'A right to 'read' for machines: assessing a black-box analysis exception for data mining'

摘要

This panel looks into the impact of the current copyright framework in the European Union on text and data mining (TDM) and discusses the impact of introducing a TDM exception in EU copyright law. A design of this exception is proposed for the panel, and is partially based on findings in the Horizon 2020 FutureTDM project. This project aims to improve uptake of text and data mining (TDM) in the EU and, in that regard, has studied the legal barriers to TDM and will be developing and recommending a policy framework in the future. Part of this policy framework will consist of possible actions to be undertaken by the - European and national - legislators. A TDM exception is considered to include in the recommendations and we therefore broach the topic to discuss the possible legal, economic and practical impact of such an exception with experts from the field. The TDM exception, as proposed for this panel, is inspired by the "black-box analysis" exception from the Software Directive, which allows lawful users of a program to perform any of the acts of loading, displaying, running, transmitting or storing the program to "determine the ideas and principles" underlying it. The authors of the panel believe that this underlines the general principle of copyright law: namely, that ideas and facts are not protected. Therefore, proposition to be discussed is that a similar exception should be introduced for copyright law in general, that would allow reproductions to be made of works for the sole purpose of extracting facts and ideas underlying them. This would allow TDM activities, where machines 'read' lawfully accessed works just as the human reading of works does not require further authorization from the copyright holder.