Interactive fiction is a young art form and so it should come as no surprise that there aren’t many examples of shared worlds. A shared world is a setting which is used in common by multiple authors over several different works, where the works together form one canon. In other mediums, this is a very common model: television shows often have different writers from episode to episode, and film series often have different directors. In print fiction there’s the Wild Cards and Thieves’ World shared settings and this model is the norm for the major comic book series. Interactive fiction’s visual cousin, the graphic adventure game has a shared world in the Adventure Game Studio Reality-on-the-Norm games.
What value is there in having a shared world? There are several possible answers. There are some ideas that are just too big for one person to fully explore, and are intriguing enough to warrant fuller exploration. Where a shared world is successful, the experience of the overall collection of works is greater than and irreducible to the sum of its parts. Allowing other people to share in a setting allows aspects that are of less interest to the original creators to be explored. Further, as in any shared endeavour, there is the joy of co-creation, of working with others to create something that no one person could achieve on their own.
While shared worlds are thin on the ground, interactive fiction has a rich history of successful collaboration. The Zork trilogy arose out of a vast collaborative mainframe game; there have been several paper-telephone style collaborations in the IF Whispers series; and a contribution model has lead to stories as deep as Alabaster or as exhaustively amusing as Pick up the Phone Booth and Aisle.
An illustration by Erik Temple for an upcoming release of Kerkerkruip
Recently, Kerkerkruip has emerged as a collaborative project based on continually developing a single game. The draw for Kerkerkruip bluntly isn’t the setting or characters, but its nature as a game. A roguelike in text, Kerkerkruip placed well in the 2011 IFComp but unlike entries, its development didn’t end there. Gijsbers kept adding items and monsters and opened up the development process to allow others to contribute ideas, code and art to the game. In this way, Kerkerkruip follows a pattern familiar among roguelikes which typically have development cycles lasting years, sometimes decades. The usual model for roguelikes is to have a single developer or small team that eke out a game over many years while a community of modders develops mods in tandem with the main development. Sometimes these modders take over developing the main game long after the original author has moved onto other projects. Where Kerkerkruip differs is that it has opened up its own development to others so that the game itself becomes a shared-world in continual development. Parallels can be drawn with the development of the original Zork, which was collaboratively developed over several years (though with less of an eye to internal consistency, or indeed, player experience).
As Kerkerkruip is a game before it is a story it is well suited to continual development — less so more narrative-based projects. From another story in the IFComp 2011 an alternative model can be seen. Andromeda Awakening placed 17th of 38, but Marco Innocenti’s vision didn’t end there. Midway through 2012 he launched a minicomp, The Andromeda Legacy™: The Expansion of a Shrinking Universe, inviting others to create stories in the same shared universe as Andromeda Awakening. Two stories were entered in the competition: Tree and Star by Paul Lee and Andromeda Dreaming by myself. These were followed up by Innocenti’s Andromeda Apocalypse which won the IFComp 2012, which among other things portends well for IF shared worlds.
Both works adopt the rich setting of Awakening as their starting point and take it in different directions. Andromeda Dreaming dove-tails its plot and time-line tightly with Awakening, but differs chiefly in its game-play style. While both stories involve the player enmeshed in a web of conspiracies, Awakening involves exploration and puzzles while Dreaming involves conversation and sleep. Tree and Star, in contrast, shares the same basic plot as Awakening (a man struggling against the odds to reveal a terrible truth) but transports this concept to a much earlier point in Andromeda history. Like Dreaming, perhaps in part due to the time constraints inherent in a competition, Tree and Star is more linear than Awakening, and yet it involves more character interaction. Essentially, both stories remain true to the spirit of the progenitor work while expanding the social and historical aspects of the universe. If more Andromeda stories are made, undoubtedly the likelihood of contradictions and weaker entries will grow, and like other series, there will eventually come a point where the setting is creatively exhausted.
In a way, the Andromeda Legacy competition acted like the publishing of a collection of shared world stories, though one without an editor. Though they had the same starting point, as luck would have it, the authors picked different time periods to focus on and the resulting stories are mutually consistent. It’s quite possible that had there been more entrants, the entries could have been inconsistent with one another, and issues of canonicity would emerge. As it was, Innocenti declared all the stories canonical.
Marco Innocenti has produced cover art for every work in the Andromeda universe
Even in a shared world where stories are developed sequentially, rather than in tandem, there can be issues of canon. In Reality-on-the-Norm, an eleven year cycle of adventure games sharing the same setting (as well as often the same sprites and backdrops), one game controversially killed him off the central character and subsequent games had to take a stance on whether that event had actually happened. Similarly, most major comic book universes, in focusing on shared characters rather than setting, are failed canons as each series inevitably undoes or retcons the developments of previous series.
Developments in a series can typically be split into two categories: canonical developments and fan fiction. Zork Zero by Steve Meretzky and Zorkian Stories 1: G.U.E. by Marshal Tenner Winter are examples of these two categories. The Zork series of games and their spin offs and pastiches are probably popular largely because the Zork games were so predominant in the formative years of the medium. If the big text adventure hit of the 80’s was an epic quest to rescue a princess from demonic turtles, then we’d probably still be seeing Turtle Quest spin-offs. Questions of taste aside, the Zork universe was rich and accessible enough to support sequels in a way that the other Infocom games didn’t, perhaps because the focus was on the locations rather than characters or story events.
A story can’t support a sequel if the most interesting potentialities inherent in the story have already been explored by that story. So a work will best support follow-ups only when it doesn’t exhaust its own narrative potential. This failure to exhaust may arise due the breadth of the work’s setting, or its vagueness: Tolkienesque fantasy is so rife because the lore Tolkien created was so ripe with potential that even his own mammoth offerings could never exhaust the potential of the setting; Zork and Reality-on-the-Norm in contrast have settings so non-committal that almost any content is appropriate so long as it shares a family resemblance to the games that have come before it.
Interactive fiction is still young and in the years to come we’re sure to see more shared worlds arise. The most successful model is likely to be the successive-work model, like the later Zorkian games. The collection model of organising a setting beforehand and editing a collection of stories in that setting is possible, and is most common in print medium, but requires a greater burden from its organisers. The Andromeda Legacy method of encouraging new works based on an existing work is equally promising as it is based on existing interest in the setting, but issues of collective consistency are likely to arise. However as a successful example of an interactive shared world, with it interest for other shared worlds can only flourish from here on.