4 ms·
It's funny. He gives an example of how by grouping data and functions that operate on that data (in a class) is a great example of COP and different from OOP...
by e12e 11y ago
It's funny. He gives an example of how by grouping data and functions that operate on that data (in a class) is a great example of COP and different from OOP...:
http://mollyrocket.com/casey/stream_0019.html http://mollyrocket.com/casey/stream_0019.html
I can only conclude that he spent ten years in the world of toy class-oriented "oop", premature oversized architecture and other nonsense that follow from going from "have no idea about the code/function; will model as strict class hierarchy" -- rather than doing the sane thing BDD/TDD/throw-away prototyping -- followed by refactoring / reimplementation.
It makes for a great sound-bite -- but he's kind of attacking "straw-man" OOP.
- yoklov 11y agoThere are several large differences you're missing. Many of these are elaborated upon in other posts, and some in his twitch stream. His methodology basically completely eschews dynamic dispatch in favor of static dispatch. No function pointers or vtables, just conditionals. Abstractions are built up slowly, only after you're reasonably certain its the abstraction you want. The initial pass of the code has almost no abstraction and favors large functions. The reasoning here is that its usually far easier to add abstraction than it is to remove it, and starting with abstract code is a good way to lead yourself in the wrong direction. Most interfaces between systems are based around data layout, and not functions/method sets.