I took part in the TV Tropes Game Jam last week, and used it as an opportunity to try out my idea of livestreaming myself as I developed a game. A few folks, like the organizer, checked out the videos, but I didn’t really have anyone watching them. That’s fine, though, since it gave me an opportunity to get comfortable in front of the camera and still be able to make mistakes.
Thanks to this exercise, I’ve gotten a lot more comfortable with my own voice, figuring out how to set up my posture and projection. I’m using a cheap microphone, so I need to crank the gain up to 300% before I’m audible, but that seems to work just fine. I’m mostly intelligible except for the instances where I get really excited and start running my words together. The problem with me isn’t really so much in how quickly I speak, but the fact that I have a tendency to try to start saying a new word before the current one is finished. I’ve been doing some exercises where I read passages aloud with the intention of enunciating each syllable before moving to the next one. There’s a psychological challenge to this, since it feels like you’re going slowly, but in the same way that it feels like you’re going slowly when you drop from going 65 on the highway to 55. I’m only speaking slowly in my own perception of myself, not in anyone else’s.
As for the content of the video, I had the opportunity to demonstrate an authentic programming project and was very happy that problems cropped up and where swiftly dealt with. The first video was an excellent bootstrapping video that showed off how to start a Python and Curses project, but in the second one, I was plagued by a simple bug that took up 15 minutes of video.
The thing is, that’s exactly what I was hoping for. Part of the reason for doing the livestreaming is to capture myself during these moments of debugging so I can show everyone how I recognize a bug, how I figure out where it is, and how I go in and solve it, even when it’s something embarrassing. While I’m unhappy that it takes up valuable time from recording (especially when I limit myself to just an hour each night), I’m happy that I’m able to find the bug, fix it, and offer a recap in how I went about it. I imagine that if I had an active audience, just like in the classroom, there would be some bright viewers who can see and point out the bugs so that I can solve them. If that was the case, spotting bugs and fixing bugs would be much faster, and I would still have the ability to bring up where the problem was. I do wonder if that would take from the authenticity of the experiment, but it’s not really fun to watch someone spinning their wheels.
It was an interesting experience having to program the game in hour-long spurts. I was only given the theme the night before, and while I had some ideas for what I wanted to do, my ideas were very big compared to the coding I would actually do. I think that making each video a self-contained “goal” was very healthy for the game development process, since it stopped me from getting distracted and doing other things (sometimes). When I was off-camera, I did the less glamorous parts of development, such as refactoring and commenting (thought to be fair, that also belongs in a video, just not when I only have a week). One thing I wish I did was come up with what was going to go in the videos earlier. On the fifth video, I was very unhappy with how much I tried to “wing it”, often having no idea what I was going to do. The code had become a mess, and I didn’t feel like refactoring it for a game jam. I finished it, made it done, and moved on.
Going forward, I think I’d like to try this again on a longer time scale. Maybe start developing a game in January and recording the development process. I have a few games I’d like to try to make, and even if I lose interest in the game itself, I would hope that the video would motivate me to do a good job regardless. I want to think that the reason for losing interest in the game jam was due to the timeline, deadline, and Thanksgiving. The only way to find out is to make a bigger game and see how well it works.
All of the videos I recorded are available on my Twitch.tv channel as well as on my Youtube Channel. If you decide to take a look, let me know what you think! The quality for the Youtube videos is pretty terrible, but Twitch is able to broadcast at 1:1 pixel resolution, which I think is pretty important. Since I’m not actually playing any games, I just stream at 5-10 FPS, which is enough to cover the miniscule changes in code that happen.
The completed game can be found on Github.