Using the evidence gained in the surveying phase experiment with the use of semantic technologies in a digital library environment to ascertain the extent to which the perceived value of these materials might be enhanced and to consider the sustainability implications of using semantic technologies.
The above passage is one of the core objectives of STELLAR: to explore semantic technologies and how they might be used to support preservation activity. One such technology is the Learning Resource Metadata Initiative. This post gives a quick overview of LRMI and the potential implications for STELLAR.
A (very!) quick introduction to LRMI
The Learning Resource Metadata Initiative is a relatively new initiative that has been getting many in the metadata and online learning worlds rather excited. Developed jointly by Creative Commons and the Assocation of Educational Publishers the LRMI is a metadata schema for describing learning resources on the web. There’s nothing new about that of course, people have been using Dublin Core to describe web objects for ages but what is different about LRMI is the way it adds metadata to webpages.
LRMI is a subset of Schema.org which is a metadata development co-driven by the ‘big three’ of web search – Google, Bing and Yahoo! It uses microdata tags in HTML 5.0 to describe common internet items like music, videos, people and so on. Search engines then read these tags and are able to determine the semantic qualities of the data on the page. So, for example, someone building a webpage about a film would be able to tell a search engine that what they are describing is a http://schema.org/Movie with a name, actors, directors and so on and which is currently being shown in a http://schema.org/MovieTheater with a name, address and so on.
LRMI has added vocabulary terms to Schema.org to specifically describe learning objects, attainment levels, age ranges and more. It will eventually allow content creators to provide more detailed search results allowing users to filter and narrow results according to the metadata. You can see this in practice through the Google Recipe platform. It allows users to filter their results by cooking time, calories and recipe and gives ratings information and images in the link description.
What does this mean for STELLAR?
So the question is – how could LRMI/Schema.org, as a semantic technology, enhance the sustainability of Open University learning materials?
Well first of all it’s important to remember what LRMI is and is not. It is a metadata standard for learning resources on indexible web pages which uses a universal vocabulary from one source – Schema.org. Unlike Linked Data (another approach to structured data on the web) Schema.org/LRMI is not flexible to particular applications and does not come from as diverse a range of different communities of practice and expertise.
From a preservation point of view LRMI would only preserve relationships within the lifecycle of a webpage. However as web archiving is becoming an increasingly common preservation activity this could be considered a strength of LRMI given that it would preserve the semantic relationships within the metadata of the page. Furthermore one could argue that because the ontologies used by LRMI/Schema.org come from fewer sources there is less chance of inconsistency in maintaining those ontologies and less chance of organisations losing interest.
Finally on a practical level it’s easier to ‘sell’ LRMI to the people creating learning content because it takes relatively little technical knowledge or time to incorporate metadata into pages.
Conclusion
LRMI and Schema.org represents a form of semantic technology which has the potential to expose learning resources to a wider audience via major search engines. Furthermore it will allow users to easily apply more detailed semantic queries to search results. Beyond this neither offers a semantic solution as flexible as Linked Data. For STELLAR both LRMI and Schema.org will likely be the metadata ‘cherries on the top of the cake’ with Linked Data providing the semantic structure underneath.
