How would you evaluate the necessity of code formatting now contra code formatting in the yearly 2000s?
The reality early 2000s:
- Huge code bases due to monolithic nature of application
- High level of reusability , on low level due to developers being obsessed with coding shared libraries and frameworks to compensate things that did not existed.
- Products were written in single dominant language.
- A microservice rarely exceeds 5000 lines of code (probably that is too much).
- We have multi language setup with microservices written in Java, Kotlin, Golang, Python.
- High level of redundancy. Code is duplicated much more often.
- Code is disposable, very easy to just throw it in the garbage and write something new.
- The code being disposable means we will not end up with a maintainer having to maintain it for many many years as it is with monoliths.
These facts made me to ask myself the question if enforced code formatting is that important now as it was back then.
We have 2 entities:
Many would argue that code formatting does things easier to the Maintainer. But today the Maintainer will have a set of Microservices to maintain; maybe one maybe more. He will not have a huge code base of hundreds of thousands of lines. And what if he has 1 service in Kotlin, 1 in GoLang, 1 in Java? Does it really matter how the code is formatted as long as the formatting is adequate?
All the statements above should be taken under the presumption that we have adequately formatted code; even if we don’t have it questions the necessity of unified format enforcement.