Lessons Learned in OZWSTing (2014 Edition)

Posted: July 9, 2014 in Test Events, Testing Skills & Education

Held in the occasionally sunny climes of Brisbane that are my backyard, the Australian Workshop on Software Testing 2014 was too great an opportunity to pass up (even though I almost did).  This small, relentless and incredibly valuable peer conference lived surpassed the high expectations that had been set by my first visit to this conference down in Sydney last year.

Organised yet again by Dave Greenlees, facilitated by Rich Robinson and having the discussion guided by James Bach, the Saturday and Sunday spun by in a kaleidoscope of risk, incisive questioning and animated (but for the most part orderly, thanks to the inconspicuous attentions of Rich) discussion.  While I could write a lot more about it from a personal perspective, Joey Corea’s blog post summed up the conference very nicely (first-timer or otherwise).

As I discovered in last year’s post-mortem, there were simply too many comments, lessons and scribbles jotted down in barely legible fashion to expound upon.  With such an embarrassment of note-based riches, I’ve decided to list eleven (being one more than ten, which was so last year’s OZWST post) of the many things I took away from the weekend.

  • A risk profile provides a model with which to communicate product risk more effectively.  It’s a seemingly straight-forward yet great idea, as risk is something I’ve always struggled to explain (on the rare occasion when I’ve had to).
  • Take a step back, de-focus, and change the perspective of your analysis.  Attacking a problem from different angles can yield a metaphorical truck-load of new insight and information.  I was to hear a lot more about this in the Rapid Software Testing (RST) course I recently attended, and is something I’d like to be much more aware of.
  • Using visual thinking to look at testing in advance, in order to foresee potential problems and risks.  This is an area I see a lot of potential in, and one to explore much further.
  • Using feature-level and story-level risk analysis to provide comprehensive risk coverage.  While there may be an overhead going into such depth, it provides new angles in which to attack risk from.
  • When looking for potential risks, calling the help line of the web-site being examined can provide a rich source.  Our small group discovered this first-hand when Kim Engel got on the phone during the risk exercise we’d been given.
  • Using “disposable” and clearly communicated assumptions can allow risk analysis on even the vaguest specification or object.  I particularly like the scaffolding metaphor: these assumptions can act as scaffolding for the constructs that are our ideas and thoughts based around those assumptions, and can then be dismantled when they are no longer useful or relevant.
  • Surprise can help you re-evaluate your models, opinions and risks.
  • Focus on exploring risk, not coverage.  An interesting perspective I’ve perhaps unconsciously done once or twice, but never as a deliberate or focused activity.
  • The nature (and not necessarily the number) of bugs can indicate serious underlying problems in the software.
  • Analytical lag: coherent thoughts and insights can lag immediate impressions and activities by some time, even up to a couple of days.  A prime example of this was given by Anne-Marie Charrett during her risk analysis of a password protection application.
  • Testers can and are used as “blame shields”, as Mags Dineen starkly demonstrated in her experience report.

If you hadn’t noticed the 14 occurrences of the word to date, the conference theme was on risk (specifically, product risk).  While it was a difficult definition to pin down (as James put it, “how do you come to understand if the product is good enough?”), it made for a lot of fascinating discussion throughout the week-end.

As things turned out, yet another benefit of OZWST was the fact it provided the perfect complement to some of the concepts visited in the Rapid Software Testing course that followed less than a fortnight later.  But that is a post for next time…

Advertisements
Comments

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s