[mlyle]

Yah, but if you needed 2 hours less of sleep, there'd still be myriad advantages.

Times when it's not possible to get enough sleep: you'd be less sleep-deprived.

And, if it's truly without negative side effects: maybe you get 20 minutes more useful time per day and increase your ratio of conscious relaxation to useful time. I'd call that a pretty big win.

[naming_the_user]

It's interesting but I feel as if the discussion is a bit flawed because it feels very binary, i.e. sleep 8 hours = have 16 hours non sleep, sleep 6 hours = have 18 hours non sleep, and we assume that each hour is the same regardless, but it won't be.

It feels a bit like - well could I engineer myself to require 1500 calories a day instead of 2000. Maybe. But as a result there would definitely be some downside, maybe I'd be less physically strong, or have less endurance, or have a bit less processing power, or have to rest more, etc. If we straight up designed a more efficient body, then I'd rather keep eating 2k and be "more powerful", or maybe I'd go to 4k!

Maybe I can sleep for 2 fewer hours and retain the same or more productivity. Or maybe I could sleep for the full amount of hours but be more well rested and as a result be more productive in the awake time. That sort of thing.

[mlyle]

Yah -- note that I didn't posit that things would be the same at the margin (120 minutes more conscious time without other downsides yields only 16% of "useful" time and assumes the rest goes to minimally useful relaxation and recreation).

> There is no free lunch.

Selective pressure and evolution are pretty good at optimizing... but not perfect. What they optimized for, also, is not quality of life in the modern world. It's likely there are a whole lot of free lunches available, or at least big wins with relatively low opportunity costs.

It's certainly not great to discard potential improvements because of an assumption that what we have must be optimum.

[naming_the_user]

Sorry, I edited my post a bit so what you've quoted isn't there any more but I agree fully with what you're saying.

I guess what would be cool is being able to play with the sliders. People already do this in other ways, for example people at the extreme end of bodybuilding or strongman are almost certainlhy explicitly shortening their lifespans for a shorter term benefit, calorie restriction looks like it lets you go the other way, etc.

Maybe you could even fiddle and have 16 hours of sleep and overclock your brain for the other 8 being super-intelligent, lol.

[jvanderbot]

It might make sense to just drop the term optimization when discussing natural selection.

It's not like "Natural Selection" was given an ecological niche and optimized from scratch the being for that niche. That mental model implies there's not much to be done and any changes would produce a less useful product.

A more apt term would be "Refactored" or even "Patched".

It's more like a giant ugly legacy software system was added to, and the result was just enough to keep going in the new system.

I like that analogy better because it implies: 1. There is room for optimization, and 2. Any minor change is likely to break things far away due to the continuous re-patching and legacy cruft. Both of which seem correct.

[iterateoften]

> "Natural Selection" was given an ecological niche and optimized from scratch the being for that niche

I think you are conflating natural selection the process with organisms that are a result of natural selection.

Natural selection isn’t a process that is intentionally made by anyone, but a way we describe a simplified model slew of complex processes.

Also conflated is what you are optimizing for. The only thing natural selection optimizes for is survival. That’s it. Nothing else matters expect which individuals survive long enough to pass their genes to the next generation. As a result niches develop and as each generation survives compared to their peers their survival strategies are optimized.

Optimization perfectly describes what is happening to the survivability and reproduction of certain genes.

You appear to be working backwards. That there was a niche and evolution somehow crafted an organism to fit that niche. That is misleading. Genes replicate, and in a certain context some genes survived and replicated better than in some other context. So genes evolved in that new context creating a niche.

[jvanderbot]

You're right about the mechanism of course, but the language and mental model I see over and over is that it a bespoke optimization that ignores the fact that it was incremental changes, and that some adaptations are no longer beneficial because they were created for an environment that no longer exists.

I'd like to deprecate that mental model.

[mewpmewp2]

Unless there's some huge evolutionary inefficiency that can be artificially hacked. But it's a big if. If there was it would be possible to get some benefits without significant or no downsides.