Legal Framework

There are many legal issues with regard to the reuse of Language Resources (LR), particularly in Machine Translation (MT) and Machine Processing (MP) settings. As a result, the QTLaunchPad consortium has provided a simple report on permitted acts that makes key policy suggestions for relevant legislative bodies.

LR processing requires the use and re-use of information of various kinds, in a variety of ways, by different types of organisations and, as a result, it involves a wide range of legal regimes. The primary issues are related to Intellectual Property Rights (IPR), but may also involve Personal Data Protection, Public Sector Information and Geodata Regulation.

In order to assess the conditions under which LR re-use in MT and MP may lawfully take place, we need to appreciate the acts it involves, the degree to which they are regulated by different types of laws and the permissions that someone needs to obtain in order to perform such acts. The report starts with the key methodological premises, explains how the flows of rights are to be treated in a paradigmatic MT & MP scenario, and what the key emerging issues are. It presents the core concepts of the main legal regimes that influence the LR-based MT & MP and elaborates on the key legal issues featuring in each of the different kinds of legal regimes. It also presents the various legal issues related to LR-based MT & MP in terms of (See Figure I):

  • IPR issues
  • Data Protection Issues
  • Public Sector Information Issues
  • Confidentiality and Other types of rights Issues

It then proposes different ways in which LR-based MT & MP are to be treated, exemplifying them by a set of use cases that may be used as prototypical scenarios illustrating different aspects of the use of LR for MT/MP purposes.

The full report (“Legal framework of textual data processing for Machine Translation and Language Technology research and development activities”) is available online as a wikibook.