l33t gm haxors?

Friday, 7 February 2003

How do you convince people that genetic modification (GM) is a good thing? Richard Dawkins evidently believes that you do it by telling them that the opponents of GM are luddites who don't get it but that their children probably will. Genes, he says, are just like software subroutines so, although there are a lot of Luddites whose gut hostility to GM is based on ignorance, their children, who understand computer software much better than their parents, will easily recognise the resemblance and welcome it.

Dawkins develops his argument with the analogy of a rocket guidance system developed by lifting sub-routines from (say) a financial spreadsheet program. If the rocket guidance system needs a sub-routine that calculates square roots and the financial package has one, why not just lift it from the latter and paste it into the former? A square root is a square root is a square root after all. Dawkins admits that this analogy is a little unfair: suppose the routine to be lifted comes close to doing what the guidance system requires, but isn't quite right. All we need do then is to tweak the routine a little. And we can do the same with genes.

At this point, experienced computer programmers might note that Dawkins' doesn't understand computer programming as well as he understands genetics.

It's unlikely that a rocket guidance system would be written in the casual way that he suggests. However, a lot of other software has been. Especially on the Internet. If you're building a reel kewl website for the Nar-Nar-Goon Chapter of the Satanic Knights of Death Metal and Quake Arena with a message board and other kewl stuff, there are plenty of places where you can find free CGI stuff. In a few months, when the site goes irretrievably pear-shaped, it won't be your fault: it will be the fault of the idiot who wrote the form to E-Mail script and those bastards in the Perl newsgroup who keep saying "RTFM" when you ask how to fix it.

My generation isn't without fault here either: many of us have wreaked havoc on corporate IT budgets with the help of certain Windows based programming tools, encouraged by the attitude that professionalism means being paid a lot of money and wearing a suit to work.

Considering this, Dawkins' casual assurance that GM isn't very different to computer programming is more a case for Luddism, rather than a case against. I doubt that he is seriously suggesting that GM is going to be applied as carelessly and ineptly as the creation of the website of the N-N-GCSKDMQA but, if he is, I'm with the Luddites.