h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

Two journal postings in one calendar year. A new record for me!

A quick side note before I get started: Between writing the last entry and this one, I read DNA's last book, "The Salmon of Doubt". I was delighted to find that many of the ideas I'm describing here are explained in his own words in the book. I highly recommend getting yourself a copy (and reading it, too!) if you haven't already.

As promised, here's Douglas Adams' second observation on technology that changed my life: computers are just high-tech modelling devices. A computer is nothing but a lump of plastic and silicon without programming, just as clay is nothing but a formless lump without a sculptor's hands to mold it. This means computers are very adaptable.

However, the adaptability of computers is a double-edged sword. In the Good Ol' Days, people used to spend a lot of time designing and building things. For example, if I bought a radio, there was an electronic circuit inside designed by a human being. And the radio worked quite well. That's because if there were anything wrong with it, the designer would be in a heap of trouble...thousands of units would have to be recalled, scrapped, and replaced.

With software, the "designer" is a programmer. But there's one major distinction between the designer of an old radio and a computer programmer: the programmer isn't under as much pressure to avoid mistakes, and so most computer programs are essentially unfinished prototypes. If a programmer screws up, he can just tell his customers to wait for the next version. The computer itself doesn't need to be recalled, and the expense of issuing updates and fixes is much lower to the developer...and consequently much higher to the consumer, who is forced to rely on a shoddy, unreliable, unfinished product. Desktop computers as we know them are unreliable...they crash, contract viruses, and destroy the document you worked days to type. That's because there's little economic motivation for programmers to create a reliable product.

This means we'll always notice our computers, and we'll never stop thinking of them as technology...which, if you believe my last journal entry, means they will never really become useful. Or, more to the point, they are precisely as useful as their programming is finished.