Briefing Paper: Learning Content Discovery & Exchange

This paper provides context for the Learning Content Discovery & Exchange activity within the Technical Standards for Digital Education project. It outlines key technical challenges for discovering and sharing learning content between Australian schools. Existing repositories of learning content offer many potential benefits for learners, however there remain a number of interoperability challenges including discovering content using the language of the classroom, dealing with the variety of technologies used to implement content, and accessing content across jurisdiction and learning system boundaries.

The aim is to provide background and provide a framework for eliciting feedback from the focus group on business requirements, possible technical approaches, and priorities for engagement with standards and specifications bodies.

This document is a work in progress: it will be expanded in collaboration with the focus group for this activity.


© Copyright 2009 University of Southern Queensland

Business drivers

  1. The Digital Education Revolution assumes that teachers can discover and use digital curriculum content relevant to their classroom.
  2. Curriculum content comes from multiple sources (the web, publishers, jurisdiction content development projects, cultural agencies), is hosted in multiple places (the web, jurisdiction repositories, classroom file systems) and exchanged both within jurisdictions and between jurisdictions.
  3. Some teachers wish to adapt digital curriculum content for their classroom.
  4. Some teachers wish to share the digital curriculum content they create and adapt.
  5. Some teachers wish to use social networks to discover and share content.

Interoperability challenges

The challenge is to produce mechanisms that allow teachers to:

  • discover content using the language of the classroom,
  • discover and recommend content using their social networks,
  • to use that content in their learning environments despite the variety of technologies used to implement content,
  • and to access content across boundaries imposed by different jurisdictions, learning systems, and licensing regimes.

The mechanisms should be seamless: a fragmented experience distracts busy teachers. As much as possible, teachers should be unaware of the boundaries imposed by technical and legal decisions.

Scenarios

The following scenarios present two situations which highlight the need for technical interoperability for content discovery and sharing across the schools sector and beyond. The second scenario motivates the issue of sharing content and lesson plans across jurisdiction boundaries.

Discovery & use

Jo, a teacher, is preparing a lesson for year 6 class to meet a Science curriculum outcome. She has been focusing on amphibians in her class recently, so wants to theme the lesson around frogs.

Jo searches her jurisdiction’s portal for digital curriculum content on frogs. She gets lots of hits, but the results on the first page are all for high school students. She refines her search to just show results relevant to year 6.

The portal shows digital curriculum content materials including images, videos, interactive content, and other teacher’s lesson plans. Jo is unaware that the materials are originally sourced from museum websites, the broader web (via an EdNA search), and her jurisdiction's repository of vetted learning materials. Some of the materials were produced by national & state curriculum development projects and fellow teachers produced some of the materials.

The portal shows Jo what other teacher opinions of the content through ratings and comments. Jo uses this information to make a short list of content she will use in her lesson. As Jo examines each piece of content, the portal recommends other content that might be relevant ("teachers who used the content also selected these other pieces of content").

Jo collects the materials she will use by writing a word document. She downloads a frog pond image and places it at the start of the document to grab the kid’s attention. She places web links to interactive materials in the word document. The links are to content hosted on the web and within her jurisdiction portal. Jo saves the document on her school intranet.

The next day Jo uses the content in her classroom by displaying the document on a smart-board and discussing / interacting with the material in front of the class.

Sharing

Jo’s lesson is a roaring success. At the end of the day Jo discusses the lesson with her colleagues Sam & Walter via an instant messaging service. Sam & Walter are also teaching year 6 and want access to the lesson. Jo places it on a lesson sharing part of the jurisdiction portal. The next day, Sam accesses the lesson and uses it in her classroom.

Walter, however, cannot access the lesson because he teaches in a different jurisdiction and does not have access to Jo’s jurisdiction portal. He contacts Jo and encourages her to put the lesson in a national lesson sharing portal. Jo does so and sends Walter a new link to the lesson.

Walter access Jo’s saved lesson, but still has problems using it. He can access the web hosted content in the lesson, but can’t access the learning content hosted by Jo’s jurisdiction. Although frustrated, he knows that the content is also hosted by his jurisdiction, so searches for the content in his jurisdiction portal and updates the links in his copy of Jo’s lesson one by one.

Interoperability and technical analysis

There are many possible approaches to realising the scenarios described above. The Discovery & Use scenario might be realised as shown below.

graphics1

In this model, the teacher discovers content using a jurisdiction portal. The portal uses content indexes at either the jurisdiction or national level. The content indexes are maintained by periodic harvesting of metadata about the resources. The teacher creates a lesson plan in their classroom. The lesson plan references content hosted by the jurisdiction, nationally, and potentially internationally.

This model presupposes a choice of harvesting metadata into a central registry, rather than performing federated search over different repositories. A solution can also be driven by federated search, although experience has shown that scalability can be an issue.

