Kubernetes Is in Hospice

linkedin.com

17 points·gerbilly··3 comments

This is not a very good article, there does not seem to be any actual facts in it. It seems to be a rant about something with no first hand experience or concrete exampls to back it up.

The article appears to be written by someone who is the CEO of a company that is directly competing with container industries.

Seems to be a lot of smoke generation.

More than a lack of facts, there's a lack of any real depth of opinion. E.g. "{containers, k8s} were an attempt to solve problems X, Y, and Z. Here's where we believe they fail, and where future directions lie." This could be an opinion-based approach, subject to debate, but have value as a point of conversation. And could we get any deeper into Appeal to Authority fallacies: cherry picking tweets, bashing k8s because Google itself, with scale almost beyond reckoning is bound to another tool, etc.

This article lacked any useful insight into Kubernetes and the trade-offs involved in using it. My personal experience is this: Kubernetes itself is complex to operate. However, it's very easy to run applications on top of it. The question is whose life you want to make simpler: a small number of admins or a large number of developers?

Attentive readers will note that this is the same design trade-off that public clouds make. (There's also the cap-ex vs. op-ex thing but Kubernetes enables you to manage that as well.)

Reply on news.ycombinator.com