That's What He... is Sorry For

Two recent blog posts have called me on my bullshit and I owe everyone an apology.

First, Jessamyn Smith wrote about Fighting Sexist Jokes the Geeky Way, and then Katie Cunningham—whom, though we’ve never met in person, I consider a friend from the Django community and Twitter—wrote about being told to Lighten Up.

Neither of these were, as far as I know, directed at me, but I feel like they still effectively called me out on contributing to our culture the wrong way, with scottbot.

Reading Jessamyn’s post (go read it, I’ll wait) I couldn’t help but wonder if the original bot had been scottbot. And—selfish as this reaction may be—that didn’t make me feel good.

I’m not usually proactive about helping make open source more diverse and more friendly to diverse groups. I could make excuses but ultimately, even though it’s important to me, it’s never something I’ve put time into. But I’m not going to be proactive in making it worse, either.

Scottbot was an interesting programming question. “Can we make a bot that learns to make these jokes better?” I didn’t think about the broader implications. Not thinking about it is a big part of the privilege hegemonic groups have. I can’t go back in time and stop myself from writing it but I can recognize it, call it out, and take action.

• First, I’m sorry.
• Second, I’ve removed the scottbot repository from GitHub. I can’t, and wouldn’t, do anything about the existing forks, but I don’t have to keep distributing it. The code is so simple as to be really boring, and while the bot could be edited to say something else, why? There’s a limited set of use cases for a Bayesian response bot.
• Third, I removed the scottbot package from the npm registry. With the recent password reset stuff on the registry, it took me a little while to figure out how to log in again. But it’s gone now.

Scottbot was my copyright under the WTFPL, so people who have it can still distribute it, but I’d ask them to think about it, first, and think if they’re contributing to the right kind of culture but using or distributing the code.