At DevOpsDays India 2018, I opened with a look back at nearly 10 years of devops. The picture I showed from John Allspaw’s famous Velocity 2009 presentation was essentially the start – developers and operations sitting on separate islands with a tiny route between them. Over the years, that bridge got wider. Network ops sparked up then died down then came back. Storage emerged. Database was still reluctant. We even got integrated with project management. But ask any person what devops is and you still get a different answer.
The CAMS framework – Culture, Automation, Measurement, Sharing – was the summary of the early DevOpsDays. After 10 years, I noticed people were forgetting the culture and sharing parts, focusing almost entirely on automation and measurement. Looking at the conference schedule itself confirmed it. When companies start implementing devops, they go straight to extending the delivery pipeline. But to make automation succeed, you have to overcome structural problems: operations valued by uptime, developers valued by features, and nobody aligned on the same goals.
One of the more interesting moments in devops history was the trademark attempt. Someone tried to trademark the word in the US. The reason the court rejected it was not about automation or measurement – it was because the community had already shared the knowledge so broadly that it qualified as community property. That felt like validation of the sharing principle.
There was also a fundamentalism phase where people started dictating how devops should be done, which was ironic for a movement that wanted less structure. And when Amazon launched Lambda, the “no ops” narrative resurfaced – the same claim that operations was going away. I happened to be the speaker right after that announcement. The reality is that operations never goes away; it just changes shape. The practices evolve, the boundaries expand, but the need for feedback loops, shared understanding, and collaboration remains.
Watch on YouTube – available on the jedi4ever channel
This summary was generated using AI based on the auto-generated transcript.