Silicon Valley Software Quality Association (SSQA) Invited Guest Speakers Home | Speakers | Officers | Bylaws | Links | Jobs | Membership Management | Directions

SPEAKER INDEX

SPEAKER DETAILS

April 8, 2008

The SQA Approach on the Mozilla Project - How Firefox gets Tested

QA is a challenge in any organization, but open source development adds extra dimensions to that task.  Come hear Mozilla's Director of QA speak about testing in the creative world of open source software and how the Mozilla Project combines the effort of 22,017 test engineers and community volunteers to bring together specific SQA strategies, tools, and infrastructure.  Our speaker will cover Mozilla's approach to developing and executing both manual and automated tests, focusing on the testing of Firefox, Mozilla's award-winning web browser.

March 11, 2008

Testing a Cool Internet Technology called "Ad Serving Systems" (think Google and Yahoo!)

This presentation will address testing one of the latest internet technologies, "Ad Serving Systems".  Using a hypothetical Ad Server System as an example, the speaker will describe the goals, main functionality, key modules, and complexity of a contemporary Ad Serving System.  The speaker will describe the QA challenges faced and discuss solutions (test tools and test frameworks) employed to address those QA challenges.

February 12, 2008

Clichés, Metrics, and Methods: A Discussion of the Quality System and its Role in Contemporary Software Development

With several well-known software development clichés as a starting point, this presentation looks at the primary forces shaping a software development project and how these forces are typically balanced in a commercial software development project.  We'll discuss the implications for quality goals, system, and methods.  The talk will discuss the need for lightweight and flexible processes, combined with efficient, effective and quantifiable defect containment.  The presentation looks at methods and techniques to enhance efficiency and proposes simple metrics to monitor in-process and overall defect containment effectiveness.  A section of the presentation (MICRO Methodology) looks at procedures and tools generally needed for any software development project, independent of development methodology, and with a potentially huge impact on team performance.  The presentation takes a broad view of software development, in general, and quality assurance, in particular.  Rather than attempting to provide complete and final answers to specific problems, the goal of the talk is to spark discussion, and maybe make us stop for a moment and think about how we do what we're so passionately doing.  If time allows, we'll consider some slightly more philosophical aspects of development methodology and take a quick look at how Isaac Newton, Albert Einstein, Niels Bohr, and Brian Greene (superstring theorist) might have gone about developing software.

January 8, 2008

Exploring an Expanded Model for Software Under Test

Many testers think of the test exercise using the simplest model for the SUT (Input/Process/Output).  More complex models facilitate the design of better tests, help us understand the nature of what we're testing, and interpret the observed results.  Doug has developed an expanded model, which represents the influences on SUT behavior and the domains for possible outcomes.  The talk will be a presentation and interactive discussion of the model and some of its implications.

December 11, 2007

No SSQA meeting this month.  Happy Holidays.  See you in 2008!

November 13, 2007

Security in the Software Development Life Cycle

The best way to incorporate better security in any software development life cycle is to have a well-defined security process in place.  Yet, the reality in the current marketplace is that there is a heavy emphasis on security tools while ignoring the principle of incorporating security processes.  The bottom line should be "Process comes first, then tools."  Better testing processes like parameter validation (using the cardinal principle that "all input from users or external systems is evil until proved otherwise") and diligent code reviews (to catch logic bombs and poor coding practices that lead to vulnerabilities) will provide more bang for the buck than spending money on security audits of production-ready systems.  Understanding the fundamentals of setting in place better processes in the SDLC will be discussed in detail.  Using existing open source tools to perform security analyses on the code base will be demonstrated if time permits.

October 16, 2007

Life is not static, so why are your test cases?

Protocols are curious beasts.  They're the ultimate interface of any hardware/software product to the external world, be it file formats, APIs, communication protocols, RPC or command line.  They're all intimately connected through re-use of constructs and patterns of vulnerabilities.  Testing protocols is an effective way to both unit test as well as system test a hardware/software deployment.  More importantly, by analyzing the attack surface exposed by these protocols, one can hone in on the bugs that matter for expedited remediation.  In the connected world, there's not much difference between a bug and a vulnerability.  This talk will discuss these constructs, vulnerability patterns, and how a voltage regulator that's IP-enabled has much in common with web services.

