Dataverse
Description
The Dataverse[1] is an open source (code is available on GitHub) web application to share, preserve, cite, explore and analyze research data. It facilitates making data available to others, and allows you to replicate others' work.[2] Researchers, data authors, publishers, data distributors, and affiliated institutions all receive appropriate credit via a data citation with a persistent identifier (e.g., DOI, or Handle).
A Dataverse repository hosts multiple dataverses (see diagram). Each dataverse contains dataset(s) or other dataverses, and each dataset contains descriptive metadata and data files (including documentation and code that accompany the data - see diagram).
Provider
Institute for Quantitative Social Science at Harvard University, along with many collaborators and contributors worldwide.
Licensing and cost
Apache 2 License – free.
Development activity
Version 3.0 was released in May 2012. The current version (in August 2013) is 3.5.1. The software is continually development, as revealed by an active issues tracking page. The project is Harvard-sponsored, and appears to have support for the foreseeable future.
Platform and interoperability
The Dataverse Network makes use of the following components: Java Server Faces (JSF2); Enterprise Java Beans (EJB3); PostgreSQL; Lucene; and R and Zelig. Prerequisites for installation include Sun/Oracle Java JDK 1.6+, a “virgin” installation of Glassfish v2.1+, preferably as part of the NetBeans Web Development bundle, PostgreSQL v8.3+, and R. The software was designed to integrate reCAPTCHA, Google Analystics, ImageMagick, Handle, and LOCKSS if the installer so wishes.
Functional notes
DataVerses can be configured for two levels of access. An Open DataVerse allows all registered users to edit their own studies; a Wiki DataVerse allows users to edit all studies, although only a curator or administrator can release the changes. A DataVerse will accept any format, but will only give full support to tabular data. SPSS and STATA are the preferred formats; data in these formats will be eligible for subsettable features, multiple formats for download, and a Universal Numerical Fingerprint (UNF). GraphML is recommended for network data, and will be eligible for subsetting and pre-defined measurements. A DataVerse Network has the capability to be registered as a handle.net server, which allows the Network to assign persistent identifiers to data sets.
Documentation and user support
The website contains extensive software documentation, including user, installer, and developer guides. A new Users google group appears to be reasonably active. Aside from a webform, contact information is not prominently advertised.
Usability
The DataVerse software provides a web-based interface for both administrators and users. The package includes an installer, which is run through the command line; basic install is designed to be very quick. Comfort with command-line interface and general systems knowledge appear to be crucial for configuration and installation of any add-ons.
Expertise required
To take full advantage of the archival management features in the software, users should have a firm grasp on the metadata expectations for their field.
Standards compliance
The software supports numerous metadata standards, exporting as XML records in DDI, Dublin Core, FGDC, and Marc formats. The software is Z39.50 and OAI-PMH compliant, and has the ability to register information to Handle.net.
Influence and take-up
Current installations include Dataverse Networks at Harvard IQSS, ICPSR, the University of the Thai Chamber of Commerce, and the Utrecht University Library. The software's sourceforge page reports nearly 7000 downloads.
User Experiences
Development Activity