"While it has been a faithful companion, your Companion Core cannot accompany you through the rest of the test."
It's funny but to me your analogy works the other way in my mind. The "evolution" of the car you describe has one thing that drives every change, intelligence. Either the auto maker "decided" to try something different, or the consumer "chose" one vehicle over another. I think the fact that you came up with this analogy points to the fact that humans by nature have a sense that something is making all this madness into some kind of sense. Hard to shake no matter how hard you try, which is why I think honest teachers are having this problem.
Now find me a Lamborghini in a crater on Mars and I'll reconsider your point.
That's what I thought when I read the title.. Oh well, I guess I'll keep waiting.
Yeah I'm already skeptical that this would produce good results.. Not having any samples doesn't help much.
Seems like a lot of people commenting in the article were confused about this. I think the temporal aspects throw people off. They think, well one is born first, it would be the gambler's fallacy to say that has any effect on the other. When you think of the problem in terms of a finite set of possibilities, and you are choosing one, it becomes clearer. Here's another example of the problem:
You have a bunch of marbles, each is a color: green or red, and each is either solid or dotted. Given any set of two marbles, there are 16 possible unique sets:
If one marble is red/dotted, what is the probability that the other is red?
Here are all the sets in which one marble is red/dotted:
Out of these 7, 3 of the other marble is red. Thus a probability of 3/7 that the other is red. However, if you were to randomly draw a marble from the bunch, that marbles color or texture would have no effect on the next marble drawn from the bunch.