Compliance of National TDM Rules with International Copyright Law: An Overrated Nonissue?

Abstract

Seeking to devise an adequate regulatory framework for text and data mining (TDM), countries around the globe have adopted different approaches. While considerable room for TDM can follow from the application of fair use provisions (US) and broad statutory exemptions (Japan), countries in the EU rely on a more restrictive regulation that is based on specific copyright exceptions. Surveying this spectrum of existing approaches, lawmakers in countries seeking to devise an appropriate TDM regime may wonder whether the adoption of a restrictive approach is necessary in the light of international copyright law. In particular, they may feel obliged to ensure compliance with the three-step test laid down in Art. 9(2) of the Berne Convention, Art. 13 of the TRIPS Agreement and Art. 10 of the WIPO Copyright Treaty. Against this background, the analysis raises the question whether international copyright law covers TDM activities at all. TDM does not concern a traditional category of use that could have been contemplated at the diplomatic conferences leading to the current texts of the Berne Convention, the TRIPS Agreement and the WIPO Copyright Treaty. It is an automated, analytical type of use that does not affect the expressive core of literary and artistic works. Arguably, TDM constitutes a new category of copying that falls outside the scope of international copyright harmonization altogether.

Mineração de textos e dados na pesquisa em saúde: reflexões sobre direitos autorais

“…Alguns defendem que sequer seria preciso uma limitação para mineração de textos e dados 24 , 32 , por ser um uso não-expressivo 33 , 34 ; ou, porque o que se extrai das obras - padrões e elementos factuais, por exemplo - não integraria o escopo de proteção dos direitos autorais 24 , 32 , 34 . Contudo, múltiplos projetos já recorreram ao uso direto de conteúdo protegido por direitos autorais e causaram disputas judiciais, mesmo quando o material havia sido acessado legalmente 33 .…”

Text and data mining in health research: reflections on copyright

“…Some argue that it would not require limitations to text and data mining 24,32 , because it is a non-expressive use 33,34 or because what is extracted from works -factual elements and patterns, for example -would not be covered by the scope of copyright protection 24,32,34 . However, multiple projects have already resorted to the direct use of copyrighted content and caused legal disputes, even when the material had been legally accessed 33 .…”