Metro-SQA

A common misconception is that strong technical skills or actually having the strongest technical skills in your chosen profession will propel you to greatness. Unfortunately, that is not the case! Technical skills are one the core foundations of your career, but not the only one. Wayne will engage in a presentation and dialogue regarding both the technical and business skills that are important to managing your testing career in today's economy.

Wayne Rampey is one of the owners of The Insource Group, a successful technical consulting and staffing firm. Wayne is also the owner of a project based Speech Recognition development firm which he started, sits on the Board of Directors for the Metroplex Technology Business Council, and also serves on the advisory board for the E-Business curriculum for Texas Christian University (TCU). Wayne has a BSEE and a MBA with 30 years industry experience.

Call it SOA, or just call it "integration," but the trend toward services-based architectures within medium to large enterprises is not a fad - it will represent more than 80% of all development budgets by the year 2010. Client-based UI testing and code-level testing will become less relevant, as the software we test becomes increasingly heterogeneous and federated across organizations. Aside from having to learn new technologies, QA will be asked to play a more strategic role in ensuring quality on a continuous basis in this new paradigm for software delivery. Customer examples of how development and QA organizations have handled the shift to SOA will be briefly discussed.

Please bring your questions and challenges and we'd love to talk!

Chris is an expert testing and architecture strategist, with a 17 year background in computer software development, product management, and sales support for enterprise software. As Product Manager for iTKO's LISA SOA Testing software suite, Chris applies both project management and development experience to refine LISA features to best suit customer quality needs throughout the software delivery lifecycle.

Chris was previously a Retail and Manufacturing Industry Manager at the enterprise software platform firm webMethods, overseeing requirements, customer presales and training for the $16M annual group. At supply chain software provider i2 Technologies he worked in i2's infrastructure group, with responsibility for the release of business process, workflow and monitoring engines. Prior to i2 at Software AG, Chris specialized in cross platform product installation and administration. As a software engineer, project manager, and solution architect, Chris has a great breath of industry knowledge from working with companies like Citi, TD Ameritrade, Lenovo, Tandy, Rubbermaid, TI, and TxDoT.

This talk will increase your awareness of the different mentality that developers use when unit testing, and give some insight into why trying to get developers to start unit testing is not as easy as convincing them to simply use black box testing techniques on a smaller scale. Test Driven Development is a form of developer testing where test code is written before the code under test. Test Driven Development is a great practice, but it's not all about quality. Another major goal of Test Driven Development is to hone the system and get feedback from the design. This talk discusses unit testing from the developer's point of view, and will show how "testing as design activity" is a very different approach from how a typical black box tester designs tests.

The world of the modern test-driven developer involves exotic concepts such as mocks, stubs, fixtures, and hecklers. Some developers are also using code coverage metrics, which have turned out to be not very useful for many black box testers. Adam will touch on these concepts, and also how some developers are shifting mindsets to Behavior Driven Development, which focuses users on "testing as design activity". This presentation will give testers will have a better idea of what kind of testing occurs on the other side of the cubicle wall.

Adam Keys is a software developer and writer in Dallas, TX. He's worked in a variety of domains and technologies, from pre-silicon tests written in C to web apps in Java. He's spoken at user groups around the metroplex, in addition to the first RailsConf in Chicago and the SD Forum Ruby Conference. These days, he's quite excited to work with Ruby and Rails on a daily basis.

This presentation will cover the basics of a structured risk-based / quality-centric approach to software testing.

Because, realistically, everything can't be tested or can't be tested to 100% depth of coverage (because of limited time, people, and/or money), how does the test team and the client decide what can be tested while working within those limitations?

By having the client identify and prioritize risks related to testing (and/or the project) and agreeing on and prioritizing the industry standard quality characteristics which are most important at each level of testing, as limitations change, sound business decisions (rather than emotional or abitrary decisions) about what to test and to what depth can be made.

This approach focuses on mitigating the greatest product, process, and project risks during testing, while emphasizing the quality attributes required by users for acceptance of the system under test.

Craig Mayer offers over twenty-five years domestic and international executive level experience in management, information technology, and operations in a diversity of industries. He has played essential roles in business reorganization and reengineering, as well as in all phases of the software development lifecycle. His skill set includes expertise in business process integration and effectiveness, Sarbanes-Oxley - CobIT - ITIL assessment and compliance, SOA (Service-Oriented Architecture), kaizen and gemba kaizen (continuous improvement), and risk-based / quality-centric software testing. As well as leading the Sogeti Texas Test Practice, Craig has also contributed to three Software Control Testing books. He is also a PMP certified project management professional.

Testers encounter stresses from all directions. You're delivering bad news about the products you're testing, and fending off attacks against the accuracy and completeness of your testing. All the while, you knew that disaster was coming, because you could see so many parts of the process that needed to be improved. Testers sometimes move into a QA role where they try to improve the processes, but that often just leads to a dead end and more frustration.

This talk won't show you how to fix everything in your organization, but it will help you center yourself and get a better understanding for why the organization isn't improving. When you can more calmly assess the situation, you'll be in a better position to make improvements both in the organization and how you work within in it.

Danny is the founder of the Metro-SQA group and an independent software consultant helping organizations manage the quality of their software. His Fort Worth, Texas-based consulting practice is Tejas Software Consulting.

This talk describes a theoretical model for defining test cases, known by its acronym "EPDAV". It shows how the model was derived and how it can be applied to any System Under Test (SUT). Variations on this model are illustrated, with practical examples from the author's testing experience at Texas Instruments. If time permits, I will also present some of the challenges of selecting or building a Test Management System (TMS) and an Automated Testing Framework (ATF) that supports this model.

