SERC CMS Features

Many of these features are explained in more detail in Building Educational Portals atop Digital Libraries published in D-Lib.

CMS Features

Support for Distributed Authoring of Websites

The documentation for authors provides more detailed information about these features.

• All site content can be edited within a web browsers allowing authors to work from any computer with an internet connection
• Authors generate content within pre-defined templates customized for the site.
• System supports a wide variety of formatting options including support for images (including thumbnail versions) and downloadable files (pdf, word, excel, etc...), embedded video (flash, quicktime) and embedded java applets.
• System template include both a simple blank slate format as well as highly structured formats that enforce consistency across groups of pages.
• Past versions of content changes are retained allowing for easy 'undo'. This includes versioning of files and images.
• All sites have both a development and live version allowing authors to keep in-progress work out of the public eye. Synchronizing these sites, either on a page-by-page, or module-by-module basis is a one-click process
• The system contains many refinements based on author feedback, which streamline its use. (e.g. shortcut urls, and quick access from a site page to its editing interface)
• A robust permission mechanism allows many projects to co-exist within the CMS with appropriate delegation of authoring responsibility.

Support for the Creation of High Quality Websites

• Many current best practices in information architecture and web authoring are embedded in the system and so are realized without any need for author expertise
• The system attempts to create pages that meet web standards, including accessibility. The authoring tools are constructed with an explicit goal of guiding authors away from standards/accessibility-breaking constructs.
• The system automatically generates navigation elements (navigation menu, previous/next links, breadcrumbs) based on simple author choices about local pages. Sites are organized into inverted tree structure of modules which provides a consistent site architecture.
• The system default urls are human readable and reflect the underlying site structure.

Integrated Digital Libraries Tools

A complete set of digital library tools support the creation, use and reuse of metadata describing external web and print resources, as well as web pages created within the CMS.

• Project specific cataloging queues track and organize the workflow as resources are identified for cataloging, catalog records are created, checked for quality assurance and integrated into the CMS-based web sites.
• Web-based tools allow catalogers to create metadata records that comply with Dublin Core standards. CMS pages can also be automatically exported as Dublin Core XML for use by external digital libraries.
• OAI-PMH tools for harvesting and ingesting metadata are integrated to allow records created within the CMS to be shared with other repositories to allow CMS authors to take advantage of metadata in other repositories.
• Links to external sites within CMS web pages can be augmented (through a semi-automated process) by references to human-readable version of the metadata record describing the site.

Support for Mini-Collections

The tools specifically facilitate development of sets of parallel pages with a common theme and format (e.g. a group of activities, a group of course descriptions).

• Templates enforce common look and feel as well as codify common elements within each mini-collection page. This ensures that each page presents the same sorts of information in the same order which is crucial in enabling efficient exploration of the collection.
• Mini-collections can be built through online community contribution and exposed through faceted search (see below for more detail on both of these).

Support for the Development and Use of Controlled Vocabularies

• Tools support development of hierarchical controlled vocabularies. Special consideration is given to creating small, application-specific vocabularies to suit the needs of individual projects and communities.
• Controlled vocabulary terms can be applied to web pages within the CMS, uploaded files and images, as well as metadata records describing external resources (including metadata imported from other collections).
• A mapping systems allows semantic relationships (and especially equivalences) to be identified between terms across vocabularies. This help minimize the need to 're-tag' resources that have been associated with terms in a given specialized vocabulary if they are to be reused in a context where a different vocabulary (often with a different organization but with some specific term-by-term equivalencies) is more appropriate.

Support for Faceted Search

• System supports easy creation of global and specialized search interfaces across sets of catalog records, CMS web pages files and images. A specialized search can be limited to web pages in particular areas of the site (e.g. a 'search this site' for a particular project or sub-project, or search across a single mini-collection) as well be constrained to meet any criteria related to controlled vocabularies (e.g. a search that shows all catalog records which are classroom activities dealing with hydrology (assuming classroom activity and hydrology are terms within controlled vocabularies that have been applied).
• The faceted search interface combines full-text search with a browsing interface based an controlled vocabularies. This approach exploits the value of educator-oriented controlled vocabularies and encourages collection exploration by providing better information scent than traditional 'advanced' search interfaces. The full-text search covers internal web pages, content from external sites referenced in catalog records, as well as text within uploaded files (Word, RTF, PDF).

Support for Community Contribution and Discussion

• Email listservs with archives support public discussion as well as project development.
• Online discussion boards integrated into CMS web pages support general discussion and focused community sharing events (e.g. online workshops).
• The browser-based web authoring system includes the ability to create arbitrary web forms including the standard text,menu, checkbox, radio-button and file-upload options.
• Contributions through these forms aggregate in queues where individual contributions can be examined or groups of submissions can be exported to spreadsheets for further manipulation.
• Input from multiple forms can be linked. (e.g. to support unified view of both application and registration information in a site supporting a workshop)
• Form fields can be flexibly mapped to web page templates allowing contributions from online forms to be transformed with a single click into a new CMS web page
• People who contribute materials through these online forms can self-register and gain editing access to the web versions of their contributions.
• A simple interface allows these external contributors to track materials they have contributed across the CMS and enhance and update the material with oversight from the internal authors responsible for that area of the website.

Support for Distributed Ownership/Management/Development of CMS-based Projects

Although the CMS technical infrastructure is centralized it supports a distributed model of authorship and ownership that encourages individual projects using the CMS to oversee their own site.

• Automated email notifications about updates to site material allow managers to keep tabs on a site without having to checkup on the site regularly
• Private project-oriented online document archives facilitate develop work by project members.

Support for Reuse of, and Connections to, Information Across the CMS

• Metadata records, created in the CMS and imported via OAI are available to all projects for direct reference or exposure through customized search.
• Manually and automatically created cross-site related links provide users with connections to information across the site (often from other projects) that relates to their current page. Automatic links are generated based on controlled vocabularies.
• Cloning feature allows content in one part of the site to be repurposed in other areas (for other projects) in ways that integrate with the navigation and look and feel of the new area.

Built on Standard Web Technologies, with Proven Performance and Reliability

• Uses a standard LAMP stack with MySQL and PHP at it's core.
• Uses Lucene library for high-performance searching.
• 'Shared-nothing' architecture allows for linear scaling.
• Careful use of caching ensures performance under high load (single server handles 1,000,000 page views/month at 10% load).
• Successfully used by dozens of projects, hundreds of authors, and over a million visitors each year.