If You Build It, Will They Come? Lessons from 20 Years of Industrial Use of a Code Transformation DSL
TL;DR
An experience study of RcScript, a proprietary code-transformation DSL used in production at Raincode and Raincode Labs for more than 20 years, and of the long-lived transformation platform built with it to modernise PACBASE-generated COBOL, which shows a consistent pattern of features that improve local readability are used, whereas features that hide errors, are hard to read, or depend on weak editor support are avoided.
Abstract
Legacy modernisation at industrial scale requires automated source-to-source transformation rather than manual refactoring. This paper reports an experience study of RcScript, a proprietary code-transformation DSL used in production at Raincode and Raincode Labs for more than 20 years, and of the long-lived transformation platform built with it to modernise PACBASE-generated COBOL. The platform has been applied to more than 500 million lines of legacy source code. We highlight one engineering question within that industrial ASE solution: which optional language features of the implementation DSL are actually adopted in practice, which are avoided, and why. The study combines static analysis of two production configurations with an interview with the lead maintainer. The observed feature use varies, yet the results show a consistent pattern: features that improve local readability are used, whereas features that hide errors, are hard to read, or depend on weak editor support are avoided or used only sparingly. From that, we distil practical lessons for the design and evolution of industrial DSLs and long-lived automated transformation platforms.