home | calls | dates | committees | submission | program | registration | venue | news | contacts |  colocated events | publicity material
Program
Overall Schedule
Main Conference Technical Program
Keynotes
Social Program
Keynote Speakers at WCRE 2006

Donwload a printable version of the WCRE 2006 program: pdf

Hausi A. Muller
University of Victoria, BC, Canada
Home page

Lionel C. Briand
Carleton University, Ottawa, ON, Canada
Home page

Title: Bits of History, Research Challenges and Autonomic Computing Technology

October 26, 2006; 9.00am-10.30am

Over the last fifteen years, the software reverse engineering community has produced many software engineering methods, tools, and techniques that have had significant impact in the software industry. Trying to assess the impact of various approaches and results is a difficult but worthwhile exercise.

The lessons learned from such exercises lead directly to research challenges for the field, and may shape the focus of the community. After a research area has evolved for 10-20 years, it can easily fade away due to narrow focus, overgrazing, or lack of impact. It is also useful to investigate cross-fertilization with related areas. Autonomic computing is a new field with plenty of exciting technologies that could be employed for software reverse engineering.

In this keynote presentation, we will highlight selected achievements of the reverse engineering community and their impact in software engineering practice, present research challenges in this rapidly expanding field ofreverse engineering, and illustrate how autonomic computing technology can be leveraged for reverse engineering and program understanding tasks.

Title: The Experimental Paradigm in Reverse Engineering: Role, Challenges, and Limitations

October 25, 2006; 10.00am-11.30am

Most reverse engineering processes follow a common pattern: A program is analyzed through static or dynamic analysis and the collected low level program information is abstracted into some higher level representation. This representation is usually deemed to be more usable for some specific task than source code as it captures certain program features in a form which is congruent with the needs of the task.

Most of the time, the process of abstraction is only semi-automated and involves human analysts. This is to be expected as not all the information contained in abstract representations can be readily derived from the source code-- otherwise the pertinence of the reverse engineering technology being developed might be put into question. Though the abstraction process usually requires the analyst to make certain decisions, this process should be minimized and facilitated. This is where the core problem of most reverse engineering solutions lies.

Given that most reverse engineering processes are going to be human-intensive, a number of issues clearly arise. How do we know that the abstraction process is correct and efficient? How can we compare two or more alternative solutions for a given reverse engineering problem? How can we determine to which extent empirical results addressing those questions are valid and generalize?

There is no simple answer to those questions, but it is important that the reverse engineering research community strive to provide guidelines and develop standard empirical procedures and benchmarks. This will necessarily be a lengthy, stepwise process but it is crucial to facilitate the efficient development of a body of knowledge based on the careful assessment and comparison of research works.

Last updated on October 18, 2006
Sponsors Research Centre on Software Technology University of Sannio IEEE Computer Society TCSE The Reengineering Forum