So I suppose we can finally put to rest that tired "debates don't really matter" nonsense, right?

10 Have a big incident. 20 Assume your systems are simple and fail for simple reasons. 30 Determine the root cause of your simple system failure. 40 Repair your root cause with a permanent fix. 50 GOTO 10 RUN

This is the first time I've ever seen a company live-stream a public incident retrospective. Props to Microsoft Azure here. youtube.com/watch?v=APiK8C… (Note: this incident is unrelated to Crowdstrike)

A hard-boiled egg is bigger than a fried egg and nobody will ever be able to convince me otherwise.

In CrowdStrike's defense, anyone who has ever worked with RegEx knew that someday, somehow, it would be responsible for the end of the world.

There is a restaurant in Yakima, Washington called New York Teriyaki that sells sushi and has a 4.8 rating. I have not yet decided how this information is going to effect my evening.

I've had some time to think about it and I would totally download a car.

A salad bar sneeze guard but if people over 3 feet tall existed.

Step 1: Develop robust incident taxonomy. Step 2: Apply it only to unicorn incidents. Step 3: ... Step 3: ... Step 3: ... Step 4: "Why aren't we getting better?"

If your effectiveness as a leader is 100% coupled with the physical location of your employees, why should I trust you as a leader? If you are ineffective at leading remote workers, why is that a workers' problem and not a leadership failure?

I voted.

The part that I hate most is having to reconcile that for the next four years this asshole is going to find ways to benefit me as an older white male while simultaneously destroying me as a human.