Today’s guest blog is written by Martin Lea who works in a technical team at Ordnance Survey. Martin has been working on a collaborative project that aims to deliver a new API platform called “DataSpring”. DataSpring is a one-stop-shop, providing developers with easy access to authoritative public sector environmental data.

The project has finished an initial phase of development and is now ready for user feedback. So, if you’re a developer who’s keen to get early access to the platform and provide feedback, please read on!

Martin writes:

“We need to move away from government’s reliance on bulk data sharing and create an economy of APIs”

–       Matt Hancock MP

I hope we can all agree that, for the Big Data revolution to take off, many more datasets need to be available live and from source via APIs. Gone are the days when hard-media or email delivery will suffice, data consumers don’t want the overhead of managing their own copy of a dataset but need the data to flow from the experts who are capturing it.

But these data publishers have a problem, many “raw” APIs don’t have some critical functions such as:

• Security and encryption to protect sensitive or commercial data
• User rights management to differentiate between users
• Throttling to enforce fair use and manage server strain
• Usage tracking and potentially charging by usage
• Analytics to improve the service

For this reason, publishers employ API Management Systems, a relatively new class of software that provides these layers of control around raw APIs. One of the biggest advantages of these systems is that they can manage all APIs within an organisation and present a unified frontend to the users.

This is great, from the perspective of the developer community, as all APIs from an organisation can be accessed from a single interface, however, developers seldom use APIs from just one publisher. This can leave developers with the added headaches of:

• Working with multiple usernames and passwords
• Different security and encryption protocols
• Multiple licenses to sign and track
• Separate payment systems for premium datasets
• Micro-managing changes to multiple back-end systems

These are the issues that the new DataSpring service aims to solve for both publishers and developers. The Environmental Science to Services Partnership (a collaboration between Met Office, BGS, OS, CEH, EA and NERC) is developing DataSpring as a one-stop-shop for authoritative public sector environmental data. Key aims include:

• A discovery interface to find the right data API
• Plenty of metadata and technical documentation
• Single sign-on
• Cross-API security token access
• A set of standard licenses
• Usage statistics for both publishers and developers
• In time, an easy payment interface with itemised billing

DataSpring will only succeed if it fulfils specific business needs and if its iterative development is driven by user community feedback. To this end a number of open datasets are already available in an alpha version of DataSpring and we are hoping to engage with publishers and developers around their requirements.

If you would like to participate, please contact me at martin.lea@os.uk and I will send you a short Getting Started Guide and a feedback questionnaire. I hope you will join us and help release the potential of government data!