| ▲ | rvz 4 days ago | |
> ... Because it came from Google? Nope. None of what you said are any of the reasons given that it WAS written in Java already [0] but rewrote it all in Go explicitly because of its performance, concurrency and single binary distribution characteristics. Those were enough technical advantages to abandon any thought of a production-grade version of k8s in Java. [0] https://archive.fosdem.org/2019/schedule/event/kubernetesclu... | ||
| ▲ | gf000 3 days ago | parent | next [-] | |
> rewrote it all in Go explicitly because of its performance Because someone wanted a new, shiny toy. | ||
| ▲ | vips7L 3 days ago | parent | prev [-] | |
> the anti patterns weren’t enough we also observe how Kubernetes has over 20 main() functions in a monolithic “build” directory. We learn how Kubernetes successfully made vendoring even more challenging than it already was, and discuss the pitfalls with this design. We look at what it would take to begin undoing the spaghetti code that is the various Kubernetes binaries built from github.com/kubernetes/kubernetes It seems to me that perhaps it wasn’t the languages fault but the authors. | ||