If devs see their jobs as "plumbing" then that is a problem with their current culture & environment.
Most devs should be focused on delivering business/mission value. This means custom code, unique ip, algos, GUIs, etc.
However, the current state of devops has thrown all sorts of engineering challenges at devs that lose sight of the actual value that should be continuously delivered. It becomes plumbing when we get caught up in redundant tasks such as dependency management, tests, logging, containers, runtimes, integrations, cloud provider specifics for scaling/ha/sre, and connecting them all to create "the platform" that is supposed to enable true continuous delivery of value. We end up more focused on this internal platform development even though the tooling is pretty advanced. We still have to configure/orchestrate/manage that advanced tooling, and this, in turn, creates the plumping effect. We greatly underestimate the work involved with all of these things.
It is not plumbing when we are focused solely on the continuous delivery of value. We remain lean and agile, constantly testing a hypothesis and developing our "uniqueness".
Very few developers deliver direct business value or revenue.
For every rockstar dev who gets paid a lot of money to make unique solutions and use his brain, there’s dozens of unsung heroes who do the grunt work reading from docs and setting up plumbing so the rockstar can deliver his value effectively.
These developers are not working in cool languages or solving new problems, and many of them will never do so in their life. They are just expendable workers swinging hammers and writing JavaScript.
But isn't it a problem that most devs are setting up the plumbing vs already being enabled to deliver rockstar/valuable solutions?
i.e I've worked at a large enterprise and we all (product devs & ops) became so focused on our grand platform (continuous delivery, heavy integration with AWS, container orch, all using OSS) that product development stalled. Though we were doing "rockstar" things with cool tech and our resumes have a bunch of cool words, but it's true that most of that work was plumbing. In the end that tooling became a bch to manage and continually developing upon. The product continues to stall.
Wouldn't it be better for all if we weren't so caught up in the plumbing of these advance/cool tools? -- that's the thought that actually leads into something like building your own platform, vs something provided. Ops then become more like an overseer sre type of role, and most engineering shifts back to direct value.
Most devs should be focused on delivering business/mission value. This means custom code, unique ip, algos, GUIs, etc.
However, the current state of devops has thrown all sorts of engineering challenges at devs that lose sight of the actual value that should be continuously delivered. It becomes plumbing when we get caught up in redundant tasks such as dependency management, tests, logging, containers, runtimes, integrations, cloud provider specifics for scaling/ha/sre, and connecting them all to create "the platform" that is supposed to enable true continuous delivery of value. We end up more focused on this internal platform development even though the tooling is pretty advanced. We still have to configure/orchestrate/manage that advanced tooling, and this, in turn, creates the plumping effect. We greatly underestimate the work involved with all of these things.
It is not plumbing when we are focused solely on the continuous delivery of value. We remain lean and agile, constantly testing a hypothesis and developing our "uniqueness".