The dangers of believing in the improbable...

[David Franklin]

Ya, computer science degree and evolution supporter here. Indeed, at school I wrote an evolution simulator in Visual Basic 3 (ha). In my simulator, the success of evolution was strongly linked to the starting conditions and the "genetic language" I used. Also, in many setups the outcome was dramatically influenced by chance.

Yeah - one experiment I'm familiar with is writing code to simulate the evolution of the shape of the eye. The basic idea is you have a scoring function which is "how much light is focussed onto the retina" and then you make random small changes to the shape of the lens, selecting those that maximise the light onto the retina, until the shape evolves to a perfect parabola (or whatever the right shape is - I don't remember, and it's not actually important).

The interesting thing was how rarely it actually worked. What usually happened is that you'd get a "multiple lens" - the lens surface would effectively be 2 or more lenses, each of which focussed light onto the retina. This was less efficient than a single lens (primarily due to loss where two lenses met), but once you had two lenses, it was very difficult to get to the single lens solution because you'd have to evolve quite a bit away from the good 2 lens solution before you could start evolving towards the 1 lens solution.

[bigdjiver]

... I'd like to proffer my opinion that the human race has stopped evolving, at least in the conventional sense.

wrong

Evolution is about survival of the fittest

wrong, it is about natural selection. The meteor strike does not select the unfit.

so that the fittest transmit their genes to the next generation.

the survivors do.

With the technology and medicines in use nowadays most humans can now survive illnesses that in the past would have killed them.

These people (who would have died without the technology) are then able to transmit their genes to the following generation, ensuring that it's not just the "superior" genes that survive and shape the future of the human race.

It is the gene pool belonging to those societies that have developed medicines.

These medicines and technologies have only appeared in the past 150 years or so, a totally negligible timescale in evolutionary terms so the effects will take some time to be felt. However as it stands we are heading for a race of inherently ill and unfit for purpose beings who will only survive by being propped up by ever more sophisticated technology. In a few thousand years if this support system were to be removed (a major war?) millions of people would die over the next couple of generations, leaving the fittest once again to shape the following generations.

This is a backtracking mechanism, so that stable "solutions" are destroyed and destroyed enabling better "solutions" to be found. Isaac Asimov floated the idea that human beings developed from failed apes that had a flaw in their maturity gene.


*rummages in the cupboard, puts on his tin hat and dives behind the nearest cover.....*

[Barry Shnikov]

No. To repeat myself, I have looked at the source code, and my description of what the program does is accurate. Bluntly, you are not qualified to argue this point, and continuing to do so is only making you look stupid.

What qualification do I need to ask you why you chose the label 'good'? As opposed to, for example, 'optimal'?

[Barry Shnikov]

So what? The only "mechanism" you're showing is essentially a "God" one: for no obvious reason, mutations towards a "chosen form" are selected for. This is more of an argument for intelligent design than one against it. You end with a model of evolution with no blind alleys, no backtracking, no suboptimal solutions that become dominant and effectively fixed (c.f. the design of the eye). That's not what we see in the world.

In real life, the selection mechanism is a key factor when it comes to evolution. It would be great if we could say only mutations that get us closer to a final goal are selected for, but it doesn't work like that. There are countless unsolved problems in computing that could be trivially solved if we could get a selection mechanism even one millionth as accurate as the one used in the applet.

Would you mind dealing with my example of the model of inversion in human vision?

This is one of the reasons I take task with you. You seem to think "who cares if it's a nonsense, unscientific argument, so long as it convinces people". I'm sorry, but as far as I'm concerned, that lumps you in with the creationists et. al. who want to convince people by holy writ (be that writ from the Bible, or from the latest pop-sci book on evolution). Browbeating people isn't science.

I regret that I 'seem' to think that. What I do think is that possibly you are unaware of the sophistry of the arguments you are putting forward. The lack of prime suitability of the selection mechanism chosen by this particular programmer is NOT the point. The point is that it has a selection mechanism and blind chance doesn't. As with the vision model, the crudeness of the illustration does not invalidate the premise.

