Student Blog: Time-Based Media

For the last eight months I have been the Conservation Intern at Small Data Industries. Located in Brooklyn, New York, Small Data Industries is a private practice that specializes in addressing the conservation and preservation needs of the objects, artworks and related materials that are generally classified within collections as time-based media (TBM). We work with a wide range of museum and archival collections, as well as private collectors, artist estates and foundations, and living artists themselves. The company's mission is to "support and empower people to safeguard the permanence and integrity of the world's artistic record," which it achieves by providing a variety of services including condition assessment, collection surveys, conservation treatment, technical documentation, acquisition counsel, digital repository management, coaching, and consulting.

The volume and diversity of on-going activity has afforded me the opportunity to gain an astonishing breadth of experience, which has furthered my exploration and deepened my understanding of the conservation practice while exposing me to new practical techniques that have exercised my skills and expanded my proficiency. Over the course of the last several months, I have completed the physical and digital examinations of collection materials, written treatment proposals, performed condition assessments, conducted artist interviews, and produced written, photographic, audio, video and technical forms of documentation for the hardware and software associated with TBM artworks and archival materials.

​Like the materials at the center of every conservation specialty, TBM often requires highly specialized custom-made tools to comprehensively examine, document, and successfully treat its constituent elements. Some of the most exciting projects I have been involved with have incorporated the development and construction tools designed to address the unique complexities of TBM by performing tasks impossible for a person alone to accomplish.

Recently, the need for such a tool became evident during a condition assessment of a kinetic sculpture examined on view in a private collection by Ben Fino-Radin, my supervisor and the Founder of Small Data Industries. An examination of the artwork's circuitry revealed that its operation was governed by an integrated circuit (IC) chip that had been programmed by the artist. Unlike most analog circuits whose function can be understood as a series of electrochemical interactions and replicated by reconstructing the circuit's configuration using generic electrical components, the exact processes underlying the artwork's behavior were more opaque being a direct result of the IC's programming. The IC, like all electrical components, is inherently vulnerable to degradation that would compromise its functionality and risks the corruption of its programming. Whereas a passive component might fail and be replaced without affecting the overall functionality, the failure of the IC threatens the survival of the artwork fundamentally by risking the loss of its programming. Thus, acquiring the program's source code to ensure the preservation of the artwork in the event of such a failure became a treatment priority. After exhausting alternatives for obtaining it, accessing the program directly off the IC became necessary.

We decided to investigate using a device invented in 2012 by computer scientist Trammell Hudson called a PROMdate and I was tasked with its construction and programming. Since the treatment would be taking place at the artwork's installation site, I also needed to assemble the software required to operate the device and install it on a virtual machine that could be used in-situ, in addition to updating, testing, and verifying that it would be able to read and download the program safely off the artwork's particular IC model.

The PROMdate was designed to read and download the data stored on ICs as it exists there. The resulting file is the IC's program compiled as binary machine language. Without knowing things like what programming language was used to write the original software, it may not be possible to convert the binary code into something intelligible. Although in this case the binary source code would not greatly contribute to a description of how the artwork's software influences its behavior, it would be sufficient to program a replacement IC that would reliably elicit that behavior in exactly the the same way as the original, should that one fail. Thus, it satisfied the primary goal of the treatment.

Trammell has provided some images of the PROMdate, the source code for programming it, and some minimal instructions. The initial construction went smoothly, but I began running into problems when I started programming it. I consulted Trammell, updated the PROMdate’s source code, and found some alternative programming instructions developed by an anonymous github contributor. After adding some of my own work-arounds and bug fixes, I finally got it working. I tested the new tool by writing a known program to a sample IC we had in the lab. Then, using the PROMdate, I attempted to follow my own written procedure for retrieving the IC’s contents, and verified the accuracy of the delivered file by comparing it to the program I had originally written to the IC. The process of verification uses a tool that highlights any differences it discovers when comparing the two files. I don't think I have ever been as excited to watch nothing happen.

Ben was able to take my PROMdate, use it to successfully treat the artwork, and obtaining the program file, which will now be stored in a secure in a digital repository following the standards of best practices for the preservation of digital material.

—Nick Kaplan, WUDPAC Class of 2019