DevOps: Time to Go Back to Basics

(From LinkedIn post)

It's been over a week since attending DevOpsDays Montréal. During that time I've reflected on some of the amazing presentations and conversations I've had in the open spaces, and this single conclusion has emerged for me:

Regarding DevOps: it's time to go back to the basics.

Over the past decade, our industry has enjoyed huge technological advancements: container orchestration, CI/CD pipelines, Infrastructure as Code (IaC), mature cloud vendors, observability platforms, etc. Whatever you need to operate, monitor, secure, or ‘shift left’, a SaaS vendor will happily sell it to you.

And yet, we experience the same problems as when we started: misaligned incentives, dysfunctional team dynamics, and ineffective leadership- all leading to subpar customer experiences, stressful on-call shifts, and unfulfilling work environments.

Our systems are now even more complex. The pace we need to maintain to keep on top of even a fraction of the technology and ‘best practices’ out there is frantic, and we have to change our titles every few years to ensure we’re properly compensated for our work. Heaven forbid our job title reads ‘cloud engineer’ or ‘operations’ in a sea of ‘SRE’ and ‘platform engineers’ (or the new job title du jour). Let’s be honest: most job descriptions for those fancy titles all boil down to being on-call and writing YAML.

Finally, despite all this progress, running production systems has become more expensive. Anyone experiencing sticker shock from a recent bill from their observability platform knows this well.

So let’s take a step back for a moment, take a deep breath, and acknowledge that we’ve all gotten distracted by an endless procession of shiny objects. That’s ok. It happens.

We need to remember why we’re here: the people. Our task is to keep paying customers happy while ensuring our teams are productive and motivated. Success for many of us means paying the bills while being able to spend time with friends and family during nights and weekends.

Good news! Many at the conference understand this problem and are trying to be a part of the solution.

As DevOps practitioners, we must start with listening, empathy, thoughtful metrics, constant retrospection, effective process, strategy, and then finally- using technology to force multiply our efforts.

What do you think?