Mark has over 20 years of industrial experience in software testing specialising in the area of Software Testing Tools and Test Automation. This includes the development of a test execution tool and its introduction into general use within the organisation. Mark has given keynote talks and presented papers at international conferences and seminars. He has also published papers in respected journals and has served as Chairman for the BCS working group developing the draft standard for software component testing. He is co-author of the book Software Test Automation with Dorothy Graham of Grove Consultants.

Test Automation has come a long way over the past 20 years. IN that time many of today's most popular test execution automation tools have been brought into use and a range of implementation methods have been tried and tested. Some of these have been developed further in an attempt to drive down the build and maintenance costs of automated testing and to increase its benefits.

Many organisations that have been successful with test execution automation have employed a data-driven approach. This in turn has been developed further into more sophisticated approaches that are often termed keyword-driven (though other terms are used).

While a simple data-driven approach is easy to describe, keyword-driven is a little more difficult to pin down. There are now many implementations of keyword-driven test execution automation and some are hard to tell apart from the more advanced data-driven versions. So what is keyword-driven test automation? Perhaps we should ask "What isn't keyword-driven test automation?".

This presentation provides an objective analysis of keyword-driven test automation. By considering the various components of implementations, advantages and disadvantages of each together with benefits and pitfalls within different contexts are identified. The presentation concludes by trying to answer the question of what is, and what is not, keyword-driven test automation.

Anyone involved with test execution automation should find this presentation of interest. For those with little or no experience of keyword-driven approaches then the presentation will give a solid understanding of what can be achieved and in outline, how. For those already familiar with keyword-driven approaches then the presentation will bring greater clarity and possibly some new ideas for how they can improve their own keyword-driven approaches.

Harry Robinson is a Software Engineer in Test for Google in Kirkland, Washington. He coaches teams around the company in test generation techniques. His background includes ten years at AT&T Bell Labs, three years at Hewlett-Packard, and six years at Microsoft before joining Google in 2005. While at Bell Labs, he created a modelbased testing system that won the 1995 AT&T Award for Outstanding Achievement in the Area of Quality. At Microsoft, he pioneered the test generation technology behind Test Model Toolkit, which won the Microsoft Best Practice Award in 2001. He holds two patents in software test automation methods, maintains the site www.model-based-testing.org, and speaks and writes frequently on software testing and automation issues.

Software is getting more complicated, and the standard methods of testing it are not doing a thorough job. Manual testing is labor-intensive and does not always find the deeper, serious bugs in an application. Traditional test automation, on the other hand, is costly and fragile.

This is where model-based testing comes in. The idea of automatically generating tests from models of an application's expected behavior has been around for more than thirty years. It is only in the past few years, however, that model-based testing has gained credibility in the wider commercial software world. Using simple methods such as state machines, grammars, and sets, testers can automatically generate and execute millions of tests, but there is more to setting up a model-based testing plan than that.   We must re-think many core test concepts, from "what is a test case?" to "when should I start testing?"

This presentation will focus on how to do model-based testing, and how test generation is changing the world of software quality. What is model-based testing? Why is it so different from most testing today? Where is model-based testing pushing the industry, and where will it end up? And, most importantly, how can you prepare yourself and your team for the changes that are coming?