The definition of terms designed to focus on their technical meaning and thereby clarify specification is the goal of this paper. Three terms--resilient, robust, and survivable--are its focus. It is too bad that the authors miss their mark and do not rely on the term “trustworthy,” which includes the need for a software system to be available, secure, and safe. I wonder why they didn’t check the literature for these terms.

The last thing software engineering needs is another paper exploring technical terms qualitatively. Unless we can compute a quantitative metric for each design term that characterizes a system, we cannot access its architecture.

The paper is well written, but not particularly useful.