权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

EAGER: DMP Roadmap: Making Data Management Plans Actionable

EAGER：DMP 路线图：使数据管理计划切实可行

基本信息

批准号：
1745675
负责人：
Guenter Waibel
金额：
$ 27.52万
依托单位：
University of California, Office of the President, Oakland
依托单位国家：
美国
项目类别：
Standard Grant
财政年份：
2017
资助国家：
美国
起止时间：
2017-09-01 至 2021-08-31
项目状态：
已结题

来源：
https://www.nsf.gov/awardsearch/showAward?AWD_ID=1745675&HistoricalAwards=false
关键词：
EAGER DMP Roadmap Making Data

项目摘要

The California Digital Library (CDL) will work with an extensive coalition of national and international collaborators to convert static data management plans (DMPs) into machine-actionable documents useful for structuring the course of research activities and communicating with other systems. Open data policies are proliferating worldwide and researchers are now required to submit DMPs with most grant proposals that describe the data they will produce and plans for sharing and preserving it. Researchers do not always know exactly what data they will produce at the beginning of a project, however. Furthermore, they have no incentives or easy methods for updating a DMP to keep things organized over the course of their research, which can lead to poor data practices and chaotic, unusable data shared at the end. DMPs in their current, static form pose similar challenges for other stakeholders across the research ecosystem, e.g., funders who must monitor compliance manually. This project will solve this problem by building out DMPRoadmap, a new, internationalized platform to reposition DMPs as true hubs of the networked research ecosystem. The principal impact of the proposed project is to transform one component of the increasingly digital research enterprise, the DMP, into an actively updated and machine-actionable hub to document and disseminate the products of research activity. Converting free-text responses to funder requirements into dynamic, verifiable data feeds will vastly improve the entire toolchain for all stakeholders and increase the velocity and availability of information across all disciplines. The software outputs of the pilot will be shared publicly with an open source license so that the community can continue to enhance and reuse the technology, extending the scholarly cyberinfrastructure in a scalable manner. All supporting data collected during the project will be made fully available for human and machine consumption. Since DMPs are rapidly becoming a global phenomenon, the outputs of the project will be of great interest to the entire research community. The breadth of the impact extends into the future of DMP policies worldwide, as machine-actionability makes change easier and continuous improvements possible. The project will provide insight into how to maximize investment in institutions and infrastructures in a manner that achieves policy goals by providing public access to government-funded research, and advances the overall public good. By advancing best practices for data-intensive research across all disciplines the project will further the goals of the scientific endeavor.The CDL will design, develop, and prototype machine-actionable DMPs based on community generated use cases; field research on DMP and data management practices; and pilot projects conducted with disciplinary and institutional partners. The work plan encompasses iterative phases of developing, implementing, and testing with various stakeholder groups using an agile methodology. Specific use cases include implementing a set of common standards and exchange protocols for DMPs to enable information to flow between DMPs and existing research information systems (e.g., offices of research, data repositories, faculty profile systems); leveraging persistent identifiers (e.g., DOIs for articles and datasets, ORCID iDs for people) to trigger push/pull notifications across systems that enable stakeholders to plan resources, connect research outputs, automate reporting and monitoring, get credit, and promote data discoverability, reuse, and reproducibility; among others.

加州数字图书馆(CDL)将与国家和国际合作者的广泛联盟合作，将静态数据管理计划(DMP)转换为机器可操作的文档，有助于构建研究活动的过程并与其他系统进行沟通。开放数据政策正在全球范围内激增，研究人员现在被要求提交带有大多数拨款建议的DMP，描述他们将产生的数据以及共享和保存这些数据的计划。然而，研究人员并不总是确切地知道他们将在项目开始时产生什么数据。此外，他们没有动机或简单的方法来更新DMP以在他们的研究过程中保持事物的有序，这可能导致糟糕的数据实践和最终共享的混乱、不可用的数据。目前静态形式的DMPS对整个研究生态系统的其他利益攸关方构成了类似的挑战，例如必须手动监测遵从性的资助者。这个项目将通过构建DMPRoadmap来解决这个问题，DMPRoadmap是一个新的国际化平台，将DMP重新定位为网络研究生态系统的真正枢纽。拟议项目的主要影响是将日益数字化的研究企业DMP的一个组成部分转变为一个积极更新和机器可操作的中心，以记录和传播研究活动的产品。将对资助者要求的自由文本响应转换为动态的、可验证的数据馈送，将极大地改善所有利益攸关方的整个工具链，并提高所有学科的信息的速度和可用性。试点的软件成果将通过开放源码许可证公开共享，以便社区可以继续增强和重复使用该技术，以可扩展的方式扩展学术网络基础设施。在项目期间收集的所有辅助数据将完全可供人和机器使用。由于DMP正在迅速成为一种全球现象，该项目的成果将引起整个研究界的极大兴趣。影响的广度延伸到全球DMP政策的未来，因为机器可操作性使变化更容易，并有可能持续改进。该项目将深入了解如何最大限度地增加对机构和基础设施的投资，通过向公众提供政府资助的研究来实现政策目标，并促进整体公共利益。通过在所有学科推进数据密集型研究的最佳做法，该项目将推动科学努力的目标。CDL将基于社区生成的用例设计、开发和制作机器可操作的DMP的原型；DMP和数据管理实践的实地研究；以及与学科和机构合作伙伴开展的试点项目。工作计划包括使用敏捷方法与不同的涉众团队进行开发、实现和测试的迭代阶段。具体使用案例包括为DMP实施一套通用标准和交换协议，以使信息能够在DMP和现有研究信息系统(例如，研究室、数据仓库、教师档案系统)之间流动；利用持久标识符(例如，文章和数据集的DOI、人员的ORCID ID)触发跨系统的推送/拉入通知，从而使利益相关者能够规划资源、连接研究成果、自动化报告和监控、获得信用，并促进数据的可发现性、重用性和再生性；等等。