Date: 17/11/2010

Release readiness (and how to assess it)

The purpose of testing is to be able to answer a question: "When will the system be ready to be released"?

The simple answers to the question are:

1. When you have adequate test coverage.
2. When you've found all the bugs you expect to find (and had enough of them
   fixed)
3. When the system is stable.

This talk goes through the means available to answer these points in a way which will convince management and is illustrated with case studies.

Speaker(s): Peter Farrell-Vinay

Date: 17/11/2010

Test tool evaluation and implementation

Even though there are countless lessons to be learned when purchasing test tools, we still continue to make the same mistakes we made over ten years ago and end up having expensive ‘shelfware' as part of our test tool kit. Many companies fail to adequately evaluate the test tools currently on the market and as a result purchase tools that only partially do the job or simply take too much time and effort to do what is expected of them. In this talk, Graham Moran explains a process specifically design for evaluating and implementing test tools and talks about some of the pitfalls that companies may find when trying to acquire these tools.

Speaker(s): Graham Moran

Date: 17/11/2010

Improve the economic results of enterprises through software quality activities that contribute to saving

In the current economic situation, when there is an urgent need to cut costs, it is smart to implement measures that contribute to savings while not introducing those that paralyze activities that cannot be waived to obtain the stability and the progress that businesses need. At this juncture, the public and private managers may stop to consider steps not yet noticed, misunderstood or improperly developed within their own areas of technology, but not through clichéd and empty talk of cost optimization as is typical in the TIC sector.

During 2009, according to data from specialized consultants, only the Spanish government administrations spent around 1.100 million Euros on development and maintenance of software applications. If since then they had done software quality testing, in 2013 - the European Union committed date to return to the Stability Pact redirecting the deficit to 3%- they would have saved 750 million Euros over this period, at 125 million Euros per year. But this is just one example of what a well-established, well-managed business assurance and quality control software could contribute to improve performance of the companies.

Thus, collaboration to improve results and reduce company deficits is achievable through the introduction of assurance and software quality control, but what to do and how to get it? During this presentation, we will give you keys to achieve this goal.

Speaker(s): Francisco Sáez

Date: 17/11/2010

Reconciling worlds with MS Test Manager

Everyone is aware of the constant fight between developers and testers; the former make coding mistakes, the latter find the resulting defects. Other software vendors have traditionally approached the development and testing market by providing completely disparate tools and environments, forcing testers and developers to work in isolation and driving the two groups further apart. Developers hate leaving their coding UI to check the status of new defects; testers are constantly faced with ever changing environments and data without warning.

During the speech, Jose will demo Microsoft Test Manager and explain how it can be used.  Jose Aracil will discuss his experiences working in these kind of organizations where testers and developers are in constant opposition. Jose will enlighten the audience with a different approach to this collaboration between testers and developers to increase the efficiency and quality of the software being developed.

Speaker(s): José Aracil

Date: 17/11/2010

Accessibility Testing Methodology for Visually Handicapped in Web Environments

Currently, the world is in the era of digital inclusion and accessibility of Web pages has become a global concern, since this environment plays a key role in the daily lives of people with special needs. This component facilitates the lives of these people, enabling them to create new forms of relationships and fulfilling activities before unviable. The Brazilian government decreed that all sites of public administration must be accessible, serving as an example for other countries. For these reasons, this work aims to develop a methodology for testing accessibility to inspect all the phases of the project, from its conception, when the test plan is done, until delivery, after examining the tests results, anticipating the errors, generating less rework, reducing project costs and causing it to be given due importance for accessibility.

Speaker(s): Virgínia Chalegre

Date: 17/11/2010

Application Transformation and Continuous Quality Assurance to help achieve the Business objectives

Every day thousands of enterprises rely on software to drive critical business applications and keep vital services running. Their success depends on software performing effectively and fulfilling the tasks for which it was implemented or designed.
But unfortunately maintaining and building software can be complicated and, therefore, senior IT decision makers face the following challenges to keep software delivery aligned with the business:

- Reduce cost
- Improve quality
- Reduce risk
- Increase agility

This presentation addresses these challenges and gives an overview on how to reduce them.

Speaker(s): Belén Bernardos

Date: 17/11/2010

The meaning of Application Quality to the Business

As a former CIO at Allianz and Paribas, Paul Bentz has had a direct and personal experience of the challenges IT Executives face to service an increasingly demanding business community.

