Though we’re well into the 21st century, it’s only been in the last few years that digital art has moved into the mainstream—things like David Hockney’s iPad art exhibition, or the GIF which recently sold for $1,300.
But while these latest digital artworks are all the rage, their predecessors—the first digital works from the '60s, '70s, and '80s—are dying. They’re dying because the antiquated hardware used to interpret the essence of the artwork—its code—is breaking down from old age and obsolescence. Once that happens, the artwork is gone forever.
Determined to reverse the degradation, one man and his team of digital art conservators are working hard to prolong the inevitable and trying to ensure that the greatest masterpieces of digital art don’t disappear like so many Snapchat pics.
"Texts, images, films, music, works of art, documents of all kinds—indeed, virtually all variants of human expression—are today generated and stored in binary code; a code that can no longer be read directly by the human senses, but only by machines which then render it in a form that is comprehensible to people."
That’s what Dr. Bernhard Serexhe, chief curator and head of collections at the ZKM Media Museum in Karlsruhe, Germany tells me when I sit down with him to discuss the upcoming publication of digital art conservation, the results of his three-year research project for strategies on conserving artworks created with computer programming and digital technologies.
The texts, images, films, music, and documents Serexhe mentions can, of course, be summed up in a single phrase: humanity’s cultural record. As Serexhe notes, it’s a cultural memory which has radically changed recorded form in the last several decades, going from the analog clay tablets, papers, canvases, and chemically processed photographs, to the digital form that is now the backbone of everything from our entertainment to defense to commerce industries.
The move to a digital cultural memory in itself is not a bad thing. After all, information and knowledge spread easier and faster because of digital technology. But while digital technologies are a boon for consuming, analyzing, and sharing copies of aspects of our cultural memory, it presents a problem for the art world where it is the original creation that matters.
Once that medium that the digital art lives on can no longer be powered on or its code can be no longer read because of outdated technology, we lose a piece of our cultural memory forever. And that’s happening, right now. "There are uncountable digital art works being lost every year due to the obsolescence of hardware and software," Serexhe says.
Given no physical damage comes to it—no fires, or floods, or collapsed ceilings—the Mona Lisa could survive for hundreds of more years in the Louvre without much intervention from human hands. Its paint may fade a little, but it can always be restored. And with constant, active care, the original Mona Lisa, in all its analog glory, will last for thousands of years more.
But while the Mona Lisa will have a life-span any human being would be envious of, the same cannot be said for digital artworks, which typically have a life-span less than that of your average house cat.
"The error that the digital is eternal persists until today," Serexhe says. We assume that whatever we put in the cloud stays there. But if digital art has existed since the 1960s—when German artists began using massive mainframe computers to translate algorithms into computer graphics drawn by the Graphomat ZUSE 64, an early digital printing machine—then why has no one focused on the unique complexities of digital conservation?
"The general overestimation in the 1990s of how durable digital technologies are together with the simultaneous rapid progress of technical obsolescence has resulted in a massive backlog of badly conserved digital artworks, and works that have not been conserved at all," he explains.
"Often it is only when the works are to be presented again that data loss and incompatibility with current hardware and software are discovered."
In other words, the conservation of digital art is only now becoming a big focus because museums are pulling this stuff out of their storerooms for exhibitions, turning them on, and then realizing a million-dollar piece of art can’t "run" anymore.
This is not something that would have ever happened if they had pulled an old oil on canvas out of storage. The irony of digital art’s technologically superior medium lasting a fraction of the time (often less than a decade) of its analog fragile-clay-pots and oil-on-canvas cousins is something frequently mentioned in traditional art conservation circles.
But Serexhe argues such a viewpoint is a fallacy. "It is not the digital artworks themselves that age more rapidly, as is often maintained, because their real substance lies in the digital code, which cannot age."
"The problems of digital art conservation stem from the hardware and software required for maintenance and presentation, the operating systems and specific applications, and the requisite knowledge of the programs, which are no longer available after a few years," Serexhe says. "What this means for digital media art is that the faster the technical developments, the shorter the work’s half-life."
The transience of technology means systems that were the latest and greatest just a few short years ago quickly become technologically obsolete, leading to a "systems-eminent" threat to art that was created on these platforms.
And it’s not something that is only affecting programing languages and systems that are decades old, Serexhe tells me. "The explanation for no longer being able to start certain artworks is as banal as it is shocking. Let me give an example from the Apple world: With the introduction of Mac OS X 10.4 Tiger that ran on Apple computers with Intel processors, Apple no longer guaranteed downward compatibility with programs requiring the so-called ‘Classic Environment,’ which included the widely used software Adobe Director."
