As I’ve worked on my project with ImagePlot and game playthroughs in the past couple weeks, I’ve been reminded of the limitations of the archive of player experience that I’m constructing with this method. Specifically, with several games (SOMA and The Talos Principle in particular) I’ve struggled to find enough publicly accessible playthroughs to get even just the few I need to start a comparison of different players’ experiences. With a given game there may not be many playthroughs that have been posted to YouTube or other video-hosting sites, and this limits the amount of comparison that my ImagePlot method can accomplish.

Of course all archives have limitations of one sort or another, but it’s important to keep these limitations in mind as we use archives in our research. This is especially true if we use an archive to make a claim about a particular trend or meaning created in a particular culture, and how and to what extent we can back that claim up. With the playthroughs I’m using, I can start to assess variation in narrative in games, but I can’t make a claim about every player’s experience with the game in every situation everywhere. That doesn’t make the claim baseless or useless, but it does mean that that claim could change with more evidence. Or that the claim could be missing something important that the handful of playthroughs I’m using don’t represent.

It could also be easy to respond to these limitations with a call for more playthroughs and more evidence, and certainly a better sample never hurts in substantiating a claim. Yet it can also be misguided to waste valuable resources (time, money, etc.) on pursuing the perfect sample that will never exist. And this is where the balancing act of the “good enough” sample comes in. I think the archive of player experience I’m working towards here needs more to become that good enough sample, but also want to be mindful of how far I’m going in pursuit of it.