SEMAT meets to define software engineering method, theory
April 30, 2010 —
(Page 1 of 2)
Related Search Term(s): SEMAT
The year was 1905. An unknown (but not for long) patent clerk named Albert Einstein wrote a series of papers, including the particle theory of light and his famed theory of relativity, that marked a revolution in physics. His work challenged the beliefs of scientists of that era, and a period of debate and argument ensued as physics theory was turned upside-down.
So in 1911, Belgian industrialist Ernest Solvay established in Brussels the first world physics conference, with the goal of defining the core tenets of theoretical physics that could be universally taught and used.
Today, a handful of notables in the software engineering arena are working toward the same goal: unambiguously defining the universals that underlie all development methodologies.
The effort, called Software Engineering Method and Theory (SEMAT), is being spearheaded by Richard Soley, chairman of the consortium Object Management Group; Ivar Jacobson, co-creator of UML and the Rational Unified Process, among other work; and Bertrand Meyer, who created the Eiffel programming language and is an acknowledged expert in object-oriented programming and language theory.
Soley explained that software engineering is gravely hampered by immature practices. “When technology is new, it’s the wild wild West. When it gets to be 50 or 60 years old, an engineering discipline evolves,” he said. To be a true discipline, it must have wide awareness, be supported in the industry and taught in the universities.
One of the challenges the SEMAT group hopes to overcome is simply getting everyone into the same room. It’s a challenge Soley faces at OMG. “We bring together vendors and practitioners, but not enough academics,” he lamented. “To put together a core theory and best practices, you need to see it from a standards perspective, an industrial perspective, a vendor perspective and an academic perspective.”
Soley pointed out that the SEMAT effort is not about a product, standard or curriculum. “It’s just a group of people trying to come up with a core set [of] kernel universal practices that can be used to describe a methodology.”