By David Greenlees
- Richard Robinson (Facilitator)
- James Bach (Content Owner)
- David Greenlees (Organiser/Logistics)
- Dean Mackenzie
- Kim Engel
- Rita Manning
- Angela Baird
- Joey Corea
- Anne-Marie Charrett
- Margaret Dineen
- Sigge Birgissson
This year the Australian Workshop of Software Testing (OZWST) headed to sunny Brisbane in Queensland, Australia; well, it was sunny for one of the two days. One of my goals for OZWST was to make it accessible to many in relation to geographical location. 2012 was held in Adelaide and 2013 in Sydney; so far, so good.
The venue and Wi-Fi was kindly provided by Software Education via the exclusive use of their Brisbane headquarters. They also provided onsite support for both days – a big shout out to both Cherie Kelly and Bridgette Bell, you’re assistance was outstanding. The catering was sponsored by the AST, and was enjoyed by all. It’s great to see that an association based so far away still willing to be involved and support these community events. Thanks to Software Education and the AST.
Due to some late cancellations, and hopeful attendees being out of the country, the numbers reduced again this year to 11. However, this actually turned out to be a wonderful thing. The size of the group helped the first time attendees (four) better manage the format of a facilitated peer conference; an often daunting thing.
We commenced later than usual on day one due to some early morning travellers, so proceedings kicked off just after 10am. With all attendees present and accounted for, I provided a welcome introduction and covered various logistical items.
We then moved into check-ins which allowed all attendees the opportunity to introduce themselves, describe what they do in their current role, and declare any noteworthy distractions.
As the Content Owner, James checked in last and then proceeded to explain the theme in greater detail.
The theme for this year was, Experiences in Acquiring a Sense of Risk:
- Testing is about reporting the truth about a product. But that truth must speak to our clients concerns about risk. So how do we come to a sense of risk? This is a vital subject, because it not only relates to our ultimate mission, but also our prejudices which lead us to test more in one area than another. A sense of risk can be a self-fulfilling prophecy. There are no formulae that tell us how to judge risk. It is a heuristic and narrative process. We are interested in experience reports (ER) about large and small ways that you have done it. Does it emerge just from bugs? From reputation? From periods of time not finding bugs? A risk analysis exercise may also be acceptable instead of an experience report.
James had worked with several of the attendees prior to OZWST to help them ground their thoughts on actual experiences. This was evident as the ERs kicked off.
I will only provide a high level summary of each ER, and will say upfront that open season (facilitated questioning) for each was fantastic. There were many great questions and discussions that followed each ER.
ER 1 – Sigge
Sigge’s ER was identified as the first approximately 5 minutes before we commenced the welcome introduction. He handled this beautifully! Sigge introduced himself as a member of the QA team at Atlassian; QA as in Quality Assistance. I was very happy to hear Sigge refer to QA in that respect and it’s wonderful to hear this more and more in recent times.
Sigge presented his ER via a mind map and took us through an experience he had on a recent project. It was great hear about Sigge’s (and his team’s) approach to risk identification and assessment prior to product build commencing. He spoke about the difficulty of testing a development-centric product, and how the use of ‘dogfooding’ was a key element to their approach. Risks are identified at not only the feature level but the story level as well, using the mnemonic STOPSU (Stress, Time, Operations, Platform, Security and Usability). Being a big user of mnemonics myself I enjoyed learning a new one.
In summary Sigge urged attendees to talk about risk early and often. All levels of a development project need to undertake risk thinking, with each level having their own particular focus.
ER 2 – Joey
Joey opened up his ER with a statement along the lines of, “I’m not great at highlighting product risk and have therefore developed a model which I’d like to share with you.” (Not an exact quote).
Joey described his experience on a project and overlayed his new risk analysis model in order to describe how it would work in practice. There were four main ingredients to the model:
- User impact
- Dev-component History
- Expected versus actual code quality
- Complete sessions versus incomplete sessions
Each ingredient is given a risk rating, which changes as the project progresses and more information is identified or received. When the rating of all ingredients is combined you have an overall risk rating for product readiness.
Joey will continue to work on the model and refine it after more use. I look forward to any output that is shared with the wider community.
ER 3 – Dean
Dean would happily tell anyone that he is far more comfortable with the written word (and that is something quite special I might add) rather than the spoken word; all attendees were pleased that he got up and shared his ER.
Dean joined a small development company mid-way through the first feature sprint of a mobile application development project. It was a custom application for the client, and the company did not have a great track record when it came to mobile application development. There was no Business Analyst and the Project Manager was fairly hands off, so it was being lead mainly by the Development Manager.
For me a key element of Dean’s ER was the required shift in focus and therefore a shift in the priority of identified risks. This was dependent on the project drivers, one example being the need to demo the product and therefore presentation took on a higher priority than functionality.
