Data Management Plans and Data Centers

Data Management Plans and Data Centers Denise DiPersio, Christopher Cieri, Daniel Jaquette Linguistic Data Consortium. University of Pennsylvania 3600 Market Street, Suite 810, Philadelphia, PA 19104 USA {dipersio, ccieri, jaquette}@ldc.upenn.edu Abstract Data management plans, data sharing plans and the like are now required by funders worldwide as part of research proposals. Concerned with promoting the notion of open scientific data, funders view such plans as the framework for satisfying the generally accepted requirements for data generated in funded research projects, among them that it be accessible, usable, standardized to the degree possible, secure and stable. This paper examines the origins of data management plans, their requirements and issues they raise for data centers and HLT resource development in general. Keywords: language resources, data distribution, data management plans 1. Introduction Data Management Plans (DMPs) and their equivalents have become familiar to researchers developing language resources over the past several years. Many funding agencies around the world now require that research proposals include a specific section detailing how resulting data will be created, shared and maintained. For instance, the US National Science Foundation’s (NSF) Data Sharing Policy states: Investigators are expected to share with other researchers, at no more than incremental cost and within a reasonable time, the primary data, samples, physical collections and other supporting materials created or gathered in the course of work under NSF grants. Grantees are expected to encourage and facilitate such sharing.1 It is the US government’s intention to broaden such sharing mandates by requiring federal agencies of a certain size to ensure that direct results of federal funding are publicly available.2 Funders’ guidelines on DMPs encourage or even require behavior that is not universal in the research community, stirring debate over the best ways to comply. In some instances, requirements are deliberately unclear, leaving it to the community to reach a resolution. For example, DMPs may suggest or require that language resources (LRs) are distributed at some minimal cost. Knowing though that data management costs cannot be completely eliminated, data centers must now identify new funding strategies. This paper describes different sets of DMP requirements principally in the United States, discusses issues relating to their implementation and assesses their potential impact on the availability of LRs. 2. Data Distribution in the Pre-DMP World The majority of LRs currently available were created before or outside of the influence of DMPs though the motivations for DMPs – broad and affordable access to 1 NSF Data Sharing Policy, http://www.nsf.gov/pubs/policydocs/pappguide/nsf15001/aag_6. jsp#VID4 digital data – are long-standing. Data centers have approached this goal in multiple ways: low cost or subsidized corpora, limited free distributions under an author-pays or funder-pays model and grants in data subsidized by the community. For example, LDC distributes several data sets at no cost either because the funding agency paid in advance for a number of copies to be distributed or because the author and LDC agreed to share costs: American National Corpus (Reppen, Ide, Suderman, 2005), AQUAINT (Graff, 2002), FORM Kinematic Gesture (Martell et al., 2004), …

Content truncated to 500 words