Waterfall (2018)

10 hours ago

Winston Royce's 1970 paper introduced a figure often mistaken as advocating the Waterfall model, but he actually highlighted its risks and emphasized the need for iteration and customer involvement.
The term 'Waterfall' does not appear in Royce's paper; it originated later, and the model's negative reputation largely stems from its adoption by the U.S. Department of Defense in DOD-STD-2167 in 1985.
Royce proposed solutions like 100% branch coverage in testing and formal customer commitment, ideas that align with modern Agile practices, and he critiqued the assumption that requirements can be fully specified upfront.
Historical research by Craig Larman shows iterative development practices, such as those used in NASA's Project Mercury in the 1960s, were common, suggesting Waterfall was an accident rather than the norm.
Frederick Brooks in 1986 argued against the fallacy of specifying systems in advance, reinforcing that software complexity makes upfront detailed specification impractical, a view shared by many experienced developers.

Hasty Briefsbeta