High(er) Reliability Software Patterns for Go

on YouTube ↗22m

" summary

Shows a concrete pattern: add go-by-contract pre/post checks for API calls, run reverse proxy contracts that defer postconditions, and send violations to Sentry reporting instead of panicking in production. Demonstrates a real integrity bug (duplicate usernames) and a working proxy implementation that catches the violation without crashing clients, controlled by an env var (DON'T_PANIC=true). Also contrasts practical tradeoffs and references formal-verification examples like Tokeneer and Ada systems.

AI-written summary · about

" chapters (8)

1. 0:00High reliability background
2. 1:10Software safety risks
3. 2:57Engineering reliable systems
4. 5:57Design by contract concepts
5. 7:39Implementing contracts
6. 10:41Production contract strategies
7. 13:57API integrity at scale
8. 17:36Q&A on software reliability

" related talks

• 28mIdiomatic Go TricksMat Ryer · Golang UK 2016
• 19mHow to Future-proof your Go APISteve Kaliski · GothamGo 2018
• 32mSecure and reliable development with GoJulie Qiu, Filippo Valsorda, Katie Hockman, Rebecca Stambler · Google I/O 2021
• 45mThe Robustness of GoFrancesc Campoy Flores · GOTO Chicago 2018
• 24m"Finding Dependable Go Packages"Julie Qiu · GothamGo 2019
• 34mAdvanced Testing Techniques in GoAlan Braithwaite · GoLab 2019
• 35mDesigning for failureItalo Vietro · GoLab 2019
• 30mUnraveling Go Anti-PatternsRabieh Fashwall · GopherCon Europe 2024