Project: The Copyright Evidence Wiki
How can policy makers and academics engage with the growing body of empirical work on the effects of copyright law? The Copyright Evidence Wiki addresses this challenge by cataloguing and indexing studies in an open, transparent and accessible resource. The project was undertaken by the RCUK funded CREATe centre, and further supported by a grant from the Impact Acceleration Account (IAA) at the University of Glasgow. The current phase of the Copyright Evidence Wiki is supported by AHRC Follow-on Funding for ‘Unlocking Co–Creative Possibilities‘ (AH/P013341/1).
Investigators: The Impact Acceleration Account award was held by Theodore Koutmeridis (PI) and Kris Erickson (Co-I). From 2014 to 2017, the Copyright Evidence Wiki was developed by Theo Koutmeridis (lead editor), Kris Erickson and Martin Kretschmer. Research assistants helped with coding entries: PhD candidates with CREATe included Kenny Barr, Megan Blakely, Jaakko Miettinen, Victoria Stobo and Andrea Wallace. From September 2017, Amy Thomas (CREATe PhD student) joined the project as Sub Editor. An Editorial Board was formed (see project page here).
What did your research aim to do?
Prior to the Copyright Evidence Wiki, there was no central source for accessing empirical studies of copyright. We aimed to create such a resource to facilitate evaluation and comparison of all existing empirical work, across a range of copyright subject matter (publishing, music, broadcast, performance, etc.) and across a range of policy concerns (unauthorised copying, exceptions, artists’ earnings, scope of protection, etc.). It may ultimately be impossible to gather the complete picture of academic studies (obvious limitations include the choice to focus on English-language studies and the use of a snowball sampling method for practicality). We currently have over 600 studies indexed in the Wiki, and we are actively exploring ways to expand the selection. For example, we are working with librarians at the University of Illinois to develop a further sample of empirical studies not yet included in the Wiki.
How did you do it?
In order for this initiative to be useful to policy making, the Copyright Evidence Wiki needed to be transparent in terms of selection and evaluation methodology. We chose the MediaWiki platform because it met the needs of transparency (by maintaining a record of edits and changes to individual pages). MediaWiki is also open source and allows for a range of useful plugins and additions, such as graphical visualisation. Finally, the Wiki format leverages peer production to generate benefits such as improved fact-checking by multiple contributors and a wider range of perspectives from distributed contributors.
The first major challenge was to decide on which data to collect about individual studies. We needed enough information to make meaningful comparison possible, without overwhelming the user by reproducing entire studies. In consultation with a team of PhD student contributors from various departments at the University of Glasgow, we pilot tested and refined the list of variables (see an example entry page here). The most significant variables relate to the data collection and analysis methods employed by studies, the sources of their data, the units of analysis used and the nature of the sample. Out dataset also permits comparison within and across different media types, over time. An initial source of inspiration for the Wiki was the comprehensive literature review study carried out by Watson, Zizzo & Fleming (2014). We wanted to extend and improve their dataset so that this type of work could continue to inform policy making.
What are your key findings?
One key finding emerged from a network visualisation of the Copyright Evidence Wiki developed by one of the teams at a 2016 Hackathon in Glasgow. By linking studies based on their references to one another (key related studies), it was possible to observe distinct clusters of research. Government reports seldom linked to academic work, but did tend to cross-reference one another. This suggested a shortfall in the scope of government commissioned reports on copyright policy.
A second finding relates to academic coverage of copyright subject matter. Sound recording and music are disproportionately represented in the sample. Enforcement and unauthorised copying dominate as key issues studied in this medium. Photography, although highly contested in recent copyright reform debates, has received far less academic attention. The Copyright Evidence Wiki can be used as a form of dynamic literature review to identify existing blind spots in research focus.
What impact has your work had so far/what impact do you anticipate it will have?
The project has been cited widely in professional and trade press (Kluwer IP Blog, IPKat, Infojustice.org). The Australian Productivity Commission used the Evidence Wiki as a basis for collecting evidence about the impact of copyright law for creative industries in a recent consultation
The Copyright Evidence Wiki was used for the 2016 EU Hackathon in Brussels, organised by N-Square Consulting in collaboration with Google. That event drew more than 30 young people from across Europe who competed during the 48-hour event on development of visualisation tools to enhance the Evidence Wiki.
The current phase of the Copyright Evidence Wiki is supported by AHRC Follow-on funding (AH/P013341/1). As part of that project, the Evidence Wiki is seeking new collaborations with policy makers and the public to explore how the dataset can be put to productive use in new and unanticipated ways.
Challenges encountered/Lessons learned
The Copyright Evidence Wiki is both a source of data for future research, and an action research project in its own right. As managers of the resource, we had to learn to work more effectively with the MediaWiki platform as well as understand how its unique semantic data structure operates. We also gained experience with distributed peer production in a contested policy area, a theme we plan to explore further in future research.
Are there additional/new research questions still to be answered in this area?
Data mining – the Copyright Evidence Wiki could provide a detailed account of the overall landscape of empirical knowledge of copyright, and its evolution over time. We think a quantitative analysis of the literature could prove valuable, by mapping the current state of academic research about copyright, to help guide future research in productive directions.
How has your association with CREATe helped to take things forward?
The Copyright Evidence Wiki draws heavily on human and technical resources from CREATe. We are grateful for the valuable input from PhD students who worked tirelessly to locate, evaluate and enter studies into the Wiki. We are also extremely grateful for the technical expertise provided by Andrew McHugh, Pete Bennett and Jesus Rodrigues-Perez. The Copyright Evidence Wiki project is an example of cross-disciplinary collaboration between economics, law, and computer science.