Ingrid Mason on Digital Humanities Research
We are now at an intersection of data technologists and humanities researchers in the digital / eresearch field. What is the cross-over between digital scholarship and digital humanities? We need to adjust the way that we work with our collections. We need to know what access an API provides us with – all data or a limited set? We need to look not only at an item level, but also at a collection level as an overview. Archivists are used to working at a collection level. When we look at bulk data we can review, arrange, and get perspective. Four researchers were acknowledged: Kenderdine, Manovich (pattern analysis), Sherratt (trove), Whitelaw (interfaces). Changing environment provides an opportunity for re-engagement. Proposal: Add a “digitise this” request button to catalogue items (e.g. National Archives). VicNode offers a datasharing platform (http://vicnode.org.au/). Proposal: give authorised users access into our systems. Researchers should be able to create formalised relationships with organisations to enable access to their data, because this is an ongoing problem for any researcher trying to access cultural organisation data.
Peter Neish on linked data at Parliament of Victoria
Triple statements enable linking between two items via a third item. They enable verification or identification of data, can be used to retrieve additional data, can help to collate information. Recipes are now using structured data markups. Ontology (http://schema.org/). Atlas of Living Australia using linked data to track changes in naming over time and across boundaries (http://www.ala.org.au/). See: http://www.govhack.org/atlas-of-living-australia-developer-information/ The motivation for linked data at Parliament was to make work easier internally. Peter includes a great layout of the process in his slides. Used http://popoloproject.com/ for Government information ontology. The process of working with linked data helps your understanding of the data set.
Deb Verhoeven / Toby Burrows on linked data in the cultural sector.
Linked data enables change such as greater impact of collection material, a re-orientation of the way we view the material, and also spatialisation, so that we can get perspective on a macro level. We need to look at new ways of working with our collections, not just collecting work but research instead. We need to gather and integrate data sets. Our data sets need to move from silos to integrated forms. We need to focus on the end use of our collections, rather than the collecting process. Huni uses 28 data sources from humanities collections. Data is not changed or manipulated during the process. Aggregated data is published as RDF. Entities are displayed as fixed terms. Huni will serve to enable discovery and exploration of data, to make connections between data, to save and share data, and to enhance curation. Data provenance is preserved through reference to the original record. Two functions have been enabled in the interface – saving a record, and also making a connection between records. The latter enables data linking. You can get a visual view of connections created. Huni will enable larger scale access to data, it will break down the silos and boundaries, it will enable sharing and collaboration. Structured serendipity enables us to find connections between almost anything. This means that data exploration no longer runs into dead ends, but should aim to open up new possibilities instead. The amphibology (fuzziness) of humanities data (e.g. Captain Matchbox). Different ways of defining items and creating connections for it open new possibilities.