[It's interesting to note that (as far as I'm aware) pretty much everyone disagreeing with you here [/QUOTE]
Um - that would be two people - o no! I'm outnumbered by scientists!

has a degree in science and supports evolution. That might tell you something...]

Unfortunately, more realistic algorithms wouldn't prove your point. As I said earlier: evolving a sentence that makes sense? Distinctly possible. Evolving a particular sentence? Very unlikely - rather like the chances of evolving a particular human being.

Well, exactly! What would be the point in a model which was able to 'evolve' a phrase in a time period which - although geological - would still be a fraction of the time required by blind chance? "Uh - by chance, this would take more than the entire age of the universe. But - um - with this model, once you click the start button it'll be merely two, three million years. QED."

[bigdjiver]

Perhaps I'm just prejudiced, but some of the most fecund and fertile people (the ones with the most children, that is) appear to be the least attractive and the least healthy. I've heard it argued that free childcare, child benefits, healthcare and social and welfare benefits are doing immense damage to the gene-pool!

If an IQ of 160 was superior it would be the norm. People who are "stupid" enough to have kids at 14 and keep on having them, and perhaps smoke and drink are probably breeding at twice (compound) the rate of the sensible "wait until I have my degree" types. The gene-pool is getting its priorities right.

[bigdjiver]

Yeah - one experiment I'm familiar with is writing code to simulate the evolution of the shape of the eye. The basic idea is you have a scoring function which is "how much light is focussed onto the retina" and then you make random small changes to the shape of the lens, selecting those that maximise the light onto the retina, until the shape evolves to a perfect parabola (or whatever the right shape is - I don't remember, and it's not actually important).

The interesting thing was how rarely it actually worked. What usually happened is that you'd get a "multiple lens" - the lens surface would effectively be 2 or more lenses, each of which focussed light onto the retina. This was less efficient than a single lens (primarily due to loss where two lenses met), but once you had two lenses, it was very difficult to get to the single lens solution because you'd have to evolve quite a bit away from the good 2 lens solution before you could start evolving towards the 1 lens solution.

House flys have eyes too, and they seem to prosper.

[David Franklin]

What qualification do I need to ask you why you chose the label 'good'? As opposed to, for example, 'optimal'?

I described how the program worked, and you argued with me. The word "good" was used in quotes because I was perfectly aware that "good" is a human interpretation: evolution doesn't care about "good". But more to the point, it makes absolutely no difference to my description what word is used, whether it is 'optimal', 'good', 'bad', 'indifferent' or 'quaxiflargoid'. That you do not understand this only underlies the futility of discussing this point further.

The fact of the matter is that the selection applet is a really, really bad illustration of how evolution works. You can argue that it's "just a simplification", but the problem is that it's a simplification that gives a completely distorted view of evolution in the real world. It's as bad as creationists saying "the odds of humans evolving are 10^10000" and then saying "well, no, we didn't consider selection pressures. But that was just a simplification".

I also note that if you were to transfer the applet's modus operandi to the real world, you would be saying: "there's an overseer who looks at us, and compares us to the ideal form. Those that are closest to the ideal are 'chosen' and get to have more children". I can't believe someone with your views on religion really wants to go down that road...

[Dreadful Scathe]

Perhaps I'm just prejudiced, but some of the most fecund and fertile people (the ones with the most children, that is) appear to be the least attractive and the least healthy. I've heard it argued that free childcare, child benefits, healthcare and social and welfare benefits are doing immense damage to the gene-pool!

H.G.Wells had the right of it - another millennia of relying on technology, pill popping and eating foods-in-a-box will see us regress to that of 6 year olds.

[El Salsero Gringo]

I regret that I 'seem' to think that. What I do think is that possibly you are unaware of the sophistry of the arguments you are putting forward. The lack of prime suitability of the selection mechanism chosen by this particular programmer is NOT the point.

That point is well made, and understood by all. (Trust me we do get it.) But the other features of the example suck so badly in reference to evolution that that factor outweighs any benefits the example brings.