Achieving Self-Sustainability in Interactive Graphical Programming Systems

Programming is fraught with accidental complexity. Software, including tools used for programming, is inflexible and hard to adapt to one's specific problem context. Programming tools do not support Notational Freedom, so programmers must waste cognitive effort expressing ideas in suboptimal notations. They must also work around problems caused by a reliance on plain text representations instead of Explicit Structure.

The idea of a Self-Sustainable programming system, open to adaptation by its users, promises a way out of these accidental complexities. However, the principles underlying such a property are poorly documented, as are methods for practically achieving it in harmony with Notational Freedom and Explicit Structure. We trace the causes of this difficulty and use them to inform our construction of a prototype self-sustainable system. By carefully reflecting on the steps involved in our specific case, we provide insight into how self-sustainability can be achieved in general, and thus how a motivated programmer can escape the aforementioned sources of accidental complexity.