a global infrastructure for sharing learning resources
Here's a great little article on working towards creating a 'global infrastructure for sharing learning resources'. The article, on the Creative Commons wiki, discusses what you, as a repository or content owner might like to think about should you want to make your content discoverable and sharable as open education resources (and why wouldn't you). Of course, it isn't necessarily limited to OER - it outlines good practice and considerations for other types of resources. Firstly, you need a consistent way of structuring the information about your resources (title, author, description and so on). This descriptive information, known as metadata (data about data) should be created conforming to one of the many metadata standards. The OER team who put the article together recommend Learning Object Metadata (LOM) or Dublin Core Metadata (DC). LOM metadata is used to describe a particular type of learning resource known as Learning Objects while DC metadata is a more general purpose metadata specification used to describe a wider variety of learning resources. Once you have described your resources you need to make them more discoverable. There are a few ways you can do this. The OER folk suggest that you allow it to be 'harvested' (collected by a special computer program and stored in a centralised repository along with metadata from other repositories). This is similar to the way Google and other search engines collect information about your website. There are standards for harvesting too. The Open Archives Initiative - Protocol for Metadata Harvesting (OAI-PMH) is perhaps the most well known specification for harvesting while more generalised Web specifications such as RSS and Atom might also be worth considering in some situations. Harvesting your data into a centralised repository is one way of making your resources more discoverable. Another way may be to participate in a real-time federated search whereby a search service will make multiple simultaneous search request over a number of repositories. The OER article recommends harvesting over federated searching and there are a number of reasons for this. Firstly, it is more efficient and more scalable, but also helps enable richer search functionality (it is easier to develop and implement advanced search features in one place than potentially many). GLOBE, another initiative dedicated to making educational resources more discoverable and sharable also recommends the approaches and specifications put forward in the OER article. GLOBE however does recognise that there are instances where there may be business rules in place that prevent harvesting and is looking at the SQI specification. A number of GLOBE partners have implemented SQI.
