Archive for November, 2010

London Tester Gathering – 2 November 2010

12 November, 2010

I really enjoyed the London Tester Gathering on 2 November.  It was good to finally meet Darren McMillan ( after several online conversations and Sharath Byregowda (

Michael Bolton ( gave us a short talk entitled “Burning Issues in Software Testing” which was appropriate with Bonfire Night being just round the corner.  As always this was an inspirational talk full of the Michael Bolton sense of humour which I – and most of the audience – appreciated.

There have been many good blogs on the night including Darren McMillan’s write-up so I will leave my own summary at that.  Can I just say, though, a big “thank you” once again to Tony Bruce for once again organising a great evening.  I am just sorry that I could not stay for longer but I was staying in an unfamiliar part of town overnight.

Until next time…


UK Test Management Forum – 27 October 2010

12 November, 2010

This is my write-up of the UK Test Management Forum meeting on 27 October 2010.  Sorry it’s been so long in coming but things have been pretty hectic of late.

As usual there were three tracks running in parallel with two talks apiece.

The first talk I went to was led by Gojko Adzic entitled “Continuous Validation, Living Documentation and other tales from the dark side”.  Gojko discussed the fact that we often use different names for the same thing or use the same word but mean something different each time.  He highlighted various examples of this and proposed some solutions which make the terms more meaningful for people.  Graham Thomas pointed out that although this process has happened before – most notably about 25 years ago in the structured software development world – we still need to keep reviewing our terminology.

Gojko is writing a book on this subject and has a website to run alongside the book.  See for more details.

We had great discussions within the session.  I think we could all see that there is a need to address the confusion that we create by our use of terminology in the industry.   As Gojko pointed out legacy technical names do confuse people and create barriers which can hold people back from embracing change and adopting new processes.

The second talk I went to was entitled “The testing challenges associated with distributed processing” and was by Mike Bartley from TVS.  Mike was talking about the challenges we face with the rise of multi-core processors.  Whilst we can write parallel-savvy code, if the hardware and software platform on which the code is running is not using a distributed architecture there will be few – if any – benefits from the parallel code.

Mike talked about two common paradigms for distributed computing: message passing (which can lead to Race conditions) and shared memory.

Mike recommended reading J.B. Pedersen’s Classification of Parallel Programming Errors book (seems to be out of print and unavailable on Amazon).

He recommended that we adopt diverse static analysis techniques and think about design patterns and policies in our tests.  From a tools perspective we should consider which tools we can use at an architectural level to gain most beneft.

As always I thoroughly enjoyed the afternoon and felt I benefitted from the talks.  Things that I will take away from the talk include thinking more about the language I use to describe the testing that I am carrying out and thinking more about static analysis as a technique for checking out our distributed code.

Many thanks Paul Gerrard and Susan Windsor from Gerrard Consulting for hosting the event.  After the main forum talks we had a discussion about the future of the Test Management Forum and more information about the things we talked about and the decisions that have subsequently been made can be found at

The Prezi and PowerPoint slides from the two talks I attended are also available from

Comments welcome!