The result of Apple’s abandonment of the Classic Environment meant a number of digital works of art written in Director on PowerPC Macs will eventually not be able to run any longer because someday no machines will exist that are capable of running and interpreting that PowerPC code for human eyes.
Serexhe argues that hardware and software companies like Apple generally do not care about the preservation of cultural heritage. This is due to the market the companies operate in: an economic system, upon the interests of which cultural developments are substantially dependent, that does not mediate between the interests of the different social actors but rather in accordance with the principle of consumerism, which constitutes the basis of our society and propels both producers and consumers into an increasingly virulent vortex of consuming and replacing.
When a museum realizes a piece of digital art they have no longer works, they usually turn to Dr. Serexhe and his team, who are made up of hardware and software engineers and whose conservation work can best be likened to a mix of clever coding and searching for scraps.
For digital works that require specific hardware that stopped production decades ago, Serexhe’s hardware engineering team literally scours eBay looking for old, usually broken, units to buy and repair, which they then use to swap out with the original machine the artist used. However, finding and repairing old computers from eBay is only a half measure when it comes to digital art conservation—as exampled by Serexhe’s team’s work to save The Legible City.
The Legible City (1989/91) by Jeffrey Shaw is widely considered to be one of the most important works of digital art. It’s made up of a stationary bicycle, which the viewer rides. The bicycle is connected to an analog-to-digital converter, which is in turn connected to a Silicon Graphics Indigo2 (IRIX) computer that processes code the artist wrote. The code turns the digital signal from the bicycle into giant 3-D letters that form words that are projected onto a screen as you cycle through various cities.
In order to save this work of art from virtual death, Serexhe’s hardware engineering team worked for over a year to reconstruct the broken analog-to-digital converter and then swapped out the original Silicon Graphics Indigo2 (IRIX) with an identical one they had found on eBay and loaded the artist’s code—which can only be read on a Silicon Graphics workstation—onto it.
But as Serexhe points out, simply relying on swapping old broken hardware with old restored hardware isn’t a permanent solution. After all, one day— Serexhe says within 10 years—all the Silicon Graphics workstations left in the world will become completely irreparable due to component obsolescence. What, then, will the the artwork’s code run on?
This is the crux of what makes digital art conservation so hard. If the hardware components of a digital work of art are its flesh, the code is its soul. When the hardware can no longer be repaired or replaced, the essence of the artwork, the thing that gives it life—its binary code—has nothing to run on that can understand it and render it in a form that is comprehensible to humans. It is at that point in which digital art dies.
The single best option for The Legibile City’s code to live on after all Silicon Graphics workstations are gone is to port it to a new operating system capable of running on modern-day computers. But unlike porting an app, for example, Microsoft Word from Windows to the Mac, where altering—or completely rewriting—large chunks of code is perfectly reasonable, for digital artwork the trick is in maintaining as much of the original code as possible.
To do this requires coders with not only the most advanced modern-day programming language skills, but also with a deep knowledge of coding in obsolete programming languages, and the ability to see the aesthetic in code. That why Serexhe turned to Bernd Lintermann, an artist and computer scientist who also created xfrog, a procedural modeling and animation software that is used by companies ranging from Electronic Arts to Lucas Digital.
"Bernd is one of the most experienced programmers in the art world," Serexhe says. "When you look at the aesthetic and functional complexity of the code in the The Legible City, no ‘normal’ coder or software engineer would have been able to do this work."
Though it took almost a year, Lintermann was able to port the original code over Linux—but it was a painstaking process. "The porting was carried out in small steps, which facilitated a comparison of the new software modifications with the original code in each step of the work, thus minimizing differences," Serexhe says.
"Given the concurrence of the processes, the dynamic behavior cannot be identically illustrated, but was approximated to the original behavior. Now the work can run perfectly on any normal on-the-shelf computer."
What Serexhe, Lintermann, and the rest of the ZKM team managed to do besides saving an important piece of digital art is show that in the future programmers will have a dual-role to play in society. They will not, as in years past, simply be coding to advance technological progress, they’ll be coding to preserve something just as vital.
The methods, proceedings and results of this three-year research project are published in: Preservation of Digital Art: Theory and Practice, edited by Bernhard Serexhe, AMBRA Vienna, 2013. You can find more information on the project’s website here.