Comment by duckkg5
5 hours ago
I would so very much love to see GitHub switch gears from building stuff like Copilot etc and focus on availability
5 hours ago
I would so very much love to see GitHub switch gears from building stuff like Copilot etc and focus on availability
The #1 priority at GitHub for this year is migrating from their own data center to Azure, any other work that gets in the way of this is being deprioritized: https://thenewstack.io/github-will-prioritize-migrating-to-a...
> It’s existential for GitHub to have the ability to scale to meet the demands of AI and Copilot, and Azure is our path forward. W
More existential than going down a few times a week?
This is an absurd state they are at! Weekly outages in 2025 and 2026. From developer beloved and very solid to Microslop went faster than I expected
They may have been Beloved before MS bought them. It takes awhile for technical debt to catch up.
I think GitHub shipping Copilot while suffering availability issues is a rational choice because they get more measurable business upside from a flashy AI product than from another uptime graph. In my experience the only things that force engineering orgs to prioritize uptime are public SLOs with enforced error budgets that can halt rollouts, plus solid observability like Prometheus and OpenTelemetry tracing, canary rollouts behind feature flags, multi-region active-active deployments, and regular chaos experiments to surface regressions. If you want them to change, push for public SLOs or pay for an enterprise SLA, otherwise accept that meaningful uptime improvements cost money and will slow down the flashy stuff.
Unless a major out(r)age forces a change of leadership, expect more slop down our throats.