Shrini, the problem is that some people consistently compare scripted testing to exploratory testing. The reason I wrote this was because I had just left a presentation in which the speaker made this comparison.

I find comparing scripted tests to exploratory tests similar to the way people sometimes compare agile to waterfall development models. (Interestingly enough, very little commercial software is developed using a true waterfall model; most commercial software is developed using some iterative model. So, the comparison is virtually meaningless except in theorectical or philosophical discussions.)

Of course, over the past few months I have been doing much more thinking about what ’scripted testing’ is and can only conclude that scripted testing is execution of written (or recorded) validation tests derived from the specification and designed primarily to verify a product meets or adheres to specified requirements.

In my opinion, all other types of testing are investigative in nature.

So, I am not sure why you think the debate is dead since so many people you know continue to use the inane comparison of ’scripted testing’ (made especially worse by the fact that ’scripted testing’ is never clearly defined) to exploratory testing (ET).

There are some recent studies comparing ’scripted tests’ (those tests derived from a specification) to exploratory testing (as described by Bach, Kaner, et. el.). And empirical data I’ve collected over the past 5 years closely parallels the data from a similar study done at the University of Helsinki. When I publish the white paper (around August) the information should cause rationally thinking individuals to question some common assumptions about exploratory testing in hopes that testers will better understand the advantages and disadvantages and learn to utilize exploratory testing (ET) to augment other approaches to testing.

Monday, May 12, 2008 6:57 PM by I.M.Testy

What actually promopted you to write this post?

Did you hear or see someone insisting on doing either scripted or exploratory test?

or it is just a example that came to your mind while discussing about Agile vs waterfall development models ….

You are right – a good testing is continuum between two extreems of scripted and exploratory testing approaches.

I think, the debate about scripted vs exploratory is DEAD … unless there is a total new twist or new angle to it …

Shrini

Monday, May 12, 2008 4:56 PM by Shrini

While we still adhere to the test scripts that we use, we also implemented Exploratory Testing, to a degree, and it was successful enough in the pilot project it was used under to warrant further use throughout the company.

This was a huge success story for us because our investigations had hit somewhat of a plateau, but when focusing on a particular area we were able to log a couple hundred more bugs before the product release.

With that said, I don’t know that I would ever rely solely ET or Ad Hoc as a QA methodology. It’s not that it doesn’t work, and it’s not that it can’t be structured (because we feel we’ve created somewhat of a hybrid that does have a certain amount of struct along with the freedom), but I think that scripted tests are a great checklist to make sure you’re seeing the forest.

What I mean by that is, I think that exploratory is a great exercise and it has its place in the test cycle, but I can also see where it could be so focused on finding deep-rooted bugs that you miss some of the most obvious ones.

Monday, May 12, 2008 11:36 AM by brentpaine