Unifying Business, Data, and Code

Chapter 11. Deploying a JSON Schema Registry

It’s not about ideas. It’s about making ideas happen.

Scott Belsky, author, entrepreneur and investor

This book has proposed a new methodology for thinking about and working with data. Data champions have applied this methodology to their organization using spreadsheets, process maps, whiteboards, diagrams, and more. After countless meetings, they felt everybody was aligned toward the bright future ahead. However, they could only go so far by writing documents and guidelines. Eventually, documents were ignored and misalignment found its way in again. As it turns out, reaching alignment only solves half of the problem. You also need a strategy to stay there.

To maintain alignment, the outcome of the unifying process has to be encoded and connected into the software systems that store and process your data, making divergence quickly detectable and correctable at its root. This is where a schema registry comes in.

A schema registry is a centralized collection of schemas meant to serve as the single source of truth for three out of the four facets of data introduced in Chapter 4: context, structure, and meaning.

By connecting all of your software systems to your schema registry, you can ensure data integrity through schema validation and, when needed, extract semantics through schema annotations, as covered in Chapter 8.

Schemas Over HTTP

A schema registry—a collection of schemas—can be made available in many forms. For example, some ...