Reduce System Complexity with Data-Oriented Programming • Yehonathan Sharvit • GOTO 2023

Reduce system complexity with data-oriented programming by treating data as data, not code, and avoiding mutation for safer and simpler systems.

Key takeaways

We separate code and data to reduce system complexity.
Data-oriented programming promotes treating data as data, not as code.
Immutable data allows for safe sharing and reduces the complexity of data manipulation.
The schema of data is used to describe the shape of data.
data can be represented as a JSON schema or a map/dictionary.
Inheritance is not necessary when using data-oriented programming.
Mutating data is a source of complexity and should be avoided.
Immutable data is safer to share and easier to debug.
Data-oriented programming aims to reduce complexity for information systems.
Principles of data-oriented programming:
- Principle 1: Separate code and data.
- Principle 2: Represent data as data, not as code.
- Principle 3: Immutable data ensures safety and simplicity.
- Principle 4: Use structural sharing to avoid complexity.
Data-oriented programming can be used in any programming language, and is especially useful for reducing system complexity.
By treating data as data, we can reduce the complexity of our systems and make them easier to understand and maintain.

More talks