Responsible Software Engineering

Abstract

Software trustworthiness today is more about acceptance than technical quality; software and its features must be comprehensible and explainable. Since software becomes more and more a public good, software quality becomes a critical concern for human society. And insofar artificial intelligence (AI) has become part of our daily lives-naturally we use language assistants or automatic translation programs-software quality is evolving and has to take into account usability, transparency as well as safety and security. Indeed, a majority worldwide rejects currently the use of AI in schools, in court or in the army because it is afraid of data misuse or heteronomy. Insofar, software and its applications can succeed only if people trust them. The initiatives towards "responsible software engineering" address these concerns. This publication is about raising awareness for responsible software engineering.

Sustainable-Security Assessment Through a Multi Perspective Benchmarking Framework

“…In addition, Fuzzy ANP-TOPSIS process allows for further comprehensive justifications of practitioners' confusing and undefined knowledge [32,33]. The number of experts is used in the Fuzzy ANP-TOPSIS procedure during combination, and the Fuzzy AHP-Average Weighted Procedure is employed only for simple average procedures of fuzzified values.…”

“…Hence, it has an impact on sustainable-security in several ways. [32]. Maintaining integrity enhances the sustainable-security.…”

Candidate Software Process Flaws for the Boeing 737 Max MCAS Algorithm and Risks for a Proposed Upgrade

“…This warning was an optional feature previously. We infer, from these promises, that Boeing has assessed the problem with the B737 Max to reside solely in the design of the MCAS algorithm, but not so much in the actual implementation thereof 11 . From these statements, one cannot infer that the the MCAS software component contains a fault or defect (IEEE terminology, see [6]), because that would imply that MCAS had made a promise that it did not keep.…”

“…A requirement, which might be imposed in the long run for MCASu, is that it will be the outcome of so-called responsible software engineering (RSE). A criterion for RSE as mentioned in Schieferdecker [11] is explicability, which goes beyond the narrow circle of software technicians involved in a project. 32 We expect that thinking in terms of the removal of software process flaws may be helpful for obtaining explainable algorithms.…”

“…In Promise Theory, this is only a promise, as there is no such thing as a guarantee in practice. 11 After ignoring pilot errors (including training and pilot certification deficiencies), and also maintenance problems (each of which which may or may not have causally contributed to either of the accidents, what remains is the algorithm's intent. 12 This is standard software practice for quorum determination in redundant algorithms.…”