May 2016 Summaries

A GraphQL query is parsed and validated against the schema before being executed. The execution process involves parsing the query into an abstract syntax tree, validating it against the schema, and then executing the resolve functions for each field in the query. These resolve functions can contain arbitrary code and are used to fetch data from various backends. The server takes the results of these resolve functions and passes them up to the root of the query, returning the final result. GraphQL's execution model is designed to be efficient and scalable, with features like batching and caching to reduce the number of requests to the backend.

Apollo Client is a new GraphQL client designed to simplify data fetching in React applications, particularly those using Redux. It aims to provide a transparent and simple data flow model, making it easier for developers to manage server-side data fetching while maintaining control over their app's state. The client integrates seamlessly with Redux and has a self-contained architecture that allows for easy installation via npm install apollo-client react-apollo. Apollo Client enables declarative data loading, similar to React's connect function, but with the added benefit of automatic tracking of query loading using Redux devtools. It also supports using Redux state in queries, making it easier to control what data is fetched from the server based on client-side state. The client provides a hands-on approach to learning and using GraphQL, allowing developers to try it out now by reading the documentation and exploring sample components.

The first GraphQL meetup was held on May 26, 2016, in San Francisco. The event featured talks and a discussion panel with experts from Facebook's GraphQL and Relay teams, Graphene, GraphQLHub, and Apollo. Lee Byron and Joseph Savona from Facebook presented their work, while Syrus Akbary from Graphene discussed his project. Clay Allsopp from GraphQLHub also participated in the meetup, along with Jonas Helfer and Plushcap (myself). The event aimed to bring together people interested in GraphQL for a social gathering and discussion.

The author, who is the director of web development at NewSpring Church, a large organization that has experienced explosive growth, was looking for a solution to integrate multiple data sources from different vendors and in-house systems into a single endpoint. They discovered GraphQL, an application query language that allows client-side applications to get the data they want, and decided to use it to unify their disparate stacks. By using GraphQL, they were able to connect multiple databases and systems together into a single endpoint, resulting in improved performance, scalability, and developer productivity. The author credits the Apollo data stack for making it easier to work with GraphQL, citing its great client library, awesome features, killer dev tooling, and simplified server setup. With GraphQL, the organization was able to build their products around their features rather than around their tech stack, providing a solid foundation for future success.

GraphQL is an application query language that allows developers to ask for specific data from the server, rather than exposing all capabilities of backends and databases. This approach makes it easier to secure GraphQL servers compared to traditional RESTful APIs. Building a GraphQL server is also relatively easy, requiring only defining a schema and resolve functions to fetch data. Various clients, including Relay, Apollo Client, and Lokka, are available for integrating with different frameworks such as React, Angular, Ember, and Redux.

Sashko Stubailo and other experts discussed the tradeoffs between Cordova, fully native apps, and React Native in the Meteor context, with a focus on Apollo as a unified data loading system. Martijn Walraven recently finished rewriting Meteor's Cordova integration to use native code for secure production mobile app updates, and is now working on enabling Apollo to work well with hybrid native apps built with frameworks like React Native and NativeScript. He aims to build fully native shells that allow iOS and Android apps to easily integrate GraphQL data. Martijn will be giving talks on progressive enhancement for mobile apps at PhoneGap Day EU and React Europe, exploring the continuum between Cordova and React Native.

The author of the text discusses the limitations of current API development tools and proposes GraphQL as a solution to create a uniform and descriptive API language that can be understood and integrated seamlessly into design and development workflows. The author highlights the potential benefits of GraphQL, including self-documenting data, deprecation as a first-class feature, and the ability to build next-level developer tools such as in-editor query autocompletion, UI design with real data, and more. The text also touches on the need for standard ways to introspect information about one's GraphQL schema and proposes the creation of a configuration file called "graphqlrc" to discuss this topic further.