When it comes down to managing the complexity and performance of an Information System a lot has been achieved over time in the areas of infrastructure and project management.

Strangely enough the internal quality of business applications has been left aside for a very long time and is becoming the focus point of management with the new sourcing models and the concern for operational risks.

Mr. Bentz has always been trying to find ways to manage risk, measure outcomes, and establish the governance for application delivery that would allow benchmarking.

As an advisor to CAST Software Mr. Bentz will explain why and how CAST is providing companies around the world the ability to immediately and significantly minimize business risks, reduce hard IT costs, and gain an objective understanding of the performance of their outsourced and in-house IT teams.

Speaker(s): Paul Bentz

Date: 17/11/2010

Business Driven Test Management at Panda Security

Panda Security, having implemented a series of improvements and practices proposed in a Product Lifecycle Management consulting project executed in 2008, has significantly  boosted the level of quality of their products and the control over the development and QA processes.

By the end of 2009, the need was identified to improve the balance of the investment in quality control with the exact needs of the business. Panda invited Sogeti Spain to study and optimize their test strategy and estimation processes. From December 2009 until february 2010, a Sogeti Consultant analyzed the maturity level of the QA divisions test management processes and stakeholder relations. Using the TPI®Next test process improvement model, Sogeti tailored and deployed a process fitting to the maturity and objectives of Panda, based on the Tmap®Next Business Driven Test Management philosophy.

To improve visibilty and enhance relations with business and technology stakeholders, the QA Divisions mission has been stated in a QA Policy and
detailed in an organisational QA strategy.

The resulting test strategy process and templates have been deployed in two projects to date, and is currently being rolled out within the remaining
certification groups, so as to be fully implemented by the end of June 2010.

Speaker(s): Miriam Serna, Ewout van Driel

Date: 17/11/2010

GUEST: Worldwide Testing Services Market 2010-2014: Key Growth Opportunities & Sector Trends

In the past, testing has often been overlooked in terms of its importance to the business as well as its contribution to annual IT costs.  At many organizations it has often been the developers that perform testing over the course of a project, which meant that the true cost of testing was often buried within the overall development budget, and testing rigour was sacrificed in order to meet release deadlines.  

Recently, several factors have helped drive Testing services up the list of priorities for IT decision-makers: the renewed pressure on IT to contribute to company-wide cost reduction programs in the shadow of the economic slowdown, and a greater appreciation of what the cost of unsuccessful, insufficient or misdirected testing can be to their business. More and more system failures have been highlighted by the media causing, beyond disruptions in activities, some prejudicial impact on image. 

Worldwide spending on testing (including both internal and external testing skills, tools and related hardware and systems) will continue to grow faster than the IT market and should reach approximately the €100bn mark in 2014.

Testing is also becoming one of the strongest areas of recruitment in the IT sector, and PAC estimates that there are now more than 100,000 professional testers worldwide.

The recent Worldwide Testing Services report written by PAC Consultants, Nick Mayes (PAC UK) and Arnold Aumasson (PAC France) gives a clear insight in the global software testing services market, one of the most dynamic areas of the IT services sector.

Speaker(s): Arnold Aumasson

Date: 17/11/2010

KEYNOTE: What Managers Think They Know about Test Automation but Don't

Recent research has highlighted the critical role that managers play in the success or failure of test automation.
Without adequate management support, well-intentioned initiatives are doomed to be short-lived and guaranteed to waste time and money.
But even extensive management support, without sufficient knowledge of the issues, may fail to produce results with lasting and growing benefits to the organization.
 
In this talk, Dorothy Graham outlines common management misconceptions about automation and how to ensure that you have realistic expectations. Understand why objectives and measurement are critical for test automation and how automation objectives are different to test objectives, how to scope, estimate and resource test automation efforts, how to measure Return on Investment, and more. Know what issues to look out for as early indications of problems, and how to get back on track to successful automation.

• understand management’s role in successful automation and how to guide for success
• avoid common mistakes managers make in automating testing
• know what to monitor and how to get back on track

Speaker(s): Dorothy Graham

Date: 17/11/2010

The Cassandra Syndrome: The Tester’s Dilemma and What To Do About It

It can be a challenge for QA and test engineers to be seen as making positive contributions to the success of software projects while engaged in essentially ‘negative' activities - looking for problems, predicting problems, preventing problems, and reporting problems. This ‘negative' focus is highlighted by software testing book titles like "How to Break Software". While others working on a project focus on how to make software work, testers focus on finding ways in which it doesn't work. These divergent perspectives have the potential to interfere with team communication, cooperation, and effectiveness, as well as impact job satisfaction. QA and test may be seen as ‘not team players' with a ‘negative attitude'.

