We’re not trying to build a clone of Linux that can run any POSIX-compatible binary. we’re doing a new thing, with very different assumptions…which is a freedom afforded by being a pure hobby project. We have ended up being influenced by a lot of ideas from throughout the history of computing, like Erlang, microkernels, Forth, et cetera. but a lot of these ideas aren’t really part of the current crop of modern operating systems. It’s kind of as though we aren’t really looking at our contemporaries (like Linux), but we are looking at our shared antecedents. Instead of writing mnemOS in a post-Linux world, we are kind of pretending we are writing mnemOS starting from the same place where Linus started writing Linux. And maybe that’s sort of like what Rick Rubin meant about “pretending heavy metal doesn’t exist”: you don’t pretend that rock’n’roll doesn’t exist, or that the guitar doesn’t exist, because you can’t create the first heavy metal album without those things. But, you do try to imagine you’re starting over from the same place where heavy metal actually did start.

I guess the concluding thought here is just that there’s value in cultivating beginner’s mind in software development…but there’s also just as much value in being able to learn from our field’s rich history. There’s a time and a place for both, and the best designer is one who is able to both approach the world without biases and look to the past for inspiration. Balance in all things.

Eliza Weisman, On Beginner's Mind in Software Engineering