“Why do we need to understand a bunch of test methods? I write test cases from user stories or requirements, automate what I can and execute the rest manually, and its fine.” If this is your situation: good for you.
If you are time crunched, if your automated tests have lost relevance, are hard to maintain or regularly miss bugs, if you do not have useful and meaningful ways to confidently measure and report coverage and risk, if you are doing what you have always done, if you document too much, or document too little, if testing is “a mess”, if the dev teams do not trust your testing, if you wish there was a better way……then arming yourself with new test methods or examining the methods you currently use will, without a doubt, be beneficial!
In our continuing effort to be the best source of information for keeping testers and test teams current, we have dedicated this issue to exploring test methods. Learning test methods is core to a test engineer. These are the skills and methods we use in the daily execution of our work. Without having enough tools in our arsenal, our job is compromised. Worse, it can be inefficient, insufficient, misleading and worst- miss bugs!
From useful and complex Linear Code Sequence and Jump (LCSAJ), to old faithful, Model-based testing, there are a very large number of important test methods. Lately I have seen a renewed use and importance of Scenario-based testing using personae for higher customer satisfaction, Real-world testing and user story validation.
Why does learning more test methods help? Test methods provide a structure for thinking. They give a framework for well understood measures of coverage and risk. For example, Model-based testing can give easy measurements of path coverage, Requirements-based testing is a common method when requirements coverage is measured or there is a need for regulatory compliance.
Different methods and techniques have different uses and goals. Different goals might be, for example, finding bugs, customer satisfaction, regulatory compliance, getting the product out as fast as you can, having confidence in the already functioning parts of the product with new added functions- all need different methods to provide the greatest confidence for these different goals.
Using particular test methods in your test strategy takes away the seemingly random nature of some test teams. I know a few teams who “hope” they find the worst bugs. Remember, hope is not a strategy! Test methods help create your strategy!
The objective of this edition is to present some new views on test methods. The goal is to give you as many tools as possible to attack your test effort and do the most effective, efficient job and communicate it effectively to the team so you can make the best most informed decisions on bug fixing, risk and release!
In this issue Brian Heys warns that without exploratory tests, the number of defects will always be higher; I’ll explain how Action-Basted Testing is a much saner way to evolve a testing project; Salesforce’s Keith Stobie reviews, “The Domain Testing Workbook” by Cem Kaner, Sowmya Padmanabhan and Douglass Hoffman; Robin Roy writes that boundary guidelines can provide a higher rate of error detection and LogiGear CTO, Hans Buwalda explains that a good test design can improve quality and the efficiency of the tests.
And, at this time of year, for those of you who celebrate Lunar New Year- Happy Year of the Horse! For those of you who do not celebrate Lunar New Year- give it a try! Its as good a reason as any to have a fun celebration.
Michael Hackett
Senior Vice President, LogiGear Corporation
Editor in Chief