Dataset discovery

The Linked Earth Wiki exposes information about datasets in both human readable and machine readable formats. Therefore, when you upload a new dataset to the wiki, you are not only making it available to your colleagues, but also to bots that may help you discover potential inconsistencies and search engines that may make it accessible to a wider audience.

There are three main ways of accessing dataset information programmatically

• Semantic Media Wiki API: Default API for retrieving the individuals that belong to a class. For example, if we wanted to retrieve all the pages under the category "Dataset" we could use the following query: http://wiki.linked.earth/wiki/api.php?action=query&format=json&list=categorymembers&cmtitle=Category:Dataset%20(L)&cmlimit=800. Normally the API limits its contents to 500 items, so we recommend having a look at the Media Wiki API documentation for more details.

• JSON-LD: Once a new dataset is uploaded to the wiki, we expose a summary of it to search engines by using Schema.org, a common annotation schema understood by search engines. As an example, the dataset http://wiki.linked.earth/A7.Oppo.2005 would expose the following description in JSON: https://goo.gl/Bx3v5s.

• Endpoint: We also have a living endpoint to which more complex queries may be issued. Several examples on how to perform these queries can be found in our tutorial