Not only are "sucess" and "failure" relative terms in software development efforts, they are also relative in research on software development efforts.
In the book "Making Software" (2010), on p.67, in a chapter by Basili on the NASA's SEL, a section header trumpets: "The NASA Software Engineering Laboratory: A Vibrant Testbed for Empirical Research."
Now see the abstract from the linked paper (2002):
> For 25 years the NASA/GSFC Software Engineering Laboratory (SEL) has been a major resource in software process improvement activities. But due to a changing climate at NASA, agency reorganization, and budget cuts, the SEL has lost much
of its impact.
And an excerpt from the paper, describing the situation from 1995 onward:
> We lost support of management. They did not understand what we did and we were not able to offer immediate solutions without understanding the environment. We lost the ability to interact with projects. Each project manager was empowered to do his or her own thing [and] saw the SEL activities of characterization and assessment as a potential overhead to their project.
The NASA SEL is a "poster boy" for empirical research in software engineering. And yet NASA apparently decided that it wasn't worth investing in any longer. In the 2002 paper Basili points to "lack of vision" from management and "poor salesmanship" on the part of SEL as explanations for its "fall".
Experienced project managers will recognize these as typical explanations for the failure of any project, and since it's hard to objectively assess "vision" or "salesmanship", these explanations are always handy and rarely falsifiable.
What I find puzzling is that Basili, writing 8 years after the "fall" of the SEL, totally fails in his overview chapter to mention these historically important aspects of the SEL's mission and its overall appreciation by the sponsoring entity.
In the book "Making Software" (2010), on p.67, in a chapter by Basili on the NASA's SEL, a section header trumpets: "The NASA Software Engineering Laboratory: A Vibrant Testbed for Empirical Research."
Now see the abstract from the linked paper (2002):
> For 25 years the NASA/GSFC Software Engineering Laboratory (SEL) has been a major resource in software process improvement activities. But due to a changing climate at NASA, agency reorganization, and budget cuts, the SEL has lost much
of its impact.
And an excerpt from the paper, describing the situation from 1995 onward:
> We lost support of management. They did not understand what we did and we were not able to offer immediate solutions without understanding the environment. We lost the ability to interact with projects. Each project manager was empowered to do his or her own thing [and] saw the SEL activities of characterization and assessment as a potential overhead to their project.
The NASA SEL is a "poster boy" for empirical research in software engineering. And yet NASA apparently decided that it wasn't worth investing in any longer. In the 2002 paper Basili points to "lack of vision" from management and "poor salesmanship" on the part of SEL as explanations for its "fall".
Experienced project managers will recognize these as typical explanations for the failure of any project, and since it's hard to objectively assess "vision" or "salesmanship", these explanations are always handy and rarely falsifiable.
What I find puzzling is that Basili, writing 8 years after the "fall" of the SEL, totally fails in his overview chapter to mention these historically important aspects of the SEL's mission and its overall appreciation by the sponsoring entity.