Chris Kline, 2K Boston’s technical director, espoused the benefits of embracing failure during a postmortem of the Bioshock development process at the Paris GDC.
"Bioshock was basically a sequence of failures and errors," he said. "But every one of them was a good thing – it forced us to look at the game and reassess what we had, which worked in our favour."
Charting the development of the product back to 2002, Kline pinpointed almost a dozen points at which wrong decisions or wrong paths were taken, ranging from bad middleware decisions right at the start of the project to the team’s original misguided decision to simply remake System Shock 2, a game that had failed commercially despite universal critical acclaim.
One of the original differential points for Bioshock was to be its self-sustained AI ecology, which Kline called "life around you, but also without you". While impressive results came from months of intensive focus on these interactions, it soon became apparent to the team that it was incredibly boring without any player involvement in this ecology. This forced the team to rethink not only the entire system but also the designs of the AI agents themselves to make them more interesting and empathic.
Kline also pointed at the team’s desire to make a ‘vertical slice’ of the game as being a failure on their part, because the need to have every system working took focus away from the high-level, differential functions that needed the most attention – such as the AI ecology, environment and plasmids – and on to problems that the team had already solved before, such as quest systems, a workable inventory and audio logs.
The core message behind the talk, though, was that for all of the hubbub surrounding new development paradigms such as scrum and other eXtreme programming practices, the real thing that made Bioshock such a compelling experience was due to listening to and carefully considering all input, especially that which came from outside the team, and being able to accept mistakes and adapt to solve them.
"Iteration was responsible for all of the major things in Bioshock that people liked," he concluded, "but I don’t think you can put any methodology behind it. The real thing that you have to realise is that mistakes aren’t always a bad thing."