September 11, 2007

A Graphical Display of Testing Status for Complex Configurations

Representing the status of software under test is complex and difficult, compounded when there are many interacting subsystems and combinations that must be tracked.  This paper describes a method developed for a one-page representation of the test space for a large and complex set of product components.  The latest project this was applied to had 10 interdependent variables and over 250 components.  Once the components are identified and grouped, the spreadsheet can be used to show configurations to be tested, record test outcomes, and represent the overall state of testing coverage and outcomes.  The paper uses a sanitized example modified from an actual test configuration.

August 14, 2007

Testing in the World of Open Source Software

QA is always a challenge in any organization, but open source development adds extra dimensions to that task.  Come hear the Director of QA at Mozilla speak about testing in the creative world of open source software.

July 10, 2007

Closing the Loop On Quality - Integrating Customer Feedback

Quality teams are effective at testing against software requirements, but they often don't get relevant data to feed back into their test development process.  This presentation focuses on lessons learned integrating customer feedback into the quality process.

June 12, 2007

Next Generation Testing with TestNG

TestNG (http://testng.org) is an open source testing framework designed to cover all aspects of testing, from unit to functional and everything in-between.  TestNG has innovative features and is geared towards professional developers in search of a testing framework that covers all styles of Java code, from mobile to enterprise.  This popular test harness offers a number of enhancements relative to JUnit.  Cédric's talk will illustrate several TestNG features that enable advanced testing techniques, such as: Multi-thread testing; Data-driven testing; Using groups for better organization of tests; Dependent testing; and much more.

May 8, 2007

Developing and Using a Defect Removal Model to Predict Customer Experiences on Software Products

Reliability is a key requirement for Hewlett-Packard's NonStop Server Systems.  The software that goes into these systems has to be of the highest quality.  A prediction of software quality can help better control development practices to achieve desired quality goals using available resources.  Several software defect models exist in the industry.  Growth models use statistical distributions to predict customer experience.  Other models predict customer experience based on past project, product, and the development organization's characteristics.  The speakers will share their experience of developing and using a phase-based containment model, where the effectiveness of defect removal activities is used to predict customer experience.  Larry and Sujoy will describe how they implemented the model in the organization, to assess product quality in each phase in a project's life-cycle, and how quality information is aggregated to make a release level prediction of what the customers would experience.  They will also share key benefits and lessons that have been learned as a result of the defect removal initiative.

April 10, 2007

The Software Project as a Journey

There have been many comparisons between software projects and other kinds of efforts.  Building a house or bridge, or some other engineering or architectural feat.  Another analogy which can also be useful is viewing a software project as a kind of journey.  You start out.  Things happen along the way.  You arrive at a destination.  So what makes any journey a success?  There are many possible criteria.  One famous example was "to get to the moon and return safely by the end of the decade."  For software projects, Brian Lawrence suggests that a worthy criteria might be "to fulfill the objectives of the sponsor."  Another could be "to make tons of money."  In this presentation, Brian will examine several journeys -- some where people traveled from place to place -- and some software journeys, which started with an idea and arrived at a destination.  All these journeys, physical and software, either succeeded or failed.  Why is it that some journeys succeed, while others fail?  What are the critical success factors?  Brian will assert that, for both software and physical journeys, some of the factors are exactly the same.

March 13, 2007

Requirements Management, an Integral part of Quality Release

Ambiguous, incomplete, and changing requirements are responsible for many software project failures.  Therefore, requirements management becomes a key component in project success and software quality.  Requirement clarity, definition (of attributes and constraints), storage, and change management are elements of requirements management that contribute to a quality product and realizing improved customer satisfaction.  Managing your requirements successfully means having complete visibility and accountability so your organization understands where and what the requirements are across the software life cycle.  Anita Wotiz will present an overview and key topics of requirements management, its role in the software development life cycle, and will describe how each element contributes to software quality and helps to meet project delivery dates.

February 13, 2007

How to Design Regression Test Automation Frameworks for System Testing

Sachin Bansal of Adobe will discuss designing modular, customizable and user-friendly regression test automation frameworks for the system testing of servers.  He will explain how to identify, design, implement, and execute complex automation frameworks involving different technologies.  Different frameworks will be presented to elaborate on the challenges and lessons learned.  Sachin will also discuss the challenges faced during integration of the bug tracking system, test case management system, performance testing system and automation system.

January 9, 2007

Keeping Score - How to Know When You're Done

Knowing when a product is ready to ship is one of the hardest questions companies have to answer, probably second only to its corollary, "How long will it take to build?"  A simple answer, some might think facetious, is "When it passes its tests."  This presentation presents a business case for doing just that -- scoring product completion via the state of its testing.  Unlike the common wisdom of looking at error rates or bug report frequencies, this approach predicts the total number of tests required to exercise the product completely and keeps score of the product's readiness through a few simple measures.  It is implemented via a novel approach to regression testing.  As a side effect, it gives the quality team's efforts significant visibility in the product development process.