This problem is not unique to software projects - it is in fact a human issue so pervasive that it was given a leading role in one of the best-know literary works of all time, Homer's Illiad, the seventh-century BC collection of Greek mythology and legends. The Illiad contains the story of Cassandra and the siege of ancient Troy. Cassandra had the gift of prophecy but the curse of having no one listen to her prophecies. Many testers have at some point felt like Cassandra. Predictions of impending project problems were not heard. QA engineers may have predicted failure to meet deadlines if processes weren't followed, but no one paid attention. The presentation will discuss what we can learn from the story of Cassandra, and what can be done to counteract the ‘Cassandra Syndrome'.

Speaker(s): Rick Hower

Date: 17/11/2010

Offshore Delivery of TTCN-3 Testing Services

The use of TTCN-3 outsourcing services seems to increase in direct-proportion to the downturn of the economy.  Organisations are now looking to outsource part of their software testing function. TTCN-3 Testing and development can be outsourced. This brings a clear financial benefit as well as a competitive advantage to organizations already using TTCN-3 and also to organizations willing to introduce TTCN-3 in their testing processes. For example, small organizations can benefit from a wealth of TTCN-3 testing expertise that otherwise will be very expensive an improbable to introduce in these types of organizations.

However, a proven methodology and approach for TTCN-3 offshore testing projects is vital for the success of these types of projects. The challenge of managing relationships will grow as the extent of the TTCN-3 outsourcing grows.  This presentation will highlight the key issues to take into consideration when outsourcing TTCN-3 projects.

TTCN-3 offshore testing services have clear financial and performance benefits which make them a serious consideration for many small, medium and large companies.

MTP has vast industry experience in offering and carrying out TTCN-3 Offshore projects.  In this presentation we are going to introduce the benefits of a methodology successfully carried out for TTCN-3 Offshore projects.

The methodology includes: continuity and control of deliverables, a flexible project to shift on-site TTCN-3 testing activities offshore, continuous checks to early detect errors, control of the budget through an early analysis, client relationship, cultural proximity, innovation and price based on TTCN-3 scripts deliverables not on development time.

Speaker(s): Raquel Jiménez Garrido

Date: 17/11/2010

Road Map to a QA Center of Excelence

Because testing is a mission-critical activity, we discuss a road map for building a competency center, starting small and building it out over time. The ultimate goal is establishing a Center of Excellence (CoE) with the ultimate promise of Efficiency, Consistency, Practicality but also a Career Path for your testing team, while significantly reducing your costs, accelerating the time to market and reducing the risk exposure.

Before we dig into the steps to build it, we will review the rationale behind taking that journey: Why should you consider a CoE? What should a CoE accomplish? Who is going to be involved?

Then we will go to the next level: When and where should I deploy it? How Should I do that? Deploying a CoE will involve people, process and technology and a number of decisions to trade off, like: Do we want to favor a centralized or a decentralized model? Through a series of pragmatic steps we'll review what the implications are to those three dimensions, as well as the fact that not two organizations will adopt it the same way.

We will wrap it up with a live demonstration of how an industry standard solution can help accelerate the adoption and realization of the CoE benefits.

Speaker(s): Cesar Franco

Date: 17/11/2010

Selecting the Most Effective Test Design Techniques

Although there is much information available on test design techniques, literature says very little on how to select the ones to use. Many testers find it difficult to select the right techniques. Very often a design technique is used for the simple fact that it's possible to do so. When selecting a technique the leading argument should be "the errors that we do not want to have in your production system". Testers should gain a better understanding of the relation between types of errors and the techniques that help finding them. This understanding will lead to a better use of techniques and will lead to asking the right questions to the stakeholders. By using the right techniques, the tester executes the right tests and is accountable for his actions and the quality of his advice. Since the techniques define the tests that are executed, they also determine the information that can be given to the stakeholders.

Selecting the techniques you are going to use is a careful process. Selecting the wrong technique may cost you a lot of time without finding many errors or useful information. Failing to select the right technique may lead to major defects not being found. I'll explain the decision model from my book, and will inverse the problem. We do not use techniques because we can, but because they help us to provide the right information. Unlike many methods we start with the errors and select the techniques that help finding them.

