Guy Threepwood
Mighty Pirate
The Zahtzee example was only given to demonstrate that evolution involves selection too, not just random variation.
I understand; natural selection of random variation. No analogy is perfect of course, but the example does demonstrate the principle- that probabilities and selection criteria do matter, (random variation +natural selection) in and of itself, does not necessarily equate to improved new generations, unless those numbers specifically support it.- the devil is in the details.
At one end of the scale, I probably play about 10 games of Yahtzee a week ( yup I'm pretty wild!) and of course with only 5 dice, 6 numbers and a 99.9% efficient selection system- a couple of Yahtzees is no huge surprise-
Then at the other end- Randomly mutate a percentage of the letter characters used in War & Peace. It is not impossible that the book would actually be improved. It's just so improbable, that even with a new generation of one million variations, even if the very best is selected to be copied, it will most likely be inferior to the original, not superior. and so on with every generation.
Where does life, digital genetic code fit into this spectrum? I'd argue that a mere book is selling it a little short, but that's the debatable part, and something Darwin could not dream of. But in every analogy; Yahtzee, books, life, cars, software- It's a question of information systems, something we are just beginning to grapple with at the dawn of the information age, - there is no easy answer these days unless we revert back to a Victorian age model where protoplasm simply does it's thing, by some presumably simple process.
There is no predetermined goal in evolution. The mutation/variation is random but the success of that mutation depends on the ability of that mutant to breed and pass the gene on. That's why the car analogy fails because their design is to a large extent predetermined.
well you make the point, that nature is at a disadvantage here, it has a far inferior selection method. We can test, review, and model outcomes of new automobile technology. All things being even- a .01% improvement in fuel efficiency CAN be selected for right? and WILL be passed onto new generations of designs
But the car/life analogy REMOVED these predetermined design goals, purely random changes, which must then go on to prove themselves in the field.. a consumer will never notice the .01% improvement, it will NOT be selected for, unless it is a very significant improvement. So too with life if you remove the goals
The photocopy analogy falls down because the copies will always deteriorate/distort i.e. more blurred or whatever the fault is, a photocopier cannot improve a picture's quality. - in evolutionary term they would not succeed and be discarded and die out
So that would be my point also, with a master copy to revert to, plus some random variation within specifically supported variable parameters (what day this year's Christmas party falls on) you have a pretty good Yahtzee like chance of success.
So too with life, a master plan + variable parameters for things like overall size, hair length. color- + natural selection- that works fine.
In all the above examples- the objective test lies within in the laws of pure mathematics, it applies to everything, life is not excepted