Over the last decade, factors affecting the sustainability of research data have multiplied dramatically. Early digital humanities projects employed databases and content management systems, which made them vulnerable to rapid internet evolution and ensuing technology obsolescence. More recent collection ecosystems support internationally recognized metadata standards and have gained a level of independence from technologies. This autonomy has, however, created a new dependency on relationships between, and the stability of standards organizations themselves - in effect, only shifting the focus of long-term vulnerability. The proliferation of metadata definition activities, together with frequent revision of standards, has actually made technology planning increasingly difficult for implementors. Despite the dramatic urgency of sustainability issues (80% of raw research data is lost within two decades), and the equally serious challenges to digital scholarship posed by growing recognition that annotation metadata management is woefully inadequate, research funding still fails to accord long-term accessibility of digital research data the priority that it clearly demands.
This lecture will present new developments in automated transformation of research data to comply with multiple and evolving standards in order to address these challenges. A collaboration between Heidelberg, Lyon, Princeton and Westminster universities and the commercial sector has produced substantial results from a strategy that rejects the notion of conclusive representation of research data and that instead focuses on effective tools for long-term migration. Such an approach enables the best use to be made of current versions of metadata standards without becoming locked in to unsupportable maintenance costs as standards evolve. This is particularly important for annotation metadata, where standards are still being developed. A number of research projects will be used to illustrate this work, including Tajik anthropology and early audio communication, in which preservation and presentation are important goals, as well as long-term accessibility of metadata.
An event of the Basel Graduate School of History in collaboration with the Institute for Media Science and the Institute for European Global Studies. Speaker: Prof. Peter Cornwell (University of Westminster), Discussant: Prof. Dr. Markus Krajewski, Chair: Prof. Dr. Martin Lengwiler
24th November 2016, 6-8 pm
Seminar room 1, Department of History, Hirschgässlein 21, 4051 Basel
Peter Cornwell is a British computer scientist and media theorist. He is Director of the Data Futures project and professor in the Institute for Modern and Contemporary Culture at the University of Westminster. He was an expert for development of the EU's first Strategic Program on Research in Information Technology (ESPRIT) work-program, has served on numerous U.K. strategy committees and, as director of the Institute of Visual Media at the ZKM Center for Art and Media Technology founded its Center for Digital Heritage. Cornwell has also received numerous awards in industry: he was manager of European R&D for Texas Instruments, Inc. and founder and CEO of California IPO Division Inc.