When depression knocks the door.
While trawling one of my favourite repos - exa, I stumbled upon something I never would have expected. A heartful message from its author when he wasn’t at his best.
This is as real as it gets. Unfortunately nobody knows about it, because we don’t talk about it.
exa v0.4.1 (basically a pre-release v0.5.0) — Mar 29, 2017
This is a snapshot of exa’s current code, which has been cleaned up a bit and at the very least been run through untry (which managed to pick apart all try!s but one). It’s being released to work around a bug in macOS Sierra.
Usually when I do a release I go through the commit history, pick out anything that looks like it could be a bullet point in the release notes, look at the contributors and see who’s there, then edit the website and upload — there are more features in here than just the openssl upgrade, and the work done so far is really deserving of a “proper” release, rather than just a beta.
But I don’t even want to learn how little work I’ve done on it in the past twelve months.
I think I need to be honest with everyone here. My interest in exa is still there, and the list of features I’d still like to add is as long as it’s ever been, but my motivation is completely shot. Compared to even a year ago it’s plummeted. Every day I go to work, bang out code on whatever it is I’m doing, leave, arrive home, then do my best to not do anything, on the computer or off it. My life’s going alright, I can still do things, I’m able to do things, but I just… don’t, and I wish I could give you a better reason than that. GitHub notifications were something I’d deal with later that day, then next week, then whenever I’d get around to it, and then routed into a folder in my inbox that I never actually checked, and then every day I’d see that number and hope it didn’t increase, and then I noticed that whenever it did increase I wouldn’t be able to concentrate on anything for another hour, so I just hid it, which hurt at first but at least gave me some time back. I’d still commit from time to time, but not mention any of this because I didn’t want to call attention to it, as though in time everyone would just forget. Whenever I saw a reply to the Sierra bug thread, I’d log in, read it, start to worry that if I replied now then I’d be admitting to everyone that I was still around and hadn’t died or something and was instead just ignoring their comments, and then what would people think of me? I even got as far as navigating to “Delete this repository” before I had to push the computer away.
That’s one push that went correctly. It’s obvious to say, but it’s stupid to delete code that people are using, just because you didn’t want to deal with the fact that you made a mistake.
I’m going to try something new: switching exa to a routine 6-week release schedule. The Rust has a release train that works over the same duration using multiple versions. I think exa (a hundredth its size) can do without that part, but the rest of it is good advice. Having an actual date for each one means I don’t get a choice in releasing anything or not, which is an incredible motivator for actually doing it! I still have to put 0.5.0 out properly, but with luck, the second release (0.6.0, which would come out halfway through May) would be a lot less daunting than the first, the third less than the second, and it would plateau from there.
The root advice here is “if you’re having difficulty doing something, then do it more often, because it gets easier”. I found actually releasing my software hard. I want it to be easier.
I hope you all find this acceptable, and I apologise again for the massive commit lacuna.