Each volume of Journal of Digital Media Management consists of four quarterly 100-page issues. Articles scheduled for Volume 13 include:
Volume 13 Number 3
The European Broadcasting Union’s AI Hub is a pioneering platform that facilitates the development and evaluation of customised AI solutions for the media industry. This collaborative ecosystem enables AI and media experts to co-create and refine artificial intelligence (AI) models designed specifically for media applications. By providing private spaces for testing AI models with proprietary content, the AI Hub guarantees comprehensive evaluations across a wide range of media content. This paper examines how these collaborative AI solutions are driving media innovation and highlights the crucial role of open source models in adapting to rapidly evolving technological landscapes. The paper will detail three key innovations: MetaRadio, which enriches radio experiences with metadata; a facial recognition system tailored for television programmes; and a Fake News Analyser designed to evaluate the reliability of news articles.
Keywords: AI; generative AI; media; user-centric; evaluation; radio programme; facial recognition; disinformation
Born-digital archival collections have proven an emergent challenge for repositories, often bucking traditional infrastructures and approaches. At Harvard University’s Schlesinger Library, recent born-digital archival work engages known gaps, resulting in a case study on technologies and procedures. Since 2013, Schlesinger has used digital forensics technology and custom tools to convert and retain borndigital materials in narrow content types using an expansive and enduring preservation repository system. Over 18 months in 2023 and 2024, archivists and librarians at Schlesinger collaboratively refined a born-digital archiving model that leverages digital forensics technology, collaborative processing tactics, custom scripting, existing repository tools and a new virtual reading room. Starting from evidence-based methods in digital forensics and digital preservation, Schlesinger’s current born-digital operational scheme is further grounded in collaboration and automation. Developments in Schlesinger’s approach to born-digital archiving have resulted in successes and new challenges. This case study begins by summarising groundwork in born-digital archiving practice and the genesis of a formal digital archives programme at Schlesinger. The following sections walk through the current state of Schlesinger’s practices for born-digital capture, processing, post-processing, access and preservation, including references to recent collections and the methods employed. A concluding section will reflect on the advantages offered by the preceding efforts and address the challenges that shape the work ahead.
Keywords: digital archives; born-digital processing; digital forensics; Python; PDF/A; digital preservation
This paper describes the exploration into and implementation of a hosted digital preservation management system at the University of Montana Archives and Special Collections. This was done to find a solution to the increasingly economically and environmentally unsustainable procedures practised at the university. Three products were analysed, and one was recommended as a potential future solution. A project-based implementation of the chosen solution using grant funds was used to make a business case for full implementation. The business case was successful in so much as an additional implementation for a single collection was completed. The paper details the roadblocks encountered during implementation and how they were successfully managed by the digital archivist and the person hired to support the implementation. The paper concludes with a discussion of next steps to maintain the project into the future.
Keywords: digital preservation; sustainability; accessibility; digital preservation tools; workflows
This paper provides a case study of the development, launch and expansion of the White House Historical Association’s (WHHA) Digital Library. The Digital Library supports the WHHA’s core mission to protect, preserve and provide a comprehensive history of the White House by storing and displaying materials for users to access. As this paper shows, the digital asset management (DAM) system that powers the Digital Library not only supports WHHA’s internal needs, but also allows team members to control access to assets in order to provide a research tool for a variety of audiences without disrupting the user experience for WHHA staff. The paper also highlights how the DAM system functions as a collaborative workspace for ingesting assets into the Digital Library, and describes the multiple methods of accessing and delivering image files to internal and external users. Finally, this case study demonstrates how organisations of all sizes can best use DAM to support their missions and make their resources accessible to the public on their own terms.
Keywords: non-profit; digital humanities; digital archiving; workflow optimisation; digital asset management system; metadata; rights and reproduction
This paper examines the challenges and solutions associated with making archival PDF finding aids accessible to blind and low-vision users, particularly those who rely on screen readers. The project, conducted at the University of Toronto, highlights the barriers posed by unstructured PDFs, which fail to meet the various accessibility standards specified in the WWW Consortium’s Web Content Accessibility Guidelines. Three methods were tested to improve accessibility: manual remediation, PDF-to-HTML conversion, and data migration into the Access to Memory (AtoM) platform. The results indicated that both the manual and automated remediation methods were either too costly or ineffective, while the most promising approach involved migrating the description data into AtoM via CSV import, enhancing both accessibility and search functionality. The paper underscores the need for ongoing funding and professional expertise to address web accessibility issues in archival settings and outlines future steps to improve PDF generation within AtoM for broader application.