Speaker(s): Derk-Jan de Grood

Date: 17/11/2010

Pentest as verification of the security level in Web Applications

Fortunately, more and more organizations are incorporating SDLC security strategies, understanding that adopting a set of best practices will help ensure effective and efficient levels of security as new developments arise. Many frameworks can be used to guide companies in implementing these strategies such as OpenSAMM (Open Software Assurance Maturity Model), BSIMM (Building Security in Maturity Model), Security Software SDL Touchpoints and Microsoft, among others.

Each application, however, requires a certain level of security, though we must also know what evidence we need, how we find it and what goals we pursue.

In addition, it is vital to include Security Testing objectives when developing secure software for this or any other related discipline (such as for the security frameworks described above).

This presentation will focus on these questions and the techniques currently used to carry out a comprehensive review of such applications.

Speaker(s): Vicente Aguilera Diaz

Date: 17/11/2010

Testing, so many problems but we have the solutions don't we?

Sir Humphrey Davy said "The most important of my discoveries have been suggested by my failures."

This is as true today as it was when Sir Humphrey said it. I have been working in the test process improvement field for many years and see the same issues coming up time after time. For example, I see testers blaming everyone else for the issues they have with requirements but not doing anything to get the issues resolved. I see many test teams arriving on a project and they seem to start from a blank sheet of paper each time, repeating every error and taking all the same risks as they have always done, it's as if they never learn or want to learn.

As a profession we need to look at our common failures and understand why they happened and how we stop them happening again, and who knows maybe we will find the testing silver bullet?

This presentation will provide an overview of the common problems testers face today (problem patterns) from various perspectives:
• Outside the test project looking in
• Test Organisation
• Test Management
• Test Environment
• Test Design and execution
• Test Automation
alongside some suggestions for how to avoid them in the future.

Speaker(s): Geoff Thompson

Date: 17/11/2010

“Software Quality Expectations and Perceptions”: Mission Impossible?

This presentation examines service quality assessment or software quality assurance in terms of two criteria: "expectations" and "perceptions."

Currently, software quality systems are thought to be "mature" in terms of their implementation as they are supported by numerous standards and good practices. However, the majority of these systems are based on abstract methodologies and are applied directly without analyzing the specific needs and expectations of the specific management teams, organizations or clients. They thus tend to be less than satisfactory.

For this reason, we would like to reflect on this issue from two perspectives: the first refers to clients' perspectives and the second to their perceptions. The difference or discrepancy between the two serves to define the level of service quality.

The key behind offering a quality software service in IT organizations is found by balancing these expectations and perceptions and thus improving the service provided. This raises several additional questions: Do we need to talk about software quality types? Should we consider objective-based software quality? And, who benefits from our actions? In effect, on many occasions the strategies required to satisfy the different information technology clients are diametrically opposed and, as such, are ill-suited to satisfy all our clients' expectations.

Speaker(s): Salvador Folgado

Date: 17/11/2010

Load Testing on a Cloudy day

We will focus on the art of load and performance testing. What can you expect for tomorrow? How you must make your load testing practices evolve to fit with the application development lifecycle?

We'll start with the basics: "Why is load testing a central part of any testing process for successful deployment". You will grasp the complexity of performance testing and know why it shall not be considered as an easy or optional task.

Then, this will give you a strong insight into the importance of using the right methodology to deal with the latest industry challenges. Sooner or later, you'll have to deal with the complexity brought by new technologies such as Rich Internet Applications or the use of the Cloud. You will understand how your recommendations can have a beneficial impact on your organization's choices.

Finally, you will be given the keys to execute Load and Performance testing to gain in productivity & flexibility. You'll deliver more value to stakeholders and gain respect in return.

After attending this presentation you will be prepared for what's coming your way. You will know the key questions to ask yourself. This way you will be able to give the right recommendations to improve the quality of the applications you have worked on and the effectiveness of your team.

Key Points:
1 Why Load and Performance testing is not just optional or to be done if you have enough time
2 Deal with complexity, the latest web applications and the Cloud.
3 Make your practices evolve and deliver more value more effectively to your stakeholders.

Speaker(s): Thomas Ripoche

Date: 17/11/2010

Best Practices for Delivering Quality Solutions in a Distributed, Agile Environment

There are many pressures on software delivery organizations to produce more software faster in the context of extreme cost pressure and growing globalization of the software delivery organization. How can these pressures be addressed while maintaining software delivery quality?

