On September 17, Rockstar Games released Grand Theft Auto (GTA) V , the latest entry in the studio’s 16-year-old franchise of sprawling, open-ended, and controversial video games. It was promptly denounced by human rights organization Freedom from Torture, called “corrosive” by England’s deputy prime minister, and praised as “the most immersive spectacle in interactive entertainment” by the New York Times. Within three days, it sold over $1 billion worth of copies worldwide, making it the largest entertainment launch—of any type—in history, according to Forbes.
Whatever one’s position on the merits or faults of GTA V’s profane, violent caricature of modern America, its popularity alone basically cements the game’s status as an influential cultural artifact of the early 21st century. With expected sales of 25 million copies, it is almost difficult to imagine that in as few as ten years, it will be impossible for historians to study the game in its original form. Its multiplayer online mode and other online features will stop working as soon as mainstream popularity inevitably dissipates and Rockstar decides to shut down the game’s servers.
Within 50 years, the offline version of the game could become completely inaccessible as well, barring significant, ongoing institutional or corporate efforts to preserve it. Simply put, video games are already raising many of the thorniest challenges in the field of digital preservation, and as games continue to become more complex, those challenges are rapidly compounding.
“In some ways, video games are the canary in the coal mine for our whole digital world,” says Jon-Paul Dyson, director of the International Center for the History of Electronic Games (ICHEG), National Museum of Play at the Strong in Rochester, NY. In this sense, studying ways to preserve games has a broader importance to the field of digital preservation in general, Dyson adds.
Henry Lowood, curator for the history of science and technology collections and film and media collections in the Stanford University Libraries, agrees. “Digital games, as software, are a form of multimedia, and almost every other [digital] media you can imagine is embedded in games—there are movies, music, there’s text here and there, animation,” he says. “There are all kinds of layers to the software, relating to other kinds of software. It’s a very complex multimedia object. So, one argument can be made for tackling digital game preservation is—to put it bluntly—if you can solve that problem, you can probably solve any other digital preservation problem. It’s that complex.”
Preserving virtual worlds
One of the most extensive recent efforts to analyze these challenges has been the Preserving Virtual Worlds (PVW) project, a collaborative research venture of the Rochester Institute of Technology, Stanford, the University of Maryland, the University of Illinois at Urbana-Champaign (UIUC), and digital entertainment developer Linden Lab, best known for the massively multiplayer online (MMO) game Second Life. The project was conducted as part of Preserving Creative America, an initiative of the National Digital Information Infrastructure and Preservation Program at the Library of Congress.
Video games “are economically, socially, and culturally important,” says Jerome McDonough, assistant professor at the Graduate School of Library and Information Science at UIUC and principal investigator for PVW II, a continuation of the original project funded by a 2010 grant of $785,000 from the Institute of Museum and Library Services (IMLS). McDonough doesn’t regularly play video games himself, but, like Dyson and Lowood, he views the study of game preservation as having a broader importance to the field of digital preservation and argues that their cultural impact has become difficult to overlook.
“When you see things like Call of Duty 2 raising more money in the first weekend of its release than just about any major motion picture that’s come out in the past decade, this is a significant cultural and economic force,” McDonough says. “And you are starting to see more and more people who are involved with gaming to one degree or another, particularly digital games. This goes across all age groups, genders…. It’s important to preserve knowledge of these games and their evolution so that people can understand their place in the culture at this time.”
When preserving any type of digital object, key considerations include the stability and viability of the storage media itself. Even when kept in ideal conditions, CDs, DVDs, hard drives, flash drives, or Nintendo cartridges, for that matter, will all become corrupted eventually. The hardware used to read these media will eventually become obsolete. The software used to access digital files can become obsolete as well, so files must be converted to new formats and migrated.
“Preservation is not a single activity. It’s not a thing you do once and it’s done,” says Lowood, who has also been involved with the PVW project. “The term that’s used a lot in the digital preservation community to capture that is the idea of a life cycle. Preservation involves a life cycle—continuous activity, some of which occurs at a moment in time, some of which occurs several times. There are different things that have to be taken care of over the course of the lifetime of an object that might involve several human lifetimes.”
Migration to new media and new formats, for any collection of files, requires auditing to ensure that data is still viable, Lowood adds. Then there are “a whole slew of access issues,” some of which involve hardware and software and others that are not technological, such as copyright and access restrictions.
“All of these things involve different kinds of activities. Some are extraction of software, some have to do with providing access to that software, some might have to do with writing up a legal agreement,” says Lowood.
