Academic content hosting and management platform Figshare on September 5 launched Figshare for Institutions, a cloud-based repository that enables universities and colleges to host datasets, papers, videos, and other research outputs and make this content publicly available.
The new platform offers researchers all of the tools available via individual Figshare accounts, including the ability to save some files in a private, Dropbox-style cloud storage folder, while making other files and datasets public, searchable, and citable using automatically assigned DataCite digital object identifiers (DOI). Any public content is free to upload and is made accessible to others under Creative Commons licensing terms. Figshare for Institutions allows universities to aggregate research outputs from their faculty at the departmental or institutional level and offers storage space for private collaboration, enabling faculty to work on files together prior to publication.
The concept for Figshare was sparked while founder Mark Hahnel was conducting research in stem cell biology for his PhD at Imperial College, London, in 2011.
“A lot of my raw data was in the form of videos—cells moving from one side of the screen to the other—or the raw datasets themselves, or graphs that were never going to see the light of day, either because they contained null data, or because they didn’t fit into a full story,” Hahnel told LJ. “I just wanted to find a place where I could make some of these research outputs available online in as simple a manner as possible…. Dissemination of content on the Internet has been solved—we’ve got YouTube for videos, Flickr for images—we came at it from that approach, thinking ‘surely there should be a very simple way to do this for academic outputs.’”
Figshare for Institutions addresses two emerging issues. First, many funding agencies are beginning to require institutions to make research outputs publicly available, not just the finished paper based on them, and these are often large collections of files. Hahnel’s own PhD research, for example, produced nine gigabytes of videos and datasets. Second, when that research is made public, academics, universities, and funding agencies are interested in quantifying the impact that these outputs are having on the researcher’s field.
“If you have a dataset that’s cited multiple times more than your top-cited paper, you should be able to get credit for that,” Hahnel said. Such a development could become a factor in tenure decisions as well as future funding applications, promotions, or awards.
To use the institutional edition, a university pays an annual licensing fee for a set amount of private storage space—ten terabytes, for example. Data is hosted using Amazon Web Services, which stores multiple redundant copies of files for security and stability. In addition, publicly available datasets and files are preserved via the CLOCKSS archive. Hosting fees for this amount of accessible storage will cost tens of thousands of dollars annually, but Hahnel notes that the more storage space a company licenses from Amazon, the less it costs per gigabyte.
“Because we already have publisher partners such as PLOS and Nature, we have a lot of content on Amazon Web Services already. So we can give you the whole service for less than it would cost for the storage” if it were purchased directly from Amazon by an individual institution.
The Figshare platform, meanwhile, keeps track of metrics such as citations and downloads. Hahnel explained that Figshare for Institutions is designed to share this data with altmetrics services provided by companies such as Plum Analytics and ImpactStory, and the service can help supplement existing institutional repositories, which are often better equipped to manage papers than other types of content.
“That Figshare accepts any file format and visualizes it in a browser is a big feature for us,” he said. “Some of these institutional repositories don’t deal very well with other types of content—videos and datasets. We do take papers, so we could take in everything and do the [institutional repository] job for them, but we appreciate that the institutional repository, as it stands, is a very important thing. We can push content to the institutional repository, we can pull content from it. The main thing we want to do is work with those librarians who have expertise in this area, and either make sure that they have one repository for papers, and this repository for everything else, or some combination of the two.”