Skip to main content

Resources

  • Extracting Lexical Data: XPath for Dictionary Nerds

    XPath (XML Path Language) is a standard query language for selecting nodes from XML documents. In this step-by-step tutorial, you will learn how to write XPath expressions in order to navigate around our XML-encoded dictionaries and select only those bits of data that you are interested in.
  • DARIAH's Role in Connecting with Arts and Humanities Researchers

    In this lecture, Sally Chambers, Digital Humanities Research Coordinator at the Ghent Centre for Digital Humanities outlines how DARIAH as a Research Infrastructure works within Europe to connect with arts and humanities researchers. She elaborates on how such a European Research Infrastructure could start to work more widely internationally.
  • Design Based Research in Educational Contexts

    In this lecture, Tony Hall examines design-based research (DBR) in educational contexts and settings. Drawing on key contemporary concepts and literature in educational design research, he focuses on how design-based research can be adapted and adopted, both to develop and deploy bespoke educational innovations and technologies.
  • What Skills, Knowledge and Workforces are Needed into the Future?

    What skills, knowledge and workforces are needed into the future? This panel discusses interdisciplines and methods, emerging data practices and ‘Humanities 4.0’. It features presentations by Professor Jean Burgess (Director, Digital Media Research Centre, Queensland University of Technology) on Digital methods and the future of communication and media research and Professor Joy Damousi FASSA FAHA (Lead Chief Investigator) on Future Humanities Workforce project and by Associate Professor Mitchell Whitelaw (Australian National University).
  • How do we Design Infrastructure that Connects?

    How do we design infrastructure that connects? This panel discusses collaborative platforms, partnerships between research and cultural sectors, and libraries as labs. It features presentations by Seb Chan — Chief Experience Officer, Australian Centre for the Moving Image and Dr Marie-Louise Ayres — Director General, National Library of Australia.
  • Transformations: What are the Big Challenges and Opportunities for Data-intensive Research?

    What are the big challenges and opportunities for data-intensive research over the next ten years? This panel discusses digital transformations in the humanities and arts, data ethics and sovereignty, and infrastructure with impact. It features presentations by Dr James Rose (Indigenous Studies Unit, Melbourne School of Population and Global Health) on Data Sovereignty in a Colonial Context: Towards an Integrated National Governance Framework for Australia, Dr James Smithies (Director, King’s Digital Lab) on Integrating DH into the longue durée: Research Laboratories, History, Methods.
  • Cowboys and Consortia: Thoughts on DH Infrastructure

    In this lecture, Quinn Dombrowski shares her thoughts on Digital Humanities Infrastructure, with a special focus on sustainability. She argues that solidarity (i.e. recognition of the interests of the larger group) is a prerequisite for the sustainability of DH infrastructures.
  • What Are We Talking about When We Talk about Data in the Humanities?

    Data as a term is too flat an ontology for the kinds of things that we are all dealing with, argues Sally Wyatt in this keynote lecture. It reduces people, events, objects to things, bits, to be imagined as impersonal, scientific and neutral. Also, she contends, the use of the word 'data' tends to assume that everything is digital. In this keynote, she explains her argument that this is wrong and asks: 'what are we talking about when we talk about data in the humanities?'
  • What Does Data Want?

    Many academic disciplines use data science to analyze contemporary culture. The question posed by Lev Manovich in this lecture is: shall we continue to aggregate big cultural data and reduce it to a small set of patterns? Or shall we refuse this dominant paradigm instead and focus on diversity, variability and differences (including tiny ones), i.e., work on big cultural data without aggregation and with attention to what is infrequent and outliers?