This presentation describes the approach taken by an externally-funded series of analytic projects in OER to “solve” the requirement, first posed by UNESCO in 2012 (D’Antoni 2013), but later taken up by the Hewlett Foundation (2013), of geographic mapping of OER initiatives, policies and other related entities. There are of course several such “solutions”, all with their strengths and weaknesses. The presentation will consider the decisions taken on technology, databases, mapping and user interface, looking both at the distribution and the collection aspects.
For the means of both widespread distribution and collection the particular solution eventually decided on, after several trials of other technologies (which will be summarised) was the use of semantic wikis, specifically Semantic MediaWiki, with the Semantic Maps extensions. This toolkit has the advantage of being both widely available (either via service providers or as an institution-installable software suite) and very similar in approach to Wikipedia and WikiEducator, thus familiar to a large community of educators. In our view, this makes it a wise choice for OER map projects.
For the curation aspects, requiring painstaking editing of metadata, it became clear that the use of spreadsheet software (be it Excel or open source) offered the best trade-offs, having an easy learning curve for many educational users yet being powerful enough to use to curate and then load even a “large” database of OER initiatives. (In the world of OER initiatives, 1000 is a large database; in the wider world of open access, 5000 is a large database. Both numbers are in fact quite modest in terms of modern spreadsheets and databases.)
Selection of relevant fields for the database was a key aspect, requiring many tactical decisions. For example, it is relatively easy to agree on a standard list of countries, but when it comes to regions and subregions there is much less stability, a fact which North Americans find challenging, used as they are to years of stability in state/province boundaries and even county boundaries. Languages, educational levels, and subject taxonomies all require careful handling if (as we required) the curation is expected to be done by the generality of the educational OER community rather than a cadre of information scientists.
On the rendering of maps there are many issues to consider, including pin clustering (an institution may have several OER initiatives with effectively the same geolocation) and the issue of “How far can pins take you: what else is needed?”
Linked Open Data is increasingly important and the decisions there will also be reviewed. The presentation will conclude with an overview of the main OER mapping solutions extant in early 2015.
Selected references (recent ones removed)
D’Antoni, S. (2013). A world map of Open Educational Resources initiatives: Can the global OER community design and build it together? Summary report of an international conversation: 12–30 November 2012. Available at https://oerknowledgecloud.org/?q=content/world-map-open-educational-resources-initiatives-can-global-oer-community-design-and-build-i
Hewlett Foundation (2013). Launching the development of an OER World Map: Phase 1. RFP released 15 November 2013. Available at https://docs.google.com/file/d/0B5FQbmPL4C6TV21uT09CRXdIYnM/edit