Column – High-performance, columnar, in-memory store with bitmap indexing in Go

Re:Column – High-performance, columnar, in-memory store with... by shkkmo on Monday June 21, 2021 @07:32PM (#27585945)

> OOP is not inheritance, just one possible trait among OOP implementations, just like FP isn't Haskell.

I didn't say it was. The key par was "state is managed seperately from logic."

> Component programming with interface separated from state

There is no such things as interfaces in ECS. Interfaces are a way of describing how state is bundled with logic. ECS does not do that.

> Those that promote ECS as not being OOP 99% of the times never read books like the one I linked on my comment.

I'll admit that I have not read that book. Your condescending appeal to authority here doesn't actually promote conversation.

Please, tell me what the "objects" are in ECS or what else qualifies it as OOP?

[ Parent | on HN ]

Re:Column – High-performance, columnar, in-memory store with... by pjmlp on Tuesday June 22, 2021 @02:32AM (#27588636)

There are no "objects" in ECS, rather Components as the name clearly states.

Funny why do most ECS frameworks use interfaces/traits/pure virtual classes/static polymorphism then, since they don't exist according to you?

Component orientend programming is a subset of OOP, as for why that is, I provided a book, feel free to educate yourself.

More CS papers are available on demand.

[ Parent | on HN ]

Re:Column – High-performance, columnar, in-memory store with... by jhgb on Wednesday June 23, 2021 @06:25AM (#27602852)

> I didn't say it was. The key par was "state is managed seperately from logic."

But state is managed separately from logic pretty much anywhere where you don't store lambdas as fields of structs.

[ Parent | on HN ]