Conversation

Exactly. FP is a good idea, but it's not the *only* good idea in programming language theory. Practical programming languages should incorporate a variety of good ideas. After all, programming is hard; programmers need all the help they can get.

Quote Tweet

Jordan Rose

@UINT_MIN

· Jan 19

More seriously, Justin’s real point is a good one: FP “won” in that more and more of its features are available ubiquitously. It also “lost” in the same way Kay’s OOP lost: most people do not want its purest form, and it’s not often used as *the* guiding principle for a language.

Show this thread

125

Nicholas Nethercote

@nnethercote

Jan 19

SimonPJ once said about Haskell that purity is important, not laziness, but the latter requires the former. As a follow-up, I suspect "aliasing XOR mutability" is more important than purity...

Replying to

and

I dunno, "aliasing XOR mutability" is incredibly useful, but I don't think it replaces being able to have control over what effects/capabilities a some code is depending on? 🤔

11:47 PM · Jan 19, 2022Twitter Web App

Replying to

and

Aliasing, mutability, effects, seem to be part of a wider program of having support for controlling for, and taking advantage of substructural properties of programs in general. See also: uniqueness, linearity, information flow, capabilities, staging, phases, resource usage, etc.