The switch, as usual. We are not talking about fitness cost in the beginning of life, and the article does not deal with the beginning of life either. We are discussing evolution, the time after the beginning of life. The problem with the beginning of life is not fitness cost, it is the absolute impossibility of arranging some millions of molecules into at least half a million DNA pairs in the exact way necessary to produce a living thing.
As to fitness cost, I already gave it to you in the post you are responding to " However, in real life there is a fitness cost of non-useful organs, DNA, etc. It takes energy, food, etc. to keep such useless things alive so there is definitely a fitness cost." Further you already agreed that even a small fitness differential will result in the more fit organisms overcoming the less fit.
And, again, for perhaps the 100th time, you have failed to produce your proof--or even any particularly suggestive evidence--as to why I should think that the beginnings of life could only have been a prokariote being instantly assembled out of organic spare parts.
You mean the switch from an interesting discussion to yet another recitation of the exact same feeble, execrable, canned arguments you recycle endlessly without showing the least sign of even enough rudimentary comprehension and courtesy to prevent your deponents from having to explain the same point again and again and again and again...?
We are not talking about fitness cost in the beginning of life,
We were, in just this previous set of posts, in fact.
and the article does not deal with the beginning of life either.
It most certainly does. As does any Alife experiment. The game of life, like the experiment of this article, ramps us up to Turing machines, from, essentially, a silicon substrate embedded with a regularly ordered matrix of charged transistor storage devices. If you can get to a turing machine, you can get to self-reproducing automata, as the Game of Life demonstrates right in front of your eyes, if you bother to look. If you can do it in silicon, you can certainly do it in sulpher bubbles, or clay silicates, or any of the dozen or so candidates for beginnnings that have been floating around. Once you have enduring self-reproducing automata, it is just a matter of mutating your brains out for millions of years to replace the silicon substrate with something else, just like the early amphibians gave up the ocean for the land, just like the early aerobics gave up on nitrogen for oxygen, just like the bisexual multicellulars gave up sucking nutrients from pond scum for hunting down and killing huge mobile packages of nutrients.
I haven't any idea how this point supports the rest of your argument, but, hey, what's new?
The point, if I might be so bold as to suggest sticking to it, was that it isn't necessary to punish failure to have evolutionary changes take place. It is only necessary to reward success. If evolution's beginnings were not a time of heavy dependance on hunting down your food to survive; if failure does not result in your genome losing the meat machinery in your possession to some other genome that's eaten you; then failure is not significant to evolution at this stage of the game.
What's important, I boldly aver, having just thought of it, is the ability to evolve as fast as possible, without stretching yourself so thin you discorporate.