What does collections care for the digital look like? The V&A’s expertise on conserving digital and media collection objects has been rapidly expanding over the past few years. When people think of a conservator that works in a museum, they likely imagine a painting or sculpture conservator caring for objects from the distant past. But our heritage is changing, and today “the past” is not so distant anymore. The materials that conservators are expected to understand and work with are changing with the objects that are acquired into the collections. But what if the material that needs conserving is, quite literally, information? The idea of a conservator who works with digital objects may raise some eyebrows, and so in this post I will be demystifying what caring for the digital looks like and share my recent work on the software-based artwork Shaping Form 14/5/2007 by the artist Ernest Edmonds.
As a practicing conservator and an embedded doctoral researcher at the V&A, I have been focusing on the conservation of software-based objects in the collections, such as mobile apps and computer-based art. Just over a year ago, I carried out additional conservation work beyond my doctoral research in the museum, which included looking more closely at Shaping Form from a conservation perspective.
Made just under 20 years ago in 2007, Shaping Form consists of a MacMini computer that connects to a liquid crystal display monitor, designed by the artist to be mounted on a wall in a similar way a painting would be. On this MacMini is a software application which, in response to movement captured through the USB camera embedded in the frame of the monitor, generates an animation consisting of vertical bands of varying thickness and colour. The software was programmed by the artist using primarily the visual programming language Max/MSP – a technology which is still incredibly appealing to media artists today. My PhD colleague Katherine Mitchell previously wrote a brilliant blog post mentioning this artwork, which I recommend to anyone interested in finding out more about this acquisition.
As a conservator my responsibility is to look after an object’s materiality and to manage its condition over time in a way that is informed by conservation ethics. In practice, this means assessing the condition of an object and performing treatments in response to what I find. This could be as simple as installing a missing dependency required to make the software run, or more complex, such as fixing a bug in the software code. At the same time, conservation is not always reactive to undesirable change or damage. My work equally consists of taking steps to prevent loss in the object or, if loss does happen, to minimise the negative effects of such loss.
Disk imaging is a preventive conservation strategy for digital objects that include storage devices such as CD-ROMs or, as in the case of Shaping Form, a computer’s internal hard drive. As the V&A is only in the beginning of setting up its collections care workflows for digital and media objects, a lot of my work involved liaising with more senior colleagues in Conservation, Design and Digital curators and colleagues over in Technology to build the foundation for the museum’s first dedicated digital conservation studio. Disk imaging is one of the capacities that I was involved in establishing in the studio, and it includes the ability to create digital, bit-for-bit copies (“images”) of said storage devices (“disks”).
Creating bit-for-bit copies is important in the long-term care of a digital object because you do not want the only surviving copy of an object to be on aging hardware, which can fail at any point and thus sabotage your access to that object. After all, electronic components comprising digital objects in the collection are of little meaning without the information they give access to.
Disk imaging is not simply a copy or back-up of the files comprising a digital object – instead it includes wider information such as the files’ organisation (file and directory systems and structures), the wider digital environment (for example, the operating system, as in computer-based works) and even a copy of deleted files. Components created through disk imaging can be used for facilitating the migration of digital objects to different hardware and, for computer-based works, to perform a treatment called emulation. This treatment includes reconstructing the entire software and hardware environment (in software form) independently of the hardware on which it is running. While treatments such as migration and emulation are unlikely to be implemented for Shaping Form until the original hardware becomes too difficult to repair, performing disk imaging now ensures that such treatments can indeed take place when needed in the future.
There are several methods of disk imaging, including connecting remotely to the computer over a network. However, I decided to image the MacMini containing Shaping Form by disassembling the artist-provided computer and manually extracting the hard drive to perform the imaging physically. Ideally, conservators shouldn’t be disassembling the computers supporting collection objects, especially those that are no longer in active circulation, unless absolutely necessary to avoid wear and tear. However, in the case of Shaping Form, because one of the internal batteries needed changing anyway, it made sense to disassemble the MacMini to perform basic maintenance and disk imaging all in one go.
