Your Ops Stack is a Category*
The core strategy of managing complexity in computer science and programming is leveraging abstraction. Category Theory is a branch of mathematics dedicated to dealing with abstraction. It sounds like a match made in heaven, since so many of our best technologies are inspired by mathematics: PageRank, routing, graphs, and computers themselves.
After reading a post on Hacker News about how categories solve a particular problem, a developer is stuck reading Wikipedia pages about homological algebra and algebraic topology. Instead of starting from a math background, this presentation will take a network diagrams (the typical arrows and objects powerpoint illustration) and demonstrate it is actually a category. Changes to our category introduce us to functors, and functors lead us to natural transformations that let us merge multiple transformations while guaranteeing nothing breaks. Applied examples will take precedence, with the theory presented when appropriate and helpful to abstracting problems.
Using directed graphs to represent complex systems has can be an effective approach without sufficient context to represent system constraints. Essential concepts from Category Theory such as categories and functors will be presented as concrete examples of systems.
Real world applications spend inordinate amounts of time managing complexities introduced by schema changes. This talk will demonstrate schema migrations in terms of categories and functors as migrations between them, allowing non-breaking migrations and fast-forward / rewinds between categories via natural transformations.
Category Theory, systems, graphs, devops, Formal Methods, complexity
Presentations in a graduate level context and some background in computer science education.
Nike via TekSystems
I’m a DevOps Engineer contracting at Nike, coming at Ops from a software development background.
I think in pictures.