The author would like to thank

Tan Zi Hao, the artist, Why Knot Studio, Chua Wei Han, Mahyuddin (Din) Chan, Muhd Syakir Juraimi, Christel Pesme.

The Singapore Art Museum: June Yap, Ong Puay Khim, Tiara Tizzanty.
The Heritage Conservation Centre Time Based Media Working Group: Run Nan Luo, Fabiola Rocco, Alex Soo,  Kezia Toh, Jo Anne Wong.

Background

1The Heritage Conservation Centre (HCC), a Singapore government institution managed by the National Heritage Board, is the repository and conservation centre for all of Singapore’s cultural institutions. There are currently nine museums, heritage interpretive centres and community heritage institutions under HCC’s care, each with their own collections, the sum of which is known as the Singapore National Collection. The variety of the National Collection means that HCC cares for a wide range of artefacts, ranging from paper, paintings and archeological objects to born-digital items.

2All these artefacts are generally housed and worked on in one physical location, HCC’s facility. This facility also contains conservation labs and the conservators who work with them, and the collections management staff such as registrars that track artefact movement.

3Currently, the Singapore National Collection is made up of over 300,000 artefacts, the vast majority of which consist of mediums such as paper, sculpture or paintings. The TBM collection, in contrast, is around 300 artworks, less than a tenth of a percent of the entire collection. Because of this, and the overall centralised nature of HCC as a whole, the existing processes and workflows to acquire, accession and conserve artworks have been developed with the majority of the artworks in mind, with prior TBM artworks being shoehorned into this mould.

4An increase in TBM artwork acquisitions is expected in the coming years - projected to be at least 25 to 35 artworks per year - partly due to added curatorial interest in the medium and the establishment of the Singapore Design Collection. Mirroring the increasing amount is the increasing complexity of the artworks themselves. While much of the current TBM collection consists of relatively straightforward single channel or unsynchronised multi-channel works, more recent acquisitions have started to consist of synchronised, software driven or installation elements. Because of this, it was deemed necessary to focus on the needs of the TBM Collection and create processes to better deal with the increased influx of future artworks. The first major step to do so is to establish a TBM-specific workflow and standardise their documentation.

The Prior TBM Acquisition Processes

5As mentioned above, prior TBM artworks were made to fit into the acquisition, documentation and accessioning processes originally meant for more traditional artwork mediums. In practice, this resulted in the physical aspects of a TBM artwork being the focus. For example, the hard drive or optical disc holding the artwork - rather than the digital file within it - was sometimes accessoned and documented as the artwork. Because they were treated like physical artworks, the files that make up these TBM artworks were generally only assessed in full when it was needed for display and not necessarily upon acquisition. Similarly, the documentation for these artworks focused on the physical. For instance, sometimes the physical dimensions of the hard drive holding the artwork was documented, rather than the pixel dimensions or aspect ratio of the video file within it.

6This disparity is most evident when viewing the overall past accessioning workflow, which consisted of thirty-three separate steps. Of those steps, conservators had a direct hand in six of them. This process worked well in traditional artwork mediums but not necessarily for their TBM counterparts. Of course, this is not also to lay blame on the work of past conservators or collections managers. As with many institutions now, they did what was best practice with the resources and expertise available to them at the time for what was admittedly a minor subset of the collection as a whole.

7Clearly, the growing amount of TBM artwork and the problems of using this process on current TBM artworks meant that a new workflow had to be developed.

The New TBM Workflow

8One of the first key steps in the creation of the new TBM was the establishment of a TBM working group, which consists of six active members from both the Photography, Time Based Media Conservation, Contemporary Art Conservation and the Collection Management departments in HCC. This group continues to meet regularly to refine the workflow and discuss specific artwork issues.

9The new workflow details the journey from the TBM artwork physically entering HCC’s premises via a hard drive, documentation and condition reporting of the files within the hard drive and its eventual storage into a backup server onsite. In contrast with the prior process, the assessment of the digital files is one of the first steps in this workflow, including more standard processes such as file fixity checks to guard against file corruption. The importance of assessing the files early was evident in the case study that this case outlines.

10With the new TBM artwork workflow came new TBM artwork terminology, reflecting an added understanding and nuance of these works. For example, artwork files are now split in either Master and Exhibition copies. Collection Objects are files related to the artworks - sometimes artist provided or generated internationally - but not the artwork itself. These designations mark the file’s different role in conservation or exhibition, its storage location and needs. The documentation around these artworks has also been a focus. TBM-specific condition reports, which have fields for aspect ratio and duration, are now the norm. An artwork’s identity, deeply important as a hallmark of TBM artworks is changes in their display, is now also formalised using iteration and identity reports.

11As a point of comparison, the new TBM workflow has forty-three steps, of which twenty seven are conservation-specific.

Case Study: The Mercurial Inscription

12The Mercurial Inscription (2022) is an TBM installation artwork by Malaysian artist, Tan Zi Hao. The artwork consists of physical elements interacting with digital and software based elements. The creation of the digital aspect primarily the role of Why Knot Studios, a Malaysian game company. As this is a recent acquisition, the new TBM workflow was used and its benefits were immediately apparent.

