Incrementally Adopting GraphQL. The Holy Grail?
summary
Argues that incremental GraphQL adoption is practical by making frontends declarative and backends able to serve generated queries that fetch exactly the right data, demonstrated with Isograph which compiles component selection sets into GraphQL/SQL and avoids waterfalls via deduplication, prefetch, and single-request fetches. Introduces Barnum, a structured language for analyzable async pipelines that lets the compiler decide where to execute fields (client or server), hoists derived fields, reuses intermediate work, and thereby makes migration incremental and low risk.
AI-written summary · about
chapters (13)
- 0:00Introduction and context
- 1:03Problem GraphQL solves
- 3:10Network waterfalls and trade-offs
- 5:15React Server Components limitations
- 7:19Loaders and selection sets
- 9:27Tree of dependencies and build systems
- 11:52Short-circuiting and performance
- 13:34Derived fields and privacy
- 16:18Client vs server calculations
- 18:45Compiler decisions and mutability
- 20:29Barnum and structured logic
- 23:17Summary of the journey
- 25:43Final thoughts and Q&A
related talks
27mWhat if ... How to Achieve GraphQL Domination
23mGradually Adopt GraphQL Without Writing any Backend Code
30mWhy GraphQL vs. Why not GraphQL
15mThe Right Size For GraphQL
23mHarnessing the Power of GraphQL
32mFrom learning to thinking in GraphQL
35mGraphQL in the Enterprise
58mGraphQL -- Flexible APIs for Adaptive Clients