Bloggo back to the blog
A view from the Chair with Michael Bolton (Volume 6)-->
This year, we’ll be trying something new to EuroSTAR. We’re going to be shortening the track presentations and keynotes, and allocating time for discussion as part of each session. This is an approach that started at small peer conferences, and has been scaled up for larger conferences like the Conference for the Association for Software Testing and Let’s Test.
In the late 1990s, Cem Kaner and Brian Lawrence set up the Los Altos Workshops on Software Testing (LAWST). LAWST began as a reaction to the larger commercial conferences, at which vendors or consultants would stand in front of a room, and speak for the entire duration of their sessions. Presentations would rarely end early, and speakers would often run out of time for questions. The presenters might make claims that experienced testers disagreed with. Some listeners wanted to ask clarifying questions, but felt inhibited due to social pressures. Others in attendance were eager to contribute comments or personal experiences to amplify on points in the presentation, but time would run out.
The LAWST peer conference model took a different tack. The meetings were small-from 15 to 25 people. The organizers of the conference (the “content owners”) would identify a theme. Each participant would arrive prepared to deliver a presentation-not a marketing spiel or a lecture based on theory, but a story of some experience related to the theme. During the experience report, a facilitator would allow clarifying questions, but would suppress probing inquiries or discussions that distracted from the presenter’s story. After the experience report would come “open season”, in which the other participants would ask questions, raise issues, and provide comparable or contrasting stories. Rather than a strict one-question-one-answer format, people might raise questions that prompted others to develop entire threads of discussion. The facilitator directed traffic, guiding the participants and maintaining order, with the goal of making sure that noisier people didn’t dominate the conversation, and that the more reticent people got a chance to speak. Conversation would continue until energy for the topic ran out, and then the next speaker on the content owners’ list would give a presentation. This system of experience reports and facilitated discussion spread quickly throughout the context-driven testing community. After a few LAWSTs, other peer conferences emerged. Some, like the Software Test Managers’ Round Table and the Workshops on Teaching Software Testing followed the LAWST protocols strictly; others, like the London Exploratory Workshops on Testing and the Toronto Workshops on Software Testing took a somewhat less formal approach.
At many peer conferences, experience reports would prompt questions, which would prompt multiple answers, which in turn would prompt multiple replies, which would give rise to multiple comments. For the facilitators, keeping track of the threads and who was responding to whom was always tricky. One of the peer conferences, the Workshops on Performance and Reliability (WOPR), presented a formidable challenge for its regular facilitator, Paul Holland. Some of the attendees were particularly energetic, passionate, and very eager to speak. Keeping track of “the stack”, as it was called, proved difficult until Paul’s wife Karen proposed an idea to make things easier for participants and facilitators alike, and the K-card system was born. Each participant was given three cards with which to signal the facilitator. A wave of the green card indicated that the holder wanted to start a new thread of discussion; a yellow card indicated that the holder wanted to speak on the current thread; and a red card signaled that the person waving it needed to interrupt and speak immediately-perhaps on a point of personal privilege, a strong disagreement, or a desire to clarify something important and confusing. (A fourth colour, blue, indicated that the holder believed that the discussion was going into a “rathole”, a conversational dead end, but peer conference culture has developed such that blue cards aren’t being used much these days.) If someone over-used his red card, the facilitator would confiscate it, which helped remind enthusiasts and persistent interrupters to be more circumspect. K-cards were adopted into LAWST-inspired peer conference culture.
Organizers of the Conference for the Association for Software Testing wondered if the K-card system could scale up to large groups and plenary sessions, so they tried an experiment. The system worked remarkably well both for track sessions and keynote talks, and afforded the opportunity for richer, more interactive discussion after the presentations. For topics that carried particular energy, the organizers provided an overflow room where conversation could continue.
For this year’s EuroSTAR, the Programme Committee and I chose keynote and track presentations oriented to the theme of “Questioning Testing”, but from the outset, we wanted the structure of the conference itself to reflect the theme too. Apropos of that, we’re expanding the role of track chair to incorporate facilitation, and we’ve asked Paul Holland provide training for those track chairs who are new to it. (Others who are interested will want to attend his track session, “How to Organize a Peer Conference”.) For each session, whether a track or a keynote, 15 minutes of the time slot will be allocated to questions and discussion. We’ll be asking the presenters to keep this in mind as they prepare their presentations, and we’ll ask the presenters and facilitators alike to be strict about the time limits. We invite conference participants to participate; to engage with the speakers and their stories; to contribute their own ideas and experiences; to question and debate respectfully. We’ll guide people to use spaces in the conference facility (including the Community Forum, the Test Lab, and open areas) to continue their conversations.
We know that this approach puts some pressure on our presenters. It’s a real challenge to deliver a compelling story or to present a novel idea in 30 minutes. We also acknowledge that this way of doing things represents a break from EuroSTAR’s traditions. Our experience with peer conferences, with CAST, with Let’s Test, and with EuroSTAR in previous years makes us confident that the EuroSTAR community is more than ready to try this experiment. After all, the most important part of any conference is conferring.