Comment by ravenstine | Hacker News Reader

Comment by ravenstine | Hacker News Reader

Parent Full thread

This answer makes two big assumptions that haven't been proven out yet.

- Understanding code without writing it is as viable as understanding code that you've worked with directly or indirectly

- Businesses care that you understand code

I really doubt the first one. Traditionally, understanding a code base in large part came from working with it intimately and building that muscle memory. The idea that understanding code by reading it is as good as understanding it from writing it, in my opinion, is not realistic.

Whether businesses care that their engineers (which they are increasingly viewing as monkeys at LLM typewriters) to understand the code remains to be seen. I don't think they particularly care whether their code runs slow and is buggy so long as it works just enough to churn out features and continue to pull income.

> The idea that understanding code by reading it is as good as understanding it from writing it, in my opinion, is not realistic.

As one of those developers who has written almost no significant code by hand since November 2025, but has produced a great deal of working software, I still understand the majority of the code I've produced just as well as if I'd typed it myself.

I may not be typing it myself, but I'm manipulating it constantly. It's not as simple as "reading" it - I'm reading it, executing it, figuring out refactorings for it, having tests built for it, having documentation built for it, sometimes writing that documentation myself, spinning up example scripts that use it, then building new code that depends on that previous code.

It's that act of exercising the code that gives me confidence that I understand it.

> understanding it from writing it

On the surface it sounds weird - why would this be?

Possibly because building a system is not a one-shot step, but a process of many iterations, each of which involves experiments in production, and gaining more learnings. So at the end of the process, you don't just have N lines of working code, but also N lessons learned along the way. So presumably with the AI process we miss out on half the value.

Now the going thesis is that this extra value is unnecessary if we take the plunge and don't look back. My gut says the answer is somewhere halfway, I guess we'll see.

Reply on news.ycombinator.com