Pycon was awesome this year. I learned a ton of stuff and got to wander around a foreign city at 2 in the morning with people I just met.
The biggest thing to come back with my from Pycon (that wasn’t some new piece of tech) is a renewed attention on culture. Like most nerds I tend to focus on the most tangible things: workflows, tooling, metrics, science. Things that are easy to “fix”.
All that stuff is super-important, to be sure: give your text editor/IDE a way to notice bugs, and you’ll have that many fewer make it into production. That’s boring, simple stuff: better feedback in most human endeavors makes for a better experience.
The hard part, the thing that can’t be easily tweaked or tuned by running an installer, is culture.
So for instance: our company culture doesn’t value testing. It doesn’t value IT. It doesn’t value experimentation; we basically either ship our prototypes or consign huge amounts of work to the dustbin. It is relentlessly negative (and arguably I’m one of the worst proponents!). We have become cynical about re-writing everything from scratch every 18 months, such that I think in the back of our minds, no one really values their work: it’ll all be rewritten anyway. We retreat to our caves and work alone, on the “silo” of code we know best: there is hardly any collaboration in design or development. Knowledge sharing is an afterthought.
It is not to say that we are any more broken than any other technology company out there. Most of them are like this. Honestly, do you think anyone at Computer Associates is insanely happy to be at work every day? Our culture is at least fixable and our work matters.
(Sidebar: At least I presume our culture is fixable. We are young, smart, competent, and thus ostensibly can be motivated to fix what are obvious problems, or at least are obvious to me. Moreover as a relatively “flat” organization, there is no one person who can stop change simply by saying “no”. One person can say “no” and halt progress but if and when others buy in, peer pressure takes over. Orgs with strict hierarchy simply cannot change except from the top.)
So that’s increasingly on my mind after Pycon as this year’s “project”: can our company culture be fixed? First among the challenges is fighting my own negativity, which screams “nopenopenope get out while you still can”. We’ve tried before in various ways, and a combination of apathy and the need to move product seemed to kill it before it went anywhere.
Also Docker was really cool.