This project gathered and analyzed seventeen linked data projects conducted by archival institutions and projects that include significant amounts of archival materials. Some of these projects are large-scale projects conducted by major metadata aggregators or national libraries, such as OCLC WorldCat, Europeana, Digital Public Library of America (DPLA) and data.bnf.fr created by Bibliothèque nationale de France (BNF). Others are smaller projects conducted by individual university libraries or small research groups, such as the SALDA project, Linked Jazz project, and the Cultural Repositories & Information Systems (CURIOS) project that developed the software platform for the Hebridean Connections cultural repository.
The author found that some projects keep their original user interfaces in addition to publishing linked data, or incorporate linked data features into a conventional user interfaces which do not require users to know linked data technologies. There is a common use of existing ontologies/vocabularies. Many projects choose to use generic vocabularies for some classes and properties, instead of specialized LAM terms in recognition of the benefits of doing so. Four types of linked data have been generated for archival materials: 1) archival descriptions; 2) archival authority files for corporate bodies, persons, and families; 3) controlled vocabularies for subject indexing and; 4) content annotations. However, most current projects primarily convert existing descriptions instead of creating original linked data. In addition, data modeling of existing projects are based on existing archival description standards instead of user needs in shifted technology and information environment. The author concluded that the archival community is still in the early stage of linked data implementation. Nevertheless, she pointed out linked data has demonstrated great potential for improving archival description and archival information discovery. More specifically, linked data will enrich archival description, make archival description more interoperable, more granular and make archival information discovery more powerful, such as directly answering user questions instead of returning documents that might contain answers.
I am an assistant professor at the School of Information, University of South Florida. I received my PhD degree from University of Michigan, Ann Arbor. Prior to that, I worked for the Tsinghua University Library in China for three years.
My research covers all areas of archives management and digital curation. Recently, I have become very interested in the curation of big data.