Wednesday, August 12: GitHub, Incident with Pull Requests and Issues

GitHub says the outage is resolved, but the real fireworks were in the comment section. For a while, people trying to open pull requests and issue pages — basically the places where programmers review changes and report bugs — were hit with ugly error messages. GitHub later said the problem came from an internal setting pointing at something that had been removed, then flipped the bad setting off and watched things recover. Calm corporate update on one side; full internet eye-roll on the other.

The strongest reaction was a familiar one: "this is why you can’t blindly trust giant cloud platforms". One commenter mocked the usual sales pitch that online services are supposed to be more reliable, adding that yes, maybe the uptime is unmatched — just not in a good way. Another dropped the devastatingly dry jab, "So much software engineering going on," which is the kind of comment that lands like a sitcom stare into the camera. Meanwhile, one practical hero rushed in with a public service announcement: if the website was acting up, you could still make pull requests using GitHub’s command-line tool.

And then came the bigger ideological food fight. One side argued Microsoft is still being unusually generous by giving the world free code hosting at massive scale. The other side was having none of it, with one person calling GitHub "utterly unreliable" and bragging their team had already escaped to self-hosted Forgejo after previous GitHub chaos. In other words: outage fixed, but the community used the moment to reopen the eternal debate over convenience, control, and whether "free" is worth the drama.