Interactive Programmatic Modeling

Modeling and computational analyses are fundamental activities within science engineering. Analysis can take various forms, such as simulation of executable models, formal verification model properties, or inference hidden variables. Traditionally, tools for modeling analysis have similar workflows: (i) a user designs textual graphical the is inferred from data, (ii) tool performs on model, (iii) visualization displays resulting data. This article identifies three inherent problems with traditional approach: recomputation problem, variable inspection expressiveness problem. As solution, we propose conceptual framework called Interactive Programmatic Modeling. We formalize interface illustrate how it be used in two different domains: equation-based probabilistic programming.