September 2022 Summaries

Apollo Router v1.0 has officially reached version 1.0, offering a lighter and faster supergraph runtime written in Rust. The release expands its capabilities to support even more use cases with optimizations for client app performance, hardened security, and robust extensibility. Entity-based `@defer` support allows apps to access essential data with minimal latency while deferring less critical data, improving performance benchmarks like first-content-paint (FCP), time-to-interactive (TTI), and total-blocking-time (TBT). Apollo Router now supports cross-site forgery protection, subgraph error redaction, and disabling introspection for enhanced security. The release also includes enhanced extensibility with a new stable v1 configuration schema, full support for Rhai scripting, and native extensions that can be used via standard YAML config and Rhai scripting. Stock router binaries and docker images are available for Linux, Mac, and Windows, along with pre-built docker images and an updated Helm chart for Kubernetes deployments.

GraphQL Summit 2022 is a three-day event taking place in San Diego, October 3-5, where graph architects can learn about best practices and help building or scaling their organization's supergraph architecture. Attendees will hear from leading global enterprises on migrating to a federated supergraph, scaling large graphs with many resolvers, and more. Sessions include "Determining the Right Architecture to Design When Adopting Apollo Federation," "State of the Federation Subgraph Ecosystem," "Sit Back and Let the Graph Grow," "How We Unified One Graph at Block," and "7 Tips for Writing Better Resolvers." The event also offers opportunities to connect with fellow graph architects, share advice, and learn from scaling a supergraph enterprise-wide. Registering early is recommended as the event is expected to be popular, with a justification letter available to help persuade bosses of the benefits of attending.

Apollo is extending its capability to ensure that changes made to a source variant's subgraphs or to the contract filtering configuration won't cause breaking changes to any downstream contract variant's schema or active clients. This is achieved through contract checks, which run two types of tasks: build tasks and operation tasks. These checks are triggered when changes are made to the source variant's subgraphs or when the contract filter configuration is edited in Studio. If any checks fail, a warning will be displayed, but the user can still choose to update the contract even if there are warnings or if checks haven't finished running. Additionally, users can decide which downstream contract variants should be blocking their source variant checks to fail by marking them as "BLOCKING". This feature is available on Apollo's Enterprise plan and requires an upgrade of the Rover CLI to at least v0.8.2 for full functionality.

Apollo Studio's new Field Insights page provides a powerful tool for graph maintainers to understand and improve their schema's usage and health metrics. By analyzing field-level data, users can identify adoption patterns, impact on user-facing apps, guide change workflows, assess field-level health, and understand federation dependencies. This feature is available to all organizations with longer data retention offered on paid Apollo plans, and it requires the use of field-level instrumentation in servers or subgraphs.

The GraphQL Summit 2022 is a must-attend event for anyone looking to navigate the technical and organizational challenges of driving supergraph adoption at their company. The summit will feature key architectural and governance insights from champions at Netflix, Reddit, and Allergan, as well as interactive sessions and workshops on fostering collaboration, developing stewardship models, and ensuring program success is self-sustaining over the long term. Key speakers include Stephen Spalding, Dan Crews, AD Slaton, Juliana Congote, Kaitlyn Taboada, and Dan Boerner, who will share their experiences and provide practical frameworks for successful supergraph adoption. The event will take place in San Diego from October 3-5, offering a unique opportunity to connect with peers and gain valuable insights into the world of GraphQL.

Priceline.com implemented a GraphQL supergraph to unify its data and services, including sensitive PCI data, into a single graph. This allowed them to selectively expose only necessary data to different audiences, such as client apps, while keeping sensitive information private. To restrict access to sensitive data, they created contract graphs using Apollo Studio, which derived a filtered subset of the supergraph schema based on tags applied to subgraph schemas and filters defined in Apollo Studio. This approach eliminated duplicative maintenance work and provided a single source of truth for definitions of every entity in their data model. With this setup, Priceline can now power nearly all of its experiences with a single endpoint while maintaining security and scalability.

GraphQL Summit 2022 is a conference focused on building applications on top of the supergraph, with sessions covering topics such as reducing friction in app development, optimizing schemas, and solving caching challenges. Attendees can connect with fellow developers to share advice on schema best practices and innovative customer experiences. Keynote speakers will discuss modularizing iOS applications, navigating the normalized cache, and handling schema updates on mobile clients. The event also features a session on isolated client development using the power of the schema to create meaningful stub data quickly. With three days of surf, sun, and GraphQL discussions in San Diego, registration is now open for developers to attend and take advantage of the benefits of attending GraphQL Summit 2022.

GraphQL Summit 2022 is an upcoming event that focuses on building, managing, and securing supergraphs at scale. It will be held in San Diego from October 3-5, offering a unique opportunity for backend developers to learn best practices from leading global enterprise organizations. The conference features sessions such as "Offensive GraphQL Security Testing" and "Server Driven UI and Batching", which cover topics like security testing and batching operations with Apollo Federation. Attendees will also have the chance to connect with fellow developers, share advice on building subgraphs, and modernize their stack while enjoying the San Diego surroundings.

A generated GraphQL schema is a machine-readable document that declares available operations and their return types, allowing backend services and clients to have a shared understanding of an API. It can be used to merge data sources from different databases into one schema, providing flexibility to clients in selecting what parts of the data they want to get back. However, using generated schemas has trade-offs, such as not allowing for changes to the types, which may lead to duplication of code and maintenance issues if the underlying database is changed or migrated. Client-oriented schemas, on the other hand, provide flexibility and abstraction, allowing clients to customize their response selections while hiding service implementation details. They can be built with a schema-first or code-only approach and are beneficial for teams that want to ship new features quickly with minimal effort. The choice of using generated or client-oriented schemas depends on the specific needs and goals of the project, considering factors such as scalability, maintenance, and client requirements.