The following is a critique of the “Preserving Virtual Worlds” digital humanities project.
The Preserving Virtual Worlds project, modeled after other digitization works such as Google Books, seeks to archive past and present video games. According to the Executive Summary in their final report, the authors found that video game archiving brought a very specific set of challenges and media related questions .
The aims to preserve what is referred to as a “virtual world.” “Virtual worlds are software artifacts, communities, and commodities” . In practice, this means archiving the worlds created by video games so that in the future there exist records and playable versions for people to experience. In a game such as “Spacewar!”, preserving the world is as simple as keeping the code and hardware. For other games, the project has made attempts to capture the atmosphere of these games by compiling screenshots, descriptions from manuals or guidebooks, and videos of gameplay. For instance, the project website<> links to this source of DOOM videos and articles. The site provides records of what DOOM is like, as described by DOOM players. This preserves an aspect of the community surrounding the game, which the project views as a vital component of the “virtual world.” These archives may help people who cannot physically play these games to appreciate their virtual worlds. They may also help give an impression of certain players’ views of DOOM’s world to people who might find the game’s primitive graphics and interface alienating.
Most of the challenges faced by this project can be split into two categories – infrastructure and disputes over intellectual property rights. The greatest infrastructure issue stems from the wide range of hardware used to run video games. Some of the oldest games the project preserved – “Spacewar!”, for example – ran on punch-card computers that are no longer in production. As a result, in order to preserve it in its original form, the computer used must be maintained, along with the actual program. Many modern games designed to run on Windows also face issues. Not all games run on all graphics cards or processors, and as a result several different system configurations must be maintained. In addition, some games won’t work on newer operating systems, and those too must be maintained. The project has assembled guides for running now outdated computer programs on modern systems to allow more backward capability.
Moving forward, the project may face difficulties deciding how to preserve a game. Many games are released for multiple consoles, often involving very different methods of play. For instance, Call of Duty can be played on a computer with mouse and keyboard, or on an xbox with a controller. Do players across systems experience the game differently? The nature of a virtual world can also change across consoles. Players using PC’s can access the source code of the game and customize content. Should these exploits, or at least the potential for such exploits, be counted as part of a virtual world? The project has not made efforts to address this concern.
There are greater questions and challenges proposed by more interesting and widely played games such as World of Warcraft. World of Warcraft is a massively multiplayer online game, and as a result has millions of characters which make up its “virtual world.” Arguably, preserving a single copy of the game does not accurately capture its essence. Instead, the millions of characters must be preserved, and still it is nigh impossible to recreate the atmosphere of the game with the raw data.
Consider the game “Journey” for PS3, where the player’s character is shrouded and only utters musical notes throughout the entirety of the game. The gameplay is exploring a desert and unraveling a deep story to give context to the adventure. While this can be preserved by simply saving the data, the game also offers for other players to join in and help other players. The players have no communication, but are able to call to one another and assist each other in reaching otherwise unreachable places. This aspect of the game’s world may be lost in preservation, as it is difficult to recreate a community atmosphere and the anticipation of random encounters.
In terms of intellectual property, the authors found that it was a challenge to acquire proprietary software from companies. For example, Linden Lab, the maker of Second Life, was unwilling to hand over the code for their servers . Without the server code, any copy of the game is useless if Linden Lab decides to turn off the game. Much of that type of software is maintained and upgraded and is therefore hard to retrieve for preservation purposes. As a result, most of the games preserved are old enough that rights holders are no longer protective of the source material. For games that rely on multiplayer components, this is a crippling limitation. Those games rely on a server infrastructure being in place, and unless companies give up a server, they cannot be truly preserved.
Much of the authors’ focus surrounded the archival of the game “Second Life” . “Second Life” is a massively multiplayer online game where players live out a “second life” through a customizable game avatar. In many ways, the experience of playing “Second Life”] is defined not by the players individually, but by the community surrounding them. The game allows players to buy property, hold jobs, and in fact features an entire, functional economy. Some players design their avatars not as they personally view themselves, but as they want to be viewed by other players. In order to preserve this world, one must capture the relationships formed between players, as well as the social trends, which spread throughout the community.
Unfortunately for the authors, all server components developed for the game were closed source. In addition, due to the Terms of Service of the game, players retained intellectual property rights to the objects they created in the game. As a result, it was impossible for the authors to simply copy the data without explicit permission from all the residents of that section of the simulation. In the end, without access to server software, the authors were only able to archive metadata representative of a Second Life island along with metadata] from players who gave permission for their objects to be used. This metadata was simply a description of the features of the objects – the general shape and layout of the island as well as player objects.
This left the project unable to recreate the defining aspects of Second Life’s virtual world. The game features a physically large world that is populated and altered by the games millions of users as they customize the aspects of their avatars’ daily lives. Although the project has made great efforts to preserve the layout of several islands in great detail, this amounts to documenting the consequences of players’ decisions, such as customizations made to houses, without presenting what it was like to “live” in this community. Playing the game on one of these islands would be akin to wandering through a ghost town. The layout of the world has been archived, but the world has not been preserved.
Though the project has its limitations, it is based on an admirable premise. As media such as video games continue to develop in depth, they can produce both beautiful visuals and stories. They are often compared to interactive movies, adding extra depth and giving the player choice. It is important to preserve these artifacts like we do movies and books, and this project is a well-envisioned first step into the realm.
In the future, it would be interesting to see these archives of games made publicly available. Part of the benefit of digital preservation is redistribution. While some of the games are still protected by copyright and licenses, it would be great to see them made available to the public – much like the Library of Congress’ classic movie library.
 McDonough, J., Lowood, H., Kirschenbaum, M., Krauss, K., Reside, D., Donahue, R., & Phelps, A., et al. (2010b). Preserving virtual worlds: Final Report. National Digital Information Infrastructure Program.Washington, DC: Library of Congress
Co-Authored by Matt Hebert and Sai Cheemalapati