Organic creativity: the Roomba process

"...because you're taking chances and doing things you haven't done before, so of course you're going to mess up quite a bit of the time, but still when you get it right... there's a freshness and vitality that is really inspiring."

That was Michael Brook (music producer, composer, inventor) this morning on NPR's Sunday Weekend Edition. He was talking about sound and music, but the notion of accepting assuming failure as part of the process works nearly everywhere. We've talked about fear of failure (death by risk aversion) a lot here, and I need to be reminded of it every day. But it was another comment he made about creativity that I especially loved.

When host Liane Hansen suggested that his work was a very organic process, he said:

"I only think one step at a time. I do something and that implies what the next step should be... I do that and then if that works or doesn't work I carry on or back up... it's like those little robots... the things that vacuum your house and they just keep bumping into a wall and then turn that way..."

"I don't preconceive it."

That was a new metaphor for me: the Roomba approach to organic design.

While this don't-plan-every-damn-thing-in-advance model has started to gain popularity in the software development world, most, um, old-style programmers like me had an almost opposite model beat into us from the beginning. The well-intentioned concern for future extensibility, flexibility, scalability led us down the design garden path... skipping along assuming that WE were the smart ones who'd be ready when the dreaded yet inevitable Requirements/Specification Changes came in. With enough upfront design and extra coding, we could make our life down the road much easier. What we lost in time-to-release now would be more than made up for later. So we said.

But then the Extreme Programming and Agile Manifesto began to challenge that idea. While not everyone has drunk the XP koolaid (and oh how I hate forced pair-programming), most modern software development teams have been heavily influenced by at least some of the XP/Agile once-edgy, now more mainstream practices.

Now fast-forward to 2006 and the world of Web Something Dot Something, to the world of lightning-fast releases and community/user-driven design. If we thought we had trouble designing it all up front before, NOW we're most definitely screwed. If we're doing it right and listening to our users, they are partly in control. Whatever we thought users would do with our products and services, they have ideas of their own. Ideas we would NEVER have imagined... things you couldn't possibly see until after it was out there in use.

The use of (and audience for) our products is evolving with or without us, organically. And if we want to evolve with it--let alone innovate and add new value that even the users hadn't thought of we have to be there like that Roomba, always trying and trying and backing up when we hit a wall and turning 90 degrees and always moving moving moving (until its time to recharge).

The poster children (or at least the most prolific in articulating it) for the don't-plan-it-all are my good friends at you-know-where (rhymes with flirty-bevin rignals). In their PDF book Getting Real (worth the $19 if you really stop and think about what they're saying), they talk a fair bit on this:

It's a Problem When It's a Problem
"Don't waste time on problems you don't have yet. Do you really need to worry about scaling to 100,000 users today...?"

Just Wing It
"Bottom Line: Make decisions just in time, when you have access to the real information you need."

"Real things lead to real reactions. And that's how you get to the truth."

Work in iterations
"Let the app grow and speak to you. Let it morph and evolve. Instead of banking on getting everything right upfront, the iterative process lets you continue to make informed decisions as you go along. The result is real feedback and real guidance..."

Sounds like Roomba to me.

Organic. Just-in-time. Real. Community/User-evolved. Clean.

(And maybe not so good for cats... a bonus!)