4 ms·

What were the arguments back then that supported the opposite bias?

• cmrdporcupine 3y ago
  "Move-fast & break things." "Test-first-design & aggressive unit testing is the solution." To be charitable, perhaps that the type system was getting in the way, physically of what they wanted to accomplish conceptually. I actually think it has more to do with object-oriented programming as it was propagated through C++ and Java. We were told (and believed) that OO abstractions would solve many software complexity problems. The focus of a lot of programming work was/is on bringing these OO design patterns into play. In large part the OO thing is about a kind of dynamic dispatch. And yet languages like Java imposed a very rigid way of doing OO, and made some of the patterns awkward to express. So if you believe in your heart in the promise of OO, but find that your OO experience is relatively handicapped by your static type system, one then blames the type system. Ruby was maybe the best example of where this line of thought was going/coming-from? I think we are getting past this by getting (somewhat) past OO as a dominant design methodology. But also by adopting languages with richer static type systems.