DevOps is for the Rich - Fight Me!

So, why are we trying to get them to do operations?

People who know how to write business logic and manage the systems they create are in short supply. When you are flush with these resources, you can build world-class customer-facing services. Let’s call these services, Facebook, Amazon, Apple, NetFlex, and Google. I came up with that one myself :) My point is that these services earn the money to pay the $500,000 to $1.2M/year required for the skill to stand up and justify Developer-based operations (DevOps). Then there’s the rest of us.

The life cycle of a typical DevOps movement goes something along these lines:

1. Create an incredible revenue-generating application

2. Have the developers who built the app maintain the application operations

3. Squeeze all of the growth out of the application

4. Build the next incredible revenue-generating application

5. Create a Site Reliability Engineering (SRE) team to handle to overflow of work from the pure DevOps model

6. Build more incredible revenue-generating applications

7. Pass along more work to the SRE team

It gets to a point where the DevOps model falls over. We don’t have enough revenue or a big enough developer talent pool to justify throwing developers at operations.

So, what do we do? We do what computer scientists do when there’s too much demand for finite resources. We pool the resources. Platform Groups are not only about improving the developer experience. That’s a thing for sure. The Platform Groups charter is also to fulfill the promise of DevOps with a limited set of developer resources.

I have some other thoughts, but this is a #CTOShort… Longer than a tweet and shorter than a blog post.