(Not) Managing Software Developers — cover art: a quiet warm workshop at evening, a senior owl in shirtsleeves listening attentively to a small worried mouse engineer, a hand gently on the mouse's shoulder while three other small animal apprentices work peacefully at neighbouring desks.

2006 · Stevey's Blog Rants · Rant

“If you have true empathy for your engineers, they can forgive almost anything. Which is good, because you will make mistakes.”
— From (Not) Managing Software Developers, May 2006
Read the essay

© 2006 Steve Yegge. Originally published at Stevey's Blog Rants.

AI Notes

The headline promises tips, and Steve spends a fair chunk of the essay explaining why he mostly won't give any: cataloguing everything a tech manager can get wrong "would fill fat books," so what's on offer is three meta-tips — advice about finding your own mistakes. The good material lives in between. There's a taxonomy of suspect motives for wanting the job (mother hens who smother, process people whose process "can be used to disguise incompetence for an entire career," empire builders quietly lowering the hiring bar), a rude one-word label for the aspirant who mainly wants the pecking rights, and a blunt rule of self-assessment: if you don't know whether you're a bad manager, you're a bad manager, because that's the "start-state" for managers everywhere. And instead of the usual leadership-book fantasy, Steve aims the whole thing at the job most readers actually have: inheriting a roomful of miserable engineers on crufty legacy code, no new headcount, where leading anyone anywhere is strictly a long-term project.

Anyone arriving for actionable technique will leave hungry. Steve's method for acquiring the one trait he says everything else hinges on is to get a dog or cat ("preferably not a pit bull") and learn to love it, which is his way of conceding it can't be taught from a blog post. The concrete tips were deferred to a promised sequel, "How To Manage," that never materialized in the twenty years of writing since. What keeps the piece from being an anti-manager screed is that Steve puts himself in the lineup: after fifteen years of managing and "a hundred managers with a hundred different motivations" observed up close, every reason he has ever heard for wanting the job, his own included, now looks to him like a psychological problem waiting to land on some unlucky engineering team.

Related listings

  • 2006

    Good Agile, Bad Agile

    The other 2006 piece on engineering culture and the management overhead bolted onto it — written four months later from the inside of the company this essay is talking around. Read together: the diagnosis, then the field notes.

  • 2008

    Done, and Gets Things Smart

    Two years on, the hiring-side of the same conversation. (Not) Managing is about not destroying good engineers once you have them; Done, and Gets Things Smart is the bar to set when bringing them in.

  • 2023

    Death of the Junior Developer

    Seventeen years later, the same instinct on a different decade. Both are pieces about the cultural and structural choices a company makes about its engineers — the long arc of Steve's writing on engineering culture.