Abstract:

For more than two decades the term "Test Case" (TC) has been widely used as
a testing work unit, metrics and documentation entity. A thorough review of
the extant literature reveals that there is no formal and agreed-upon
definition of a TC, in spite of its centrality and extensive use. Following
Kaner's (2003) assertion that "good TC is one that gives the required
information", we see benefits in formalizing a unified, well defined and
structured TC format.

Dani  will review and explore the definitions of TCs, and propose a
classification of the various definitions into four categories, highlighting
the conceptualization underlying each category. We then discuss undesired
implications of this situation, suggest four criteria for a 'good' TC
definition, and discuss the benefits entailed by formulating such a
definition.

The presentation is proceeding by suggesting an alternative comprehensive
definition of a "Test Case" addressing the above criteria. The proposed
definition includes a TC structure, activities, and data types such as:
setup/preparations, dynamic input, internal flow and logic, verification,
and output /expectations. We illustrate the definition and its benefits via
the lens of the four criteria of a 'good' TC, and discuss its use in several
applications showing that it addresses issues pertaining to various
environments.

In conclusion Dani will discuss limitations and future research trajectories
entailed by the proposed TC definition, as well as its implications for
practice.

Keywords: test cases, definition, classification

Biography：

Currently, Doni most of the time teaches and researches at Ben Gurion
University of Negev Israel.

As software developer and team lead for 5 year, managing a small (20
Experts) software house and cofounding a company called NSI – a spinoff
from Digital CO. where he served as VP for software development for the
finance market. By 2000 Joined Amdocs as director for SW development and
customer account manager.

A year and a half ago, Doni have retired from Amdocs where he last served as
director for test automation – where he built all Amdocs infrastructure for
testing including the development of testing tools, methodologies and
training for all Amdocs R&D division with the new technology. His last
assignment was to lead enormous efforts of changing all software development
strategy – moving into agile and SOA for the entire organization.

Recently he have joined the Ben Gurion University academic staff; teaching
and researching.  Most of his writings are based on his experience in Amdocs
and other organizations I worked for. He teach SW quality Engineering  and
testing,  and  research the interaction between development processes and
quality/testing  trying  to introduce to the academic world some of the
achievements and work models we have develop in Amdocs and the industry in
all. He also try to apply the open thinking and free spirit of the academy
into new paradigms trying to influence industry.