The role of catastrophic failure in software design
summary
Argument: failures teach more than wins, supported by actionable practices and tools. Claims: run post-mortems with timelines, keep unit tests run on every check-in, and use integration smoke tests, alerts and external monitors (Pingdom), plus logs classified as warning/error/fatal. Built examples: Stack Overflow and Discourse use GitHub, moved a post‑mortem server to AWS after an outage, added a HackerOne bug bounty that finds real issues. Outcome: these practices reduce repeat catastrophic failures.
AI-written summary · about
chapters (12)
- 0:00Intro
- 0:24Failure is not an option
- 5:58FAILURE IS Always AN OPTION
- 7:34SUCCESS IS BORING
- 9:25I actually ask people when hiring how many failures they've worked on, and I'm actually more likely to hire someone based on how many failures they've experienced. I think it's the best learning system. - Will Wright
- 13:24FAILURE IS INTERESTING
- 13:39To succeed in software development, you must be obsessed with failure.
- 14:13How do you get the benefit of failing without... actually failing?
- 17:4636 Classic Software Development Mistakes
- 26:05What does a safety system test look like in software?
- 28:48CHARLIE VICTOR ROMEO
- 33:401. Industry failures 2. Our past failures 3. Our future failures
related talks
34mWhy Good Software Goes Bad
30mLearning From Failure: Post Mortems
01h 06mCongratulations, you Failed!
41m‘Failure’ as Success: The Mindset, the Methods, and the Landmines
01h 01mThe Art Of Coding Disasters And Failures
43mBetter Software Development Systems through Stress
46mPlanning for and handling failures from open hardware, aviation, to production at Google
50mDevOps Project Failures - Lessons Learned and Strategies for Recovery