December 12, 2006

Book Signing and Holiday Festivities with Local Authors

This month, we will celebrate the efforts and accomplishments of community members who have contributed to the body of knowledge in software quality and testing.  Four authors will be present and each will take about 10 minutes to present the primary theme of his/her book.  Afterwards, we'll enjoy holiday food and beverages and authors will be happy to sign books.  Copies are available for purchase, but it's better if you buy yours beforehand.  All books are available for purchase from http://www.amazon.com.

November 14, 2006

Test Automation Beyond Regression

Most testers think of GUI based scripted regression testing when they picture test automation.  This is a very limited view of the potentially vast possibilities open to us when automating tests.  When we think of test automation we should first think about doing things that we can't do manually.  This talk is about the limitations and how other kinds of test automation may be much more valuable.

October 10, 2006

The 5 Percent Rules of Test Automation

Successful automation is every test manager's dream.  It will shorten time-to-market life cycles and improve quality assurance.  Testers would love it too, since it can alleviate much of the boring work of executing tests and free up time to design better tests and to better follow up on test outcomes.  However, practice is reluctant.  Most test automation tools either end up on the shelf collecting dust or are at best used for a small part of the testing.  When is test automation successful?  In this talk, Hans will argue that stable automation is only achieved if a large percentage of test cases can be executed automatically and the automation does not take too much time away from the testers.  To provoke the issue, he will challenge us with the following two rules:  (1) No more than 5 % of all tests should be executed manually, and (2) No more than 5 % of all efforts around testing should involve automating the tests.  Apart from presenting the rules and why he feels they are important, Hans will mostly talk about how to meet the rules, using actual projects to illustrate a number of key principles to drive automation success: (1) test design, (2) automation architecture, and (3) organization.  He will introduce his Action Based Testing framework as an example of a methodology with which the 5 % standards can be achieved.

September 12, 2006

Who is Responsible for Quality?

Since the internet boom, small and large companies alike are driving their teams to work faster.  Since the tech-bust, small and large companies are asking their teams to do their work with less people.  The squeeze is on, teams miss their goals, software is shipped before its time and, in the end, the customers suffer.  Whose job is it to make sure we produce quality software?  How do we build quality into the planning and development process?  How do we avoid creating adversarial situations between those trying to meet company goals and those trying to ensure quality?  Come and hear Mark talk about his experiences and answers to these questions!

August 8, 2006

Quality-Driven Build Scripts for Java Applications

Agile build scripts not only compile code but also provide important information about the product. These documents and metrics vary in their objectives, but they facilitate the attainment of a common goal: building quality software. Unlike physical goods, obtaining information about applications is relatively inexpensive and disproportionately beneficial to engineering efforts. We'll explore open-source, off-the-shelf utilities for creating quality-driven build scripts for Java applications.

July 11, 2006

Metrics, Benchmarking and Predictive Modeling

How does a company measure and deliver on customer success? In this interactive session, David will share how Cisco links Customer Lifetime Value, Customer Loyalty, Customer Satisfaction, Product / Service Satisfaction, Product / Service Quality Experience, and Product / Service Design. He will be eliciting feedback on how to set customer experience targets based on customer explicit and derived needs, competitive pressures, industry best practices, and company process capability. We will discuss the role and approaches to benchmarking in helping set these targets as well as the precision and key elements to predictive modeling needed to assure that the company delivers on these measurable goals. Please come to this engaging session to learn, share, and discuss these key elements to measuring and delivering on customer success.

June 13, 2006

Fighting the BUG WAR with Inspections and Reviews: A Success Story

Understanding the quality of your systems is best aided by proper defect classification and analysis so that the right practices are followed, the right policies are chosen, and the tools are used in the right way.  Through the use of careful logging and analysis of the inspection and review results, learn how Cadence applies the lessons learned to reduce its engineering costs.  Topics covered are:

• Strategy for fighting the bug war
• Bug profiles
• Leveraging the best bug fighting practices
• Organizational preparation
• Analysis of the results

May 9, 2006

Using Data-Driven Analysis to Increase Customer Satisfaction

Learn how the Cisco Systems Technical Services Group uses a combination of statistical tools and integrated data analysis to identify key drivers of customer satisfaction and loyalty.
Topics that will be covered are:

    * Apply various statistical methods, such as:
          o Qualitative data integration using Linkage Analysis
          o Multiple regression analysis modeling
          o Other processes borrowed from the Six Sigma toolbox
    * Present data that gets attention
    * Translate feedback into improvement initiatives
    * Monitor progress and measure success
    * Apply what we learned

April 11, 2006

Software Testing as a Career – Still Viable?

