Programming errors:
One great advantage of communicating with others is that they can point out ones own blind spots. That happened the other day on an airplane. I was looking forward to spending the trip reading, but the man in the seat next to me seemed more interested in talking than in leaving me be. Well and good, I told him what I could about life, kinship and fertility. He verified that I had said that the current sluggish economy and the Great Depression were both caused by a low birth rate. Then he asked, if babies are good for an economy, why aren’t places with a high birth rate more prosperous? All I could do was wave my hands and say that if you had an industrial society that only needed more throughput in order to increase productivity, then babies made you rich. If you had a traditional farming society in which productivity could not be increased by available means, then babies made you poor. I didn’t do a very good job at it, but I much appreciated the insightful question. I gave him my phone number, but never heard from him again. I rarely have someone who asks good questions.
So I have conversations inside my head. Sometimes tempers flare in there. The two sides can get really nasty. So far no punches have been thrown.
One of my repeated conversations goes like this.
“Harrumph. So you have this computer program that models the genes of a population?”
“Yes. Precisely. Good alter ego.”
“And the fertility behaves just like the fertility in a real population.”
“With all due modesty (and not much modesty is due) that is the case.”
“How do I know you aren’t lying?”
“Because you’re me, you idiot.”
“But how would anybody else know you aren’t lying?”
“They just have to come up with a computer program that gives the same results some other way.”
“Maybe it’s just a movie.”
“But there are a whole lot of variables. You can change them one by one. Each slight change has a slight change in the result. And you can run them over and over again and not get a repeat. That would be more movies than you could fit on a CD.”
“But maybe you’ve made a programming error that you don’t realize.”
“Hah. Hum. Yes. Err. Well. I see what you mean. After all, every time I do something I make every mistake one could possibly make. The computer program was a personal worst along those lines.”
“Yes, I know. I’m in here, too, remember. I had to live with that and all your whimpering and hysterics and histrionics and nosebleeds and sleepless nights and dry heaves.”
“Sorry.”
“It goes with the job. Anyway, you don’t really know what’s in that program. Nobody could. It’s too complex. Too complex for you, anyway. So you may be getting the right answer for the wrong reason.”
“So what’s the problem with that? It’s still a computer program. If my computer can do it, DNA can do it. The point is that DNA can generate the same curve of fertility against population as real life does and there is nothing else that could.”
“Don’t quit your day job.”
I hate that guy.
Anyway, after some years during which my virtual population mimicked real populations perfectly well I realized that there was more data in the real world that I wasn’t accounting for. The Great Depression, for instance. So I re-designed the program in several ways but to no avail. The data was there and I couldn’t account for it.
In the end I simply rewrote the program for bigger numbers and tweaked the devil out of it for weeks. I found two things. For one, I could account for the newly noticed data.
For another, I found that I had been wrong all along, just as my alter ego insinuated. It isn’t genes that are fine tuned to each other. It’s chromosomes. In fact it’s haplotypes. I made a programming error years ago and ever since then the program has given the right results based on the wrong theory. But I think I’ve got the theory right now, and it isn’t that bad. Some day, maybe when I’m through with the novel, I’ll go into the new data and show you how the model can now see it. For now I just wanted to say that there is at least one more assumption in addition to the list I gave at the beginning.
And maybe I should be a bit more modest after all. My great break through was simply an accident. But it was an accident I was looking for.
Oh, and there is an error in the “summary” part of the program, too, but the screen print still seems to be true bill.
There have been 2,662 visitors so far.