Ward Cunningham recently posted a link to Cem Kaner's LAWST on a couple of mailing lists I am on.

I found that document to be really insteresting. It seemed a little specific-purpose toward the testing workshops, so I started trying to generalize it a bit, since it seems so useful. Here is what I came up with:

• Specify the length of time for the discussion
• Select 1 - 3 tightly-defined subjects for the Agenda
• Establish a tightly-defined goal for each Agenda topic
• Select a facilitator (this can change each meeting)
• Select someone to record the proceedings (this can change each meeting)
• Discussion Format:
  1. Tell war stories to provide context. Up to 5 (this is time dependant too) volunteers describe a situation related to the topic, from their own experience. Others ask storyteller questions, in order to clarify facts, or to focus on key items. Generally, stories are success stories ("we tried this and it worked because") rather than dismal failure stories, but instructive failures are welcome. No one screens the stories in advance.
  2. General discussion
  3. Boil down some apparent points of agreement or lessons into short statements and then vote on each one. Discussion is allowed. The list of statements is a group deliverable, which will probably be published.
• Publish: No one person is generally made responsible to publish results - each can do so as they see fit.
Notes on generalization of format:
• For short time periods (i.e. 1 hour) 1 topic is probably more than enough
• The facilitator probably would have to watch the clock more closely for shorter time frames. Time should be roughly one third per each section.
• Determining Agenda items seems to be the most difficult part. Someone (the facilitator?) should take responsibility for having this ready.
• Selection of facilitator should be done ASAP after discussion is scheduled.
• The original format is meant for technical discussions and fact finding.
• Further Modifications for other types of discussion might be
  • Instead of war stories, people might indicate positions (e.g. if the goal is not to distil experience, but to weigh strategies)
  • Publication might be strictly controlled.
• Emphasis should be given to covering 100% of fewer topics, rather than some lesser percentage of more topics. Prefer to defer.