Towards a hermeneutic definition of software

The paper addresses the need for a comprehensive philosophical definition of software, arguing that its pervasive influence across economic, social, and cultural spheres necessitates a deeper understanding beyond simple algorithmic definitions. It critiques existing perspectives within media studies, citing Kittler's assertion that "there is no software" (reduction to hardware) and Manovich's counter-argument that "there is only software" (reduction to algorithms). The author points out limitations in both views, arguing that media studies often neglect technical and philosophical aspects, and computer science overlooks the ontological status of software. The author contends that existing definitions in computer science are too intellectualized, fail to explain the program-machine interaction, struggle to account for errors (bugs, malware), and underestimate philosophical implications. The introduction lays the groundwork for a philosophical approach that aims to analyze the limits of existing theories and offer a synthesizing definition, utilizing a mainly continental philosophical approach.

The paper reviews existing definitions of software, highlighting their shortcomings. It discusses Kittler's argument against the existence of software, reducing it to mere electrical circuits, and contrasts it with Manovich's view which emphasizes software's ubiquity and defines it as a combination of data structures and algorithms. The author critiques these approaches for their limited scope and lack of engagement with the history of technology. Furthermore, the paper examines Colburn's concept of "concrete abstraction" and Irmak's definition of software as an "abstract artifact," pointing out limitations in both. The literature review sets the stage for the author's proposed hermeneutic approach by emphasizing the need for a more nuanced and comprehensive understanding of software's nature.

The paper employs a hermeneutic approach, drawing heavily on Paul Ricoeur's philosophy of text. It proposes five criteria for a comprehensive definition of software: (1) software is a form of engineering with functions; (2) it is realized in written form not intended for direct reading; (3) it produces sets of written instructions (programs); (4) it connects to the functioning of a physical machine; (5) these programs can be of various types. The author argues that the written form is not merely incidental but fundamental to software's nature. Ricoeur's concept of "distanciation" is applied to analyze how software gains autonomy from its author, context, and initial readers. The paper explores the dialectic between "distanciation" and "appropriation," examining the various layers involved in software creation and execution. A detailed examination of the FDE cycle in the CPU and the analogy between Boolean logic and electrical circuits explores the ontological aspects of software execution. The methodology is explicitly philosophical, aiming to synthesize insights from media studies and computer science through a hermeneutic lens.

The paper's key finding is that software cannot be adequately defined as an object or a form of language; instead, it's a complex hermeneutic process. The five proposed criteria highlight software's written form as fundamental, contrasting it with other forms of writing where writing is merely a means to an end. The analysis reveals a multi-layered structure of software, moving from user interaction to high-level programming languages, compilers, assembly languages, machine code, and finally, the physical manipulation of electrical signals within the CPU. This process is shown to be a continuous cycle of writing and rewriting, emphasizing the inherent hermeneutic nature of software. The connection between Boolean logic and electrical circuits is identified as a regulative analogy underpinning the functionality of software, creating a bridge between abstract symbols and physical reality. This analogy, the author argues, is the condition of possibility for the hermeneutic appropriation of software. The analysis goes beyond viewing software as a tool, emphasizing its capacity to shape and transform human experiences and our understanding of the world.

The paper's findings address the initial research question by providing a nuanced hermeneutic definition of software, moving beyond simplistic algorithmic or materialist approaches. The multi-layered structure of software, identified as a cycle of writing and rewriting, explains its complexity and its capacity to shape human experience. The key role of the regulative analogy between Boolean logic and electrical circuits highlights the deep intertwining between abstract symbolic representations and concrete physical actions. This hermeneutic analysis challenges the limitations of previous definitions and offers a new framework for understanding the relationship between software, technology, and human existence. The significance lies in the integration of continental philosophy into the study of technology, leading to richer and more comprehensive understanding of software's impact on culture and society.

The paper successfully establishes a comprehensive hermeneutic definition of software, arguing that it is a complex process of writing and rewriting with epistemological and ontological dimensions. It demonstrates the usefulness of continental philosophy in understanding software's intricate relationship with human knowledge and experience. The integration of Ricoeur's hermeneutic model provides a framework for analyzing software's multifaceted nature, highlighting its role in shaping our understanding of the world. Future research could explore the ways in which software and artificial intelligence transform narrative identity and potentially the very nature of the self in the digital age.

The paper primarily focuses on a theoretical and philosophical analysis of software, with limited empirical data. The scope is confined to a specific hermeneutic framework, potentially overlooking other valid perspectives. While the analogy between Boolean logic and electrical circuits is central to the argument, further investigation could explore the limitations and complexities of this analogy in the context of software errors and unexpected behaviors. The discussion of the "algorithmic self" remains largely theoretical and could benefit from further empirical research to validate the claims.