"Chance (re)Collections: Twine Games and Preservation on the Internet"
by Tobias I. Paul | Xchanges 15.2, Fall 2020
Contents
Methods
I began my research with these curiosities in mind, seeking out Twine games that no longer existed where they once had. I perused articles, lists, and forums, following links to games and seeking out ones that were more difficult to track down. In the process, I amassed a spreadsheet of 20 games that I cataloged alongside the creator name(s), original year of publication, means of playable access (original sites of upload, the WM, third-party download, etc.), and the most recent year the game was available and/or functionally archived, if applicable.2
In this analysis, I draw on that set of 20 texts, a sample of Twine games—by no means complete—across the spectrum of style, recency, popularity, subject, and beyond. My intent in creating this sample was not to capture a representation of overall trends across all Twine creations, but rather to examine trends within the subset of games that exist in more complex states of availability. I began by looking at various aggregations of well-known/well-received Twine games, as texts that have had some kind of secondary coverage offer unique forms of partial preservation. I also aimed to include a variety of prominent Twine creators as well as a number of lesser-known ones. All of the texts are individual Twine games that have been shared and, to varying degrees, preserved online.
Overview
There are three basic categories of availability within this dataset. A Twine game might be entirely here—that is, available on the site to which it was first uploaded and playable in its original, complete form—or entirely gone—unavailable online and unplayable anywhere else—or it might be somewhere in between. Of my 20-game dataset, three games are here and three are gone. That leaves 14 games in between.
This gray area of betweenness, however, accounts for a broad stretch of the overall continuum. Five games, for instance, just moved host sites and suffered link rot, like Maddox Pratt’s Anhedonia. At the height of its popularity in 2013 and 2014, media coverage linked to the game via the website of merritt k, where Anhedonia was initially hosted. However, in 2015 Pratt moved the game to their itch.io profile—possibly in relation to merritt k’s website overhaul, possibly due to Pratt’s shift away from Twine in their creative endeavors, possibly for reasons unknown to outside observers. Now these links no longer work; an individual who did not search further than that first link might believe that the game was taken down entirely. A Google search, however, would reveal its presence on the new site.
On the opposite end of this continuum, other games are just on the cusp of being fully gone. Of the 14, four can only be accessed through one webpage: Leon Arnott’s personal Twine archive. Arnott preserved 1,638 Twine games up until 2014, offering links to their host sites but also providing a mass download option for a file (a spool, as he calls it) containing all of the games in the collection. What this means, however, is that if his website is ever taken down or otherwise made unavailable, there is no recovering these four games—and possibly many others. Of course, if someone has downloaded the spool, it lives on through their computer, to share as they see fit. In this way, the preservation of Twine games is often left to incidental benevolence and chance.
One degree away, there are three games which are preserved on both Arnott’s site and on the WM’s archive. While the WM is a valuable resource, it also relies on the goodwill of content creators to leave the archives up; anyone can request that an archived page be taken down or excluded from crawls. If that were to happen with any of these games, we would once again have to rely on Arnott’s page alone. The remaining two games exist in similarly precarious spaces, with one preserved in the Electronic Literature Organization’s collection as well as on Arnott’s site, and the other shared across a forum as well as through Arnott and the WM. On the whole, though, the majority of these games rely heavily on the latter two sources for preservation.
In the following sections, I discuss a number of specific games from my dataset, analyzing the degrees of preservation the games experience, the role of anxiety as a catalyst for preservation efforts, the after-effects of incomplete preservation, and the complexity of Twine games’ ephemerality. In understanding the implications of preservation for marginalized voices, these artifacts offer insight into the consequences of digital impermanence.
[2] See Appendix A for Primary Dataset and Appendix B for Ludography.