As we near the end of 2015, there has been a flurry of activity coming from U.S. federal agencies in the Public Access policy arena. Late last week, the Department of Transportation (DOT) released its plan for ensuring public access to articles and data resulting from its funded research, laying out a comprehensive framework to ensure access and productive reuse of its funded research outputs. And on Monday, the publisher-based CHORUS initiative issued a press release touting a new agreement with National Science Foundation (NSF), designed to supplement its current Public Access plan for its funded articles.
Let’s take a look at DOT’s plan first – there is a lot that’s new in this one.
DOT’s Comprehensive Research Life-Cycle Approach
The DOT’s plan strikes a slightly different chord than many of the other plans released to date, weaving together existing practices for publication and data sharing, with some new enhancements that should allow the agency to more efficiently track its research at a project level throughout the research life cycle.
DOT’s plan outlines a framework for tracking its funded research from project initiation to the generation of research outputs and products, treating each step as integral to the research process. The agency will establish new terms and conditions for its funded research that will require new strong licensing requirements for articles, mandatory use of Digital Object Identifiers (DOIs) for articles and data sets, as well as the requirement for researchers to secure and use a unique ORCID ID for all results submitted to DOT and for publication.
DOT Plan for Articles: Local Deposit and Management
The DOT’s plan calls for all DOT-funded researchers to deposit their final peer-reviewed manuscripts into the National Transportation Library (NTL) digital repository upon acceptance in a peer-reviewed journal, and to make them available to the public with no longer than a 12-month embargo period.
As with the other agencies, DOT will provide stakeholders with a mechanism for petitioning the agency to shorten or extend the allowable embargo period. Unlike other agencies, the DOT will limit any changes to no more than six months in either direction. The DOT is also unique among agencies in announcing that they are developing an online mechanism that will allow any member of the public to petition for an embargo period change, noting that decisions will be provided within one month of such a request.
The DOT stands out from other agencies in explicitly noting that will take a very proactive stance on copyright. The agency intends to establish new terms and conditions for all DOT funding agreements that require both the grant of a comprehensive, non-exclusive, paid-up, royalty-free copyright license to the DOT, and the submission all publications to the NTL digital repository. This puts the agency in a very strong position to be able to meet the OSTP Directive’s requirement to enable productive reuse of their research outputs.
Despite this requirement, the DOT still seems to have some hesitation around allowing bulk downloads of articles, and indicates that it will explore taking a tiered approach, distinguishing between “General Users” (allowed limited downloading and restricted crawling, but unlimited text and data mining rights) and “Qualified Users, ” (allowed mass downloads, and unlimited crawling and text and data mining rights). It will be interesting to see the agency’s calculus on the overhead required to maintain these distinctions versus the potential risks/returns of simply considering all users “Qualified.”
The DOT expresses a strong commitment to requiring the use of DOIs, ORCID IDs and funding tracking numbers for all of its research outputs. This should give the agency an advantage when it comes to compliance monitoring, as these indicators can be tracked in a variety of different indices, including the Transportation Research Board’s TRID database, FundRef, and also potentially by the SHARE database.
DOT’s Plan for Research Data: Building on an Existing Culture of Data Sharing
The DOT has a strong track record of making data generated by its intramural researchers accessible to the public1, and its Public Access plan builds on this foundation. The agency will join all other U.S. federal science agencies that have announced their plans in requiring that all extramural investigators requesting funding submit a Data Management Plan (DMP) outlining plans for managing and providing access to research data, or provide a rationale as to why their research cannot be made available. This continues the trend of effectively setting the default mode for federally-funded research data to “open.”
DOT’s DMP requirements include some fairly standard elements, such as descriptions of the data to be collected, preservation strategies, description of standards and machine-readable formats used for data collection and storage, and descriptions of any applicable protections to be used for purposes of privacy, confidential business information, national security, etc.
However, they also have some fairly unique requirements. DOT will require DMPs to include a section addressing reuse, redistribution, and creation of derivative products from DOT data, and they point to the strong preference given in OMB M-1313 for use of Creative Commons licenses as a guideline. The DOT plan states that they have a strong preference for the use of CC-BY or equivalent license on data generated by extramural researchers, and a public domain dedicated marking for data generated by intramural researchers.
All DOT datasets will be required to carry a DOI or to be deposited into a repository that provides DOIs. They are also exploring dataset identification frameworks, and are evaluating the use of DataCite, Data-Pass and the Data Document Initiative.
The DOT places a strong emphasis on making the data underlying its funded peer-reviewed articles freely available at the time of publication and effectively linked to the articles themselves. The robust use of identifiers combined with the twin requirements that researchers deposit their articles locally into the NTL and their data into an openly accessible repository will go a long way in achieving this, but the DOT plan takes this one step further.
As noted earlier, the DOT is using the Public Access requirement as an opportunity to improve its tracking of its research portfolio at the project level. They plan to connect several existing internal databases, including the Research-in-Progress (RiP) data base, the Research Hub database and the NTL Digital Library, to provide an environment that seamlessly tracks and links research projects from inception to completion. Ultimately, the agency will be able to point to one publicly-accessible record for each of its funded projects that contains a full description of the funding and project, along with links to any data or publications generated from the research, improving the transparency and accountability of the agency.
NSF Announces New Agreement with CHORUS
Along with the release of the detailed DOT plan, this week saw another notable recent development in the arena of U.S. public access policies. On Monday, the National Science Foundation officially signed an agreement with CHORUS (the publisher-backed Clearinghouse for the Open Research of the United States). CHORUS will support NSF’s existing partnership with the U.S. Department of Energy (DOE), by providing distributed repository and search services – searching the NSF Public Access Repository (NSF-PAR) that will be hosted by the DOE, and following links that point back to articles hosted on publisher websites.
This approach—linking out to the full text of articles residing on individual publisher websites— is in contrast to the DOT’s comprehensive research portfolio enhancement approach. While the use of CHORUS does facilitate the location of NSF-funded articles, questions persist.
For example, with articles on hundreds of individual, proprietary publisher websites, each with their own unique technologies and legal restrictions, how will productive reuses, such as computation and text and data mining be enabled? Will uniform rights be granted to agencies and to end-users without individual negotiations for specific re-uses? How will data be effectively linked to articles? How will the agency leverage the full collection of articles reporting on its funded research to further its scientific mission and ensure accountability when they are dependent on a third party for access to the full text of their articles?
As the last of the U.S. federal agencies report back on their plans for complying with the White House OSTP Directive on Public Access, these are important questions to keep in mind. Our community should consider whether the proposed solutions from U.S. agencies actually achieve the admirable aims of the Directive, and what actions we can take to play a positive role in ensuring that they are successful.
[1] Note: DOT notes that its intramural researchers are already covered by the requirements set out in the White House Executive Order on Open Government Data and OMB M-1313 (Making Open and Machine Readable the Default for Government Data) along with DOT Order 1351.34, (The Department’s Data Release Policy). The agency indicates that they believe these policies together effectively achieve the aims of the OSTP Public Access Memo.