  • The LORN portal developed by the VET sector was initially implemented used federated search; but it has since found that harvesting allows richer functionality and scalability.
  • The GLOBE consortium has also found difficulties in scaling federated search over a large number of repositories. They are currently investigating registry-based search that selects which repositories might contain content relevant to a user query to minimise the search throughput in their federation.
  • The ASPECT project has implemented registry-based search across a large range of publishers and departments of education in the European Union, as part of a drive to improve adoption of standards through putting them in practice through realistic context exchange.

The technical infrastructure in this model would require technical agreements and standards about

  • Search protocols for searching the content indexes
  • Query languages used to search the portal and content indexes, e.g. free text queries, search on selected metadata fields
  • Metadata about the content, covering
    • relation to curriculum, e.g. topics, curriculum outcomes, year levels.
    • technical aspects of the content, e.g. format, size, location
    • accessibility information
    • well-defined vocabularies for populating the metadata consistently
  • Harvesting protocols for collecting metadata into the content indexes
  • Content formats that can be used in the classroom (e.g. HTML + Flash, IMS content packaging).
  • Compliance testing to ensure the integrity of harvested metadata

Exposing the metadata relation to the curriculum is made easier by common curriculum frameworks (the National Curriculum), and machine-readable curricula (such as the Achievement Standards Network through RDF, being investigated by ESA); this is being investigated in the Curriculum Description Technical Standards activity.

Advanced versions of the infrastructure might use an identifier system to relate duplicate copies, versions and derivatives of content. For example, the TLF content in the jurisdiction and national repositories above would have the same identifier so that it was not displayed twice in search results. Identifier systems can also support smart resolution systems that for example allow Jo to embed a URI in her document that resolved to different copies of the content depending on whether Jo and Walter accessed the system.

Advanced versions of the portal might use a registry of content indexes to decide which queries should be sent to which content index. Also, the content indexes might use a registry of content repositories that they harvest metadata from to create the index.

The social networks of people using content should be harnessed when searching for content. Discovery is enhanced if users can access existing user ratings and user comments. The connections between people, content, and people who are expert in content are also highly relevant to search results, and may be harnessed through social networking software.

The discovery sketched in the scenarios involves multiple systems, and different modes of interacting with resources (search, author, hyperlink). In the first instance, this leads to interfaces with multiple spaces for the different tasks, run by different systems. Late engagers have less patience for such a fragmented user experience, and prefer a uniform environment in which they can carry out all the tasks described.

Context exchange presupposes that the licensing arrangements for the content permit it. This is a policy challenge, and little recent technical progress has been made on digital rights management; the emphasis has been on identifying appropriate liberal licenses through metadata.

Content exchange is increasingly informed by the Open Educational Resources (OER) approach, which calls for open sharing of content; there are advantages to this approach, including promoting reuse and marketing directly to schools and parents, but there are also difficulties with cost recovery. The approach is partly being driven by the trend in government departments to make public information openly available by default. Difficulties can result from third party materials embedded in learning content, which is not necessarily subject to the same licensing. Work on guidelines and matrices to clarify content exchange and licensing is ongoing, both at DEEWR level (IPTAAG) and at jurisdiction level.

Relevant Standards and Specifications

1. Search protocols

  • OpenSearch
    Widely used, simple format for embedding search tools in web browsers.
  • Search/Retrieval via URL (SRU)
    Standard XML-focused search protocol for Internet search queries. Widely used by library and higher education repositories.
  • OASIS Search Web Services (SWS)
    Specification in development, attempting to combine various existing web search activities.

2. Query Languages

  • Contextual Query Language (CQL)
    Formal language for representing queries, used in SRU and other specifications. Attempts to maintain a balance between simplicity and expressiveness. Allows search indexes to be defined abstractly.
  • IMS Learning Object Discovery & Exchange CQL context set
    Set of learning content specific attributes for constructing queries. Allows discovery based on learning-specific concepts such as year level, curriculum etc

3. Metadata (See EdNA Group Australian Metadata for Education)

4. Packaging formats

  • Object Reuse and Exchange (OAI-ORE)
    Specifies dynamic aggregations of found digital objects for reuse, as an alternative to conventional packaging; allows discovery of internal structure of compound digital objects.

5. Harvesting protocols

6. Identifiers and Resolution

7. Registries

  • ISO 2146 Registry Services for Libraries and Related Organisations
    Standard providing an information model for describing registries of metadata and content repositories, and their associated services, parties, and activities.
  • IMS Learning Object Discovery & Exchange registry work
    Developing a model for a registry of learning content metadata repositories. Contributed to by the project team, reflecting Australian sector requirements, with proof of concept testing done at USQ. Provides data model for search in e-learning, for describing and accessing registries and repositories (inspired by the ISO 2146 framework), and for organizing search results according to the relations between them (versions, formats, copies).

8. Social Networking Sofware

  • OpenSocial
    Open API for web-based social network applications, allowing clients to access data and core functions on participating social networks.

9. Machine-Readable Curricula

  • Out of scope for this activity. See the Curriculum Description activity.

10. Digital Rights and Content Licensing


The Technical Standards for Digital Education project is funded by the Australian Government's Department of Education, Employment and Workplace Relations (DEEWR).

Southern Cross