[comp.software-eng] ECOOP workshop on testing

marick@m.cs.uiuc.edu (06/14/90)

Here's another ECOOP workshop.  Sorry about the cross-posting.

Workshop on Testing of Object-Oriented Systems 
Monday, October 22, 1990 at ECOOP/OOPSLA (Ottawa, Canada)

The purpose of this workshop is to bring together three groups of
people: researchers who design object-oriented systems, environments,
and techniques, practitioners who build and test large object-oriented
systems, and researchers in test methods.  The overarching goal is for
each group to understand the others and to understand current and
future problems and possibilities.

Invitations will be based on position papers.  In keeping with the
theme, the position papers should address what you think the other
workshop attendees need, as well as what you can provide.  Systems
researchers might describe their assumptions about how the environment
affects testing, and describe how a particular environment improves
the testability of systems built with it.  Testing researchers might
describe how they believe object-oriented systems differ from
traditional systems, and then explain how existing testing techniques
can be adapted.  Practitioners might describe particular experiences,
draw conclusions about the reasons for the experiences, and explain
how these provide data and ideas for systems and testing researchers.

My hope is that everyone's assumptions will be challenged and changed.


Topics of interest include, but are not limited to:

What are the major testing problems facing the developers of large
object-oriented systems?  How do they differ from other types of
systems?  What testing techniques and tools do practitioners use now?

How do existing testing methods apply to object-oriented systems?
Are they adequate, or are new approaches needed?  How are theories of
testing affected?  How can they be applied?

How do language features, programming environments, and design methods
affect testability?  What changes could improve testability?  How can
the need for testing be reduced?  What are the tradeoffs between
flexibility and testability?

The earlier testing can be done, the better.  Do object-oriented
systems represent design in a more concrete, testable way?  

Does the organization of object-oriented systems affect the
organization, development, or execution of tests?


Position papers are due on August 1.

Moderator:	Brian Marick
		Motorola, Inc.
		1101 East University
		Urbana, Illinois  61801
		Marick@cs.uiuc.edu
		(217)244-0263
		(217)384-8500 (to leave a message)
		FAX: (217) 384-8550