So, for example, in 2013 there will be plenty of headaches involved when migrating a collection of WordPerfect 7 files saved in 1999 on a stack of 3.5″ floppy disks. But consider the potential copyright issues that might arise 30 years from now with a game like GTA V, which includes a sound track of 240 songs licensed from dozens of different record labels, as well as software components licensed from several third-party developers.
“Any complex modern game doesn’t have just one rights-holder. It has a bunch of rights-holders,” says McDonough, adding that preservationists are already dealing with a major orphan works problem with games from the 1970s, 1980s, and 1990s. “Microsoft does not own the rights to the music that appears in [popular sf shooter series] Halo. It licenses that from the orchestra that created it. So when you’re talking about getting rights on a modern game, in particular, there’s a whole bunch of agencies that you may have to talk to.”
Complicating matters further, many video games are designed for proprietary hardware systems, so preservation must involve a platform—such as an Atari 2600 or Playstation 3—as well as individual games. As the final report for PVW II explains, “The most obvious problem affecting these materials is the obsolescence of the hardware and software infrastructures necessary to allow software to run.”
The report notes that Spacewar!, a 1962 space combat simulation developed at the Massachusetts Institute of Technology, “currently exists in its original form stored on a punched paper tape intended to be read into the memory of a PDP-1 computer.” There is only one known functioning PDP-1 computer left in the world, the report adds, and paper tape readers have functionally vanished as well.
“The fate of the paper tape of Spacewar! is the fate awaiting all games without the active intervention of preservationists,” the report states. “A book may pass 50 years on a shelf and still be readily accessible; rapid technological change and the resulting obsolescence of the technology necessary to access software mean that a computer game will not.”
There are work-arounds available for transferring content to a neutral format. “ROM dumping” devices can copy software from old video game cartridges, while disk imaging technologies can help duplicate more recent games published on CDs, DVDs, or Blu-ray discs. Also, software emulators have been reverse-engineered to mimic many consoles or other legacy hardware, allowing games to be migrated and stored in a playable format on newer hardware, such as a modern PC.
Yet as the PVW II report notes, many of these emulators were developed by the gaming community and “are of questionable legality.” The act of imaging a disk that includes copy protection features may be a violation of the Digital Millennium Copyright Act (DMCA). And regardless of their legal status, emulators often present an imperfect preservation solution.
“One of the things that we did during PVW was look at how good of a job emulation and virtual machine technology do at reproducing the original game experience,” McDonough says. “The answer we came up with is that there’s a tremendous amount of variation. Determining the best virtualization platform to use for a particular game is very time-consuming and involves a lot of experimentation.”
One of the games studied by both PVW I and PVW II was Doom, a 1993 title by id Software that is notable for its groundbreaking game engine design, as well as its formative influence on the first-person shooter genre.
“We looked at about 30 different combinations of DOS operating systems and virtualization and emulation software to see if we could get that game to run properly,” McDonough says. “And most of them reproduced the game sort of OK, but only two of them actually got the sound right.” The original version of Doom was published before the consolidation of sound drivers under DirectX and Windows, he explains.
The preservation challenges for console or old PC games pale in comparison to the issues faced by people interested in preserving MMOs such as Second Life or Blizzard Entertainment’s World of Warcraft.
“We can get a Commodore 64 game, and if we preserve the machine, and preserve the disk, and the information is still on there, we can still play that game,” Dyson explains. “However, a game that relied on an online server that a company has shut down is, in fact, unplayable.”
Games like these simply cease to exist in a playable form once they are no longer commercially viable. Of course, it would be theoretically possible to store everything on a game’s servers and then power them back up half a century from now, but it wouldn’t be possible to study the in-game experience.
“For Second Life, the world itself consists of a database that documents what [virtual, personal] property and objects are where in the world, what texture files are associated with them, what scripts are associated with them, and who owns what,” McDonough explains. “There’s nothing there about user activity. If I preserve that database, I’ve preserved the neutron bomb version of Second Life,” meaning that much of this virtual world might still be intact, but it will have no inhabitants, no record of past activities, and no record of how the game was played.
In yet another wrinkle, many users do document and discuss their MMO activities, generating content that could help future historians understand the dynamics of a game that no longer exists, but these players generally do so in Internet discussion forums and blogs—formats that are also inherently difficult to preserve. Other contemporary information about games is preserved in consumer magazines targeted at gamers, but few universities or institutions had the interest or foresight to build collections of these titles.
“We need to save more than the games themselves,” says Raiford Guins, associate professor of culture and technology in the Department of Cultural Analysis and Theory at the State University of New York at Stony Brook and curator of the university’s William A. Higinbotham game studies collection. “Just putting a lot of game consoles into computer history museums—I value those artifacts, but they only say so much in terms of how we remember or understand or study those games. You could have a working version, 50 years from now, of World of Warcraft, but it’s going to be kind of lonely playing that, unless you have documentation of the social experience of playing that game.”
