"Why software engineering performance is hard to measure" shows a sample of dimensions we might assess developers on. It has some sample of dimensions we might perhaps use, just a starting place really, but not bad: high standards, delivery speed, reliability, team/company focus, open minded, & critical thinking. Ask yourself how many of these are easy to find good clear clean cut objective metrics we can assess people on.
https://betterprogramming.pub/software-engineering-performan...
And my god, apennwarr's recent System designs 2: What we hope we know is such a wonderful lovely intimate post/ramble touching in a number of places around how we have to embrace real complexity. Science loves control & hard numbers, and in this quest it resists seeing things as they are, as an accrual of many factors: "scientists absolutely bloody despise emergent complexity → systems design → magical thinking."
Management like scientists wants nice neat objective tools to understand the world with, without ever having to deal with emergent complexity. Most performance metrics & OKRs end up being dehumanizing, shallow, low-context. (That said, I think they can be helpful for high-context folks to better understand situations near to them & how things change over time.)
Energized motivated developers do such a better job. Caring prevents so many problems. Having the ability to do things well should keep paying off, are the first & often online line of defense, the only ward to keep the tech-base from becoming mired & stagnate.
2 comments
[ 5.0 ms ] story [ 13.8 ms ] thread"Why software engineering performance is hard to measure" shows a sample of dimensions we might assess developers on. It has some sample of dimensions we might perhaps use, just a starting place really, but not bad: high standards, delivery speed, reliability, team/company focus, open minded, & critical thinking. Ask yourself how many of these are easy to find good clear clean cut objective metrics we can assess people on. https://betterprogramming.pub/software-engineering-performan...
And my god, apennwarr's recent System designs 2: What we hope we know is such a wonderful lovely intimate post/ramble touching in a number of places around how we have to embrace real complexity. Science loves control & hard numbers, and in this quest it resists seeing things as they are, as an accrual of many factors: "scientists absolutely bloody despise emergent complexity → systems design → magical thinking."
Incredible post for a number of great reasons; this is just one. https://apenwarr.ca/log/20230415 https://news.ycombinator.com/item?id=35580776 (180 points, 9d, 27 comments)
Management like scientists wants nice neat objective tools to understand the world with, without ever having to deal with emergent complexity. Most performance metrics & OKRs end up being dehumanizing, shallow, low-context. (That said, I think they can be helpful for high-context folks to better understand situations near to them & how things change over time.)
Energized motivated developers do such a better job. Caring prevents so many problems. Having the ability to do things well should keep paying off, are the first & often online line of defense, the only ward to keep the tech-base from becoming mired & stagnate.