Much has been made of how offshoring has robbed Silicon Valley of its high tech jobs, with new opportunities springing up in such far-flung places as India, China, and Eastern Europe.  But what's the real impact today?  While the SSQA membership has seen profound improvements in employment over the past year, come hear the thoughts of Mikhail Portnov of Portnov Computer School in Mountain View, CA.  He experienced first hand the explosive growth of the 90's, where job needs went from a handful of openings to hundreds of open positions and then back down in the early 2000's.  Yet, Mikhail remains positive about the outlook for software QA and test professionals.  As the recent job market revives, his students (many of whom are well-educated and legal immigrants) and their spouses are finding QA and testing positions more quickly again.  Mikhail will cover the following topics:

    * Different venues for QA training today
    * Different niches in the QA training market
    * Specific ideas for short-term career transition for degreed students
    * Realities of today's job market
    * Thoughts on local QA resources versus offshore testing

March 14, 2006

Software Engineering: Facts or Fancy?

We hear many things that we either are or should be doing to deliver the right software in a timely and cost-effective manner.  Some of these approaches are very popular, some are less so.  But do we really know which of these approaches actually work?   And how would we know if they did?  In this presentation, Brian will offer some ideas and we'll examine some of the assumptions we make about software engineering.   Which things that you currently believe in are indeed fact?  And which are fancy?   Let's learn together.

February 14, 2006

QA Road Warriors

QA has matured, and yet in many parts of the world there is a deep need for qualified QA professionals, individuals who know how it is done right.  While Silicon Valley continues to shake itself awake after its long slumber, these two professionals couldn't wait and sought opportunities and professional excitement outside the valley.  Doug Hoffman joined SDT and brought his assessment and training expertise to companies in India, China, Canada, and France.  Claudia Dencker brought her test leadership and management skills to an electrical utility company in Edmonton, Canada.  Share in their stories of excitement, trials and tribulations of what it means to take QA expertise on the road.

January 10, 2006

The T in Quality

This talk will focus on the T in Quality.  We'll start with the definition of QUALITY and contrast it to Quality Assurance and Quality Control.  We'll also talk about the institutionalization of quality and what types of support systems need to be in place to make this happen.  Finally, we'll review some case studies in implementing training to support Software Process Training.

December 13, 2005

Quality Training: What's been your experience?

The speaker will facilitate a discussion on the audience's experiences with quality-related training in the corporate environment.  How useful has it been?  How does one measure its effectiveness?

November 8, 2005

Estimating Software Size

Accurately projecting the size of a proposed software system remains the weakest link in the software cost estimating chain.  Deriving an appropriate size estimate is neither straightforward nor trivial.  Due to the lack of definitive information during the concept and design phases of software system development, size estimates made in those phases are characterized by uncertainty, generally resulting in estimates of very low credibility or validity.  Even as systems mature in their final stages (with requirements stabilized, all data inputs, outputs, and interfaces identified, and all processing functions clearly defined), the process of sizing software is still subject to a wide margin of uncertainty.  This presentation addresses the software sizing problem and discusses the Software Sizing Model (SSM) developed by Dr. Bozoki and in use worldwide. Also discussed will be how this model can help organizations address CMMI® model requirements regarding estimation and historical data.

October 11, 2005

Logical Entity/Relationship Modeling: The Definition of Truth for Data

Logical Entity/Relationship (E/R) models, also referred to as "conceptual" or "semantic" models, define the information requirements of the enterprise, independent of the resulting implementation.  A well defined E/R model is the key to successful development of data oriented applications.  Although most frequently associated with relational databases, the logical E/R model is equally applicable to object oriented and XML implementation.  This presentation will provide an overview of the fundamentals of E/R modeling as the definition of the information requirements of the enterprise.  It will focus on the underlying concepts and notations, with a strong emphasis on the semantic content of the E/R model.

