The following is a guest post by Peggy Griesinger, National Digital Stewardship Resident at the Museum of Modern Art.
As the National Digital Stewardship Resident at the Museum of Modern Art I have had the opportunity to work with MoMA’s newly launched digital repository for time-based media. Specifically, I have been tasked with updating and standardizing the Media Conservation department’s documentation practices. Their documentation needs are somewhat unique in the museum world, as they work with time-based media artworks that are transferred from physical formats such as VHS and U-matic tape to a variety of digital formats, each encoded in different ways. Recording these processes of digitization and migration is a huge concern for media conservators in order to ensure that the digital objects they store are authentic representations of the original works they processed.
It is my job to find a way of recording this information that adheres to standards and can be leveraged for indexing, searching and browsing. The main goal of this project is to integrate the metadata into the faceted browsing system that already exists in the repository. This would mean that, for example, a user could narrow down a results set to all artworks digitized using a particular make and model of a playback device. This would be hugely helpful in the event that an error were discovered with that playback device, making all objects digitized using it potentially invalid. We need the “process history metadata” (which records the technical details of tools used in the digitization or migration of digital objects) to be easily accessible and dynamic so that the conservators can make use of it in innovative and viable ways.
The first phase of this project involved doing in-depth research into existing standards that might be able to solve our documentation needs. Specifically, I needed to find a standardized way to describe – in technical detail – the process of digitizing and migrating various iterations of a time-based media work, or what we call the process history of an object. This work was complicated by the fact that I had little technical knowledge of time-based media. This meant that I not only had to research and understand a variety of metadata standards but I also had to simultaneously learn the technical language being used to express them.
Fortunately, my education in audiovisual technology developed naturally through my extensive interviews and collaborations with the media conservators at MoMA. In order to decide upon a metadata standard to use, I needed to learn very specifically the type of information the conservators wanted to express with this metadata, and how that information would be most effectively structured. This involved choosing artworks from the collection and going over, in great detail, how these objects were assessed, processed, and, if necessary, digitized. After selecting a few standards (namely PBCore, PREMIS, and reVTMD) I thought were worth pursuing in detail, I mapped this information into XML to see if the standards could, in fact, adequately express the information.
Before making a final decision on which standard or combination of standards to use, I organized a metadata experts meeting to get feedback on my project. The discussion at this meeting was immensely helpful in allowing me to understand my project in the wider scope of the metadata world. I also found it extremely helpful to get feedback from experts in the field who did not have much exposure to the project itself, so that they could catch any potential problems or errors that I might not be able to see from having worked so closely with the material for so long.
One important point that was brought up at the meeting was the need to develop detailed use cases for the process history metadata in the repository. I talked with the media conservators at MoMA to see what intended uses they had for this information. To get an idea of the specific types of uses they foresee for this metadata, we can look at the use case for accessing process history metadata. This seems simple on the surface, but we had a number of questions to answer: How do users navigate to this information? Is it accessed at the artwork level (including all copies and versions of an artwork) or at the file level? How is it displayed? Is every element displayed, or only select elements? Where is this information situated in our current system? The discussions I had with the media conservators and our digital repository manager allowed us to answer these questions and create clear and concise use cases.
Developing use cases was simplified by two things:
1) we already had a custom-designed digital repository into which this metadata would be ingested and
2) we had a very clear idea of the structure and content of this metadata.
This meant we were very aware of what we had to work with, and what our potential limitations were. It was therefore very simple for us to know which use cases would be simple fixes and which would require developing entirely new functionalities and user interfaces in the repository. Because we had a good idea of how simple or complex each use case would be, we could prescribe levels of desirability to each use case to ensure the most important and achievable use cases were implemented first.
The next stop for this project will be to bring these use cases, as well as wireframes we have developed to reflect them, to the company responsible for developing our digital repository system. Through conversation with them we will begin the process of integrating process history metadata into the existing repository system.
As I pass the halfway point of my residency, I can look back on the work I have done with pride and look forward to the work still to come with excitement. I cannot wait to see this metadata fully implemented into MoMA’s time-based media digital repository as a dynamic resource for conservators to use and explore. Hopefully the tools we are in the process of creating will be useful to other institutions looking to make their documentation more accessible and interactive.
Can you share what company you are using to build your custom (or customize an existing) repository? Is it a DAM?
Molly – We’re working with Artefactual. They’ve built us a custom repository using Archivematica and AtoM. The code will eventually be released open source.
Molly and others, if you’re interested in learning more about Artefactual’s work with MoMA, you can see a presentation given at AMIA 2014 here: http://www.amiaconference.net/2014-open-source-digital-preservation-access-stream/
Hi Peggy, and many thanks for this post. I was wondering if you will, either now or in due course, be able to share the mapping and the final profile you have settled on for this process history metadata. I think it would be really useful for the community, so as to avoid repeating this kind of work.
Hi Tom – sorry for the incredibly late response, I didn’t see this comment until just now. I definitely think the mapping will become public at some point. I will have to explore what might be the best way of doing that so as to make it as useful as possible for other institutions, as my application was in some ways very specific to MoMA’s needs.