When researchers share data, their empirical results become more reproducible and more reusable. This, in turn, can accelerate progress while enhancing accountability and transparency. This grant supports efforts by Gary King and Mercи Crosas of the Institute for Quantitative Social Science (IQSS) at Harvard University to facilitate data sharing through continued development of the Dataverse Network (DVN), a leading Harvard-based data repository.  Working with scientists, technologists, and academic publishers, King and Crosas have launched an ambitious project to help academic journals make data submission a fully integrated part of the paper submission process, using the Dataverse infrastructure to store and manipulate data submitted by authors.  Grant funds will support several activities aimed at expanding and improving Dataverse, including convening workshops and conferences with stakeholders to develop uniform standards and protocols, crafting an application programming interface, and developing several “data widgets” that allow real-time manipulation of data uploaded to the system.

The meaning of mathematical formulae depends on how they are represented and displayed. Mathematical symbols have to be arranged and ordered precisely, lest the meaning of formulae change completely. A poorly placed line break can render a mathematical expression incoherent.  The rise of the internet has made this problem acute.  Major browsers developed by Apple, Google, and Microsoft do not support mathematical content.  As more and more content is accessed on screens, tablet computers, and smart phones, mathematicians need a tool that can rearrange mathematical expressions dynamically without distortion of meaning.  This grant funds efforts by a consortium led by the American Mathematical Society (AMS) to “semantically enrich” MathML, a markup language used by the popular, open source MathJax platform.  The AMS team aims to further develop the MathML language, allowing it to encode information about the meaning of mathematical expressions and how to display them.  If successful, the project would eventually allow browsers to treat mathematical expressions not as uninterpreted strings of symbols, but as contentful expressions whose meaning must be preserved across changes in display.

This grant provides continued support to George Mason University’s Roy Rosenzweig Center for History and New Media for the continued development of PressForward, a software platform that facilitates the creation of “overlay journals,” curated collections of scholarly materials whose contents are drawn not from original submissions, but from existing academic sources.  PressForward journals have the ability to draw material not only from existing online journals, but from the rich landscape of reputable working paper repositories like SSRN, rapid publication venues like PLoSONE, preprint repositories like arXiv, and the untidy world of blogs, posters, and other gray literature. Previous Sloan grants supported the initial development of PressForward and its deployment to a handful of pilot sites.  Funds from this grant support the expansion of the platform, the hiring of an outreach specialist to give presentations and handle online engagement, increased help desk capacity, a summer institute to train potential users, and additional software development as determined through user needs.

Developed by the California Digital Library (CDL), which serves the entire University of California system, the Data Management Plan (DMP) tool is an open source software platform that allows UC researchers to create and implement data management plans, which are an increasingly ubiquitous requirement of government and private funding for scientific research.  The existence of such platforms reduces the barriers to data sharing, allowing scientists to make their data permanently available in accordance with funder requirements without having to invest significant time, effort, or other resources in the process.  Funds from this grant will allow the CDL, which operates out of the Office of the President, to launch and implement a redesign of the user interface of the Data Management Plan tool.  Using detailed user feedback that is the norm in much for-profit software development, the CDL team will redesign its primary interface using detailed user-experience testing, letting the needs and competencies of actual users drive how the interface works. The result will be a lightweight open source software application that would be accessible initially to the thousands of scientists and researchers employed throughout the University of California system, but which will be generalized enough that it could, in principle, sit in-between users and any data repository.

Though the Sloan Foundation has funded several initiatives to make the citation of data a regular, established practice in science, data citation is itself unidirectional.  In a properly cited scientific article, the reader will know what datasets are being referenced and used, but the creator or curator of those cited datasets may have no way to know his or her data is being cited.  Yet knowing how a dataset is being used and by whom can be a crucial factor in making decisions about its value, how to extend it, and how to increase its usefulness.   This grant supports work by Sayeed Choudhury, associate dean for research data management at Johns Hopkins University, to develop a third-party service called “Matchmaker” that would independently map the relationships between articles and data, linking between existing publishing platforms and data repositories.  These relationships could be created by a number of different stakeholders in the scholarly communication process:  by a publisher, by a data archive, by an individual researcher, or even by a library.  When fully developed, these relationships would then form a "graph" that could be queried without having to repeatedly poll every repository and publisher, a complement to more traditional citation services like ISI or Google Scholar.

In 2013, a White House Office of Science & Technology Policy directive outlined new open-access expectations for research products funded by the federal government.  One question left open by the directive, however, is how exactly those materials should be managed and made discoverable, particularly for the long term.  Funds from this grant support a project by the Association of Research Libraries (ARL) to facilitate compliance with the OSTP directive by developing a platform for reporting and notifying parties of events related to the release of publicly and privately funded research.  Partnering with the Association of American Universities and the Association of Public and Land-Grant Universities, the ARL will create a multi-institutional platform, the SHared Access Research Ecosystem (SHARE), that will tie together existing university-based institutional repositories into a coherent discovery and compliance tracking system.  When completed, SHARE will function as connective tissue that will enable others to build user-facing services that build on the multi-institutional architecture, leveraging university investments in their own institutional repositories and providing a valuable resource to help university offices of sponsored research meet their reporting and compliance-tracking obligations.

The rise of the Internet and digitally enabled means of disseminating scholarly research has led to a burgeoning interest in “altmetrics,” alternative measures of the impact and importance of scholarship that extend beyond traditional measures like citation counts. Funds from this grant support efforts by Vincent Lariviere and Stefanie Haustein of the University of Montreal and their colleague Cassidy Sugimoto of Indiana propose to dig deeper into the relative value and meaning of two specific altmetric indicators:  social media tweets and “saves” by popular bibliographic reference manager platforms.  The researcher team will match bibliographic and citation data from the Web of Science (linked with the same articles as they appear in PubMed and arXiv) with these forms of altmetric activity in order to answer a set of questions about the relationship between altmetric signals and the ultimate impact of a given work as traditionally measured in citation.  Particular focus will be given to the relationship, if any, between initial attention paid to preprints or working papers and the subsequent citation of formally published versions of those same papers.

This grant provide 14 months of support for the Hypothes.is Project, a web annotation platform that aims to bring granular annotation of online scholarly materials to users through the development of an easy-to-use interface that makes web annotation fully collaborative, shareable, and searchable.  Grant funds will support continued development of the Hypothes.is platform as well as three pilot implementations, one at the American Geophysical Union, one at the arXiv preprint repository, and one at eLife, an influential online journal sponsored by the Wellcome Trust and the Howard Hughes Medical Institute.  Additional funds support a 2014 summit for Hypothes.is stakeholders to ensure compliance with current and forthcoming standards set by the World Wide Web Consortium.

Funds from this grant support a series of projects to increase the usefulness of the Isis Bibliography of the History of Science, the oldest and largest bibliography in its field and an invaluable resource to historians of science worldwide.  Using grant funds, Isis Bibliographer Stephen Weldon and his team will spearhead a series of initiatives designed to bring the ISIS Bibliography more fully into the digital era, including retrospective digitization, data extraction, and cleanup of the existing bibliography; the development of new researcher-facing tools and interfaces; “community-sourced” mechanisms for maintaining the bibliography going forward; and a mini-grant program to incentivize novel or innovative ways of utilizing the bibliography as a scholarly resource.