DART Project Work Package CR 1 Tom Denison

DART Project Work Package CR 1 Tom Denison Stefanie Kethers Monash University.

DART Package CR 1 - Move data from personal data repositories to secure trusted alternatives § CR 1 – Move data from personal data repositories to secure trusted alternatives § Locate willing researchers with personal data repositories, preferably in the areas of the demonstrators (climate change, X-ray crystallography, humanities/social sciences) § Identify barriers to them depositing these datasets into secure institutional repositories § Develop strategies to overcome these barriers and test these with the researchers § Document the findings in a research report 2

Social Sciences and Humanities: Some Characteristics • Data can be text, video, audio, … § Open to interpretation § Not necessarily in digital format (e. g. , pots in archaeology) § Can be historically specific § Need for diverse formats, schemas, models, ontologies, meta data, … • Research subjects can be people § Use of data governed by ethics agreements § Consent management required – what happens if someone withdraws consent? § The research ‘subjects’ are (or should be? ) co-owners of the data • Access by different people with different backgrounds, fields, perspectives, for different purposes § Terminologies and models can differ § Meanings / semantics can be ambiguous (esp. across disciplines) • In some disciplines, there is not a long tradition of cooperation § How to foster a culture of collaboration and trust? 3

Consequences • Data reuse is difficult • Legal implications (e. g. subpoena-proofing in the context of the Trust & Technology project) • Long-term consent management • Context needs to be represented in sufficient detail • Meta data definition quite complex • Motivate researchers by providing trusted environment • The ACLS report (2005) suggests that the technical infrastructure, while complex, is less daunting than the ‘enabling layer’ made up of institutional policies, legal frameworks, standards, tools and services, and training of users. 4

Trust Issues • Researchers need to trust the repository and the organisation(s) who provide it • Show competence, benevolence, and integrity • Researchers need to be able to define access rights • Supported by legal frameworks & experts – it should not be a burden! • Parallels to health records management? • ‘Watchful distrust’ needs to be implemented • ‘Audit trail’ for attempts at access? • Sanctions for breaches of trust / access rights • Trust issues for ‘research subjects’, as well? 5