First screen, captured 1 day ago. Open the video.

Service degradation on Microsoft 365

What HN thinks

Watching Microsoft turn it off and on again, live, with commentary from people who still have Teams but no email.

Leadership, layoffs and AI blamed
Many argue that Nadella's AI push, mass layoffs and buyouts have drained talent and left Microsoft products janky, making outages like this predictable.
Cascading failures are genuinely hard
Others push back on calling it amateur hour, noting that thundering-herd and cascading failures are hard to design and test against. Some reply that the whole software industry lacks real engineering discipline.
Status updates reveal confusion
Commenters relaying the updates hour by hour read them as frantic, trial-and-error debugging with no root cause found for many hours.

Consensus: Most agree this is an unusually severe, global outage, and that hours of vague updates without a root cause are alarming. Everyone wants a public postmortem.

Recommended in the comments

Linux Mint (product): Switchers report it works well and greatly improves laptop battery life compared with Windows 11.
GrapheneOS (product): Praised as fast and more private than mainstream phone software.
Okta (service): Called a much better IAM solution than Entra, though pricey.
How Complex Systems Fail ↗ (book): Cited as an enduring explanation of why nobody fully understands large systems.
The Machine Stops ↗ (book): Recommended as an eerily apt read for a world where nobody understands the systems they rely on.