Hacker News Reader

This should be obvious. Writing well-structured, decoupled, modular code can make you lose track of performance concerns. And since no one likes to write things more than once, so there's only "one shot" at writing the application, people tend to err on the side of "well-structured", because hey, who's gonna fore you for following best practices?

View on HN

I’d argue that getting performance in a coupled unstructured mess (I’m not saying that’s what SO made) is going to be more difficult.

On the other hand, trying to optimise too early leads is wasteful and leads to headache down the line.

Writing "well-structured, decoupled, modular code" is a best design practice for a good reason. It what leads to best results overall, where "best results" means code that is easier overall to test, maintain, modify/expand, and optimise as needed.

Regarding optimisation, something related to keep in mind is that optimisation at system level (which is often the desired end result) is not the sum of independent local optimisations.

Reply on news.ycombinator.com