August 10, 2016 · Events CAST 2016

CAST 2016: Day 1 (Tuesday)

Today was the first full day of conference activities, which seems a strange statement given that many of us were at TestRetreat some 72 hours earlier. Nevertheless, events were certainly moving on a much larger scale today, not least due to the presence of the webinar cameras, allowing attendees from around the world to interact throughout the day's sessions.

Below are just a few of my notes from the day. I captured 27 pages of thoughts within my Moleskine journal, and I'm throwing this blog post together at midnight after 17 hours of conference activities, so it's going to take significantly longer for me to digest all of my learnings!

Lean Coffee

Following the same format as yesterday's blog post, an approximate summary of today's Lean Coffee topics is as follows:

  • What do you look for when interviewing?: In particular, as an interviewee, what do you look for in a potential employer as a potential positive or a red flag? Warning signs included if the company appeared to have given less consideration to the interview than you have; not being given an opportunity to meet the team or see the office; vague answers to frank questions (e.g. "What would you change in the organisation given unlimited power"); inconsistency in answers from multiple interviewers; or your manager being unavailable or untraceable. I provided a couple of nightmare stories of my own, both as interviewer and interviewee, which I'd be more than happy to share in private!
  • Are blog posts good for discovering new information?: A lot of attendees suggested that they found it easier to read blogs than books on a day-to-day basis, although some struggled with consultancy-style blogs which were secretly trying to sell products. While blog posts (unlike books) were sometimes unauthoritative and lacking in sources/references, this could work in their favour, allowing the publishing of angles which might otherwise not make it into print. Blogs also tend to have comment threads, allowing quick feedback and information exchange with the author, although one major challenge of blog posts is that they tend to capture a "moment in time" and it's difficult to see where they fit into the bigger picture, e.g. whether a particular post reflects the general beliefs within the community when it was written, or at a later date.
  • Establishing a QA department: Practical requests from an attendee who is currently trying to setup a testing team within their organisation. Although the circumstances were deeply contextual, we came up with a bunch of questions which could suit any such situation. For example, does the organisation understand the activity of testing? What metrics do they intend to use to measure its effectiveness (and does the new QA Manager have the ability to affect these metrics)? What does the company want from a QA lead, and what visibility do they want into the testing process? Is the company open to hearing potentially-controversial statements such as "you can't test everything"? Above all, it's important to take a step back, look at your customers, projects and the time available, and attempt to put together a strategy which will serve all three of those factors accordingly.

Keynote: Nicholas Carr

This was a particularly interesting way to open CAST 2016. Given the various noisy discussions which have occurred on Twitter in recent months, regarding testers' supposed objections to automation (a viewpoint which I don't associate with), it was interesting that (as Nicholas himself admitted during Open Season) the opening keynote focused on the downsides of automation. However, as per the above tweet, I didn't take a negative message about automation from this session; rather, I gained an appreciation into how human imperfections can threaten the value of automation.

Nicholas showed how automation and robotics has gained in-roads within most of the typical typologies of our day-to-day work: for instance, action (robotic lettuce pickers), analysis (cancer detection algorithms), persuasion (fitness apps) and creativity (building/architectural designers). We also see machines making strides in particular areas of specialisation, such as self-driving cars (robotics), IBM's Watson (information recall) and AlphaGo (logic). Yet when tasks call for these skills to overlap, there's still no substitute for human talent at the cross-section of these abilities.

We were shown several case studies where automated processes were performing exactly as designed, but the human element led to failure scenarios. For example, when a waypoint error led the Royal Majesty to drift at sea in 1995, there was supposed to be a manual system to mitigate errors - a person on-deck was supposed to manually identify navigation buoys. But when the navigator failed to spot the buoy, they kept an embarrassed silence, assuming that the computer was correct and they'd just failed to spot the marker. Such automation complacency is also occasionally matched by automation bias, where the information provided by a machine isn't challenged; see examples of bus/truck drivers who fail to navigate low bridges because their GPS didn't tell them to stop (despite multiple roadside signs warning of the impending impasse).

