Was Testing to blame at Heathrow T5?

May 10, 2008

I'm looking forward to finding out at the next SIGIST conference on the 18th of June.

Posted by: Dave Evans

I think it takes a very brave man to say they are a Test Manager for the Heathrow Terminal 5 Baggage Handling system. Well, not only do Huub van der Wouden and Roger Derksen admit this, but they are getting up in front of a crowd of other testers at the BCS Special Interest Group in Software Testing, to talk about the project.

There has certainly been some high profile test-bashing recently over the opening day fiasco. Colin Matthews, chief executive of BAA, told the House of Commons Transport Select Committee that "the testing regime did not adequately replicate the first days of operations." Well, I can understand his beef, but you can't really replicate something that hasn't happened yet. There will always be a first day of operations, and it will always be different to testing for it.

I am not prepared to damn these men before hearing their story, especially as the word on the street is that the test team correctly identified the problems with the system before that fateful opening day, but that management overruled and decided to go ahead anyway. If that is the case, it is an extraordinary abrogation of quality governance on the part of the decision makers.

Whether it is the case or not, I think the T5 opening day fiasco is a good illustration of the challenge that comes from testing software systems that rely heavily on interactions with 'tricky stuff', like moving parts and human beings.

The value of any test result is limited by the differences in environmental context between testing and live use. People -- lots and lots of them -- provide the most interesting and challenging context of all.

Newsletters

Our Daily & Weekly Newsletters highlight breaking news, vital issues, expert analysis and solutions to help you stay ahead of developments in the industry.