It was clear that Dean has been able to reflect and learn from this experience.
ER 4 – Anne-Marie
Anne-Marie’s ER was prepared only a few days earlier when she was lucky enough to test a product with the help of James; this was done for the sole purpose of presenting at OZWST. Yet again another output that may not have happened if it wasn’t for the peer conference.
Anne-Marie opened up with a picture of her eldest son jumping off a pier into the ocean. What was interesting to Anne-Marie was how long it took her son to finally take the leap after her youngest son had done it with seemingly no hesitation at all. This illustrated that risk means different things to different people. Anne-Marie also noted that having an imagination for potential risks is good for testers.
Anne-Marie had conducted a testing session of KeePass, a product used to generate and store passwords. She took us through a mind map of her ideas and findings which seemed very in-depth.
It came out during open season that the notes from her original test session were much briefer than the notes delivered in her ER. Anne-Marie had experienced what was referred to as an ‘analysis lag’, whereby many new ideas had come to her after the session was complete. This was an interesting outcome and one that we believed required a name in the community; keep an eye out for that.
Check-out & Dinner
After the completion of Anne-Marie’s ER we end the day with check-out. It was wonderful to hear the opinions of day 1, and the get a sense for how all attendees were feeling at that very moment. It was evident that the challenges of the day had been beneficial, and as an organiser this is my ultimate goal.
The group then ventured out for dinner. This was also a wonderful experience as we could relax and enjoy each other’s company over some food and drinks, while solving all the world’s problems!
After check-in had been completed it was evident that the energy levels were high, which is a great thing considering it was day two after a later evening was experienced by most. James spoke to the group in relation to the content for the day and offered to facilitate a risk workshop if there was energy for it. There was! It was agreed that a practical workshop would help pull together some of the content and learning’s from day one.
I won’t detail the workshop in this report as James will continue to use it for his training in the future, however I will say that it was very beneficial. It was clear that while we were all reasonably comfortable identifying risks (after making and declaring reasonable assumptions), we were not so comfortable when attempting to explain the process we followed while doing it.
We split into 3 groups and undertook the exercise before coming back together for each group to present their identified risks. James then offered his thoughts and opened it up to the group for any further discussion.
Can you read minds?
The workshop took us all the way through to lunch, at which point there was a request from Kim to Sigge, “Get your cards out.”
From what I knew, this could mean one of two things – Art Show, or the Mind Reading Game. It was the latter. I had experienced this just a few weeks prior while at Let’s Test in Sweden where Paul Holland was very successful at reading the mind of Michael Bolton. This time it was up to Richard; could he read James’ mind?
Obviously I won’t go into detail; however I will say that it was great fun. Kim did a great job ‘managing’ the other game participants who eventually succeeded. As an organiser I was only too happy to let the group continue on after the scheduled end of lunch break, because the energy for the game was awesome… and this peer conference is just like how we test: context-driven!
ER 5 – Margaret
Margaret had a great ER to share with the group. I cannot detail any of it here due to non-disclosure; however it was wonderful to see Margaret step up and put herself out there.
No tweets were allowed during this particular ER, which was rather annoying because there were some gems!
We hadn’t done lightning talks at OZWST in previous years, so I was pleased to see the approach in action. Each speaker was given 5 minutes to present, and then the attendees were allowed 5 minutes of open season.
We had lightning talks from:
- Joey – Debunking defect detection rate predictions.
- Kim – Heuristics for predicting project success.
- Angela – Continuous delivery overview.
- Dean – Ego-less testers.
- Rita – This actually turned into a longer ER as we decided to turn the stop watch off, so thank you to Rita for telling your story.
For me this was by far the saddest part of the peer conference; it meant it was coming to an end for another year.
Listening to everyone check-out and talk about their honest opinion of the peer conference and their number one takeaway was amazing. Everyone was happy, everyone had learned, everyone had takeaways; what more could I want!
One thing I did note throughout the peer conference was the constant reference to usability. Many of you may know that I’m a big advocate for usability and the testing of it, so I was very pleased to hear it mentioned so often over a two day period.
For me, the biggest takeaway was knowing that attendees had developed very solid work items as a direct result of preparation for their ER. These work items will continue to be worked on, will be shared, and will no doubt have a positive impact on the community. As an organiser that is music to my ears. They may have never been thought of without OZWST!
Special thanks go to:
- Richard Robinson for facilitation
- James Bach for content ownership and the running of the risk workshop
- Software Education for the venue
- The AST for the catering
- All the attendees for giving up your weekend and providing fantastic content and input
Here’s hoping that OZWST 2015 is just as successful – Melbourne, perhaps?
David Greenlees, with thanks to Anne-Marie Charrett for sharing her notes which enabled a more detailed report.