Analysis of trade-offs among software quality factors needs to be based upon a clear statement of the meaning of such factors and their interrelationships. There are several instruments with this aim. In this talk, we will focus on two of them, which complement each other. Ontologies for software quality serve to bound the space of quality factors, its components and shape, and the basic properties that these elements fulfill. Languages allow engineers make the ontology actionable, declare quality requirements based upon the ontology and reason about several quality aspects, among them operationalization of quality requirements and trade-off analysis. The talk will provide the basic concepts of ontologies and languages for software quality, will survey the most widespread proposals and will analyse in depth a concrete example of each of them.

Dr. Xavier Franch. Full professor. PhD in Informatics, UPC. He is main investigator of the GESSI research group at the UPC (https://gessi.upc.edu/en). Active researcher with more than 200 peer-reviewed publications, 4.500 citations and h-index 37 (according to Google Scholar), his research interests include Requirements Engineering, System Modeling, Software Evolution and Adaptation, and Agile Software Development, among others. In EU framework programmes, he is currently project coordinator of the Q-Rapids project (H2020, RIA; 2016-2019), scientific manager in the SUPERSEDE project (H2020, RIA; 2015-2018) and partner representative in the OpenReq project (H2020, RIA; 2017-2019); moreover, he acted as RISCOSS project coordinator (FP7, STREP; 2012-2015) and partner representative in S-Cube (FP-7, NoE, UPC acted as associated node; 2008-2012). He is editorial board member of the following journals: Elsevier’s IST and JSS, Springer’s REJ and Computing, and World Scientific’s IJCIS, all of them SCI-indexed; as well as Deputy Editor of IET Software (also SCI-indexed). He belongs to the steering committee of several major conferences (remarkably IEEE RE and CAiSE) and has occupied several conference positions in major software engineering conferences conferences, remarkably as general chair (RE’08) and program chair (RE’16, ICSOC’14 and CAiSE’12). He is full member of the International Requirements Engineering Board (IREB) association (also as part of the organization’s Council). He has won several best papers awards. He has taught tutorials and organized workshops on software engineering-related topics in several major conferences as ICSE, RE, CAiSE. More details at https://www.essi.upc.edu/~franch/.

The C-suite is fed up with software disasters putting the quarterly statement at risk as they digitize the business. They will demand more accountability and force improvements in software processes that may clash with agile culture. Business critical applications have become so complex and demand for functionality so immediate that human-based quality practices are no longer sufficient. Developer capabilities must be enhanced by system-level analysis of structural weaknesses and operational risks enabled by structural quality technology and measurement standards supported in DevOps toolchains. Empirical results will be reported from research on how some of the most severe Security and Reliability flaws are distributed in business applications. Recent results from machine learning research in structural quality will be discussed along with some caveats about what to expect. International standards for measuring the structural quality of software developed by the Consortium for IT Software Quality (CISQ) that supplement ISO 25023 will be reviewed. The talk concludes with organizational requirements for successfully adopting these advances.

Dr. Bill Curtis is the Executive Director of the Consortium for IT Software Quality (CISQ), an industry consortium chartered with building international standards for automating the measurement of size and structural quality from source code. He is also Senior Vice President and Chief Scientist at CAST Software, where he heads CAST Research Labs. He led development of the Capability Maturity Model (CMM) while at the Software Engineering Institute. He has worked at the University of Washington, GE Space Division, ITT’s Programming Technology Center, Microelectronics and Computer Technology Corporation (MCC), the SEI, TeraQuest Metrics which he co-founded, Borland Software Company, and CAST. He has published 4 books, over 150 papers, and is an IEEE Fellow for his contributions to software process improvement and measurement.

Systems and software qualities (SQs) are also known as non-functional requirements (NFRs). Where functional requirements (FRs) specify what a system should do, the NFRs specify how well the system should do them. Many of them, such as Reliability, Availability, Maintainability, Usability, Affordability, Interoperability, and Adaptability, are often called “ilities,” but not to the exclusion of other SQs such as Safety, Security, Resilience, Robustness, Accuracy, and Speed. In 2012, the US Department of Defense (DoD) identified seven Critical Technology Areas needing emphasis in its technology investments. One of them was called Engineered Resilient Systems (ERS). The SERC sponsor, the DoD Undersecretary for Systems Engineering, and the lead ERS research organization, the Army Engineering Research Center (ERC), held two workshops to explore what research was being addressed, and how the SERC could complement it. It turned out that the existing ERS research underway was primarily directed at field testing, supercomputer modeling, and resilient design of physical systems, and that the SERC could best complement this research by addressing the resilient design of cyber-physical-human (CPH) systems, Some of the SERC universities were performing such research, such as AFIT, Georgia Tech, MIT, NPS, Penn State, USC, U. Virginia, and Wayne State. These universities have been addressing aspects of this research area as a team since 2013. Initially, the team found a veritable quagmire of SQ definitions and relationships. For example, looking up “resilience” in Wikipedia, the team found over 20 different definitions of “resilience,” with over 10 different definitions of a system’s post-resilient state. The leading standard in the area, ISO/IEC 25010, had weak and inconsistent definitions of the qualities. For example, it defined Reliability with respect to the satisfaction of a system’s functional requirements, but not its quality requirements. Some of the SERC universities had developed partial ontologies of the SQs, and exploration of alternative ontology structures identified found one that addressed not only the inter-quality relationships, but also their sources of value variation. The talk will summarize how the ontology can help systems engineers query, qualify, and quantify the relations among the system qualities, and better address key qualities such as Maintainability.

Dr. Barry Boehm received his B.A. degree from Harvard in 1957, and his M.S. and Ph.D. degrees from UCLA in 1961 and 1964, all in Mathematics. He has also received honorary Sc.D. in Computer Science from the U. of Massachusetts in 2000 and in Software Engineering from the Chinese Academy of Sciences in 2011. He is a Fellow of the ACM, AIAA, IEEE, and INCOSE, and a member of the NAE. While at USC, he has served as the Principal Investigator on major research contracts and grants from the Office of the Secretary of Defense, DARPA, ONR, AFRL, USAF-ESC, TACOM, NASA, FAA, and NSF. He has received industry research grants from over 25 industrial organizations. His real-client software engineering project course has successfully completed over 200 projects for USC-neighborhood clients and educated over 2000 students in an integrated approach to systems engineering and software engineering. He has published over 500 papers and books, with over 50,000 citations, and a Google Scholar h-index of 81