September 13, 2005

Automation Techniques for Enterprise Application Testing

Enterprise applications are comprised of dozens of technologies and hundreds of classes, often developed and tested by scores of dispersed teams using disparate build and test frameworks. Integrating different components of applications that are developed by different teams often results in test bases with numerous build frameworks that are inefficient and a nightmare to enhance. Through this session you will be able to:

• Identify frameworks that are inflexible.
• Seamlessly introduce automation framework in test bases with multiple frameworks.
• Create a framework that allows easy changes to supported configurations.
• Create a framework that allows easy management of administration changes.
• Create a framework that allows greater coverage on multiple configurations.

Proven testing strategies are presented and lessons are drawn from the Sun Java System Application Server SQE Team.

August 9, 2005

A Process Driven Approach for Effective Application Service Quality for IT Organizations

A characteristic of most IT organizations is that new software development project controls and procedures are very rigorous while Operations and Maintenance controls and procedures are much less structured and accountable.  However applications are constantly changing, driven by ever changing business requirements and problem reports from the end user.  A lack of solid process can drive maintenance costs sky high and create a very chaotic environment.  Keith Mangold and Q Analysts have developed a model leveraging best of breed industry standards like ITIL, CMMi, TQM and CobiT to incrementally improve application service quality for large IT enterprises.  The model prescribes standardized procedures to provide the vehicle to drive focused process improvements that improve efficiency and reduce cost. This presentation focuses on software change, but is applicable to business process, data content and infrastructure changes.

July 12, 2005

Early Testing Without the 'Test and Test Again' Syndrome

June 14, 2005

Sabotaging QA: a Primer

May 10, 2005

The State of Spyware

Spyware has come of age as computer intrusions, infections and hack attempts have increased dramatically over the past few years.  Users, ranging from the novice to the IT and QA professional, have many options open to them to better protect their systems in a game where there are no winners and no end.  Robert Konigsberg will present information on Spyware, where it is today and the various forms that it can take.  In additional, he will cover:

• Visible identifiers of spyware
• In-depth exploration of spyware installations
• The various purposes for infection or intrusion
• Legal Aspects: What "License Agreements" imply and fail to make clear
• Origins for spyware (Where does spyware come from?)
• The role of anti-virus and anti-spyware utilities
• Coping strategies for the user

Robert is the founder of Network Evaluation, and has been involved in various aspects of computer security since 1992.  He has produced magazine articles, white papers, tutorials, guides and presentations aimed at informing and educating users on various aspects of networking and network and computer security.  Prior to starting Network Evaluation, he has worked for companies such as 3Com, Computer Curriculum Corporation, and Pearson Education, as well as following the Silicon Valley tradition of trying his hand in a few startups. He has earned SANS GSEC certification, and is an active member of the Center for Internet Security.

Solaris and Open Source - Current Status

Sun has announced plans to release the source code for the Solaris Operating System under an open source license, and to open the development process to external developers. This talk described these plans, the current status, and some of the challenges involved in moving a large commercial software project to an open development process.

Andy Tucker is a Distinguished Engineer in the Operating Platforms Group in Sun Microsystems. He has been at Sun since 1994 working on a variety of projects related to the Solaris operating system, including scheduling, multiprocessor support, inter-process communication, clustering, resource management, and server virtualization. Most recently, he was the architect and technical lead for Solaris Containers, and is helping lead the effort to make Solaris available as open source. Andy received a Ph.D. in Computer Science from Stanford University in 1994.

Keith Wesolowski is an engineer in the OpenSolaris team within Sun Microsystems' Operating Platforms Group. He joined Sun and the OpenSolaris project in 2004 and has an extensive open source background, including SPARC and MIPS Linux ports and several smaller projects.

March 8, 2005

QA and Open Source - The Good, the Bad and the Ugly 

Sleepycat Software makes the Berkeley DB family of products and makes them available under a dual license model. That means that we're have both open source and proprietary licenses. Our technical team has extensive Open Source project experience and the roots of the Berkeley DB product came from the University of California, Berkeley. This presentation explored the Engineering and SQA challenges of managing an open source product in a distributed company.

David Segleau has more than 22 years of IT industry experience, oversees Sleepycat Software's engineering, quality assurance and support operations. Segleau joined Sleepycat from Visto, where he was director of Quality Assurance and led the QA partnership with Handspring for delivery of the original TreoMail product. Previously Segleau headed up customer service for Asta Networks in Seattle and was senior director of Engineering Services at Versata and senior director for quality assurance and technical support at Informix and Illustra.

Febuary 8, 2005

War Stories from the Ground Level

This February meeting was a working meeting by SSQA membership. We started out with our annual topic, War Stories at the Ground Level. We wrapped up our meeting with a brainstorming session of 2005 Topics.

