Kelsey Hightower - The thing you're trying to build should be more exciting than the tools used to build it.

So much focus is put on the tools rather than the end product. This episode was inspired by a tweet from Kelsey Hightower and why we should focus on what we are building

Put the focus back into what we are building and less on the tools underneath.

Kelsey Hightower Twitter -

Episode Transcription

Welcome back to The Byte. This episode we talk about Kelsey Hightower. He is a developer advocate at Google for Kubernetes and basically he's also a member of the Cloud Native Foundation. If you're not following him, I highly recommend it, because he gives the honest truth about platforms. From other cloud providers such as Amazon or Microsoft, he tests Microsoft's AKS, so they're Kubernetes offering Amazon's EKS offering. So, he compares them all, gives an honest opinion. He's also evaluated K3S from Rancher, which we talked about in a previous episode.

So, he really is a great resource to listen to. And in this episode we're gonna talk about one of the quotes or one of the tweets he sent out back in March 11th. He basically tweeted, "The thing you're trying to build should be more exciting than the tools used to build it." Now, think about that for a second. Because I find this many times when I'm going to customers that there's so much focus on the underlining components and all the buzzwords. So, Kubernetes and Service Mesh and all these different things. And by the time you actually get to the customer's application, what you're trying to deploy, I mean, you're already losing half the battle there. I mean, you're discussing how to build a Ferrari, but no one talks about the steering wheel. Who's actually going to drive the thing?

And that's what I want to put our emphasis back on. I mean, yes, the tools are important to maintain our applications and provide us, but do we honestly need this level of complexity? Kubernetes is trying to solve every problem, and that's kind of the issue Open Stack had is they're trying to create every possible feature for every telco and vendor out there. It just got too overwhelming.

Kelsey Hightower really recommends on the other side to look at maybe serverless. Developers don't care so much about Kubernetes and what's going on. Maybe they really like a serverless aspect where they just push their code and let someone else worry about it. Or something in between Kubernetes and serverless where it's not so complex. I don't have to learn infrastructure to push my application and vice-versa.

There's a lot of discussion around it. I mean, when I'm going to customers, it's may the discussion is around the tooling but less about their actually application.

Let's try to focus back on the application. I mean, what are we trying to build? Where is the value coming from? I mean, as us as a customer, are we making money from Kubernetes? 90% of the market, probably not, unless you're Amazon or Microsoft or Google. Our value is what we're putting on top. So, does it matter from our end customers what's running underneath? Probably not. And that's something we need to question. That's where we want to go is we need to start asking, "Why more?" And start questioning the tooling and the complexity that's sometimes we're putting on ourselves. And also the delays. I mean, when we start new projects, sometimes when you go through all these nice iterations of tooling, we go, "Oh, man. We'd love to deploy our application, but we can't until we get Kubernetes and Istio and all these different tools installed. And only then can I put in my application.

So, coming back around, I mean, let's think about how we can make our life easier and less complex. Let's put the focus back on the application and worry less about what's running underneath. Let's put the value back in our application and bring value to the end-customers.

If you look at a lot of these solos entrepreneurs, I mean, they can't afford to go into this level of depth in their application. Pieter Levels, for example, he builds applications, he has several very famous startups, and it's all running on a single server running PHP. Just to show you, you don't need this level of complexity to have something successful.

Let's continue with this story. I'm interesting for your feedback, so please give us feedback and let me know what you think about adding more value to the application versus the infrastructure.

All right. Thank you very much. Have a great one, and we'll see you in the next episode.
Brian Christner