Global and Local Implications of Computational Artifacts

This article investigates the undone science of software scalability. Software engineering literature focuses on the multiple ways that software can scale up, but too rarely on the possibilities, or necessities, of making it scale down. Drawing from geographical studies, we hypothesize that scales are constructed, rather than given, and argue that the design and use of specific computational artifacts—programming languages, data structures and protocols—are active participants in such material construction. Through a series of case studies, we discuss how computational artifacts can realize computing at both local and global scales.