WOPR4 was held in San Diego, California on April, 2005, and was hosted by Intuit.
Henry Amistadi, Jonathan Bach, Scott Barber, Bill Barnett, Ross Collard, Julian Harty, Paul Holland, Chris Johnson, Philip Joung, Rauli Kaksonen, Michael Kelly, John McConda, Jude McQuaid, Mike Pearl, Gail Rutherford, Robert Sabourin, Bob Sklar, Rolend Stens, Keith Stobie, Mike Taute, Brian Tolly, Brian Warren, Nick Wolf
Theme: Testing Systems for Robustness
This workshop explores how to test systems and components to see if they fail, and whether and how they recover. We invite first-person presentations in the form of experience reports on this theme, from people who have participated in robustness testing. Research papers will also be considered, though the audience members mostly are working practitioners rather than academics.
For the purposes of this workshop, the following definitions will be utilized:
Robustness: Ability of a system to “take a licking and keep on ticking” because of overload, equipment malfunction, inadvertent operator error such as invalid input, malicious act, natural disaster, etc. If the system does fail, robustness includes its recoverability (ability to recover). May be measured or predicted in reliability terms like MTBF and MTTR. Other terms used in this context are dependability and survivability.
Robustness Testing: Checking whether a system adequately prevents, detects and recovers from operational problems such as downed network connections, data bases which become unavailable, equipment failures and operator errors. Also called rainy day testing (as opposed to sunny day testing).
Some examples of potential experience reports include, but are not limited to:
– Testing for Stability, Security, Recoverability and Survivability when faced with accidental, intentional and/or malicious/ acts of invalid input and/or security related attacks.
– Testing to uncover failures due to inclement environmental or operating conditions:such as equipment or (software) component failures, errors returned by other components or systems, power outages, etc.
Please note, there is no place for marketing or tutorials at WOPR. Also remember that the audience will be asking detailed questions about your experience and typically expect to see actual data.