Hacker News Reader

Comment by eerikkivistik | Hacker News Reader

I find it a refreshing experience to look at your own code from 10 years ago, or even 5 and think "Who wrote this shit?". If I ever look at some of my old code and feel good about it, that means I have failed to improve.

View on HN

I find that half the time I'll wonder who wrote it, realize I wrote it, start to rework it, and then realize why I wrote it that way in the first time, add a comment and move on.

Same here. Sometimes I add comments like "don't change this shit!" to remind myself that there is a reason why it is like that.

Those comments would help if future me would listen to them. "Huh, I wonder why I put that in here... let's see what happens if I change that shit."

"I'm certain I've learned enough in the last 10 months to solve this problem now."

"I did not."

I prefer to add comments which contain some rational, like "Changing this causes race condition headaches in the frobulater." Or "Ugly... but it handles the edge case of XYZ in the foobar config"

If I don't then the comments either have the effect of a) scaring future me, or b) getting ignored because I think that I can do better now.

Could have meant that you've reached the approximate upper bound of excellence in that particular piece of code.

There are objective good code that are timeless, implementations that does not need another look or left few/no space to improve.

Granted, it's a lot harder to do at larger scale.

I suppose I have a long ways to go still :)

Bonus points if you first ask yourself "who wrote this shit?", then open blame, then see your own name.

I love this

Reply on news.ycombinator.com