “lowest” level languages offer ff the least abstraction from the machine processes, which typically indicates fewer conceptual groupings of processes. In machine languages, for example, instructions go directly to the microprocessor. A highlevel language, such as Java, needs to be compiled, or translated into processor instructions. High-level languages are marked by greater levels of abstraction, and a subset, including BASIC, COBOL , and even SQL , aspire to greater legibility to human readers. Some of the high-level languages, such as Inform 7, which is used to write interactive fiction, a genre of interactive narrative, can accept statements that read like natural language, such as, “The huge green fierce snake is an animal in Mt King” (Crowther and Conley 2011) (see inter active narr ative). The ontological status of code has been the subject of much debate, particularly whether code can be described in Austinian terms as a performative system, as language that makes things happen. For example, N. Katherine Hayles has argued that “code has become . . . as important as natural language because it causes things to happen, which requires that it be executed as command the machine can run” (2005, 49). However, Wendy Hui Kyon Chun (2008) has warned critics not to confuse source code with executed code and not to treat the code as if it is the hidden essence within the object. Meanwhile, Alexander Galloway stresses the importance of “protocols” over code, arguing that “code only ‘matters’ when it is understood as being the substance of a network” (2006, 57). Such a point complements Friedrich Kittler’s (1995) pronouncement that “there is no software,” but instead a set of electrical signals coursing through the hardware. In that sense, there is also no code. Nonetheless, though code may not be able to claim the concrete physical status of hardware, code studies has developed around the material trace, the par ticular instantiation of an algorithm that is code (see algorithm). ( Johns Hopkins University Press)