Explore Chicago Collections integrates materials from a wide range of libraries, all of which use varying standards and practices to describe their archival collections and digitized images.

The Chicago Collections consortium of libraries has dedicated significant effort and technical development to overcome the technical challenges behind uniting access to these materials. In 2013, the Andrew W. Mellon foundation provided funding to build both Explore Chicago Collections and the administrative software for contributing materials to this site.

Components

This site uses the eXtensible Text Framework (XTF) technology developed at the California Digital Library to provide indexing, search and browse capabilities. XTF was selected after an extensive review of options by Chicago Collections members because it provided the flexibility needed to index metadata that doesn't conform to one strict standard.

The Explore Chicago Collections interface was built using the Django framework, rather than relying on the default XTF interface.

The administrative tool used to deposit materials, the "Metadata Hopper", was developed at the University of Illinois at Chicago, with input from the Chicago Collections portal committee, and was funded by the Andrew W. Mellon foundation. The Metadata Hopper enables contributing members to create maps for each metadata standard that they use, so that as materials are deposited, the system is able to integrate them. The Metadata Hopper also allows users to 'tag' materials with a shared vocabulary, so that people using Explore Chicago Collections have a single, consistent vocabulary for browsing these materials. The system provides some automatic tagging of materials as they are deposited, so that tagging does not need to be done entirely by hand.

The Metadata Hopper then generates Dublin Core files associated with each original metadata record. This ensures that all of the original metadata (including all of the original subject terminology) is indexed and searchable, while the associated Dublin Core file provides the enhanced metadata for the site, including Chicago topics and neighborhoods.

The Metadata Hopper was also built using the Django framework, and was designed to be used independently of the specific Explore Chicago Collections context. The code for Metadata Hopper along with all of its associated documentation can be found in the Metadata Hopper Bitbucket repository.

Design Principles

The Chicago Collections Portal Committee established the following principles to guide the implementation of this project:

  • The system will provide full access to finding aids in their entirety, rather than simply accepting brief descriptions of collections. This enables researchers to find even obscure connections between related collections from different institutions.
  • The system uses a 'deposit' rather than a 'harvest' approach to contributing content; member institutions must not be required to maintain any additional local infrastructure in order to participate.
  • The system will not require any institution to change its local metadata practices or to maintain a second copy of materials that Explore Chicago Collections can accept.
  • The system will accept both the direct deposit of image media files or the deposit of metadata files that point to an image media file.
  • The system must be able to expand to support new types of media and metadata.