The MIT OpenCourseWare (MIT OCW) technology solution supports a complex publishing process. This is a large-scale digital publishing infrastructure that consists of planning tools, a content management system (CMS), and the MIT OCW content distribution infrastructure. The current technical solution has been in use since 2003 with a technical support team managing all aspects of this infrastructure.
The planning tools used by the MIT OCW team to assist faculty in publishing their course materials include a custom application of FileMaker Pro, and several checklists and documents. For creating and managing content, we use several desktop tools (file conversion tools) as well as the CMS, a customized version of Microsoft Content Management Server 2002 that fully supports our publishing process. Our content delivery infrastructure includes a sophisticated publishing engine, content staging server, and a content delivery network utilizing Akamai's EdgeSuite platform.
This diagram demonstrates how the faculty's content flows from their original source (various MIT homegrown course management systems, electronic materials not incorporated into Web sites, and hard copy materials which the MIT OCW team transcribes and digitizes), into the publishing environment, and then to the MIT OCW archive, which is the MIT Libraries' DSpace project.
FileMaker Pro
The FileMaker Pro application manages the upfront planning portion of the MIT OCW publishing process, including identification of candidate courses, review and selection of courses for inclusion, faculty contact management, tracking of intellectual property (IP) rights concerns for each course, coordination of strategies concerning IP rights, and development and management of specifications for course construction.
CMS
The CMS is the central (and single) store for all content in the system. All content created in the system (regardless of the interface used to create it), is stored in the CMS. The repository ensures security and integrity of content. The MIT OCW Web site is published dynamically out of the CMS.
Metadata
MIT OCW has implemented a comprehensive metadata specification that was developed in collaboration with metadata specialists from MIT Libraries. This SCORM-compliant metadata specification was implemented using XML data structures (stored in SQL Server). A Web interface (implemented using C# and .NET) is provided for ease of creation and maintenance of metadata.
Metadata is captured at the course, section and resource level within MIT OCW. Where possible, metadata is defaulted by the system based on available information (various sources: course structure, workflow assignments, and configuration parameters, for example). A sophisticated inheritance engine further simplifies metadata capture by automatically inheriting appropriate metadata from courses to sections to resources within the sections.
Content Delivery Network
Akamai's Edgesuite content delivery network serves content to the external world. Content is cached within Akamai's worldwide content network. Akamai pulls content to refresh its cache from a production UNIX server (a.k.a "origin server"). The functionality that requires dynamic interaction (e.g.: Search and Feedback) is implemented using Java, on a Tomcat application server running under Apache Web server.
For more information on the MIT OCW publishing environment or technology, please contact the MIT OCW at ocw@mit.edu.