Graduate student of Saint-Petersburg State Electrotechnical University.
SqrtMinusOne
Karma: 3
If we imagine removing the easiest prediction tasks (which evolution has already solved) from the human sciences:
I think a lot depends here on what it means to for a task to be “solved”.
After all, we already have “folk physics”, just like we have “folk psychology”. If we’re to throw a rock, we can sort of visualise a parabola-like trajectory, for instance. If we’re to release an apple from a hand, we can roughly predict where it will land. Much like with human sciences, evolution has already solved these tasks for us.
Well, perhaps evolution hasn’t given us numerical solutions, i.e. we can’t intuit the parabolic trajectory equation using our bare minds, much less the coefficients. But neither can we intuit the equations for what makes a joke funny. Like with folk physics, we only have this black box-ish understanding of funny things.
So, can we really remove the humour/emotion/etc. prediction task from the set and keep the trajectory prediction task? Perhaps the problem here is that, after all, the human equations are inherently a lot more complex to produce physics-like solutions.
Perhaps some legislative action to forbid OSes from doing stuff in the background would be reasonable. This regime can’t be optimal: the system downloads uninspectable binary blobs, crafted by an entity with questionable incentives, and there’s no opt-out.
Security is an issue, but we can have auto-updates only for the security component, and leave the rest to the user.
Not to mention how annoying it is that whenever I start my Windows VM (like once per 1-2 months), the system starts downloading some huge files with no way to stop it, runs out of space, etc.