The session concluded with an explanation of the Yerkes-Dodson Law, where (when experimenting with mice navigating a maze) it was observed that an overly-powerful electric shock was as ineffectual a corrective behaviour as zero/minimal shocking. Overly shocking lead to a stress scenario; too low resulted in complacency, wheres the best results were seen with moderate interruption which created a sweet-spot of engagement.

This was the first time I'd had the benefit of hearing from Nicholas, but given the wide array of relevant books that he's written (see the reading list below) it certainly won't be the last that I hear from him.

Carol Brands & Katrina Clokie: Creating Bonds Across Disciplines

I've had many previous encounters with Katrina and Carol in the past, through previous conference attendances and via my work with Weekend Testing, so I was particularly looking forward to this session, not least because it reflected many of my current struggles with trying to communicate the role of testing across multiple lines of business.

Katrina and Carol alternated throughout their session to provide their experience of creating bonds with various areas of their business, including BAs, developers, operations and support. There were common themes among each, such as creating organic opportunities for collaboration and knowledge-sharing, and not waiting for people to come to you for questions, as we're generally too busy for this to happen.

I had a bunch of questions to ask at the end of the webinar, and frankly I held a few more back as I wasn't comfortable asking questions about my current organisation on a recorded webstream. Thankfully both Carol and Katrina have proven to be open to further discussion in private (I already chatted a lot with Carol during TestRetreat, and am making the most of the remaining breaks to compare further notes with Katrina!)

Nevertheless, it certainly didn't go amiss that - as well as my Twitter coverage of this session - I was making my presence heard on the web feed too!


Just to prove that there's no downtime during CAST, not even during lunch, I took the opportunity to witness the lunchtime broadcast of the CAST Live webinar. In previous years, this has been my only visibility into the world of CAST, watching from my own desk in the UK, so sitting in the room as the show went out over the airwaves was a real buzz! It also fed my inner broadcast nerd, as a student of journalism at university. I enjoyed hearing from both Dawn Haynes and Richard Bradshaw, but frankly I was spending more time watching the mixing desk, cameras and microphones to see how the show was put together!

Keynote: Anne-Marie Charrett

Anne-Marie is a speaker who I've always had a lot of time for, as she speaks with both a great deal of experience and honesty in dealing with the role of testing in large organisations. So it was no real surprise that this was a session chock-full of quotable advice. Frankly I can't hope to replicate all of her words of wisdom here; I'd highly recommend viewing this session as soon as it's available on YouTube playback.

In summary, Anne-Marie shared her thoughts about how traditional test management roles (from the days of waterfall methodologies) are being supplanted by the need for each of us to have our own agency and to be test leaders. This means allowing our testers to think for themselves, and to potentially make their own mistakes, and to remain supportive of them throughout this process.

At the end of the keynote, we also received an exciting surprise announcement that a CAST-themed event (chaired by Anne-Marie) would be hosted in Sydney next February. This is a particularly compelling announcement for me, as I've been talking for a long time about Sydney being the next place that I'd like to visit on vacation; it might prove to be slightly too soon for my budget to manage, but I'm going to make massive efforts to get to this one!

Pete Bartlett: Create the Change You Want

This was hands-down both the most surprising and rewarding session that I attended today, and perhaps the most valuable talk that I've seen this year. Pete is a graduate of the Speak Easy programme, designed to encourage new voices on the conference circuit, but if he hadn't mentioned that this was his first talk then I would never have known. He spoke with a maturity and honesty that went well beyond his years of testing experience.

Pete was sharing his thoughts about how to enact change within an organisation, at a time when others might have left such a company. He spoke volumes to my own current personal situation, and (as the session wasn't being broadcast live over the web) I was able to start several lines of questioning during Open Season which allowed me to massively rethink my own career plans. There's a good chance that, wherever I find myself in the next twelve months, Pete may have been one of the biggest players in that decision.

The main revelation that Pete offered was regarding the means by which we acquire reputation through our time within an organisation, and that rather than attempting to leave (and undergoing the overheads of job-hunting, interviewing, and re-integrating into a new team) we can spend this reputation to aid ourselves in shaping organisational change. Pete offered us several challenges and thought exercises, to help us shift the focus of our test teams to provide better services within our company, and I'm going to be putting these into practice as soon as I'm back in the UK.

