You've performed unit, integration, functional, performance, security, and usability testing. Are you ready to go live with this new application? Not unless you've performed end-to-end system testing. What's so important about endto-end testing? It is the only testing that exercises the system from the users'point of view. Marie Was presents a case study detailing the introduction of a new insurance product in her organization.
STARWEST 2006 - Software Testing Conference
Some agile methodologists claim that testers are not needed in agile projects--all testing is done either by developers or users. Chris Hetzler has seen the effects of that approach, and they are not pretty. When customers find
In 1976, Glenford Myers listed a set of testing principles in his book Software Reliability. Computing has changed dramatically since those days! iPods have more computing power than the Apollo spacecraft. Testing has even been
recognized as a profession-but testing approaches have not changed substantially since Myers' book. Erik Petersen examines classic testing principles to help us understand what still works and what doesn't. He compares some of
For many organizations, creating a testing environment to replicate every combination of hardware and software that their users have is cost prohibitive. If your organization faces this challenge, the solution may be to create an infrastructure that is based upon virtual machines. Virtualization allows a single physical server to run the workloads of many different servers. Virtual test environments save time and money and support sophisticated test cases that are not possible in a traditional physical environment.
At Royal Bank Financial Group we are building a testing factory. Our vision is that code enters as raw material and exits as our finished product--thoroughly tested. As a roadmap for our work, we have used the IT Information Library (ITIL) standard. ITIL is well known throughout Europe and Canada but has yet to make inroads in the United States. It defines four disciplines: service support,
service delivery, the business perspective, and application management. These
So, you have solid automated tests to qualify your product. You have run these tests on various platforms. You have mapped the tests back to the design and requirements documents to verify full coverage. You have confidence that
results of these tests are reliable and accurate. But you are still seeing defects and customer issues-why? Could it be that your test automation is not properly targeted? Solid automated testing can be enhanced through runtime
Lloyd Roden as he unveils his list of the top ten illusions that we may face as testers and test managers. One illusion that we often encounter is "quality cannot be measured." While it is difficult to measure, Lloyd believes it can and should be measured regularly, otherwise we never improve. Another illusion Lloyd often encounters is "anyone can test." Typically when the project is behind schedule, inexperienced people are "drafted" to help with testing.
Software testing is tough-it can be exhausting and there is never enough time to find all the important bugs. Wouldn't it be nice to have a staff of tireless servants working day and night to make you look good? Well, those days are here. Two decades ago, software test engineers were cheap and machine time was expensive, demanding test suites to run as quickly and efficiently as possible. Today, test engineers are expensive and CPUs are cheap, so it becomes reasonable to move test creation to the shoulders of a test machine army.
After applications move into production, it is vital that subsequent additions or modifications are thoroughly tested and that the entire system is re-tested to ensure that it still functions after these changes. This process, called final regression testing, should be repeated for every new release. Many organizations that have attempted to implement a final regression test process have discovered that it isn't as easy as it sounds.
Are you contemplating moving from totally manual testing to automated testing? Andy Redwood shares a case study of a leading financial organization in the UK that did exactly that. Their goal was to automate testing using the
latest tools across multiple projects. They have just finished the first year of the project and have learned some valuable lessons. Andy will describe this