Writing Better Errors
summary
Errors target different audiences and should be split: log full backtrace and params to a service like Rollbar or Airbrake so developers get actionable diagnostics, return use matching HTTP status codes and a clear response body for API consumers, and show localized, concise messages with exact next steps and front‑end validation to app users. Differentiate transient failures (suggest retry) from data or payment issues (offer support or promise proactive follow‑up) and surface frequency metrics to fix UX hotspots.
AI-written summary · about
chapters (7)
- 0:00Start
- 7:04User Errors
- 7:52Server Errors
- 12:48Developer - Input Errors
- 13:49API User - Input Error
- 18:16Developer - Server Error
- 20:22User Server Error
related talks
32mWriting Better Errors
06mDX of Error Messages
32mError handling: doing it right!
22mYes, Your Error Messages Can Be Beautiful And Developers Can Love Them
31mHow Many Ways to Say I'm Sorry, Error Handling in React
23mDeciphering Alerts – Why is it so hard to make sense
45mThe design of error
34mHow to Teach a Service to Report Errors So They're Clear to Users and Machines
