Change = Mass x Velocity, and Other Laws of Infrastructure
summary
Argues that change must be handled as convergent, desired state automation rather than heavy process: specifies a GPS-like idempotent model implemented in Cfengine that computes routes from any starting state and corrects for unexpected disturbances. Claims autonomous per-machine agents yield self-healing infrastructure, loose coupling to tolerate cultural and physical diversity, and faster safe change. Notes ~20 years of research and that Cfengine already embodies these ideas in production, improving speed without losing predictability.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:33IT Change Management
- 2:55Mass and Velocity
- 4:17Risk
- 8:09Three Laws of Robotics
- 9:32Change Management
- 11:14Scale and Complexity
- 13:57Handling Complexity
- 15:19Knowledge
- 16:52The Challenge
related talks
42mBuilding Evolutionary Architectures
33mSystems Automation: Past, Present and Future
35m"Change Management: A Scientific Classification"
40mHarnessing Change for Competitive Advantage
56mBuilding Evolvable Infrastructures
01h 19mThe Business of Configuration Management
19mInterview
33mBuilding Evolutionary Architectures