Do you know any studies about the percentage of boilerplate code?

Question

Hey guys. Do you know any studies about the quantity of boiler plate code in different project types? I don't seem to find any actual numbers. I know this is subjective and relative to the language, project type and frameworks used, but there must be an actual study somewhere that covers at least a niche. Thanks!

Answer 1

In order to do a study about quantities of boilerplate code, you'd first need a definition of boilerplate. If you want the study to be completed during your lifetime, the definition will need to be something that can be checked by a computer. You'd also need access to a lot of code. However, not all code is publicly available, so there's a possibility of selection bias. Finally, what would be the benefit of the study to justify the cost?

That said, I would think the amount of boilerplate code would be more tightly correlated to the programming language used than the type of project. A website created using Lisp, for example, would have a lot less code than a website created using COBOL.

Answer 2

IMO, Boilerplate code is a symptom of a poorly written application. Why not factor out that boilerplate code, instead of repeating it everywhere? If that's impossible, then it is a symptom of a poorly designed programming language.