January 11, 2005

As more software IT work goes offshore, some local IT professionals are evolving their jobs to address the critical need for global project managers. In January we heard from three project leads/managers who are actively managing global [offshore] teams. They shared some of their success factors, challenges and concerns as they work within the new business model directed by executive management.

Yana Mezher - test lead, Software SETT Corporation. Over the past eight years, Ms. Mezher has focused on developing online QA courseware, supporting testing projects with mentoring, team training and hands-on project management. She is currently managing a team based in Bangalore, India.

Dave Weir - consultant with Calavista managing several offshore QA projects for start-ups in Silicon Valley. Over the past 18 years Mr. Weir has worked with companies such as Keynote Systems, XUMA, KPMG/Triton Container International, Pacific Bell and Tandem Computers. He is currently managing an offshore QA partner based in Pune, India.

Dave Liebreich - QA Manager at Yahoo. Over the past 20 years, Mr. Liebreich has been involved in test management, test engineering and system administration working on a wide range of technologies and products. He is currently managing teams based in Sunnyvale and Bangalore, India.

December 14, 2004

Outsourcing in Software Engineering

In the field of software engineering, outsourcing various software projects< requires developing, implementing, and managing methodologies that ensure that the job gets done and produces results.  This talk presents a high-level overview of outsourcing and the countries that are the big winners in this new business model.  It will also discuss the overall impact and ramifications related to the software and technical professions.

Sean Nihalani, DSc, is the Director of the Engineering and Technologies Department at UCSC Extension in Silicon Valley. Dr. Nihalani has designed, presented and managed engineering, IT, and management courses at many universities and corporations. His 18 years of experience includes design, development, troubleshooting of LANs, WANs, hardware, software, network security and databases, as well as managing various engineering, IT, and financial projects.

November 9, 2004

Part 11 - Electronic Records and Electronic Signatures: Review of the Regulation and a Discussion of Issues

This presentation provided an overview of the Part 11 regulation. Major areas of non compliance routinely found in non specifically developed Part 11 compliant software were discussed. Issues that organizations face in complying with the regulation were outline. Several compliance strategies were presented.

Tim Stein founded Business Performance Associates (BPA), a Cupertino based consulting firm, in 1994. Tim has helped over 100 clients achieve Part 11 compliance, validate systems, implement business applications, or develop compliant quality systems. Tim has recently finished a manuscript for a book titled: Computer System Risk Management and Validation Lifecycle. The book will be published by Paton Press and scheduled for release next year. He is a frequent speaker on the topics of Part 11 and software validation.

October 12, 2004

Topic: Why Performance QA is Broken and How to Fix It.

Abstract not available.

Damien Farnham is a Senior Manager for the Sun Microsystems Solaris Performance team in Dublin, Ireland.

September 14, 2004

Topic: SNMP: A primer

This presentation covered:

• The basics of SNMP
• Example wireless sessions
• Example wired sessions

William R. Estrada II has over 25 years experience as a System Programmer (Main frame and PC), Network Admin, Lab Manager, System Programming Manager, and Computer Operator. His on-the-job experience spans a wide range of operating systems (MVS, VS1, OS/2, DOS, Windows, Linux and Free BSD). His specialties are problem solving, automation, scripting, networking and last, but certainly not least, SNMP.

August 10, 2004

Topic: Staying Relevant in a Competitive Market

This presentation covered:

• Getting started in revitalizing your QA career
• Jumpstarting change in three areas
• The importance of soft skills

Peter Yarbrough is a QA professional with over seven years experience working on many large-scale IT projects for Fortune 500 companies. He has acted as QA lead or manager as well as QA engineer testing web applications and shrink-wrapped software products. Most recently he moved into technical support working with a small team acting as liaison between development and on-line support. As a QA professional who has had to adapt to a changing technical landscape, Peter brings a unique perspective to staying employed, staying engaged and relevant in the QA profession. He studied Engineering at Santa Clara University and holds a degree in Technical Communications from De Anza College.

July 13, 2004

Topic: Market Based Job Searching

This presentation covered:

• Current labor market information for job seekers
• WIA services and other free or low-cost resources for job seekers
• Hiring trends and factors for job search success

Merrin Donley is a Career Management Specialist with over twelve years experience working with diverse industries in Silicon Valley. Most recently she is working for the Silicon Valley Workforce Investment Board at Campbell One-Stop assisting recently laid-off workers in their job search.

