Book Idea #1: Old genes vs. new genes
Posted by aogFriday, 27 July 2007 at 10:16 TrackBack Ping URL

This post at the Daily Duck made me think about what a world with human genetic engineering might look like, and how one could create a plot for a novel from that.

In that vein, I think that we will discover that while fixing genetic problems will be easy, making improvements will be hard. The difference is when fixing a problem, we can find examples of the “right” genetic sequence in other people. We don’t have to have much understanding of why the right sequence is right, as we can see it works better by direct observation. When you want to go beyond that, then you need to know a lot more about what you’re doing. And, as has been noted, the design of the human genetic code is not a nice, clean, modular one. It’s far more akin to what you’d get by incomprehensibly jiggering stuff until you got lucky and it worked1. Having worked on such systems, making forward progress is very difficult, because any little change can break things anywhere.

Given that, how will progress be made? The cynical part of me2 thinks that wealthy first worlders will experiment on poor third worlders, to see what happens to various proposed improvements. Most of the time the third worlders will get hosed. But what happens if the improvement works? One could postulate a world on the brink of instability3 where genetically enhanced clans of the formerly poor are contesting for dominance as homo novus with the “old gene” clans who have the money but haven’t bred themselves superior yet. And what if the children of the Old Genetics think they have more in common with the similarly enhanced New Genetics? Seems likea nice brew for some cyberpunk writing.


1 We called this the “peturbation theory of programming” as we (the teaching assistants) watched students try to get the programming problem correct by cloning example code from the book and then crossing it with other example code until it passed enough of the tests to get a passing grade.

2 99 44/100%

3 If it isn’t, why write about it?

Comments — Formatting by Textile
Bret Friday, 27 July 2007 at 17:11

Regarding fixing versus improving: even if you know the “right” algorithm, it can still be extremely difficult incorporating that code into a complicated program, especially when you might have to touch hundreds, or thousands, or more pieces of code. In other words, what happens to be the right sequence in one person might be disastrous in another.

If you look at luminaries like Kurzweil, etc., they’d probably claim that we’ll be able to model the entire human body fairly soon so we can just use, well, genetic algorithms to see what would work the best.

Annoying Old Guy Friday, 27 July 2007 at 19:18

But you don’t have to know the algorithm, only the (literal) code. Obviously you can replace entire alleles with little problem, or human reproduction wouldn’t work. So you can swap in a good allele for a bad even in the worst case.

Bret Friday, 27 July 2007 at 19:31

Maybe, but you may be making it sound a bit simpler than it actually is. I suspect we might live long enough to see some such tinkering, if not in humans, then in other animals.

Michael Herdegen Saturday, 28 July 2007 at 05:48

That sounds like a pretty good setup, especially the part about experimenting with third-worlders - without their knowledge or consent, of course.

I suspect we might live long enough to see some such tinkering, if not in humans, then in other animals.

Assuming that I avoid getting hit by a bus anytime soon, I’m completely certain that tinkering with humans will occur within my lifetime, which is to say, beginning sometime in the next 60 years.

And I too expect it to be difficult to expand the range of superior human performance. Perhaps starting with faster reaction times might be fairly simple, and give a good return on investment ?

Annoying Old Guy Saturday, 28 July 2007 at 08:03

I think re-writing the DNA for zygotes won’t be much of a challenge. That’s just mechanics.

There’s also the issue of patience. You could do the human work relatively safely, if you were willing to spend very long periods (dozens of generations) do very small tinkering. That’s the setup in on of the cultures in the Miles Vorkosigan series.

Post a comment