In a paper called "Trying Not to Build the Same Old Spacecraft: Structural and Political Issues in Design Inheritance" (the pdf article is damaged so see the slides from the talk), Charlotte Linde explores the concept of "design inheritance", i.e. "how people and institutions attempt to plan for inheritance or to avoid being locked into the consequences of previous design decisions". In a sense, as she describes it's "No one wants to be responsible for the next Y2K problem" or "More thoughtful designers don’t want to be responsible for the next QWERTY keyboard". Working at the NASA, she takes the "spacecraft" as the example in her discussion. The interesting thing about spacecraft is that is both "iconic of the future" that it requires a strong inheritance of past design (because it's expensive/complex/risky and also because of the low number of prior examples). Using this example, she then shows how the social and economic factors forcing design inheritance.
"Factors Retarding Change
- “They got it right the first time”
- Installed technology base
- Amount of risk acceptable: Old technologies with known failure parameters can be preferred to possible improvements with unknown failure parameters + Are there social mechanisms for learning from failures?
- Length of mission (as compared to rate of change of technology)
- Rate of change within spacecraft design subsystems"
Why do I blog this? because it's an important topic, as I am regularly pissed by how the wheel is being reinvented all over again. The talk seems to scratch the surface about how to "plan to plan" through spiral development or modularity ("Attempts to leave room for the insertion of not yet developed technologies").