A community of preservationists has been grappling with these problems for several years, participating in projects such as PVW and building collections in technology museums. ICHEG alone has a collection of 50,000 video games, consoles, and related materials, most of which are kept in working order. Private donations, combined with a $113,000 conservation grant from IMLS in 2011, have enabled ICHEG to test the functionality of a growing number of these games. During these tests, ICHEG staff members are capturing ten to 20 minutes of gameplay video for each title—another strategy to ensure that footage remains, even if a game later ceases to function.
“One of the reasons the video capture is very important is that it allows us to at least get a record of this,” Dyson says. “It’s almost as if you have video of a baseball game from 1947. You can’t replay that game again…. But video capture allows us to keep a record of these very ephemeral digital experiences that otherwise we might not be able to preserve.”
At Stanford, Lowood works with the Stephen M. Cabrinety Collection in the History of Microcomputing, a collection of 12,000 to 15,000 software titles and dozens of consoles, many of which were preserved in mint condition when the family of Cabrinety, a former student, donated the material to Stanford in the late 1990s. In partnership with the National Institute of Standards and Technology’s National Software Reference Library, Stanford is currently cataloging, disk imaging, and digitizing a large portion of the collection.
Lowood has also written extensively on the topic of machinima—cinematic productions created within a video game, such as the long-running, Halo-inspired YouTube comedy series Red vs. Blue. Lowood currently curates a collection of machinima that is preserved by the Internet Archive and sometimes works with students proactively to record significant events within MMOs, such as capturing footage of the permanent shutdown of EA-Land, aka The Sims Online, in 2008.
“The closing event of the virtual world was really quite interesting,” he says. “People were exchanging contacts, talking about what the world meant to them in these last moments as the ship is sinking, basically. What is it like to be there when a virtual world that has been in existence for years—that hundreds of thousands of people were involved with—just ends?… It’s an error message. The world ends with an error message.”
A growing number of software developers have begun to embrace the use of machinima and the social sharing of game footage, with many companies retroactively changing their terms of service to allow the recording and distribution of creative content produced within their games. Many newer games, such as GTA V, include provisions in their licensing agreements enabling users to post machinima or simpler in-game footage on social media sites or YouTube, provided the footage does not include spoilers.
As well, some game developers are beginning to warm to the preservation movement itself. Notably, Linden Lab participated in PVW directly, and id Software gave researchers extensive access to the source code used to create Doom. A budding retro gaming movement could also inspire these corporations to invest more of their own efforts in ongoing preservation.
“There are positive signs,” McDonough says. “The International Game Developers Association now includes a preservation special interest group. You’re starting to see more awareness of these issues. And you’re also starting to see a growing awareness among game companies that having materials preserved can mean additional money in the long run. Nintendo, in particular, has made a fair amount of money over the past few years digging into its archive of past games and resurrecting them for access online through the Wii [console]…. But [the company] discovered that [it, itself], had not preserved these games and had to rewrite many of them from scratch.”
Digitization and EMPs
It’s not just video games that present preservation challenges; under very extreme conditions, all digital content could be at risk. It may sound like a scenario from a war game simulation, but if a nuclear weapon were detonated 200 miles above the center of the continental United States, the resulting electromagnetic pulse (EMP) would destroy almost all electronic devices and electrical transformers nationwide, in addition to wiping electronic storage systems clean. Citing sources including a 2010 analysis by U.S. Air Force major Colin R. Miller, Pratt Institute student James Bradley explored this topic in a paper for an LIS course taught by LJ’s John N. Berry III. The following are excerpts from that paper:
“Miller sees EMP weapons as ‘one of the most likely and potentially devastating’ threats facing our country in the near future. Of course, one may ask if it is truly the archivist’s place, along with the military strategist, to entertain such grim, perhaps paranoid-seeming speculations?… The case of EMPs may seem extreme, but it is possible…and it is a possibility that certainly affects information professionals at the heart of their domain.
“The next logical question to ask is: What is to be done?… According to Miller, there are two basic ways to protect electronic equipment from an EMP. One is to shield the environment in which the equipment is located, and the other is to shield individual circuits. The former method involves the integration of a metallic shield into a building’s structure, a kind of ‘retrofitting’ similar, in some ways, to that which is done in old buildings in earthquake-prone zones…[at] between one percent and five percent of a building’s total cost…. While digitization offers many wonderful benefits for the future of libraries, archives, and other information depositories, it also comes equipped with its own built-in Achilles’ heel.”—James Bradley