But in normal healthy people we see enough variation in abstract reasoning to have a profound effect on their performance in our brand spanking new technological environment of the last couple of thousand years; less than an eyeblink in evolutionary terms. And not just in competitive terms, but in absolute ability. A lot of that advancement can be simply be attributed to better nutrition and better mental and language models. I have little trouble believing that fiddling and tuning up the brain's electrical environment might might have some comparatively striking effects. (I can also see it possible to do great damage as well)

There are a lot of low-hanging fruits nature has missed because nature only cares about how many copies of genes there are, it doesn't care about human goals. Evolution is a poor optimizer for what it does, it has no optimization for what we want done.

I have to verify your other numbers sometime, because I am pretty sure I have heard of cases where mutations took hold faster (lactose tolerance perhaps?)

A good question to raise, but it has a lot of plausible answers. For example, maybe having better cognitive ability causes people to spend more time doing science and less time procreating/looking after their offspring.

If nature is a woman (rather than a collection of stochastic processes), then she is a strong argument against letting women into the same building as a computer.

Most biological organisms are designed incredibly badly. The body is full of spaghetti code, duplicate functionality, dead code that still causes bugs, and systems that evolved from a completely different purpose and didn't get the full rewrite they needed.

The nature methodology makes waterfall look like a perfect process:

1) Make a random change.

2) If conversions go up, commit. Otherwise revert.

This along with the following paragraph come across as somewhat misogynistic. I'm sure you didn't mean it to. Saying "person" instead of "women" would have avoided that.

I liked your description of nature's methodology as a development process.

So yes, I did mean to make it sound somewhat mysogynistic.

You know, I hear tech people saying stuff like this all the time but the question I would ask is: designed badly compared to what? I mean, if you're comparing to software code made by humans (is there any other kind?) then I think biological systems work pretty damn well. The type of systems that humans write code for are utterly trivial compared to what biology has to deal with -- and we STILL have all the anti-patterns and problems you mentioned with respect to biology. Until human beings can write code of comparable complexity to life that lasts millions of years (forget about billions) I think you should show the woman some respect. :-)

Compared to what it would look like if a human designed it. I took a course in math biology, and learned how a few of these subsystems work. A particular gene regulatory network I studied (can't recall the details, it was >5 years ago) looked like the following:

    B, C = (False, False)
    A = get_input()
    if A:
        B = True
    if B:
        C = True
    if C:
        do_something_interesting()

    A = get_input()
    if A:
        do_something_interesting()

The debate over intelligent design is clearly over. If you want to argue for creationism, you need to argue that god is a drooling retard who starts a project, forgets what he is doing, and uses the pieces to finish a completely different task.

Just saying, that kind of redundancy might be an advantage for evolution, smoothing it's path.