Objects vs Values: Value Oriented Programming in an Object Oriented World
summary
Argues that converting many mutable objects into copyable value types and exposing a single set_channel style API with an in/out relationship cuts entanglement: four UI signals per channel become one call, layout stays copyable for threaded solve functions, and large buffers become move-only image values. Uses Qt-style signals/slots and React-style store ideas, plus future/promise semantics for lifecycle. Claims this change simplifies call graphs, makes unit testing practical, and measurably reduced coupling in the projector alignment code.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 1:43Defining good code
- 5:13Value types vs objects
- 8:36Practical object usage
- 10:51Model View Controller
- 16:50The problem of entanglement
- 20:58Refactoring for separation
- 24:27Rethinking the API
- 30:54Reifying relationships
- 37:00Locating responsibilities
- 39:43Conclusion and Q&A
related talks
01h 42mValue Oriented Programming Part V - Return of the Values
59mThe Most Valuable Values
01h 01mValue Oriented Programming. Part 1: You Say You Want To Write A Function
01h 29mValue-oriented design in an object-oriented system
11mThe Value of Objects -
58m“Value Semantics: It ain't about the syntax!, Part I"
29mSingle Value Objects: where DDD and Functional Programming Share Forces
58mOO Considered Harmful