June 8, 2004

Topic: Test Variables Impacting Wireless Applications

Here is a familiar test scenario. You are logging into an account. You enter in the correct account number and password. The logon authentication fails. You try again but this time you are careful - you are absolutely certain you have entered the correct account number and password. The logon fails again.

Did you check your headset? Headset! What does a headset have to do with a logon failure? In a wireless world, accessories can sometimes cause distortion or interfere with the sending of correct tones through your telephone or other wireless device. There are many other unique factors that can impact results when testing applications designed for a wireless environment.

This overview can help you enrich your test scripts and be prepared for the surprising results you sometimes get when testing wireless applications.

Gail Lowell is the Product Manager and acting QA Manager for the Unified Communications solutions business unit of InPhonic, Inc. InPhonic is a leading provider of communication software and services. Gail has several years experience managing the introduction and release of complex software applications in the US and internationally. She has industry experience in Unified Communications, Wireless Phones, Warehouse Logistics, Semiconductor Manufacturing, Human Resources, and Insurance software applications. Gail has been working with unified communications and wireless devices for the last four years.

May 11, 2004

Topic: Security Testing

In these uncertain times, security has taken on a new significance. Homeland security, white collar crime, corporate espionage, and new legal regulations drive forward the increased need for security. Customers, users, and the government expect and may contractually oblige you to deliver a secure application.

What is a secure application? What makes something "secure"? Will security testing keep my companies' name unsullied? Will security testing turn my staff into a group of evil hackers? How can I plan for, execute, and validate security testing efforts?

This discussion overviews what security is and how you test it. The presentation provides examples of what to look for and why. Security at both the conceptual and technical levels is covered.

Jason Reid is a test engineer at Sun Microsystems working in the Software System Test group. He has also been an SQA engineer in the Developer Tools group. Prior to joining Sun, Jason worked at the Purdue University Computing Center as an UNIX system administrator while obtaining his BS in Computer Science.

April 13, 2004

Topic: Realistically Estimating Test Projects

Question: When will the system testing be completed??? (Asked by an eager pest -- your boss -- with a tone of great anxiety.)

Note #1: At the time he asks this question, you do not know (a) the final scope of the functionality, (b) when the developers will deliver the final system for testing, and (c) what test resources you will have available.

Note #2: The boss wants a definitive answer and a drop-dead commitment from you in two minutes anyway.

Answer: Take a wild guess and multiply by two.

Question: What do you do when the boss cuts your agreed-on test duration by 85%??

Answer: Remind him that you thought he really understood that quality is important.

Let's face it: developing realistic and credible estimates is a critical survival skill for test professionals and managers. The word "estimate" is actually short for the saying: "Establishing Sloppy Time Intervals Makes for Agitated Test Engineers". This discussion overviews techniques which can help improve your estimating.

Ross Collard is president of Collard & Company, a consulting firm located in Manhattan, New York. His consulting assignments have included strategic planning for technology, managing large software development projects, improving software engineering practices, and software quality assurance.

Early in his career Ross was a hot-shot software engineer for Citibank in New York City. He first became interested in quality issues when he stayed up 48 hours straight trying to find a bug in his own code. During these same 48 hours the operational failure caused by his bug cost Citibank approximately 1,000 times Ross' annual salary. Fortunately for Citibank, this same loss only amounted to approximately 5 seconds worth of the bank's profits. Also fortunately for Citibank and the rest of the worldwide business community, Ross does not program much any more. But he sure does know how to test -- which is perhaps a more challenging skill than programming.

Ross Collard has conducted seminars on business and information technology topics for businesses, governments and universities, including George Washington, Harvard and New York Universities and U.C. Berkeley. He has lectured in the U.S.A., Europe, the Middle East, the Far East, South America and the South Pacific. He has a BE in Electrical Engineering from the University of Auckland, New Zealand (where he grew up), an MS in Computer Science from the California Institute of Technology and attended Stanford University's Graduate School of Business. He is writing a series of books on software testing and QA, at a gruesomely slow pace.