This presentation will explore the changes in software delivery models that have taken place in recent years, and the ways in which overall software delivery quality can be managed in the current environment. In particular, we will discuss several case studies where large commercial organizations have achieved significant improvements in software quality when adopting more agile practices. We will conclude with a set of best practices for adopting "agile at scale" that can be applied to your own organization to provide measured improvements in quality in your software delivery model.

Speaker(s): Alan Brown

Date: 18/11/2010

Estimation of the resources involved in a Testing Project

In this talk we'll present an estimation method of the necessary resources in a Verification & Validation Project for a Critical & Complex software system.

In this way we'll define:

• The scope of the tests in a software development project,
• The basic structure of a testing team members, and the activities to realize,
• The ratios and metrics suitable for the estimation of the resources to use, such as:
  • Number of system requirements
  • Test planning definition
  • Number of the needed tests for requirements (using historical values in related projects)
  • Number of designed tests by Test Engineer
  • Number of executed tests by Test Engineer
  • Number of executed tests with anomalies
  • Number of regression tests
  • Number of awaited anomalies
  • Percentage of changes in the system requirements documentation
  • Impact of those changes in tests

 Through the talk we'll show both estimated values and real values, obtained from a practical case of an Air Traffic Control System.

Speaker(s): Juan Luis Valera Juanas

Date: 18/11/2010

Service Quality Assurance: A Global Approach

Getting back to basics, one frequently asked question is: "Why test?," followed perhaps by: "What to test?"

In most organizations, testing pays off because of the great benefits it provides to ensure key business application and service functionality, availability and performance. These business services can have their functionality compromised due to pressures from many fronts: new applications, new application versions, consolidations, infrastructure migrations/upgrades, architectural changes, etc. What and how many of these changes are tested? How? Why? Why not?

From this perspective, we will review our experiences and real customer cases running every kind of test, not only during functional testing phases, but also when running diagnostics, system integration tests, and infrastructure migrations and consolidations. With a hands-on approach, we will analyze the key elements to guarantee a successful testing process, such as test configuration management and test data management. Additionally, we will review the best practices and success stories from our customers in the deployment of testing CoEs, packaged application deployment, infrastructure optimization, datacenter high-availability, etc.

Finally, we will review Oracle Application Quality Management's integrated capabilities to provide tangible benefits to business service lifecycle management, providing solutions for multiple phases within the application lifecycle.

Speaker(s): Iván Menéndez

Date: 18/11/2010

TPI Next: The next step in test process improvement

Test Process Improvement (TPI®) methodology, has been instrumental in helping many organizations improve their overall test processes, an essential initiative to ensuring the quality of information systems and critical business processes. Providing insight into the maturity of the test process, TPI has offered logical and practical steps to enhance test efficiency and effectiveness, and moreover to instil a belief in achieving a permanent improvement cycle.

Speaker(s): Augusto Sagrario

Date: 18/11/2010

Help! We have a QA problem!

This is about a real case of too many developers feeding too few testers, causing a testing backlog of half a year, with many angry customers waiting for too long for solutions to their problems. One senior tester just had left the company. There was only one senior and one junior tester left. They were facing this huge backlog of work and didn't know where to start.

We will show how empowerment of the testers, careful planning and involvement of the developers allowed the testers to catch up in about 9 weeks, systematically making customers happy one by one along the way. The senior tester learnt how to plan the work of the testers effectively and efficiently in sync with the developers, so that there were no backlogs ever since. Trust by customers who were in the process of abandoning the supplier was restored causing turnover to grow enormously since.

We will first show how we used Evolutionary Planning techniques in this particular case. Then we will discuss in more general terms the elements of this planning technique. The answer to the question "Who is the customer of Testing?" usually causes quite a shock to most of the audience. However, this is usually a shock of recognition.

Speaker(s): Niels Malotaux

Date: 18/11/2010

KEYNOTE: Growing our industry: cultivating testing

Although software testing is a relatively young discipline, immaturity is not the only reason we are still developing our methods, professional qualifications, trade associations, and position in the software industry and society. All successful professions must continuously evolve and grow. Membership of a profession provides an expectation of behaviour, qualification and experience, regulated by a professional body. We need to understand whether it is desirable or possible to be a profession, and how we should regulate ourselves for the benefit of our customers, the industry and practitioners.

