I’ve come to expect completely inaccurate writing from InfoWorld, which is almost as bad at ZDNet. I cannot believe people read and believe this shit. People are so gullible. They actually think that stuff they read in InfoWorld is completely free of bias and couldn’t possibly be influenced by the powerful corporate interests at play, after all, no one makes money if everyone does their own Kubernetes and people make it easy for them. You know, the same reason the single more recommended Kubernetes tutorial I have heard about is the completely irrelevant “Kubernetes the Hard Way” from Kelsey Hightower which tells you on the first page that you need to pay $5/day just to use the Google Cloud Platform that the entire tutorial completely depends on.
Or how about the quote from Joe Beda, cofounder (and chief sales person) for VMware, as irrelevant as UNIX SCO in the cloud-native world at this point — especially Tanzu (pfffhahaha, I couldn’t help it). Or how about Deepak Singh, VP or AWS, or Brendan Burns, CVP for Azure. What a perfectly unbiased group of people to choose to base your article on. Really people, are you this fucking stupid not to see through this? They do not care about you. They want your money. Shills like this are not people you should interview, InfoWorld, how about interviewing people from actual companies facing this challenges instead of the vendors paying you under the table to give them a “influencer” platform akin to the Fyre festival (don’t worry, no one is going to be doing six years in a Federal prison here, they are just nice white lies, that will cost companies billions to fix later).
I wish they had used bold for the statement: “Bloomberg still runs about 80% of its Kubernetes workloads on-premises, and it has invested heavily in developing the in-house skills to reliably manage that environment and an internal developer platform on top of it.” The truth is, that competitive advantage remains the on-prem reality it has always been, not unlike 3D animation shops who create their own proprietary software, such is the case for large-scale cloud-native applications — especially in the high-performance and machine learning space (you know, where everything is fucking headed).
The fact is I work for a very big company that would just laugh at such things. They are motivated, for all the right reasons, to do stuff on-prem and in house, mostly because they have specific high-performance computing needs that GKE can’t even dream of providing, yet.
So for this stupid article to claim that “no one wants to manage Kubernetes” is just categorically wrong. Just ask the authors of the Wall Street Journal (a much more reputable source, btw) article about the inability to find enough people, because companies still also want the option to not depend on one of three providers, a world-scenario that is downright dangerous for humanity. But the real problem is not having enough people who can do it.
Centralization is dangerous and bad for humanity, not just because it is fucking insecure (and often prohibited by large enterprises because of it) but because of lock-in and the massive central points of failure such centralization creates. Imagine if all the power in the entire world was provided by only three companies. Having skills for at least large organizations to fulfill their specific, on-prem Kubernetes needs is of paramount importance for a sustainable future. Just because it is hard to find engineers who understand the power grid is no reason to consider such consolidations. We should be encouraging turn-key, on-prem solutions that share a common skill set and work to reduce the complexity of those solutions, not fucking lock someone into a remotely managed, subscription service. They only people who benefit from that are the monopolists pushing it in filthy rags like InfoWorld like Oxycontin drug reps.
Related: