100% agree because there’s a lot of value in understanding how and why past code was written. It can be used to make better decisions faster around code to write in the future.
E.g., if you’ve ever wondered why code was written in a particular way X instead of Y then you’ll have the context to understand whether X is still relevant or if Y can be adopted.
E.g., easier to prompt AI to write the next commit when it knows all the context behind the current/previous commit’s development process.
But that's not what is in the whole context. The whole context contains a lot of noise and false "thoughts".
What the AI needs to do is to document the software project in an efficient manner without duplication. That's not what this tool is doing.
I question the value in storing all the crap in git.
E.g., if you’ve ever wondered why code was written in a particular way X instead of Y then you’ll have the context to understand whether X is still relevant or if Y can be adopted.
E.g., easier to prompt AI to write the next commit when it knows all the context behind the current/previous commit’s development process.