Software testing is not the only trade or industry to be working through what is meant by being a profession; the horticulture industry is going through a similar discussion of whether profession status is desirable, achievable or necessary. Horticulture has been practiced for about 8,000 years longer than software testing. During those millennia, horticultural practices have continued to develop, supported by accidental discovery, increased scientific understanding, and improved technology. Just like horticulture, software testing is a multi-discipline, science- and technology-driven industry with political, sociological, and economic implications.

Both horticulturalists and software testers need to consider:

• How regulation has been used in other trades and industries as they developed into professions
• The role of professional bodies in controlling and regulating professionals
• The role of standards, qualifications and bodies of knowledge in controlling and regulating entrance and practice levels for professions and trades
• The role of monitoring and regulation in maintaining standards of behaviour and practice.

Presentation track record: highly praised keynote at EuroSTAR2007, received with equal success at Eriksson Conference 2008, London SIGiST 2010.

Speaker(s): Isabel Evans

Date: 18/11/2010

How to Successfully Outsource Testing

When outsourcing all or part of your testing efforts to a third-party vendor, you need a special approach to make testing effective and controlled. Martin Pol explains the roadmap to successful outsourcing. How to define the objectives and the scope, what tasks should be outsourced and what tasks should not-at least not yet. He explains how to select your supplier and how to migrate, implement, and cope with organizational issues. Martin discusses contracts, service level agreements, compensation issues, and monitoring and controlling the outsourced test work, including specific metrics. The good news for testers is that outsourcing requires more testing-not less-and that new testing jobs are coming into existence. Testing the outsourcing is becoming a very important control mechanism for outsourcing in general.

Speaker(s): Martin Pol

Date: 18/11/2010

Lean Test Management: ban waste and gain efficiency

Cost reductions and the quest for more efficiency are more evident in today's business world. It also follows that our testing processes will ultimately be affected.  Many management theories speak about "Lean" as being one of the solutions. One of the key steps in using "Lean" is the identification of which steps add value and which do not. This presentation will explore the use of "Lean" within testing and more specifically within test management.

As a guideline, the presenter will follow the "Lean manufacturing process": the generic process management philosophy derived from the Toyota Production System. It is renowned for its focus on reduction of "seven wastes" in order to improve overall customer value. Everything not adding value to the customer is considered to be waste.

This presentation will focus on the various elements as mentioned above. Also Six Sigma as being one of the more popular theories that introduces the concept of "Lean" will be explained. Several examples of Lean Six Sigma techniques applied to testing will be given during the presentation.
As opposed to test process improvement with reference models like TMM , using lean six sigma will start improving by thoroughly investigating your own companies processes and improve from there. This will all be supported with techniques that immediately enable you to show the added value for the organization. This presentation is especially of interest to business managers, IT managers, QA managers and test managers that are involved in improving the quality of test management processes.

Speaker(s): Iris Pinkster O'Riordain

Date: 18/11/2010

Building the system right is not enough, build the right system. Requirements and tests into IRQA

Requirements and tests are key artefacts for quality, both are needed, along with their traces, to build the system, test it and finally, accept it. Tests are there not just to show us that we have build the system right, but also to let us know if we have build the right system. During this demo, it will be shown how IRQA, leader requirements engineering tool, supports the definition of requirements, tests, the creation of traces between them and how we can check the specification using IRQA. Additionally, during this demo a scenario will be set up to show how acceptance tests can be obtained automatically from use cases, how we can manage the executions of tests and how we can check tests and requirements through traces, reports and dashboards.

Speaker(s): José Manuel Muñoz

Date: 18/11/2010

How to improve software in order to facilitate its maintenance?

It is well-known that the cost of software application distribution and maintenance (developmental, corrective and preventative) constitutes "the lion's share" of the total cost of the applications for the organisational user.

Standards such as ISO 9126 define maintainability as a key characteristic of software quality. Unfortunately, the recommended metrics for those standards are usually obtained ‘ex post facto': they indicate the efficiency of the maintenance process, but they lack the capacity to explain what characteristics make the software difficult and as a result, they do not make it easy to take concrete technical measures that reduce this maintenance cost.

The presentation, presents different models oriented to measure directly (by means of product metrics) the maintainability of the software and its subcharacteristics: stability, ease of analysis, change and test. It analysis the properties of these models and proposes one that achieves the key objectives (technology independence, measurements that are easy to compute, explanatory ability) that permit its application to the deliverables of a software development project.

The presentation shows the application of this practical model in different scenarios of software development (internal development, external development, software houses and factories) and technology and how this allows the identification of preventative measures that improve the future maintainability of the software system, in the development phase as well as in the maintenance phase.

