Project Overview

Equinox Partnership

Equinox is the successor of the data delivery system formally know as the Internet Data Library System (IDLS). Featuring a bilingual user interface and metadata content, Equinox reflects the accomplishments achieved through the partnership between the Faculty of Social Science at the University of Western Ontario (Western) and the Conference of Rectors and Principals of Quebec Universities (CREPUQ).

The partnership was announced through a Project Press Release on August 31st, 2007. In the release, the parties agreed to partner together to expand the existing IDLS system, developed by Social Science Network and Data Services (SSNDS), to provide the same level of information in both French and English for files released through Statistics Canada's Data Liberation Initiative (DLI).  This goal was realized by migrating metadata from SHERLOCK (CREPUQ's bilingual data delivery system) into IDLS.  The metadata transferred from SHERLOCK into IDLS was enhanced to provide equivalent functionality for French and English users of Equinox.  This allows the partners to develop higher quality services for students, professors and researchers of the universities and colleges across Canada which previously used either IDLS or SHERLOCK.

There were four primary goals of the partnership:

  1. Develop a new bilingual user interface that was intuitive to use through the primary searching and browsing functions.
  2. Provide access to both French and English metadata records with content ported from SHERLOCK.
  3. Provide access to the large collection of aggregated data managed by CREPUQ.
  4. Implement a data delivery system that would become compliant with DDI metadata standards.

Team Members

The roots of the partnership began with the initial investigation into a system replacement for SHERLOCK, the data delivery service used by CREPUQ. This occurred in 2005. The CREPUQ group investigated and compared data delivery systems that were currently in place at various academic institutions in Canada. The decision was made to choose IDLS and partner with Western through SSNDS in the Faculty of Social Science to build a new bilingual data delivery system using the content and technical skills that the two teams could offer. The initial exploratory work into the feasibility of the project was carried out jointly by:

Western Team

Once the project was agreed upon in principle, the Western technical team comprised of personnel from SSNDS designed and implemented the new Equinox interface and underlying bilingual functionality. The Western team worked closely with the CREPUQ team to achieve the final implementation and launch of the Equinox system. The Western team members included:

CREPUQ Team

The CREPUQ team were instrumental in helping to organize the partnership, develop and input bilingual content into Equinox. As well, the two teams worked closely to achieve the overall functionality, look-and-feel, and design of the Equinox system. The CREPUQ team members included:

Origins

The Equinox system is based on a long line of data delivery and retrieval services developed by SSNDS that date back to the 1970s when services utilizing magnetic tape storage were first deployed. Prior to the current Equinox implementation, IDLS was a web-based implementation and evolved as a logical next step to the Network Data Library System (NDLS). NDLS was first designed and developed in 1986/1987 by the Social Science Computing Laboratory (SSCL), now Social Sciences Network and Data Services (SSNDS).  The goal of NDLS was to provide a computerized system which would allow students, faculty and researchers to retrieve data subsets from large machine readable data holdings.

Unlike NDLS which required a custom, platform-dependent application to retrieve data, IDLS was implemented as a web-based system to provide a machine-independent mechanism for retrieving large datasets, codebook information and dataset descriptions for statistical software. This retrieval system allowed data to be requested from virtually any location on the Internet without the requirement for specialized software or necessity of consulting a data librarian to obtain data. IDLS was first developed in 1996, and the second version of the system was released in late 2000. It took advantage of the distributed nature of the Internet and enforced security restrictions to enable authorized users to use the system from remote locations. The second version of IDLS was built upon a very different structure than the initial version, using instead a flexible database foundation as described below.

With the development of the Canadian Association of Research Libraries (CARL) Data Consortium, and the subsequent introduction of the Data Liberation Initiative (DLI), a need was recognized for universities to share development activity, to avoid duplicating services where ever possible. With the permission of Statistics Canada, SSNDS decided to make the NDLS system available by subscription to members of the Data Liberation Initiative via IDLS.

IDLS Team

The original design of the Internet Data Library System (Version 2.0) that was released in 2000, was completed by the following team at the University of Western Ontario:

The team changed in 2001, with Heather Stevens, replacing Bill Dowhaniuk as Systems Programmer/Analyst. The system continued to be maintained by the core team, identified above and forms the basis of Equinox.

Technical Description

IDLS (Version 2.0), and the current successor Equinox system, have two major components that contribute to the data delivery system exposed to users. The front-end, or user interface, is implemented as a web site following XHTML standards and provides data users with options for browsing or searching the extensive data holdings. Delivery and display of metadata content is facilitated by InMagic® WebPublisher software which is used to retrieve information from text databases maintained internally with InMagic® DB TextWorks. A text database, known commonly as a textbase, is similar to a database except that it can easily manage large amounts of textual, numeric, and image data. Equinox textbases are used to house the descriptive metadata information relating to the data file holdings. This descriptive information includes, but is not limited to the file abstracts, variable descriptions, value descriptions, frequency information, and information that is used to build SPSS, STATA and SAS program files to support end-user analysis. Raw data, or microdata, for each file is generally stored in a Microsoft SQL Server database. The system utilizes queries from InMagic® to launch the data requests which in turn formulate the basis for a query against the SQL Server database to create a retrieval package consisting of either a subset of the microdata file or the entire file, supporting documentation, and statistical program files. Once finished, the package is available to be downloaded directly to the user's local computer as a compressed ZIP file.

All data in Equinox is organized in the form of files, each with their own unique title. For example: The 1991 Census of Canada: Long Form Profile: B9105 (Atlantic region). For each file, users can select the specific variables required by creating a subset. File and variable descriptions can be displayed during the selection process. In addition to receiving the raw data, users may select supplemental files including codebook information, and dataset descriptions in SPSS, STATA and SAS formats.

Through IDLS, SSNDS first introduced a mechanism by which Ivation's Beyond 20/20 data files are accessible to subscribing academic institutions across Canada. If a data user has the Beyond 20/20 software add-in installed on their local machine, clicking on a link to a Beyond 20/20 file will download the file, and open it using the Beyond 20/20 data viewer from within the user's web browser. Later, this capability was extended to include Microsoft Excel files. Unlike microdata files, users cannot subset Beyond 20/20 files through Equinox. Instead, data is presented to users as a series of tables allowing them to use the Beyond 20/20 browser software to perform data analysis on the contents.