CAST 2016: Day 2 (Wednesday)
Four days after we kicked things off with TestRetreat, we've reached the end of the conference far too soon! An exhausting but rewarding week drew to a close with a day which delivered yet more surprises, concluding CAST 2016 with some deep emotional insights that I don't think we were quite all prepared for. It's been an exceptional event, and now it's time to begin the real challenge of spreading this learning outside the venue, to enliven our own teams and rejuvenate our own personal paths.
Tuesday night was a long one; we didn't wrap-up CAST discussions until 8.30pm, and after grabbing some food/drink and writing 3000 words about Day 1, I was suddenly manically short of sleep. So I didn't make it to Lean Coffee in time for the 7.30am start, but after stopping for the world's largest Chai Latte from Starbucks, I managed to join the end of another lively in-progress session.
Much of the discussion was regarding ways in which we could continue to keep in-touch with each other after the end of CAST. Twitter is of course one way, but we talked about creating a dedicated gathering point for alumni (possibly a Slack channel). We also began to plan a series of recurring virtual Lean Coffee sessions, to be held via Google Hangouts; Alex de los Reyes (@cyber_decker when he's on Twitter) appears to be taking the lead on this, though I'll be looking to offer my support given my previous experiments with running this format in Weekend Testing.
Keynote: Sallyann Freudenberg
Holy crap. I don't think any of us were quite prepared for this. It speaks volumes about the quality of Sallyann's content that this is going to be the #1 must-see webcasted session which I insist that my team watch. This is despite the fact that Sallyann openly admits to never having been a tester, and having given this talk at events hosted by organisations such as BCS which are seen as being the polar opposite to AST. This stuff matters, regardless of your beliefs/ideals or those of your company, yet this sort of session is rarely seen anywhere near a testing conference agenda.
The talk centered around Sallyann's experiences with autism in her family, her attempts to build an understanding of how her son viewed the world, and how she recognised many of the same patterns in her own behaviours. We learned about powerful world figures who've succeeded despite an often-late autism diagnosis and a world which is configured in ways which seem destined to prevent success, such as Steve Silberman, Temple Grandin and Jamie Knight. Beyond autism there were also discussions about ADHD, bipolar disorder, depression and other mental health issues, which struck notes with a clearly emotional speaker and audience.
I've got six pages of notes from this session, the most from any that I attended, but I really want to rewatch once the talk is published to YouTube, to regather my thoughts. You can be certain that I'll be promoting this talk, and publishing a standalone article about it, once that happens.
Nancy Kelln: Lessons Learned in Implementing Exploratory Testing
Nancy's talk came in the form of a series of stories about organisational change, and - like a lot of the best sessions - open admission of failures and mis-steps. Among the scenarios that we were introduced to were:
- After an initial introduction to Rapid Software Testing, taking the decision to "throw out all the test cases!" with no coherent plan as to what the alternative should be. Sweeping change was ultimately rejected, although Nancy found success by introducing small, manageable pockets of change within willing teams.
- Being promoted to a test lead position after previous heads were removed, supposedly for incompetency, though it fast became clear that there were actually systemic problems within the organisation which meant that anyone in the role was doomed to fail. This included requests to commit to unachievable ship dates, though Nancy recognised that the person requesting this information was themselves under pressure to commit from their own manager, and looked to find common ground.
- Another more successful attempt to remove reams of manual test cases and estimate a possible ship date, by creating a low-tech product dashboard showing a list of important product areas, the stakeholders for those areas, and lightweight information about those stakeholders' concerns.
- After overcoming the panic of release, attempting to drive widespread change by providing the entire test team with Rapid Software Testing training. Despite absorbing the material, the team ultimately showed little improvement, and Nancy discovered that this is because the team felt that one dogmatic approach (test cases) had been replaced with another (heuristics, mind-maps et al). The failure to understand the reasons for these changes led to fear/panic from people who were comfortable with structure in their day-to-day role.
- When next hiring a new team from scratch, rather than sending the entire team on a training course, Nancy instead hired interested, engaged testers who were interested in being drip-fed the equivalent material over a longer period, giving them the chance to absorb/adapt in a less drastic fashion.
Nancy also spoke about the "sustainability problem": how to ensure that the initiatives that you're driving will survive long after you leave the team/organisation. She advised that creating a visible written commitment to the role of testing (and testers) within your organisation, and displaying this prominently, can help to form the basis for sustained improvement.
As with several of the speakers that I've encountered in Vancouver, Nancy was also willing to remain way beyond the end of her talk to field questions from attendees such as myself. Anyone who knows me will probably appreciate that I've been through several similar situations in recent years, and once again I left the room with the feeling that somebody had been secretly shaping this conference exactly to fit my needs!
