Developing MicroPasts: methodological, technical and ethical challenges

Developing MicroPasts CAA, 24 April 2014 Chiara Bonacchi UCL Institute of Archaeology with Andrew Bevan (UCL) Daniel Pett (British Museum) Adi Keinan-Schoonbaert (UCL) Crowd-sourcing in archaeology • Crowd-sourcing as – the practice of seeking information, services or funds in small chunks from large groups of people, over the internet (definitions discussed in Dunn and Hedges 2012) • Increasingly explored for supporting public audiences interaction with archaeology – Participation in research – Participation in micro-financing Crowd-sourcing projects • Diverse range: – Inspecting imagery for archaeological features – Transcribing papyri – Interrogating built architecture – Public recording of metal finds… Crowd-funding projects • • • Ranging from excavations to dissertations Using existing or new dedicated platforms Varying degrees of success Observations • • Mainly contributory models Little evaluation done Contributory Collaborative Co-creative Hosted Models for participation: Public Participation in Scientific Research Project, further elaborated on by Simon Introducing MicroPasts • • A collaboration between UCL and the British Museum Funded by the AHRC, Digital transformations in community research co-production Aim • To develop and test an online space where mixed groups of archaeological enthusiasts collaborate to: - produce innovative open datasets via crowd-sourcing (e.g. CC0, CC-BY) - develop new research projects into archaeology, history and heritage (sometimes involving crowd-sourcing) - micro-fund those new collaborative projects via crowd-funding Launch Day! April MicroPasts website micropasts.org Component 1: crowd-sourcing platform crowdsourced.micropasts.org Crowd-sourcing applications • • 4 applications Focused on British Prehistory Help cataloguing • 30,000 index cards of all known Bronze Age metal artefact finds in the UK from 1800 to 1983 Transcription and geo-referencing Help creating 3D models • 3D SfM models of palstaves recorded in the British Museum Bronze Age Index (Photoscan) Image masking Component 2: community forum community.micropasts.org • • Discourse (https://github.com/discourse/discourse) For research and platform co-design Component 3: crowd-funding platform • Neighbor.ly (https://github.com/neighborly/neighborly) • Catarse (https://github.com/catarse/catarse) • Micro-funding of projects codesigned via the forum or externally • 3 seed projects initially – London’s Lost Waterway – Mapping waterway sites, and transcribing relevant documents Evaluation: aims • How do online communities of interest in the human past form and develop through the MicroPasts platforms? • How do different contributors engage with archaeology and the past via the MicroPasts platforms, through time, and what is the value of that engagement for community members including institutions? • What is the sustainability of the MicroPasts platforms, and the applicability of a similar model in other countries? Evaluation: methodology • Approach – Quantitative / qualitative – Focus on MP platforms and social media / control cases amongst target audiences – Online / offline – Link info on: contributors profile, opinions, behaviour; data produced; their re-use – Taking time into account • Methods (at different stages) – – – – – – Online surveys Talks / meet-ups Google analytics Pybossa statistics Diary study Social media data analysis (cultural interests and practices) – Text analysis and SNA (Very!) initial data: 16-23 April Coming from… • Entry survey on completion of first crowd-sourcing task – Circles , whether works with archaeology / history as part of main job, age, email • Forum, Google analytics, Pybossa statistics (Very!) initial data: 16-23 April Suggesting that… • Community building: – 195 registered members – UK and US focus – 55% of respondents not working in archaeology / history – 24% within our immediate network • Engagement: – Photo masking: prevalent entry application – Transcription: fewer people, but the most dedicated ones • Number of tasks • Anonymous contributors: 23-35% on transcription, 53% on masking – Need for more guidance / visualisation Next steps • Guidance and purposiveness – [3D model viewer] – Information on the Bronze Age Index / British prehistory / 3D modelling – Development of badges • Tasks – Transcription of discovery cards • Crowd-funding platform Challenges • Time in relation to the complexity of the platform – Long development times and little space for front-end evaluation – Need to co-design the platform as we go along / challenge of not losing users in the process • Adoption of new funding practices within a university environment (crowd-funding) • Planning an evaluation that – Does not disrupt people s engagement – Is discrete but open and ethically compliant • Being ready to adapt the evaluation plan in response to people s interaction while maintaining coherence Developing MicroPast Thank you! [email protected]