Adapters for legacy computer equipment had to be sourced in order to connect to the interface of the hard drive (which has been superseded by other types of connections) and read data off the disk. Coming back to what I said earlier, conservation ethics demand a particular approach to objects that safeguards their original, intended condition during any intervention. Data integrity and fixity (that is, the idea that the bits of a digital object have been unchanged) is a requirement that digital conservation practice frequently circles back to. To guarantee integrity during the disk imaging process, more equipment such as write-blockers were also procured from a specialist, digital forensics supplier. These are physical devices that ensure no accidental writing to the storage device takes place, acting as a bridge between the device itself and the machine performing the disk imaging.
Disassembling and then reassembling the artist’s MacMini for the purposes of disk imaging provided an opportunity for me to perform additional basic preventive measures to ensure the hardware lasts for as long as possible, such as removing the build-up of dust in the internal electronic components.
Having created a back-up of the artist’s software and its environment, running the object natively using original, artist-provided parts presents less risks meaning I could finally assess their condition. Because the artwork is interactive, I assessed the object’s condition in both its “dormant” and “activated” states. With computer and software-based works such as Shaping Form, both its hardware and software components should be considered. For example, when assessing whether the artist-written code runs as intended, I used tools targeted at software developers for debugging their code, with “debugging” referring to finding errors, or “bugs”, in computer code. I made a note of any errors on the Max/MSP console when running the code and assessed whether they were critical to the functioning of the object, or simply false negatives. Conservation documentation, such as a treatment proposals and preservation plans – the latter of which is becoming regular practice in V&A’s conservation department for objects of particular complexity – have been written in response to this assessment of Shaping Form.
One of the most exciting parts about being a conservator is contributing to the ecology of understanding the object through conservation practice. It is not uncommon for a conservation intervention to lead to new discoveries about an object in the collection, and in fact these discoveries have the power to impact meaning-making in curatorial contexts and beyond.
As part of my conservation assessment for Shaping Form, I was able to further analyse the software components comprising the artwork and include this in institutional documentation. Of particular interest was the source code for artist-written software, which besides Max/MSP code and the additional Jitter framework that allowed for the manipulation of graphics, included code written in another programming language: Java. This is significant as it demonstrates an even greater intentionality in Edmonds’ artistic practice. Programming in Java meant that the artist was able to implement additional algorithmic routines to further define the visual output the software produces.
The fact that this software source code is even available in the first place is great luck. Source code is often not available to conservators because the software is acquired in its final, machine-readable form only. Reverse engineering, which is the process of deriving human-readable source code from machine code, is possible but not always ethical nor legal. Not only does the source code help study the program, but it allows minimal and controlled changes to the code to be made during treatments such as migration or emulation.
Another interesting discovery was the deleted files that I was able to recover during disk imaging, including multiple versions of documents outlining the high-level logic of the algorithms behind the source code. Edmonds previously mentioned the existence of these software architecture documents in an interview I had with him last year. Together, the source code and software architecture documents provide precise insight into the program functionality and artistic intention which needs to be respected in any conservation attempts. For curators, the multiple versions of the source code are important historical evidence and a valuable resource for studying the artist’s process of making.
More is to come at the V&A regarding conserving and preserving the digital, and the museum has in this past year appointed its first Digital Preservation Manager: Glenn Cumiskey. Watch this space!
With thanks to the artist Ernest Edmonds for his support, as well as Corinna Gardner and Pedro Gaspar for their feedback on the text.
Digital conservation raises unique questions about preserving intangible work that exists in unstable or shifting technological environments. It requires not only historical sensitivity but also technical structure to ensure long-term visibility. In exploring how institutions maintain digital artifacts, I came across mentions of computer vision development services in conversations about building archival-friendly development frameworks. It’s a reminder that conserving the digital world relies on careful, future-oriented development.