LINKS

ORGANIZERS

Position Papers

These position papers have been accepted and will be presented in the first part of the workshop.

Objectives

The aim of this half-day workshop is to bring together researchers within the field of clone detection to critically assess the current state of research, and to establish new directions and partnerships for research.  A software clone is generally defined to be a copy or near-copy of a portion of code appearing elsewhere in a system.  Clones are frequently created when programmers "scavenge'' code, i.e., reuse code by copying and pasting a code fragment, often with modification.  The problem of detecting clones in systems is an established software engineering problem known to occur in many contexts, including during pattern detection, software refactoring and perfective maintenance, system quality evaluation, code compaction, and class library and web reengineering.   Various techniques have been proposed for automatically and semi-automatically detecting clones and refactoring them.  Work comparing and evaluating such tools is ongoing.

This workshop expands upon the First International Workshop on Detection of Software Clones, held in conjunction with ICSM'2002 and SCAM'2002 in Montreal in October of 2002.  The primary aims of this workshop are to:

• bring together researchers within the field
• clarify and assess the current state of research
• establish a list of new directions and open and critical research questions, and
• generate new research collaboration partnerships.

The scope of the workshop is the general field of clone detection techniques, theories, and applications.  Relevant topics include but are not limited to:

• software similarity models
• taxonomies of clone, redundancy, or duplication types
• clone detection techniques and methods
• tool and technique evaluation or comparison
• empirical studies of clones in systems or of clone detection
• benchmarking and benchmarking issues, including data format issues, benchmark types, subject system selection, etc.

 Workshop Format

The workshop will be a half-day workshop, broken into three sections roughly as follows:

• 60 min: position paper presentations and discussion
• 90 min: brainstorming session
• 30 min: summary and wrap-up session

The brainstorming technique used will be essentially the same as the one used successfully in a recent Dagstuhl workshop on architecture reconstruction.  First, index cards are distributed to the participants, who are asked to fill as many cards as they desire.  Participants write an issue or question on the front of each card, and indicate on the back whether the question is "solved''.  Cards are then collected and collaboratively clustered, revealing points of consensus, disagreements, and directions for future research.  These clusters are used to seed another round of discussion, which may lead to refinements in the clustering.

The summarization and wrap-up session will allow participants to raise outstanding questions and collect together any new insights that emerge from the prior two sessions.  The aim is to provide the workshop participants with a summary overview of the important emergent issues in the field, and to identify potential future research collaborators.

Contact

________________________________________________________________________________