A dataflow pattern catalog for sound and music computing

This article describes a set of patterns the authors have seen emerging during years of experience developing assorted applications in the sound and music domain and receiving influences from theoretical models, existing systems, and colleagues. Those patterns aim at offering a generative pattern language that falls within a generic data flow architecture. Regardless the audio domain origins of the patterns, we foresee that they have applicability in other domains. Contributions of this paper are: General Dataflow Patterns, that address problems about how to organize high-level aspects of the dataflow architecture, by having different types of modules connections; Flow Implementation Patterns, that address how to physically transfer tokens from one module to another, according to the types of flow defined by the "general dataflow patterns". Tokens life-cycle, ownership and memory management are recurrent issues in those patterns; and finally, Network Usability Patterns, that address how humans can interact with dataflow networks.