Keywords: accessibility; AtoM; finding aid; screen reader; PDF
Newzroom Afrika — a relatively new player in the broadcasting landscape of South Africa — seeks to be a true reflection of the diverse digital and cultural landscape of Africa. To this end, it provides an innovative, multimedia news service that transcends the traditional linear television experience, engaging the audience through interaction across all news and information-sharing platforms. This case study explores the channel’s journey in navigating the challenges and opportunities faced in implementing a suitable and effective digital archive service and media asset management system tailored to the needs of a growing media organisation with a unique African viewpoint. In addition to discussing the obstacles this growing channel had to overcome in order to realise this ambitious goal, it also describes meticulous organisation of people, infrastructure, resources and systems required to establish an archive infrastructure and service that are sufficiently robust to manage, organise and store all generated content for future reference and utilisation. After a discussion of how Newzroom Afrika Archives has continued to adapt to meet the demands and challenges of a dynamic and rapidly evolving newsroom environment, the study concludes by highlighting the key insights gained from these experiences.
Keywords: media asset management; MAM; news production/archive workflow; broadcast news archive; metadata mapping; metadata schema; African broadcaster; 24-hour news channel; multi-lingual news
This case study compares the processing of two collections with born-digital content at The New School Archives and Special Collections. One of the collections, the Parsons School of Design School of Fashion records was hybrid, containing both analogue and born-digital content; the other, the Laura Auricchio faculty and administrative files, was entirely born-digital. Decision points and challenges encountered in processing these collections may prove useful to others in similar scenarios, and offer examples of the tension between aggregation-orientated archival arrangement and description, and item-level digital preservation. For each collection, an imperfect solution was arrived at to resolve this tension, with one more successful than the other. Success in this instance represents an approach that can be replicated for other collections, and that leverages extractable born-digital metadata and the existing arrangement of files to reduce overall processing time.
Keywords: born-digital archival collections; hybrid archival collections; born-digital processing; archival arrangement and description; archival processing; finding aids
This paper describes a project undertaken at Worcester Polytechnic Institute’s (WPI) Archives and Special Collections to migrate more than 12,000 digitised TIFF images, representing 0.6 TB of data, from multiple local network folders into a digital repository that could generate appropriately sized JPEG derivatives in an effort to alleviate performance issues. This paper also describes how improved workflows were implemented and altered in response to unforeseen challenges and describes collaborative efforts with campus Information Technology Services and open source community developers to integrate technical systems, notably Blacklight’s Spotlight plug-in, within the repository. It concludes with a reflection on how these new tools have improved the processing of born-digital and digitised materials and their discovery by patrons.
Keywords: digital preservation; digital repositories; file migration; Hyrax; Samvera; Preservica
Volume 13 Number 2
Over the past quarter-century, massive digitisation programmes have reenergised research, giving rise to new areas of studies, collectively known as the digital humanities. The large swathes of digitised content have an even broader impact as they are used in machine learning and artificial intelligence. The rapid transformation of content, dominated by collaborative programmes among for-profit and nonprofit organisations, have accelerated deep-rooted challenges related to copyright, silos and equity. Programmes such as the National Digital Newspaper Program, the Global Press Archive CRL Alliance and Reveal Digital provide positive examples to be evaluated, improved and replicated in future collaborative digitisation programmes. This paper presents an overview of the digitisation landscape and notes key challenges that exacerbate inequities. Case studies are presented to highlight a few programmes with unique models worth iterating. The paper concludes with recommendations of potential opportunities to model and the impact they may have to make for a more equitable digitisation landscape that will help focus efforts on more diverse content supporting underrepresented communities.
Keywords: digitisation; collaboration; digital humanities; library vendors; archives; special collections
This paper discusses the use of artificial intelligence (AI) applications for the creation and enrichment of descriptive metadata at meemoo, the Flemish Institute for Archives. We begin by explaining why we use AI tools for metadata creation and enrichment, and which AI applications we specifically employ. We then describe our evolution from small-scale pilots to large-scale projects, and how meemoo aims to transition from a project-based approach to a structural operation in using AI applications for metadata creation and enrichment. We provide details on our approach, results, lessons learned and future steps, and describe how the use of AI applications poses not only technical challenges but also raises a series of legal and ethical questions. This paper highlights our journey into AI as a service organisation supporting over 180 organisations in the cultural, media and government sectors in Flanders — not in isolation but in close collaboration with our partners, leading to shared solutions.
Keywords: cultural heritage collections; metadata enhancement; artificial intelligence; facial recognition; speech-to-text; named entity recognition; named entity linking
The Smithsonian Institution has collected, produced and exhibited audiovisual media for well over a century. This paper discusses how, during the last decade, the institution’s concerted efforts at inventorying and cataloguing the hundreds of thousands of analogue media objects in its collections have led to the development of the centralised Audiovisual Media Preservation Initiative (AVMPI). As a pan-institutional service initiative, the AVMPI’s project team, media digitisation and conservation laboratories, and best practice workflows are advancing collections care and digital access activities across the Smithsonian’s 22 museums and research centres. Software such as Airtable and a new accessibility-friendly streaming video platform built by the Smithsonian are integral to these efforts.
