Hopp til søk

Devops Link Apr 2026

This disconnect created a negative feedback loop: Ops resisted frequent deployments, leading Dev to bypass formal processes, leading to brittle deployments, leading Ops to increase resistance further.

| Aspect | Development (Dev) | Operations (Ops) | Resulting Conflict | | :--- | :--- | :--- | :--- | | | Rapid feature delivery | System stability & uptime | Misaligned incentives | | Risk Tolerance | High (willing to change) | Low (fear of change) | Deployment friction | | Environment | Local/development | Production | "Works on my machine" syndrome | | Success Metric | New functionality | Mean Time To Recovery (MTTR) | Competing KPIs | Devops link

Feitelson, D. G. (2015). From Design to Deployment: The Role of Operations in Software Development. Communications of the ACM , 58(2), 50-57. This disconnect created a negative feedback loop: Ops

The evolution of software delivery from monolithic, annual releases to distributed, daily deployments has exposed a critical vulnerability in traditional IT structures: the chasm between development and operations. Developers (Dev) prioritize feature velocity and functional change, while operations (Ops) prioritize stability, uptime, and security. Historically, this tension resulted in what Nicole Forsgren, Jez Humble, and Gene Kim term “the warring tribes” (Forsgren, Humble, & Kim, 2018). DevOps directly addresses this conflict by providing the conceptual and practical link to transform adversarial relationships into collaborative partnerships. (2015)

Prior to DevOps, the “throw it over the wall” model dominated. Once code was deemed complete by Dev, it was handed to Ops for deployment. This link was weak, asynchronous, and document-heavy.

Etsy’s transformation from a monolithic, quarterly-release platform to a continuously deployed service exemplifies the Dev-Ops link. Initially, deployments caused site downtime, leading Ops to freeze changes during holiday seasons. The link was forged by embedding operations engineers into development teams, creating shared dashboards (e.g., “Code as Craft”), and automating infrastructure with tools like Jenkins and Kubernetes. The result was a reduction in deployment times from days to minutes and a 99.99% availability rate, proving that a strong link improves both speed and stability (Feitelson, 2015).