TCFT Vice Chair Candidate

Chuck Weinstock


Chuck Weinstock received his Ph.D. in Computer Science from Carnegie Mellon University (CMU) in 1976. He currently is a senior member of the technical staff in the "Research, Technology, and System Solutions" Program at the Software Engineering Institute (SEI), also at Carnegie Mellon University. Prior to joining the SEI he worked at Illinois Institute of Technology, SRI International, and Tartan Laboratories.

At SRI International, Chuck was the co-project leader and main software designer and implementer for SIFT, a seminal distributed software-based fault tolerant system for safety critical flight control. SIFT led to many advances in fault tolerant computing including the first formulation of the Byzantine agreement problem by Chuck's colleagues, Leslie Lamport, Rob Shostak, and Marshall Pease.

At SEI/CMU Chuck initiated a focus on dependable computing. He co-created and led a software dependability working group which led to several workshops and the establishment of the Center for High Integrity Software System Assurance at the National Institute of Standards and Technology (NIST). He also began SEI's research into model-based verification which was a direct predecessor of current work on model-based engineering. Chuck's ongoing research explores assurance cases for safety and increasingly for security.

Chuck has had an active role in the dependability community since 1980. Since 1998 he has held organizational roles in many dependability-related conferences and was General Chair of the Dependable Computing for Critical Applications (DCCA) conference in 1999 and of DSN in 2003. He has been a member of the Technical Committee's Steering Committee since 2003. He is a Senior Member of the IEEE and a member of IFIP Working Group 10.4 on Dependable Computing and Fault Tolerance.


Position Statement

Computer-based systems have become ubiquitous. They are embedded everywhere and many have 24/7 service requirements leading society to recognize the need for them to function dependably and securely. Our Dependable Systems and Networks (DSN) community, that was largely instrumental in developing the fundamentals of this area since its inception (as FTCS) in 1970, should be reaping the rewards of our prescience. However, the lack of strong growth in DSN indicates that we still have a long way to go.

This is not a new observation. It's not that researchers aren't working in the area. Rather it's that the definition of dependability has become broader and as it has, the community of researchers has become more fragmented. In the past we could expect that anyone working in the area would naturally gravitate to us. Now they have a myriad of specialized communities available to them. The question for them is do they want to work in a smallish segment of the broad DSN community, or in a largish segment of a more narrowly focused community? For many the answer is clear and is not the answer we would like to hear.

* There are a growing number of conferences that now embrace areas that we believe should be integrated and cross-fertilized with the larger dependability community--our community. Most of these conferences are much larger than DSN but the dependability-related portions tend to be much smaller. In some of these conferences concepts developed by our community end up being re-discovered. In others they are ignored. This is to the detriment of both our community and the other communities. The advantage the other communities have is that they can focus topics to a specific domain. The advantage our community has is that we can hope to achieve cross-domain synergies that they cannot. There is a place for both. We have tended to address this problem by holding co-located specialized workshops along with DSN. We actively need to explore more options for enhancing our role and impact as a community.

* As with any community we need to work harder to keep ourselves fresh and vibrant. Students form our future and we need to make our research areas more attractive to them and to nurture their enhanced engagement into DSN. We need to proactively explore means of making this happen with the students, their advisors and also with the wider multi-disciplinary community.

* Historically our community had significant engagements with industry (IBM, AT&T, NASA ...) that also led to some of the key current technologies. Lately, we have somehow tended to be more academic. For the growth and relevance of our discipline, we need to attract and work with practitioners from industry. We have largely addressed this through practical experience reports at DSN, but I think there needs to be much stronger additional outreach to industry, perhaps via additional industry-themed topics and workshops under our sponsorship.

* Members of our community work with each other throughout the year, but aside from periodic emails via the FTTC mailing list there is no general mechanism to facilitate broader relationships in between the annual DSN's. This could be as simple as a mailing list for dedicated topics or DSN-associated blogs. Or it could be a series of targeted workshops throughout the year. The point is that a community becomes stronger as it becomes more cohesive and we need to be more cohesive.

Addressing these concerns will not be simple. As someone who has been involved with the FTCS/DSN community since 1979, who has a composite academic and industry background, and who has held most major organizing roles for DSN since 1998, I think my background puts me in a unique position to develop the DSN community. I believe that the incoming chair Neeraj Suri strongly believes in the need for focus and community building. This will be a multi-year effort with a consistent agenda to steer our dependability community to a leadership position. I plan to work closely with the Neeraj and the steering committee to help the dependability community grow and be of increased relevance to both academia and industry.