Martin Taylor has been a software developer for 30 years. For the past 6 years he has specialized in the development of Automated Testing Frameworks. He is currently the Sr. Test Automation Specialist in the Engineering Services group at Texas Instruments calculator division.

The Outsourcing / Off-shoring presentation will discuss the following:

• What is outsourcing and off-shoring?
• Why do we outsource or off-shore?
• Are we outsourcing or off-shoring for the right reasons?
• Common issues, i.e. Not the Silver Bullet Solution
• Managing Internal Expectations
• Selecting the right vendor
• Common Failure Points
• Improving the process

Mas Kono has over 14 years of development and software quality assurance experience. He currently works as the Director of SQA for Texas Instruments - Education and Productivity Solutions, where he is responsible for managing internal and external SQA resources, as well as outsourcing/off-shoring relationships. During his career, Mas has held various software development and SQA positions working for Embedded Software Product companies, primarily involved in Handheld Devices and Medical Devices. His background includes working as a Senior SQA Lead for Good Technologies and SQA Manager of International Products for Palm.

One of the most expensive software testing tools is a tool for load testing client/server applications. Web application performance is an especially critical issue, and companies regularly spend tens of thousands of dollars on licenses for a tool to help them stress test their systems. Most of the time these licenses are sold based on the number of virtual users, but a new open source tool called "OpenSTA" is providing an inexpensive alternative to the commercial products.

This presentation will discuss some of the capabilities of "OpenSTA" for performance testing web applications, including:

* Recording user sessions
* Generating up to 1500 virtual users from a single machine
* Distributed testing on multiple load generators
* Monitoring and capturing server performance statistics
* Reporting results

Jeff Jewell is a Senior Consultant with ProtoTest, a Quality Assurance and Testing services organization. Jeff's paper, "Navigating the Minefield of Open Source Test Tools" was recently honored "Best Paper" award at the SQE's Software Testing Analysis and Review (STAREAST) conference in Orlando, Florida. Jeff has been involved in software testing for more than 10 years, and he has held lead and management positions in software quality assurance for companies as varied as small web startups to developers of ERP software. His experience also includes working as a Test Instructor. He has a great deal of experience with a number of commercial and open source test tools, including test management, automated testing, and performance testing tools. In the last year, Jeff has successfully worked on several performance testing projects using "OpenSTA".

Any task, regardless of whether it is related to IT or not, is basically composed of three parts or variables - Process, Tools, People.

What is the process that is going to be followed in order to complete the task? What tools will be utilized in order to complete the task? What and how many people will be needed to utilize the tools and follow the process in order to complete the task?

With regard to Testing:

The overwhelming majority of the discussions, conversations, and proposals that take place between customers and tool vendors or customers and consulting companies focus on the attributes and benefits of automated test tools or the values to be gained from improving testing processes and practices.

Little if any of the discussions focus on the people involved in the testing process and the diversity of talent that is found within those groups. This presentation focuses on recognizing the importance of this variable and approach for utilizing the diversity that exists within most testing organizations.

Mr. Riley has over thirty-five years in the information technology and consulting industry with extensive experience in development, business and system analysis, project management, and testing. He has developed, modified, and implemented a wide range of solution based methodologies across a diverse range of industries encompassing a wide-range of life cycle development approaches. For the past nine years, he has been with Systemware Professional Services, initially as a senior level consultant and then as their Technical Director. For the past six years, he has been responsible for the development and growth of Systemware's testing practice taking it from inception to a multi-million dollar solutions-based service.

Objectives include major technological decisions that were encountered/made during the deployment of a test infrastructure, as well as feedback/discussion of experiences others have had in architecting and implementing test infrastructures, and notable decisions and trade-offs that were made. He will also discuss that a test infrastructure can be a backbone for completely automated tests, manual tests, or a mixture thereof.

Gary has been with HP for five years. He currently works in the High Performance Systems Lab, but has been involved in embedded systems development and testing for the HP Superdome and mid-rage cellular lines during his entire tenure with HP. Additionally, Gary has published on "Invention Disclosures" for test execution parallelization technology, as well as holds multi-patented (cannot disclose, as a result of the various states of submittals). Gary has a Masters in Computer Engineering from the University of Missouri Rolla (Thesis on application of software testing techniques on FPGA implementations is bound and available from the UMR library; Undergraduate in Computer Science).

Craig Chambers, presently lead performance test engineer at Pepsi Business Solutions Group at PepsiCo, is the co-author of seven computer and testing related books, and several technical articles and publications. He has presented at national and international conferences, including Mercury Worldwide User Conferences.

The test automation industry has finally confessed - two decades late - that capture/replay is not a viable solution. Instead, a variety of approaches are emerging that seek to address the three key challenges to success:

* Need for development skills, which excludes analysts and SMEs
* Lengthy implementation effort, which frustrates faster delivery cycles
* Extensive maintenance overhead that offsets productivity gains

These new approaches are referred to as "frameworks" whose architecture ranges from data-driven to keywords to class libraries. In this presentation you will learn:

* What the basic elements of a framework include
* What each of these architectures offer
* Advantages and disadvantages of each approach

Linda is the CTO of WorkSoft, Inc., developer of next-generation test automation solutions. She is the founder of three software companies including AutoTester, the first PC-based test automation tool. A frequent industry speaker and award-winning author on software quality, she is a columnist for Computerworld, Datamation, and StickyMinds.com.