Keywords: audiovisual archives; digitisation; digital asset management; digital workflows; metadata management; user story
This paper provides a real-life case study of how three Irish tourism agencies joined forces to create an award-winning tri-agency digital asset management (DAM) solution. The paper describes the wide-ranging elements of the DAM programme, with examples of how it was realised while maintaining a shared vision regarding its status as the single source of truth. The paper discusses the challenges, learnings, innovations and newfound best practices along this journey in order to provide readers with valuable insights to help implement or improve their own DAM solution.
Keywords: tri-agency; digital asset management; tourism; metadata; collaboration; digital assets
Brown University is the source of some of the earliest, most experimental and innovative works in the field of born-digital literary arts. Faculty and students at Brown regularly produce work in a variety of experimental formats designed for digital environments, such as neural-net and model-based aesthetic text-generation programs and fully immersive three-dimensional poetry, and have done so since the early 1960s. Like many institutions, however, Brown University does not yet have models or strategies to preserve many of these works that push boundaries of electronic literature. This paper provides a case study of the university’s New Frameworks to Preserve and Publish Born-Digital Art project, which approaches preservation by focusing first on the works that are difficult to preserve in order to determine what approaches will work best for them, as well as to encourage artists working in these areas to pursue technically innovative art with the knowledge of how to go about sustaining their work for future generations. The paper outlines the preservation frameworks developed for five innovative works that rely on dependencies including Adobe Flash (a technology that recently reached its end of life), other ageing libraries, browser extensions and broken custom code. The frameworks outlined aim to provide a sustainable future for accessing these five works, as well as a guide for institutions and artists alike in how to preserve difficult, technologically innovative art for future generations.
Keywords: digital preservation; digital humanities; electronic literature; digital literary arts; preservation; libraries
Brigham Young University library houses thousands of unique audiovisual recordings within special collections. The library has been managing an audiovisual digitisation programme since 2019 to provide access and preserve audiovisual content in perpetuity. To address the preservation of cultural heritage video content, a stratified approach has been developed according to video content characteristics. This stratification is useful in determining appropriate digital preservation. An argument is made for the use of unconventional lossy video compression codecs in video preservation practice to responsibly balance the needs of digital preservation and resource realities. This paper outlines the underlying reasoning, careful consideration, specific procedures and some results of the implementation of this approach. Similar questions and procedures should be considered when significant video preservation projects are undertaken.
Keywords: digital video preservation encoding specifications; Brigham Young University library; audiovisual; digitisation; audiovisual content; cultural heritage; digital preservation
Volume 13 Number 1
The process of tagging descriptive metadata to digital assets remains a time-consuming and largely manual process for digital asset management (DAM) system administrators. This paper explores the potential of OpenAI’s GPT-4, a large language model, to automate product descriptions, keyword tagging and alt-text. The research team developed six generative AI prompts that instruct GPT-4 to draft one-sentence descriptions and ten keywords for sample product images from six categories of household brands, namely: bicycles, food & beverage, home goods, office furniture, footwear and tools. Using an assessment framework that measures accuracy and precision, the team evaluated GPT-4’s performance by prompt and product category. GPT-4 demonstrated the highest accuracy when describing food & beverage images and the highest keyword precision when tagging footwear images. GPT-4 struggled to be accurate when images displayed low colour contrast or partially obstructed text. It also struggled when attempting to correctly identify gender, relationships and settings. However, GPT-4 showed surprising aptitude at identifying product materials like carbon fibre and species of wood. An analysis of prompts revealed that changes in persona, task description and specifications significantly influence accuracy and precision. The highest average accuracy score and highest average precision score among the prompts suggest that GPT-4 requires careful human oversight when generating keywords, product descriptions and alt-text for accessibility. Even so, it likely saves time for DAM administrators and professionals in marketing and e-commerce.
Keywords: generative AI; digital asset management (DAM); GPT-4; keyword tagging; product descriptions; metadata
In the dynamic landscape of digital asset management (DAM), managing the life cycle of digital assets extends beyond mere storage and retrieval. This paper emphasises the critical importance of content tracking for safeguarding brand integrity and compliance. It outlines the complexities involved in managing digital assets, from pre-production to postdistribution, highlighting the necessity of tracking usage rights and expiration dates across various platforms and geographies. The piece explores advanced technologies such as visual matching and rights management systems that use artificial intelligence to automate tracking and compliance analysis, thereby mitigating legal risks and enhancing operational efficiency. The paper discusses practical scenarios where content tracking is indispensable, such as managing expiring content, avoiding rights violations and maintaining brand consistency during rebranding efforts. It concludes that in a world where digital content predominates, a robust content tracking system integrated with a DAM system is essential for maintaining control over digital assets and ensuring their compliant and effective use. This not only protects the brand but also optimises content performance, ultimately contributing to the brand’s market success.
