I enjoyed the CAST2009 event last week. It was interesting and helpful. Although the conference was not focused on load testing, there were sessions specific to performance testing.
There were very bright testing professionals from all over the world speaking and informally sharing their experience throughout the event.
For example, Edgardo Greising, Director of Centro de Ensayos de Software in Uruguay presented some good information about why performance testing is not only useful but necessary. His presentation, entitled “Helping Managers To Make Up Their Minds: The ROI of Performance Testing”, explained how many IT managers try to save money by not budgeting for load testing in a development project – but they don’t end up saving money because of the costs associated with production issues.
My favorite part was learning a new game that is played with dice. You are the player, and you are playing the role of a software tester. You are given 5 normal six-sided cubes like you would use in Yahtzee. You are told that you must figure out what the dice represent by testing my ideas and the other player would tell me if I was correct. The other player decides before we play what the algorithm is for the dice representation. You’ll see…keep reading.
And then the tricky part: The only rule is that you can break any of the rules.
So I started out in a state of confusion, but out of habit I rolled the dice. I counted all the pips and said that represented 23 (the sum of the pips). The other player said “that is a 12”. This continued for several rolls. I got a piece of paper and started writing down each roll, my guess, and the response.
Eventually, I figured out that I didn’t need to roll them, nor deal with all 5 cubes at once. My thinking became more creative. I tested theories by setting down 1 die and guessing the number on it. This led me to the conclusion he was only counting a die if it had a pip in the very middle. Specifically, if a 1 or 3 or 5 was on the die, then he would add that to the sum of all the dice (no matter how many I presented to him).
This process took me probably about 15 minutes to get the correct answer. It was eye-opening to me because it forced me to stop thinking about normal expectations and break some habits. I had to stretch my imagination for testing new ideas and experiment with out-of-the-ordinary procedures. It was a tremendous learning experience relative to how the brain works relative to our exploration of software as we look for bugs.
Thanks Ben and James.
My least favorite part was the arrogant behavior portrayed by some of those that have been in the AST for years. Several of the new members that I ate with or chatted with during breaks told me how “put off” they were by this haughty and rude attitude. There were condescending presenters that insulted audience members when basic questions were asked because the presenter thought such ignorance shouldn’t be dragging down all the other high and mighty elitist testers in the room.
One person at the final lunch commented to our table that “John Doe” (name changed to protect the guilty) had an ego that wouldn’t fit in the room. It was appalling to many of us, and the discussions in our small groups tended to include “I don’t think I’ll be back at CAST again, nor will I renew my membership next year.
Please don’t me wrong, I met many gracious and intelligent people in the organization. Those are the ones that may not renew. But the self-absorbed, cocky, so-called leaders in the group spoiled it for the rest of us.