GraphQL — Usable API or Just Another Hype?
summary
Compares GraphQL and REST with concrete tradeoffs: schema generation tools (PostGraphile, Prisma) save manual mappings but can duplicate schemas; N+1 database query problems appear and are resolved by batching with Facebook DataLoader; single-endpoint GraphQL breaks Varnish-style network caching; file uploads work via multipart handlers (Apollo/Express helpers); GitHub’s rate-limit model (points per hour) shows realistic limits and that workarounds add complexity.
AI-written summary · about
chapters (15)
- 0:00GraphQL introduction
- 1:38Data formats vs query structure
- 3:47GraphQL variables and fragments
- 6:20GraphQL directives
- 8:54Custom directives and interfaces
- 10:22Schema duplication mismatch
- 12:56N+1 query problem
- 16:12Caching and performance
- 18:02File uploads and versioning
- 23:39GraphQL security limits
- 27:29Throttling and API attacks
- 31:22Pagination and sorting
- 34:37When to use GraphQL
- 40:47When not to use GraphQL
- 42:13Conclusion
related talks
37mGraphQL — A New Look at the API
23mHarnessing the Power of GraphQL
27mA no nonsense GraphQL and REST comparison
30mWhy GraphQL vs. Why not GraphQL
01h 00mGraphQL - A query language to empower your API consumers
49mGraphQL, different way to think about APIs
33mGraphQL - the What, how and why
49mGraphQL - When a REST API Is Not Enough