Managing mistakes
Originally a thread on Twitter.
Because of this, I'm watching @jlericson's career with some interest: he's moved into a team management role in his new company, and it is fascinating to watch as he works to bring this experience to bear on the problems that he faces - both within his new communites, and team.—Shog9
I'm gonna take this as an opportunity to chat about what it can be like to have total authority over a small part of someone else's life and how it's not nearly as hard as it might seem. Or if it seems like having this sort of authority is easy, how it's paradoxically impossible.
I used to tell moderators that there were only a handful of irreversible decisions they could make and they were all merges. Merging user accounts turned out to be so difficult to undo and so potentially harmful, we took that power away from mods (who can delete users and posts).
Once you see that decisions can be reversed, it's incredibly freeing. I'm ridiculously cautious by nature, but I learned to take action rather than over-analyzing everything. (I mean . . . I still over-analyze. I just do it after making a decision rather than before.)
Of course fixing bad decisions takes work. If I suspend an innocent person, it might take months of effort to restore trust. That's a powerful incentive not to screw up. The other impediment to fixing mistakes, of course, is failing to recognize them in the first place.
Stack Overflow has had a culture of arrogance. Obviously the community is seen as arrogant. But the company has (had?) it worse. "We only hire the best" is something a lot of companies claim. In Stack Overflow's case, that sentiment was true. I saw a lot of imposter syndrome.
There's no way I will ever work with as much talent as when I was with Stack Overflow. But it came at an enormous cost. Invariably we'd make mistakes and not learn from them. The less technical the work, the greater the odds the company assumed it knew best.
Now that I'm a manager of other employees, I can see how treacherous that could be. If I use the wrong syntax, a compiler will notify me within seconds. If I screw up in management, I might only find out after it really is too late to fix. People fear contradicting their boss.
Mixing "we only hire the best" with problems that are way trickier to solve than anything you might come across as a developer results in absolute terror of admitting fault. Apologies come in the "I'm sorry I hurt your [fragile] feelings" variety rather than "let's fix this".
Or, if there is an attempt to fix the problem, management expects everyone to "be a team player". Otherwise they can't maintain the illusion they know what they are doing. (Needless to say, this doesn't work so well when the problem is technical in nature.)
By the way, there's no functional difference between people who think they know what's best and people who don't know what's best but are afraid to let anyone know. Neither is capable of digging themselves out of their own hole. They'd rather let the world burn than admit fault.
There's a video I saw a few years ago that completely spoils "Tinker Tailor Soldier Spy", but hammers what I love about that movie and the book it was based on. George Smiley navigates the world of espionage with humility.
I'm making management mistakes pretty much constantly, but that's fine as long as I'm fixing them as fast as I can. In order to do that, I must be constantly alert to my own failings. I must be open to the idea that other people might know my job better than I do. (Sometimes.)
If you want to talk with me about community management schedule a meeting!