13The artwork depicts the Terengganu Inscription Stone, an archeological object on which was inscribed an early testimony of Islam’s presence in Malaysia. The Stone is shown in two ways: As a digital object projected onto a screen in the gallery space and as an physical aluminium stone on a plinth in front of it. Audience members are encouraged to touch the physical stone. Doing so activates an animation on the digital stone. This animation is of the stone dipping into water and the text on it changing to one of twelve languages or transliterations. After a few seconds, the digital stone rises up from the water until the physical stone is interacted with again.

14Functionally, the touch is detected via a sensor on a jumper cable connected to an Arduino within the stone. Once the stone is touched, the Arduino registers the touch on a Unity program in a computer hidden in the plinth below. The Unity program triggers the animation, which outputs onto a projector and speakers connected to that computer

The Issue

15According to the artist, The Mercurial Inscription’s Arduino had to be calibrated prior to each exhibition as it is sensitive to a variety of sources, particularly electrical waves. To do so, the artist provided an in-depth guide that required the installation of two programs on a computer that the Arudino had to then be plugged into. The sensitivity was then tweaked on the computer in accordance to the specific needs of the exhibition.

16As part of the new TBM workflow, any digital files have to be assessed as soon as it enters HCC’s care. For simpler artworks, such as a single channel video, this assessment may mean completely watching the artwork to check for errors, or to do fixity checks to see if the file is corrupted. In The Memorial Inscrption’s case, which has the software elements such as the Arduino and Unity programmes, it required creating a conservation installation of the artwor. Alongside contemporary art conservator, Fabiola Rocco, this installation was done using the guide provided by the artist. It was during this step that a major issue was identified.

17The Arduino simply could not be detected by the computer meant to configure it. This persisted over using multiple test computers, both windows and Mac, and the backup Arduino. With the help of the Why Knot Studio’s programmer, Chua Wei Han, it was found that the issue likely stemmed from the setup of the computer itself instead of the Arduino. In its past iterations, a computer that was known to work with the Arduino was used. However, this computer was not acquired. At the time, it was assumed that any computer would work if the right programs were installed using the artist instructions. The conservation installation proved that it was not the case. The value of the new workflow was immediately apparent as it required early testing of any TBM workflow, and not only when its exhibition was required as was stipulated by the old workflow.

The Fix

18Once the issue was identified, a solution was sought to address any potential computer incompatibilities by acquiring a main and backup computer dedicated to the artwork. This was done with the help of another artist, Din Chan, who was recommended by Zi Hao as he had extensive experience with touch interfaces. Din suggested creating a new Arduino that can be plugged into the computer but have its sensitivity configured and tested all within the Arduino itself. This eliminated the need to install the two separate programs, minimising any software obsolescence issues in the future. This solution was approved by the artist, curators and other stakeholders.

19Din also suggested changing the wire connecting the Arduino to the physical stone, the older version of which the artist stated was especially prone to damage. In fact, the original acquisition consisted of ten backup wires to account for each’s fragility. A more robust system, used upon the advice of Din, used alligator clips.

20The functionality and parts of the Arduino has been documented and a backup is currently being made to undergo more testing, but the key point is that the artwork consistently ran using the new Arduino without the need of installing new programs or using the old calibration method. To ensure that this is a more permanent solution, a master and exhibition computer will also be bought and tested.

The Results

21Early access to the artwork by conservations and collaboration between the artist and programmer ensured that a conservation install could be done, leading to the discovery of the fault with the Arduino. Once the issue was found, the early intervention meant that the artist and programmer were still contactable, still had detailed knowledge about the artwork and were available to help find and approve a solution. Armed with this knowledge, additional items were acquired to better preserve the artwork, as shown by the chart below.

Initial Acquisition Items	Items After Testing Items in bold were acquired after conservation intervention
Aluminum Sculpture	Aluminum Sculpture
Arduino Touch Sensor + 1 backup	Upgraded Touch Sensor + 1 backup
Touch Sensor Wire + 10 backup	Upgraded Touch Sensor Wire + 2 backups
MicroUSB cable	MicroUSB cable
Unity Artwork	Unity Artwork
Artwork Documentation	Artwork Documentation
	Upgraded Touch Sensor Documentation
	Mac Mini – Master Copy
	Mac Mini – Exhibition Copy

22Another advantage was doing the condition check prior to formal accessioning the artwork. This was deliberately done as part of the TBM workflow. Had the artwork been accessioned and formally added to the Singapore Collection Management System, it would be far harder to modify the record after the fact, especially if specific components or parts needed to be reclassified or deaccessioned. It also freed up resources with, for example, our photography unit, who would not need to retake any documentation for replacement parts.

Future Considerations

23While obviously effective in this instance, the new TBM workflow needs to be tested further and refined, especially under different conditions, such as when dealing with commissioned artwork, if the artist is deceased or otherwise uncontactable or a third party was instrumental in the artwork’s creation yet that is unwilling to provide adequate documentation. As noted prior, this process is highly specific to a small part of the collection, and still has to sit within the current reporting and CMS processes that cater to more traditional artworks. Its effect on work processes for those who are more used to the existing workflows will also need to be assessed. Lastly, his workflow needs to be tested to see if it can be maintained effectively when faced with an increasing rate of TBM acquisitions.