This is a topic which is seriously tough (and potentially compromising) for me to discuss right now, but I think that Pete's words are going to reverberate with me and my team for some time to come. Hopefully I'll soon be able to look back on my own situation and continue to share this advice with others in our industry.

Anne-Marie Charrett: What Developers Have Taught Me About Testing

In the day's final scheduled session, Anne-Marie Charrett took centre stage again (apparently not content with solely performing a keynote today)! Technical issues with projectors and microphones meant that the session was re-shaped on-the-fly, though with no detriment to the content. I unexpectedly found myself at the centre of a free-form discussion about the nature of trust and collaboration - which, once again, is absolutely at the centre of my current working environment. I almost felt embarrassed that today seemed to shaped exactly to my own personal needs!

The discussion focused on some of Anne-Marie's recent experiences, which were clearly still raw and emotive, in attempting to justify the role of testers within her own teams and organisations. The dialog centered on Anne-Marie's own trust (or perceived lack of trust) of colleagues and managers, and how such discussions could have been better handled. Again, I'll apologise for ambiguity here, but Anne-Marie opened up about some difficult personal experiences which I can't hope to fully express here.

The bulk of this session consisted of a discussion of the nature of trust: what is it, what does it mean to have it, and how can we hope to form it in situations where we're buried in distrust? We looked at the four pillars of trust (Sincerity, Reliability, Competence, Care) and how we can grow these to build bridges within our teams. This included a revealing thought experiment where we identified somebody who we don't trust within our own organisation, listed what we valued that this person didn't seem to identify with, stated our fears about what this person might do to these values, and outlined how we might avoid being harmed in such a situation.

Evening: Member Discussion on Ethics

Twitter fans might already understand why this is a difficult session to talk about, and was an even harder session to live-tweet. It was a surprisingly frank discussion, both from a panel perspective and in terms of the respect shown by the 100-ish attendees who'd stuck around for a late night chat about what it means to be a respectful and conscientious tester in today's marketplace. Sure, the k-cards were a bit frantic at times, and facilitator Rich Robinson had his hands full with handling the rules of engagement (or implementing new ones on the fly) but the words spoken in the room were surprisingly mature, given the childish muck-flinging previously displayed by all sides on social media.

Nevertheless, they were only words, and there was clearly still some cynicism in the room, which I can totally understand.

And actually, it was after this panel discussion that several sub-groups broke into more detailed and frank discussion. (I was personally involved in a very interesting discussion about whether the BBST courses are that far removed from the likes of ISTQB, which - given it was held by a group of people who held both qualifications, and had stood as interviewer and interviewee with those credentials - appeared to display a level of frankness and conciliatory language which maybe the ethics panel lacked.) Regardless, it spoke volumes that we were still at the conference venue at 8.30pm, each pursuing topics which were close to our heart. I understand the jaded nature of Richard's words (and had a good chat to him within this same session) but I was left much more optimistic than I've been in a long time, and I hope that this optimism can extend beyond the conference walls.

Lightning Talks

This is my third conference in a row where I've jumped straight into the opportunity for lightning talks, though for me this is potentially the scariest so far - not just because I've opted for a potentially controversial topic (addressing much of the angst that's been displayed on Twitter in recent months) but also because this is my first time meeting most of the conference audience.

It's mitigated by the fact that the talks aren't going out live - though I get the feeling that maybe they should be, and I'll talk to an AST representative tomorrow to see whether we can livestream via Periscope. I'm also reassured by the earlier discussion during the Member Ethics session - I think that what I've got to say is supported by the prevailing winds of the community. But if the worst happens, at least it's only five minutes and then it'll be over!

Reading list

Here's an (under construction) list of all of the book recommendations that I've received thus far during CAST. We've talked about creating a whiteboard at the conference venue to record these, but if it exists then I haven't found it yet...

I've only included books which I hadn't already read; the true list of recommendations is much longer! In roughly chronological order of when I heard about them during the conference, the books I need to take a closer look at are:

  • LinkedIn
  • Tumblr
  • Reddit
  • Google+
  • Pinterest
  • Pocket
Comments powered by Disqus