Learning From Accidents and a Terrorist Attack.
My essay “Software That Lasts 200 Years” has proven very popular. It
discusses software industry structure that could be helpful for a
particular type of software. It also suggests learning from other areas
of engineering to find principles to follow when creating long-term
software. I've just finished another essay that is a step towards
coming up with some of those principles.
I look at the study of major accidents in societal
infrastructure such as nuclear power and process plants. I also look at
the 9/11 Commission Report to examine the handling of the immediate
results of the hijackings and crashes. These are examples of systems
being stressed in foreseen and unforeseen situations. Terrorism is
relevant because it can result in forced changes to the configuration
of a system that were unanticipated by the original designers and
brings in a wide range of participants. From all this I glean some
principles that can be applied to design in general.
As part of examining the 9/11 Commission Report, I list a
series of excerpts that relate to communications, unplanned activities,
and the role of regular people caught up in the situation. These are
areas that relate to things that interest me and that I've been
covering for a long time. Whether or not you are interested in “200
Year Software”, I think the information there can be very important to