10th September 2007 in room 78.622 (General Purpose South) at 2pm

Afternoon tea will be served at 1:45.

Title:

Interactive debuggers considered harmful

Speaker:

Andrae Muys

Abstract:

Despite their extensive deployment and inclusion in most if not all IDE's, it has been my experience that interactive debuggers are invariably a hinderance to debugging. Commercial tool developers addiction to them has almost certainly delayed and stunted the development of useful debugging technologies. In over ten years working on multiple projects ranging from embedded systems to enterprise applications, from low-level systems programming to GUI interface development, I am aware of only a single instance where an interactive debugger was an appropriate tool to assist with debugging.

I propose a short rant, followed by a brief discussion on what _does_ work, what I use, and what I would like to have.

Format

Each session starts with a short talk (~25 min.) by a developer, which describes current or past work, and gives a personal critique of the usefulness of the software engineering process employed.

An open discussion will follow the talk, during which the attendees will be able to ask questions, make suggestions, or debate the merits of various tools and techniques.

Proceeding each talk, coffee and tea (as well as biscuits) will be available to improve attention, and encourage attendance.

Goals

The central goal of the discussion group is to facilitate sharing of software engineering expertise between developers and provide exposure to new methodologies and techniques. Our aim is to increase the level of software engineering expertise among complex systems modelers.

Talk topics may vary widely: the use of a design notation; the effectiveness of a particular tool; a description of current or past programming work; or a personal critique of the usefulness of a software engineering technique used.

We Need You

We are calling for people interested in presenting to the group to come along to the meeting on Monday or send us a short abstract and likely availability.

We also need interested participants to come along and share in the discussion.

If you have any questions or comments please contact:

Daniel Bradley <daniel@itee.uq.edu.au>

or

Leighton Brough <brough@itee.uq.edu.au>