🎙️New Betatalks: Why Developers Still Matter in the Age of AI. @rickvdbosch and @oscarvantol talk with @IAmTimCorey about learning, fundamentals, and why understanding your code matters. They also discuss AI as a tool for developers, not a replacement. https://t.co/tgg0xOdBCp
I mostly agree, but I think there is more to it than just that the world is changing. Yes, the load on their servers is astronomically higher. That is going to cause scale issues. And it is.
However, the reliability issues with GitHub didn't start when the additional load of AI hit. They started way before then. Back when Microsoft took over, added more functionality to Actions, added Codespaces, etc. That reliability was always touchy. The added load of AI made it exponentially worse.
But then, you also have the issue of how GitHub has handled the issues. The removed the uptime percentages on their status page (they are back now) and they don't show an overall uptime status because it looks bad. But even now, their status page lies. Their own incident pages show downtime when their status page shows no incidents for that day. It isn't an isolated issue. Multiple days have this issue.
Also, not all of these issues that they've had recently are scale issues. The missing PRs from yesterday was a scale issue (although they say it was a bot attack), but the merge queue issue from the 23rd was a code issue.
At the end of the day, companies rely on GitHub. GitHub is failing to do their job. It isn't all their fault, but it is absolutely a problem that needs to be addressed. It has been bad for months. Something needs to change.
Sat down with @IAmTimCorey the master of practical, no‑nonsense developer teaching. We dug into how devs can keep learning in a world reshaped by AI. Clear insights, real guidance.
https://t.co/cUovbcVIe3