Collaborative Research: EAGER: Characterizing Research Software from NSF Awards

软件是过去60年许多国家经济进步的基础，从提高生产率的更好的天气模型，到降低材料使用和成本的更好设计的产品。这些进步大多源于研究软件，最初是通过算法和许可，最近则是通过开源代码。尽管取得了如此多的成功，但在科学和工程研究中产生的许多软件是不可持续的——它们不是以能够有意义的重用的方式共享、维护或开发的。缺乏可持续发展阻碍了研究和经济进步。使用自我报告和管理数据的组合，该项目系统地调查了由国家科学基金会资助的研究项目中产生的软件的开发和维护。本研究收集的数据被用于开发一套研究软件可持续性规划的模型。这些模型支持科学家和工程师规划软件的可持续性；帮助研究资助机构在资助申请中更好地准备评估研究软件，并使支持软件开发和维护的机构能够支持有影响力的研究，从而产生有影响力的研究软件。该项目的目标是：1)通过收集资助研究项目的数据，了解影响软件可持续性的因素；2)描述当前的可持续性规划模型，并提出可能增加实现长期软件可持续性可能性的潜在新模型；3)开发评估研究软件可持续性的应急方法。为实现这些目标而收集的数据包括调查和采访研究人员，这些研究人员生产的软件是nsf资助的奖项的一部分。这个项目还使用新兴的方法来分析研究软件代码库，以便了解软件开发中的哪些活动与可持续性相关。本研究有三个预期的影响：1)为使用混合方法对研究软件可持续性进行大规模分析提供概念证明；2)修改现有的标准化软件健康指标，使其能够用于评估研究软件的可持续性；3)创建一套初始的可持续性模型，研究人员可以用它来更好地规划研究软件项目，资助者可以用它来做出更好的奖励决策，机构可以用它来更好地分配内部资源。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Software underlies many of the national economic advances of the last 60 years, from better weather models that enable more productivity to better-designed products that lower material usage and costs. Most of these advances originated in research software, initially via algorithms and licensing, and more recently via open-source code. Despite these numerous successes, much of the software produced in science and engineering research is not sustainable - it is not shared, maintained, or developed in ways that enable meaningful reuse. This lack of sustainable development hampers both research and economic progress. Using a combination of self-reported and administrative data this project is systematically investigating the development and maintenance of software produced in research projects funded by the National Science Foundation. The data collected for this study is being used to develop a set of models for research software sustainability planning. These models support scientists and engineers planning for software sustainability; help research funding agencies be better prepared to evaluate research software in grant applications, and enable institutions that support software development and maintenance to be equipped to support impactful research that in turn, produces impactful research software.The goals of this project are: 1) to understand what factors influence software sustainability by gathering data from grant-funded research projects; 2) to describe current models of sustainability planning and suggest potential new models that could increase the likelihood of achieving long-term software sustainability; and 3) to develop emergent methods to evaluate research software sustainability. Data collection to meet these goals includes a survey and interviews with researchers that have produced software as part of an NSF-funded award. This project also uses emerging methods in analyzing research software code repositories in order to understand what activities in software development correlate with sustainability. This research has three intended impacts: 1) to provide a proof of concept for large-scale analysis of research software sustainability using a mixed-methods approach; 2) to modify existing standardized metrics of software health so that they can be used to evaluate the sustainability of research software; and 3) to create an initial set of sustainability models that researchers can use to better plan research software projects, funders can use to make better award decisions, and institutions can use to better allocate internal resources.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.