Low-Code?

In my last lesson in the "Modern Embedded Systems Programming" series on Embedded.com, I discuss visual modeling and automatic code generation.

Based on human psychology, I present a case for visual modeling, which allows us to process visual information at a much higher bandwidth and retain it better than any other form of information, particularly complex textual representation. But to take advantage of our "visual super-intelligence," we must choose visual representations that are intuitive yet precise enough to convey the information unambiguously. Out of all the UML, state machine diagrams stand out as the most constructive (most useful for effective code generation).

I also present a case for code generation based on the economics of software development because the ROI (Return on Investment) of modeling is negative unless the models are used to generate substantial portions of the production code. Indeed, modeling without code generation has been tried and failed every time because it violates the DRY principle (“Don’t Repeat Yourself”). Models that don’t contribute to the final production code are a duplication of effort, become outdated, and get abandoned sooner or later.

This leads me to describe the real pragmatics of modeling and code generation, which turns out to be quite different from the initial reservations expressed by developers first exposed to the method. I argue for a new generation of "low ceremony" modeling tools like QM, which are simpler, code-centric, and relatively low-level compared to the traditional "high ceremony" modeling tools of the previous generation.

All this is suspiciously similar to the lately fashionable “low-code” approach, which seems like the most recent reincarnation of visual modeling and code generation. To me, a simplified, “low-ceremony” approach to modeling seems as close as we can get to the “low-code” vision.