March 9, 2004

Topic: War Stories at the Ground Level

These short descriptions represent real events as related at this meeting:

• Incrementing build ID's at the decimal level so the builds wouldn't seem so many: .01, .02, .03, etc.
• Dividing test cases into smaller chunks so that the volume of work would seem higher on the metric reports
• Testing for the absence of a feature without knowing how to turn it on/off
• Testing based on a schedule-driven date - why bother and not just release
• Submit a ton of bugs into the bug base to help get more time for testing or to delay the release
• Failing of tests at the same point every day due to the sun coming through the blinds and overheating the computer
• Failing of hardware due to software tests
• Exploding toner which required the tester to be rushed to the hospital due to toner in the face, in the nostrils and eyes
• Conducting extensive rework to make up for organizational shortfalls before even starting the task of testing
• Publishing the results of the BEST load test run even though every run produced a separate set of results
• Releasing a product with the limits set at test levels, not the full production limits. Several examples were cited, but the most memorable one was Excel which, in a past release, was released with a 10x10 array, not the usual 65,000 x 65,000 array.
• Testing new hardware away from all the other equipment in the lab and away from a sprinkler head so that if it is catches fire or smokes, every piece of lab equipment doesn't get destroyed by water
• Not understanding why a specification is a key input to test scope

Bio not applicable

February 10, 2004

Topic: Roll Your Own .NET Automated Tests

Despite a variety of commercial graphical user interface (GUI) test tools on the market, programmers often find themselves resorting to manual testing of their GUIs. Adopting commercial GUI-based regression tools requires developers to learn a whole new development environment and language. Furthermore, these tools are often expensive and may be overkill for what developers need. Fortunately, there is an alternative for programmers who need to test a .NET GUI: reflection. By using reflection in the .NET framework, programmers can send events to user interface elements without a separate, specialized tool. In this talk, Elisabeth Hendrickson demonstrates how to use C# with nUnit to simulate user events to test .NET applications through the GUI.

This talk is a preview of the talk Elisabeth will be giving at Software Development Conference and Expo West 2004 on March 18 in Santa Clara, CA.

Elisabeth Hendrickson is an independent consultant specializing in software quality, management, and testing. An award winning author, Elisabeth has numerous published articles and is a frequent speaker at major software quality and software management conferences. She has worked with and for leading software companies since 1988. You can reach her at esh@qualitytree.com and read more about her ideas on quality and testing at www.qualitytree.com.

January 13, 2004

Topic: A Case Study in Best Practices in Software Process Documentation: Space Station Software Project Measurement and Analysis

The Software and Data Systems (S&DS) Team of the (International) Space Station Biological Research Project (SSBRP) has recently achieved a CMMI Maturity Level 2 rating. Key to this achievement was mastery of the Measurement and Analysis (MA) Process Area -- one that did not exist in the previous CMM. S&DS applied the Practical Software Measurement (PSM) approach to tackle the MA process area requirements. This presentation illustrates best practices in the area of Process Documentation through examples from S&DS's Measurement and Analysis process. Included are insights into Process and Plan template creation and use, tailoring, and compliance. And of course there's the measurement process itself -- which has garnered praise from the CMMI appraisers and from measurement experts at the SEI.

Rob Robason is a Senior Process Engineer with Intrinsyx Technologies at NASA Ames Research Center, supporting Ames' SEPG, and the Software and Data Systems (S&DS) team of the Space Station Biological Research Project. S&DS achieved a CMMI Maturity Level-2 rating last October, and Rob was singled out and recognized for his contributions to the team's accomplishment, including his definition and implementation of the S&DS Measurement and Analysis process, critical to the CMMI rating. Rob has also improved software processes at Cisco Systems, Accugraph, and Hewlett Packard. He also has experience as a Systems Engineer and Software Development Engineer at HP, and as the Quality Manager at Accugraph. Rob earned a BS in Electronic Engineering at Cal Poly, and has done graduate work in Computer Science at Colorado State and UC Santa Cruz and in Industrial Engineering at Texas A&M. Rob can be reached at rob AT robason.net.

December 9, 2003

Topic: Capability Maturity Model for Software (CMM)

This presentation will provide an overview of the SEI Capability Maturity Model (SEI/CMM). Topics will include discussion of the purpose of the SEI/CMM, the five levels, the importance of metrics, and how the SEI/CMM framework can be used to understand and improve the quality of software development.

Jeff Jacobs has over 20 years experience in software development, with a focus on software development methodologies and practices. His management experience ranges from leading development for startups to overseeing multi-company efforts for communications satellite systems. He has consulted to numerous companies and has trained over 3000 students in Information Engineering and various modeling techniques. Jeff is the author of numerous papers and is a frequent presenter at technical conferences. He received his B.S. in Information and Computer Sciences from U.C. Irvine, where he was one of the co-authors of UCI LISP. Jeff is a consultant providing services in software process improvement, methodology adoption and tailoring, business/systems analysis and modeling. His web site is http://www.jeffreyjacobs.com.

Last Updated:  .  For comments, contact Webmaster