Monday, 15 November 2010

Lowering The Price of Failure

Back in the 60s my mum was a secretary. She spoke a few languages and worked in some cool places, and she used a typewriter. When my sister was born, she gave up work and spent the 70s doing a fair imitation of the Good Life mixed with Abigails Party (think black forest gateau served with home-made wine…).

When we'd all decamped to college she wanted to go back to work, and picked up her typing again. Initially bamboozled by word processing, she was amazed with spellchecking and backspace. Make a typo - no problem: just lean on that left arrow button and hit "Del". No more tippex or white ribbons fed into clackety typewriters.

When Henri Cartier Bresson took photographs in the 30s, he would compose a shot, check the exposure, choose the moment, and press the shutter. Then he'd walk away - the shot done, the moment captured. I read on a blog yesterday about a guy who has an ambition to get 10,000 photos on his Flickr account by Christmas. And he's not even a photographer.

So the thing that has changed is that the price of failure has just dropped.

We're starting out on building a pretty ambitious bit of software. I've been agonising over features and what to include, but the great news is that we don't have one shot. For example, Flickr and Wordpress put out multiple versions of their sites every day. Matt Mullenweg (Wordpress) blogs about this here. If a new feature, or some "optimised" code doesn't work, they just revert back. The price of failure is low.

The price of building software and not iterating it can be very high.