@Maikel — Yeah, I think the concerns about consistency and randomness are fair and well-meaning, but ultimately counterproductive — and you're certainly not alone in voicing them (
@Kartik Agaram said something to the effect of "
fucking mice, don't touch my damn cheese!" which is wildly out of character for him but yeah that guy never ceases to surprise me).
I am not going to be able to do anything in my lifetime that
stops anyone from writing business software, as much as I would love to. I'm also not, in a million years, going to try to design a programming system that stands up to the demands of having business software built in it. That camp is
over there, and I'm
over here, and never the two shall meet.
But there's more possibility space here than just
business software and
research faff. One of the more serious domains that does seem to have some overlap with my desires for aliveness is electrical engineering. EEs bend physics, and physics is the biggest cheese-mover in the room. When they get deep enough in the sauce, EEs need to
grapple with all sorts of wildness. I think it'd be possible to make something that feels alive while being truer to the laws of physics than most software is, in a way that'd make EEs feel at home. (I'm not interested in doing this, but it's worth thinking about.)
@Dave Ackley is also a big proponent of getting away from worries about consistency and determinism, and I think his reasoning is rock solid — when you start doing things spread-out in space (or time), you need to deal with some inherent, inescapable nondeterminism, and the only scalable choice is to just lean into that and see what you can do with it, see if you can dance with it and bend it into a beautiful bow, weave it into your hair and come up with a meal plan where you balance nondeterminism with other kinds of fibre. (His exact words, of course — and totally in character at that!)