Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

> For a knowledge or creative worker it's similarly easy to verify expected performance.

No it really isn't! Coding performance metrics are a wasteland of failed techniques.



Hence, I think, the rise of judging programmers on business outcomes. Revenue increases aren't a great metric for programmers (no one has a damn clue how to quantify tech debt and futureproofing), but at least they aren't actively destructive. Similarly, "story point velocity" and the like aren't tied to anything intrinsic, but at least they can sometimes be used as a predictor of quality.

Meanwhile the history of the field is one of judging programmers on bugs fixed, lines written, commit frequency, or a hundred other standards that are only perverse incentives. Most aren't even Goodhart's law issues, because they weren't predictive before they became targets!


ObDilbert: http://dilbert.com/strip/1995-11-13

"I'm gonna write me a new minivan this afternoon!"




Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: