My last post was a very long time ago. First I lost track of time, second blogging does not seems to be my cup of tea. In the past few weeks I have been working on rebooting my teams efforts, simplifying things to allow them to be more approachable. From a high level view it appears that my team and rest of the development teams are not ready for more complex next generation type of CICD platforms. This forcing my small to care the bulk of the load which is not possible. Plus even my own team has trouble understanding some of the theories I am trying to prove out.
An example of this is, we use XebiaLabs DevOps Platform XL-Deploy and XL-Release, in recent version they have come out with a very slick DevOps As Code (desired state) way to control their platforms. I started to work with and loved it, very K8S like where you defined a state of the system in code (yaml) and applied them. I thought this was a big game changer for our DevOps pipeline automation but it has been very hard to explain and push these concepts forward.
Part of it, is most do not understand a large scale deployment platform like XLD so trying to code it makes it that much more difficult. Next, XebiaLabs has not done a very good job of documentation and their As Code VSCode plugin missing the mark. Thus it is very hard to figure some of more deep dive work on DevOps As Code.
This leads me to my final thought, it is hard to find people who like to explore technology and figure it out. While this is one of my favorite things to do, it is not for most teams I’ve worked with. This means I need to rethink my direction, think about pivoting towards simpler well know paths that limit what you can do but allow you to do enough.