Dissimuler ou disséminer ? Une étude sur le sort réservé aux résultats négatifs

Auteures/Authors : Marie-Emilia Herbet, Jérémie Leonard, Maria Santangelo, Lucie Albaret

Une enquête composée de 34 questions a été adressée à des chercheurs en chimie, physique, sciences de l’ingénieur et de l’environnement, en vue d’identifier leur rapport aux résultats de recherche infructueux ainsi que les freins et les leviers de leur diffusion.

L’étude se fonde sur 310 réponses complètes émanant de participants affiliés à des établissements de recherche et d’enseignement français. Menée dans le cadre du projet Datacc, porté par les bibliothèques universitaires de Lyon et Grenoble Alpes, engagées dans l’accompagnement des chercheurs à l’ouverture des données de recherche, notre étude permet de combler le déficit de données sur le sujet au regard des disciplines concernées.

Elle relève que 81% des chercheurs interrogés ont déjà produit des résultats négatifs pertinents et 75% se disent prêts à publier ce type de données. Pourtant, seuls 12,5% des répondants ont déjà eu l’occasion de le faire dans une revue scientifique. Ce contraste béant entre l’intention et la pratique soulève des interrogations sur les obstacles en présence et les solutions potentielles à apporter.

URL : Dissimuler ou disséminer ? Une étude sur le sort réservé aux résultats négatifs

Original location : https://hal.archives-ouvertes.fr/hal-03371040

Open Data Challenges in Climate Science

Authors : Francesca Eggleton, Kate Winfiel

The purpose of this paper is to explore challenges in open climate data experienced by data scientists at the Centre for Environmental Data Analysis (CEDA). This paper explores two of the five V’s of Big Data, Volume and Variety.

These challenges are explored using the Sentinel satellite data and Climate Modelling Intercomparison Project phase six (CMIP6) data held in the CEDA Archive. To address the Big Data Volume challenge, this paper describes the approach developed by CEDA to manage large volumes of data through the allocation of storage as filesets.

These filesets allow CEDA to plan and track dataset storage volumes, a flexible approach which could be adopted by any data centre. CEDA utilise the implementation of the Climate and Forecast (CF) conventions and standard names within archived data wherever possible to overcome the challenge of Variety.

Collaboration from the international science community through contributions to the moderation of CF standard names ensures these data then adhere to the FAIR (Findable, Accessible, Interoperable and Reusable) data principles.

Utilising data standards such as the CF standard names is recommended because it promotes data exchange and allows data from different sources to be compared. Addressing these Open Data challenges is crucial to ensure valuable climate data are made available to the scientific community to facilitate research that addresses one of society’s most pressing issues – climate change.

URL : Open Data Challenges in Climate Science

DOI : http://doi.org/10.5334/dsj-2020-052

A Discussion of Value Metrics for Data Repositories in Earth and Environmental Sciences

Authors : Cynthia Parr, Corinna Gries, Margaret O’Brien, Robert R. Downs, Ruth Duerr, Rebecca Koskela, Philip Tarrant, Keith E. Maull, Nancy Hoebelheinrich, Shelley Stall

Despite growing recognition of the importance of public data to the modern economy and to scientific progress, long-term investment in the repositories that manage and disseminate scientific data in easily accessible-ways remains elusive. Repositories are asked to demonstrate that there is a net value of their data and services to justify continued funding or attract new funding sources.

Here, representatives from a number of environmental and Earth science repositories evaluate approaches for assessing the costs and benefits of publishing scientific data in their repositories, identifying various metrics that repositories typically use to report on the impact and value of their data products and services, plus additional metrics that would be useful but are not typically measured.

We rated each metric by (a) the difficulty of implementation by our specific repositories and (b) its importance for value determination. As managers of environmental data repositories, we find that some of the most easily obtainable data-use metrics (such as data downloads and page views) may be less indicative of value than metrics that relate to discoverability and broader use.

Other intangible but equally important metrics (e.g., laws or regulations impacted, lives saved, new proposals generated), will require considerable additional research to describe and develop, plus resources to implement at scale.

As value can only be determined from the point of view of a stakeholder, it is likely that multiple sets of metrics will be needed, tailored to specific stakeholder needs. Moreover, economically based analyses or the use of specialists in the field are expensive and can happen only as resources permit.

URL : A Discussion of Value Metrics for Data Repositories in Earth and Environmental Sciences

DOI : http://doi.org/10.5334/dsj-2019-058

Facilitating and Improving Environmental Research Data Repository Interoperability

Authors : Corinna Gries, Amber Budden, Christine Laney, Margaret O’Brien, Mark Servilla, Wade Sheldon, Kristin Vanderbilt, David Vieglais

Environmental research data repositories provide much needed services for data preservation and data dissemination to diverse communities with domain specific or programmatic data needs and standards.

Due to independent development these repositories serve their communities well, but were developed with different technologies, data models and using different ontologies. Hence, the effectiveness and efficiency of these services can be vastly improved if repositories work together adhering to a shared community platform that focuses on the implementation of agreed upon standards and best practices for curation and dissemination of data.

Such a community platform drives forward the convergence of technologies and practices that will advance cross-domain interoperability. It will also facilitate contributions from investigators through standardized and streamlined workflows and provide increased visibility for the role of data managers and the curation services provided by data repositories, beyond preservation infrastructure.

Ten specific suggestions for such standardizations are outlined without any suggestions for priority or technical implementation. Although the recommendations are for repositories to implement, they have been chosen specifically with the data provider/data curator and synthesis scientist in mind.

URL : Facilitating and Improving Environmental Research Data Repository Interoperability

DOI : http://doi.org/10.5334/dsj-2018-022

Data Sustainability and Reuse Pathways of Natural Resources and Environmental Scientists

Author : Yi Shen

This paper presents a multifarious examination of natural resources and environmental scientists’ adventures navigating the policy change towards open access and cultural shift in data management, sharing, and reuse.

Situated in the institutional context of Virginia Tech, a focus group and multiple individual interviews were conducted exploring the domain scientists’ all-around experiences, performances, and perspectives on their collection, adoption, integration, preservation, and management of data.

The results reveal the scientists’ struggles, concerns, and barriers encountered, as well as their shared values, beliefs, passions, and aspirations when working with data. Based on these findings, this study provides suggestions on data modeling and knowledge representation strategies to support the long-term viability, stewardship, accessibility, and sustainability of scientific data.

It also discusses the art of curation as creative scholarship and new opportunities for data librarians and information professionals to mobilize the data revolution.

URL : https://arxiv.org/abs/1803.01788