Data standards on the web: CoverageJSON
Every day, about 2.5 quintillion bytes of data are created across the world. As well as a wealth of opportunities this creates, it also presents a challenge for anyone who wants to consume some of it! In order to make it easier for data to be used, and have relevant context applied to it to mature into valuable information, communities form around specific types of data and standards emerge to enable interoperability. These standards are often overlooked by providers of spatio-temporal data (belonging to both time and space), but in a world where attention spans seem to be getting lower and time-to-market is front-and-centre, if someone can’t use your data with ease and demonstrate how it adds value to their product they will quickly move onto another provider.
CoverageJSON is a format for encoding geotemporal data (related to location of a thing at different times) by the geometry of the spatio-temporal domain — it puts the raw data into context the user understands and makes it easy to visualise in web-based applications. The full specification can be found here and below are a few examples of the types of data supported by CoverageJSON.
POINT
The simplest type of coverage. A single point in space and time, with at least one attribute/parameter relating to it. Such as the Sea Water Potential Temperature value of 23.8 below, defined by 90 lines of CoverageJSON, giving the user all the relevant metadata to plot the point on a map.
POINT COLLECTION
You guessed it, a collection of single points. Useful for displaying locations of sites that might sit within a UK based observing network. This information might be contained in a 21,500 line JSON object, but the standard format means writing back-end logic to generate these files and front end logic to read can be handling by open-source libraries.
TRAJECTORY
Something I had fun with recently was defining some 4D trajectory data, in this case latitude, longitude, time and depth. Knowing that an API will provide easy-to-visualise data in complicated domains is a life-saver for end-users and makes it far more likely, should they discover your data (that’s a problem for another day probably), that they include it in their product.
So data standards on the web are something you should care about if you provide or consume data that references things in time and space. CoverageJSON is the leading standard in this area and for anything else you should head to the CovJSON website. The Playground section has become a daily tool for me and super useful to see what types of data are possible to structure in CoverageJSON and test that your data adheres to the standard before releasing to the world. Good luck and see you in the standards-compliant data world of the future!
