What's your code for?

Solving problems. Business or otherwise.

Perfect code that doesn't solve the problem is useless. Perfect metrics that don't serve the user are a distraction.

Or as Ward Cunningham said on the world's first ever wiki:

  1. Make it work
  2. Make it right
  3. Make it fast

In that order. A fast program that isn't right, useless. A perfect program that doesn't work, useless. A fast program that doesn't work ... impressive. But useless.

Happy weekend ❤️

Cheers,
~Swizec

PS: this approach helps you shift left and catch bugs, bad specs, and misunderstandings early. Before they're expensive to fix

PPS: it's okay to write perfect code without a goal in mind for fun; to me it gets boring fast