The presentation will finish showing how this model has been implemented in checKing QA, the dashboard of quality of Optimyth Software.

Speaker(s): Luis Rodriguez

Date: 18/11/2010

Best Presentation Awards & Draw of the Delegates Contest

Raynald Korchia will give expo:QA'10 prizes to the best speakers of the conference: the Technical Committee Award and the Best Presentation Delegates' Choice Award.

An iPad will be drawn between the delegates who participated in the Vota y Gana (Vote & Win) Contest.

Date: 18/11/2010

How can you demonstrate savings from software tests?

Software testing is generally accepted as something good for IT organizations, leading to a direct increase in their systems quality. At times, however, it is also seen as an expensive and redundant activity.

The main problem is that Testing Management activities are not often backed by clear objectives nor is there measureable data demonstrating how good these activities are for system health or how much money and time they actually save IT organizations.

Speaker(s): Ignacio López Carrillo

Date: 18/11/2010

Model Based Testing

Test design is an important yet time consuming part of software testing. Many techniques exist and are used in practice. However, the use of formal techniques is not widely implemented in today's organisations.

It can be observed that requirements and system design are lacking detail and are subject to frequent changes, and test design is started late in software projects, with the following consequences:

• Late detection of specification omissions (huge negative effect on Time
• To Market)
• High chance on human errors
• Difficult to provide proof of test coverage
• Expensive test design due to (manual) rework
• High level of test (design) knowledge required

 

The model based testing approach presented in this paper aims at providing testers with techniques and tools to formalize and anticipate test design, so as to eliminate some of these constraints, and reduce significantly the effort related to test design.

The tools presented help create test designs that are easily maintained, create test cases that are "automation ready", and help testers give visibility of the coverage of their design.

Speaker(s): Pilar Blanco

Date: 18/11/2010

Quality & Testing Benchmarking: Measure, be aware and react

In this session Sopra Group will comment the results of a market study done by 30 Spanish companies measuring the quality level of their applications and the maturity of their testing processes.

The study answers questions like:

• Is the software development done in Software Factories different because of its quality?
• Are there any differences between companies with specific testing teams?
• How old is the software that has been measured by its quality?

 

Measuring the software quality levels and the maturity of the testing function, the companies have obtained a real knowledge of their situation. It also has helped them to be aware of what they really need, allowing them to set improvement objectives in the short and long term.

Sopra Group will also comment the best practices that some companies have adopted after the measuring phase. They have put in place some policies that have help them to reach important improvements, such as the reduction of the maintenance costs around 10%, savings of 6.000 man/days in software projects or the error reduction in the production environment of around a 40%.

Speaker(s): Luisa Morales

Date: 18/11/2010

Functional Test Automation in SCRUM

Functional test automation plays a crucial role in ensuring the quality of applications that are developed using agile processes such as SCRUM. Intermediate releases in short periods of time require the execution of regression tests more frequently, rendering manual execution unworkable. In turn, for automation testing to be successful, the selection of scenarios to automate becomes critical, as does the planning and management of the generated artifacts during the process.

This presentation will show how the activities related to the automation of functional tests were incorporated to the SCRUM team, ranging from the standardization of the automation process to the training of the testing team in the incorporated technologies. The presentation will also address how the generation of progress reports within the process was automated.

This experience demonstrates that agile processes do in fact allow for planning and the implementation of best practices, which in turn give visibility to the work done by the testing team to all players.

Speaker(s): Claudia Badell

Date: 18/11/2010

The Software Testing Ethics Debate

A "House of Commons" debate on the topic 'Ethics within the testing profession'. In this sixty minute session stimulating statements are given
that will guaranteed lead to a meaty discussion and push (y)our tester's morale to its limits!

The hosted 'House of Commons' debate is derived from a Dutch television show whereby the goal is to bring current affairs to your living room in both an informative and entertaining fashion. The statements are introduced by the hosts. One host will be in favor of the statement and the other host will be against. The audience will have the opportunity to react on the pleas made. The hosts will encourage the audience to take their side and philosophize on the chosen point-of-view.

Here are a few examples of statements you that might be raised:

• You can break the law to meet your test goals
• You must always tell the truth
• You must always be able to use Privacy sensitive data to test
• As a tester, you should not have a professional code of conduct, because it hinders good testing
• Testing and ethics don't mix

Do YOU dare to let go of your standards and philosophize what could be when one is not so honest? Are some statements really that black and white? And what are those nuances to be applied? Share your opinion and perhaps we will arrive at the basis of the "Testers code of conduct"!

