3 ms·

But what if you have already reach optimal growth and optimal energy efficiency? If you have ever played with evolutionary algorithm, you have seen that it is

by Iv 7y ago

But what if you have already reach optimal growth and optimal energy efficiency? If you have ever played with evolutionary algorithm, you have seen that it is very easy for a system to just "win". Want to optimize a thing that heavily depends on a mass/surface ratio? Oh, a sphere is the perfect solution. Evolution can't beat it. Yes, maybe there can be various colonies of photosynthetic organisms competing over millionth of percents of efficiency for billions of years. That's what I call a dead end of evolution.

Isinlor 7y ago
You should keep in mind degrees of freedom of whatever you were playing with as well as complexity of the fitness function. Simple optimization problems have simple solutions. Physical, sustainable self replication is anything but simple. As far as I know humanity is not able to engineer anything that would be self replicating to the point where evolution could kick in. We have limited success at reverse-engineering nature but that's about it. > Yes, maybe there can be various colonies of photosynthetic organisms competing over millionth of percents of efficiency for billions of years. That's what I call a dead end of evolution. I would call that a plateau, because that was the state of Earth for about one or two billion years. I think overall we agree - transition from single cell life to current level of complexity is not an easy one. We are significantly more likely to find single cell life in the universe, than anything that would resemble animals. But you should also remember that Earth was not uniform in space and time, photosynthesis does not work equally well everywhere and overall conditions change a lot over millions of years. So even single cell life will be damn complex and dynamic on a scale of planet and evolution.

3 ms·

But what if you have already reach optimal growth and optimal energy efficiency? If you have ever played with evolutionary algorithm, you have seen that it is

by Iv 7y ago

Isinlor 7y ago
You should keep in mind degrees of freedom of whatever you were playing with as well as complexity of the fitness function. Simple optimization problems have simple solutions. Physical, sustainable self replication is anything but simple. As far as I know humanity is not able to engineer anything that would be self replicating to the point where evolution could kick in. We have limited success at reverse-engineering nature but that's about it. > Yes, maybe there can be various colonies of photosynthetic organisms competing over millionth of percents of efficiency for billions of years. That's what I call a dead end of evolution. I would call that a plateau, because that was the state of Earth for about one or two billion years. I think overall we agree - transition from single cell life to current level of complexity is not an easy one. We are significantly more likely to find single cell life in the universe, than anything that would resemble animals. But you should also remember that Earth was not uniform in space and time, photosynthesis does not work equally well everywhere and overall conditions change a lot over millions of years. So even single cell life will be damn complex and dynamic on a scale of planet and evolution.