“The Challenger was no accident—it was an example of a predictable surprise," said Colonel Mike Mullane, a retired NASA astronaut, in the opening keynote at the Software Test & Performance Conference (STPCon 2009), which kicked off Oct. 19 at the Hyatt Regency in Cambridge, Mass. Mullane, who flew in three space shuttle missions, shared insights into the dangers of the “normalization of deviance” as it relates to the 1986 Challenger disaster, as well as to testing in general.
What is “normalization of deviance?” “When there is no negative repercussion for taking a shortcut or deviating from a standard, you come to think it’s OK to do it again,” he explained, “and that almost always leads to a ‘predictable surprise.’ “
How to avoid this type of outcome? Mullane recommended the following:
- Realize you are vulnerable to accepting a normalization of deviance. “We all are. We’re human.”
- Plan the work and work the plan, and be aware of the context. "When you’re testing software, use it as the customers will use it, not as the developers designed it.”
- Be a team member, not a “passenger.” If you spot a problem, point it out—don’t assume others know more than you do.
- Lead others by empowering them to contribute to the team.
Making the Leap
“NASA is just like us,” and deals with many of the same challenges, including multiple and conflicting vendors, changing vendors midstream, and changing rules and priorities," said DevelopSense’s Michael Bolton, who presented the afternoon keynote, "Testing Lessons Learned From the Astronauts.”
Bolton discussed two dominant myths: Scientists are absolutely special people, and scientists are simply sophisticated chefs. NASA is an example of these myths in progress, he said.
Bolton introduced the idea of heuristics, which he defined as a fallible method for solving problems or making decisions. “It’s part of the world of being a tester to find inconsistencies," he explained. "Heuristics are valuable in testing because they help accomplish this.”
“Testing is about exploration, discovery, investigation and learning," he added. "Testing can be assisted by machines, but it can’t be done by machines alone. Humans are crucial.”
Speed Geeking, Anyone?
In addition to more than 30 breakout sessions held during the conference, covering topics from agile testing to test automation, performance testing, test management and the latest test tools, technologies and trends, STPCon hosted its first (and very lively) “speed geeking” session, led by ST&P Magazine contributing editor Matt Heusser. Participants included Scott Barber, executive director of the Association for Software Testing; exploratory testing experts James Bach and Jonathan Bach; David Gilbert, president and CEO of Sirius SQA; Justin Hunter, president, Hexawise; and Michael Bolton. Each discussion leader had five minutes to cover one aspect of software testing, followed by two minutes of Q&A.
More than 50 test and QA professionals also took part in two new STP training courses at the weeklong conference: Agile Testing with Bob Galen, director of R&D and Scrum Coach at iContact and principal consultant for RGalen Consulting Group, and Test Automation with Rob Walsh, president, EnvisionWare. Galen covered just-in-time test ideas, exploratory testing, all-pairs testing, and ways to drive collaborative agile testing. Walsh addressed how to make the business case for automated testing and introduced participants to a variety of test automation tools. Open source expert Eric Pugh delivered a half-day session on the automated continuous integration process using the open-source Hudson system as an example.
New one-day STP workshops included one on test management, led by consultant Rex Black with CA’s Bob Carpenter, and another on performance testing, led by Scott Barber, joined by a panel of performance testing experts including James Bach, Dan Bartow, Ross Collard, and Dan Downing.
The Value of Exploratory Testing
Brothers James and Jonathan Bach presented the closing keynote, sharing a newly revised model of the structure of exploratory testing. “It’s wrong to call exploratory testing unstructured testing," James Bach said. “The structure of the test becomes evident through stories.”
Exploratory testing involves learning, test design and execution as activities that run in a parallel process, he explained. It emphasizes "the personal freedom and responsibility of testers to optimize the value of their work.”
More Online
Check out blog posts and tweets from the conference at www.stpcollaborative.com /community and photos by ST&P Magazine contributor Joel Shore at www.stpcon.com.