Scored "Outstanding" at EuroSTAR 2009 (9.00 of 1-10 scale).

Speaker(s): Mystery Guest, Nathalie Rooseboom de Vries van Delft, Ewout van Driel

Date: 18/11/2010

Non-Process to Process: QA CMMI Level 1 to Level 5 in Less Than Four Years

Keep it simple. That's the philosophy of software testers who made this CMMI Level 1 to Level 5 possible and repeatable in less than 4 years.

Learn and see the templates, approaches, checklists, metrics, QA estimation model and other supporting documents that, once implemented and standardized, helped this company elevate its testing organization to CMMI Level 5 in less than four years. CMMI (Capability Maturity Model Integration) is touched upon lightly but the focus is not a process improvement specific model (CE, CMMI, IEEE, ISO, ITIL, OSA, SEI, etc.); rather, this presentation is on testing processes that satisfy the business need first and, as a added bonus, lead to robust testing process model that has been appraised to satisfy CMMI Level 5.

Many companies spend many years and many dollars to put process, control and repeatability into their organization. This presentation focuses on only one part of the organization, the test group. Additionally, it is focused and developed with what works best for this business.

Attend this session for the road map from zero process to a robust process that will work for you.

Speaker(s): Jan Fish

Date: 18/11/2010

Agile Validation for Medical Device?

In a Medical Device environment, software validation is requested by regulatory bodies. Validation is expected "to provide a level of confidence that the quality of the software is appropriate to support Public Health" (FDA). To be able to certify that the software fulfils its mission for public health,
an inspector needs to know what the software does, that it works properly and that it does not imply risks for the public health. In "SW wording" it becomes: having SW functional description, test process reports and a risk management report. Validation is then a broad concept for Medical Device; it is the assurance that the process of software design, construction and verification is planned and controlled.

Is a specific SW methodology requested? There are standards that regulatory bodies suggest, but any approach is accepted, as far as it provides the
inspector with what he needs. Agile is accepted then, with those adjustments which allow manufacturer to provide the documentation requested.
There are principles in Agile approach, such as more emphasis on communication than documentation, on productivity than reliability, that may generate gaps in regulatory requirements: gaps must be overcome. Other principles, such as continuous testing, customer feedback, improvement, are in full agreement with the regulatory goal of minimizing defect insertion and maximizing defect removal from the beginning. Other Agile practices fulfil regulatory expectations, for example the Individual ownership and accountability for quality.

In conclusion, an Agile approach that considers an enhanced Life Cycle including initial overall plans and requirements, Final acceptance Test and
documentation cleaning, controlled release, appears as a good option, fulfilling both business needs of fast, easy and "low cost" adaptation to
changes and regulatory requirements (that Medical Device manufacturers are part of the business).

The presentation will focus on broad Validation concepts and Agile fulfilment or needed adjusts.

Speaker(s): Celestina Bianco

Date: 18/11/2010

A difficult meeting

Tuesday morning, rush hour was a nightmare and you didn't have enough time to prepare the impossible, a presentation detailing how to test an application for an error free release. You know the drill, show your ID, go to the meeting
room, connect your laptop and wait for people to show up... 10 minutes later you're finally facing a new customer; he is an "old school" manager, he always thought things will work if you scare developers enough to work 14 hours day to
deliver code. If performance is not good, we'll add more hardware. If end users are not happy with the tool, we can always get another company to carry on with the development, and we might even save money...

Join Jose and Alvaro in this role playing presentation to learn how software testing meetings work in real life.

Alvaro and Jose will cover the questions and situations they have faced, on some of the most difficult meetings they have been through over the last 10 years as consultants, and how they successfully overcame them.

Situations included in the presentation:

- Jose Vs "The evil manager"
- Alvaro Vs "Mr. CMMI"
- Jose Vs "The (I want it all) manager"
- Alvaro Vs "Mr. Automation"

Speaker(s): José Aracil

Date: 18/11/2010

Automation Testing Activities

Automation testing is not only functional testing automation. We will have a look over different testing processes and give some ideas on how automation can help us in our testing processes. We will try to dismantle some automation myths and to present a different vision of the testing automation processes.

We will talk about: TICs current frame, SQA and Testing Activities, Activities Automation, How to approach the automation, Automation Myths and Realities and Conclusions.

Speaker(s): Miguel Castaño