Keywords: DAM; DRM; content tracking; rights management; compliance; brand integrity; automation
This paper provides a case study of MEDIAS, the ‘living’ digital archiving solution at the Österreichische Mediathek (OeM) — Austria’s national archive for sound and video. The paper describes the origins of this model, the challenges faced during its development and rollout, including multiple data migrations, and changes in storage and naming conventions; the development of a customised search engine to monitor the objects held in the digital archive; and how the system is used on a day-to-day basis. Although MEDIAS is not commercially available, this paper suggests that the concept could be successfully adapted by other institutions with digital archives.
Keywords: audiovisual heritage; long-term digital preservation; search engine; monitoring digital archival objects; Big Data technology; data analysis; data intelligence; migration
Founded in 1963, Lands’ End is a digital retailer of clothing and home products available through online marketplaces and retail stores. In 2019, the company transitioned to a new enterprise digital asset management solution for its in-progress and final assets. This paper summarises how the new system has expanded into a content hub for a wide variety of digital assets, including images, video, catalogue pages, e-mails and integrations with enterprise systems, and describes the challenges, benefits and opportunities.
Keywords: digital asset management; content hub; searching; integrations; metadata
This paper presents a comprehensive case study examining a large-scale video migration project carried out by the USC Digital Repository involving oral history interviews from the Television Academy Foundation, captured in high-definition video (HDV). Unlike the more-widely compatible MiniDV format, HDV poses unique challenges due to its distinct format standard, rendering many open source toolsets geared towards capturing DV content incompatible. Consequently, this leaves archivists to rely on unsupported, proprietary tools, exacerbating the complexities inherent in large-scale migrations. Before presenting the troubleshooting process, this paper provides an overview of The Foundation’s Interviews collection, which includes over 900 interviews spanning over 3,000 hours of catalogued and publicly accessible first-hand accounts of television industry pioneers and professionals. Subsequently, the paper delves into the technical intricacies of the HDV format and the challenges of migrating data from HDV. It examines the strategies considered for identifying and addressing corrupt data, alongside implemented solutions aimed at balancing data authenticity with usability in the face of inevitable corruption occurrences throughout the collection. This study not only sheds light on the challenges inherent in preserving HDV content but also offers guidance for future endeavours seeking to safeguard valuable cultural and historical materials in similar formats.
Keywords: digital preservation; digital video; HDV; miniDV; videotape; audiovisual preservation
This paper investigates the use of virtual machines for the preservation of video games. The study uses Rand and Robyn Miller’s classic adventure game ‘Myst’ (1993) as a sample record to evaluate the performance of four different emulation options for the Mac OS 9 operating system — SheepShaver, QEMU, Apple’s ‘Classic Environment’ and Yale University Library’s emulation-as-a-service-infrastructure platform. Serving as the control for the experiment, Myst is first documented running natively on an original PowerMac G4 and iMac G3 at MoMA. The native performance is then compared with each emulation software. Finally, a fully configured virtual machine is packaged as a single file and tested in different contemporary computing environments. Due to both its performance and accessibility, SheepShaver is found to be the optimal choice for Mac OS 9 emulation for the purposes of accessing classic Macintosh computer games. The paper concludes with a call for further research into the issue of emulator obsolescence. More generally, this project clarifies the risks and challenges that arise when using virtual machines for the long-term preservation of computer and software-based art.
Keywords: emulation; video game preservation; virtual machines; case study; preservation project
The Reformatting Committee at Brigham Young University Library actively manages and facilitates the digitisation of library materials, playing a pivotal role in advancing larger digitisation projects. The committee comprises various stakeholders in the digitisation and reformatting workflow, each with unique concerns and requirements, allowing the committee to effectively navigate the complexities of the digitisation process. This paper discusses the motivations for the committee’s formation, explains its documentation and describes the benefits gained. As this paper explains, the Reformatting Committee exemplifies a balanced approach to meeting various stakeholders’ expectations and building bridges between departments, thereby advancing the library’s mission in the context of the digital era. The paper also explains that stakeholder involvement is crucial, that the structure of the committee enables improved communication and workflow, and, although not all challenges may be anticipated in projects, that the committee is set up to work together to resolve them. The paper concludes that institutions with a history of siloing in their digitisation and reformatting work should consider adopting a similar model, in order to break down such barriers and facilitate communication and collaboration, thus improving the efficiency of digitisation and reformatting projects.
Keywords: digitisation; reformatting; projects; workflow; committees; collaboration
