The iterative writer

I’m very interested in the intersection of fiction (written or filmed) and design, and while science fiction icon Bruce Sterling said recently that [“design has more to offer fiction at the moment than literature has to offer design”](http://magicalnihilism.com/2009/10/14/icon-minds-tony-dunne-fiona-raby-bruce-sterling-on-design-fiction/), I’ve found it extremely valuable as a designer to learn from both writing and the writing process.

The other day I caught up with a friend who had recently completed a screenwriting class. He described his writing process to me and it had many parallels to the conceptual design process.

First, he said, he determines where the story starts and ends. From these two points, he can build up the rest. It’s tempting to tell yourself “I just need an idea, a start or a finish, and I’ll figure out the rest later.” But without a start *and* a finish, you don’t have a unique story yet and your work is likely to wander aimlessly. His first project had only a beginning, and he ended up scrapping the script because it didn’t have anywhere to go. It’s also much easier to try out very different stories when they are simply a pair of beginnings and endings than when you are writing a full script for each version.

In concept design, design principles are analogous to the script’s start and finish. They provide a high-level view of what you want the experience or product to be like, or to accomplish. They are much easier to experiment with than fully-realized products or even prototypes, and without them your solutions will be all over the map.

Next, my friend creates the first section of his story, from the beginning to the main character’s first turning point–for instance, when Luke Skywalker first discovers Princess Leia’s plea for help while cleaning R2D2. Sitting down, he tries to bang out a treatment for this first section in a single sitting. This helps him see if the setting and characters are good enough to support action, dialogue, and getting the plot to the desired end point.

After finishing, he sets that treatment aside and does something else for the rest of the day. The next day, he returns to his station, but instead of picking up the draft from before, he begins a brand new version. Without trying to consciously repeat *or* avoid anything from the first version, he simply writes another draft of the scenario, then sets that aside as well. After 3 or 4 versions, the best elements have generally persisted while the extra stuff disappears, and he has a solid section of the story. He repeats this process for each subsequent section until he has a full draft.

Concept design benefits from a similar process. Apple famously [uses a “10 to 3 to 1” process](http://www.businessweek.com/the_thread/techbeat/archives/2008/03/apples_design_p.html) where their designers create 10 separate, equally-detailed and -viable design concepts for each new feature. I talked yesterday to Ford’s head of interior design, and he cited a similar process for physical prototypes–“so many different versions you can’t even count”. Later in the development phase, of course, you want to drive toward convergence, consistency, usability, and feasibility. But early on you shouldn’t be stingy with the concepts–keep them fast and loose, and make sure you cover as broad a range as possible.