My lightning talk was bumped to the top of the list, so I was a little intimidated by the prospect of opening proceedings, but organiser Dawn Haynes and stopwatch monitor Justin Rohrman ensured that the whole event ran tightly to schedule, so I knew that I only had a maximum of ten minutes (five for questioning, five for open season) to navigate!
So, yeah. I said some things that I felt needed saying. I said a few (deeply personal) things that I've never said before. I also probably said a couple of things that I shouldn't say. My talk was titled "Can't we all just get along?", but it was a more general discussion about the nature of arguments on the internet. It particularly focused on my own unwillingness to get drawn into discussions, allowing me to maintain a degree of sanity, but potentially threatening my credibility to those who see me self-excluding. I'd also had my views challenged by several people I'd met at CAST, who'd previously been on my Twitter block list (purely because they were heavily engaged in heated debate) but whom I discovered are perfectly reasonable people.
I'm not going to write any more about this right now - the entire session (minus about the first 30secs of my talk) was videoed, and will be made available online in the near future; I'd rather wait until the full session is uploaded before I talk further, as - much like the petty squabbles that we see online! - I think it benefits from the context of understanding all of the perspectives that I put forward.
Curtis Pettit: When You're Evil
Another session which was right on-the-money for the circles in which I currently move! And, therefore, another webcasted session where I hogged a bit too much of the open season - apologies to those who were sick of my face by this point :)
Curtis was talking about how we build credibility within our teams, how this has to be earned, and how our actions result in this credibility rising or falling. I've previously referred to this in bug advocacy presentations as "the credibility exchange", but I liked Curtis's reference to "wagering your credibility", as there is a certain amount of gambling involved when we put our credibility on the line.
Despite his promises of being evil, Curtis proved to be a surprisingly amiable gent, albeit one who was happy to share a few cunning tricks to help quickly increase your pile of credibility chips. These included:
- Finding ways to earn "in-group bias" by getting members of other teams to consider you as part of their team. These might include finding common interests, speaking to their desires, or learning about their kooky niche hobbies (everybody has one!)
- Positioning yourself so that you and a colleague have a common enemy, by showing them a problem that you are both facing and working to overcome that problem together.
- Using the rule of reciprocity: do a favour (even a small one) and you're likely to be rewarded with a favour in return.
- Knowing what makes particular developers tick, for instance what sorts of bugs they're commonly interested in (Curtis mentioned his team had "a colours guy" with a deep interest in how the application renders at different colour depths) and giving them the heads-up on issues in that area.
During open season there were questions from the floor and from online viewers about whether the approaches which Curtis proposed were ethical or could sometimes cross boundaries. I sided with Curtis, suggesting that the most important factor was that of integrity: underhand attempts to build relationships are ultimately well-intentioned, and their success provides rewards for all of the parties involved.
On top of the books listed in the Reading List section of my day 1 report, the following additional books were added to my backlog today (books I haven't read previously):
- Edwin Hutchins - Cognition in the Wild
- Scott Page - The Difference
- Temple Grandin - The Autistic Brain: Thinking Across The Spectrum
- Maister/Green/Galford - The Trusted Advisor
- Dale Carnegie - How To Win Friends and Influence People
- Robert B. Cialdini - Influence: Science and Practice
Post-conference: first thoughts
I'll be writing a separate post-CAST article once I'm back in the UK and the jet-lag has cleared, which might not be until early next week. But here are a few of my early thoughts, which should also serve as a trusty reminder for me when I come to write that blog post!
- As my first North American conference, I came into the event having only previously met a handful of the attendees. However I was able to meet a lot of Twitter connections for the first time, and above that I met a huge number of people with whom I'd never crossed paths at all before. I'm proud to refer to these people as colleagues now; I'm not going to start name-checking, but if we shared a meal at any point, you're on it (and you get double points if the meal included poutine).
- I did a lot of live-tweeting during conference sessions, which was a great (and frankly cheap/easy) way of becoming known to a lot of people in a short space of time. I'm not saying that being called "the Twitter guy" is my dream job, but it got me into a lot of conversations very quickly this week.
- There were a few recurring themes that I noticed running through many of the sessions that I attended. These included topics of trust, respect, responsibility, credibility, reputation, collaboration, agency, ethics, and health. Of course there was testing-themed content, but it struck me that I need to communicate the learnings from CAST to a much larger audience than simply my test team.
Most importantly, as I said in today's introduction, we need to make sure that the positivity and desire for action which we expressed at CAST 2016 doesn't get abandoned as soon as we get home. I've found my own personal spark to make change in my organisation, and I'll be creating an action plan once I get home to get things going.
CAST 2017 in Nashville is straight into my calendar; I'm already looking forward to reconnecting with my new-found colleagues and hope that I'll be bringing stories of how CAST 2016 was responsible for driving measurable improvement in my test output. Until next time: farewell!