CSEE&T 2005 Program

Conference at a Glance — Click on a Session For Details

WiFi access will be available for participants in all meeting rooms at no extra charge throughout the conference.

Session by Session Schedule

Please note that the following is tentative. Details may change for various reasons, including speaker conflicts. Click on highlighted sessions for more details. There will also be several other co-located meetings, such as the Working Group on Software Engineering Education.

    Sunday, April 17 2005

    Monday, April 18 2005

    Tuesday, April 19 2005

    Wednesday, April 20 2005

Keynote Speakers

The following people will give keynote addresses:

• Debra J. Richardson, Dean of Information and Computer Sciences, University of California, Irvine

  Informatics: Contextualizing Computer Science and Software Engineering Education

  Keynote address, Monday April 18th, 9:15 (Click to see it in the schedule)

  Abstract: Forthcoming.

  

  Bio: Debra J. Richardson is the Ted and Janice Smith Family Foundation Dean and Professor of Donald Bren School of Information and Computer Sciences at the University of California at Irvine. She received her PhD in Computer and Information Science at the University of Massachusetts in 1981. She joined the UCI faculty in 1987.

  Dr. Richardson pioneered research in 'specification-based testing', whereby formal specifications and methods are employed to guide and evaluate software testing and analysis. She has been investigating software testing for over 15 years. Her current work focuses on enabling specification-based testing technology throughout the software lifecycle, from requirements and architecture analysis through operation and evolution. She has developed leading edge tools, and has worked with several companies in adopting technology to improve the quality of critical software systems.

  Dr. Richardson has worked with several companies in adopting technology for improving the quality of their software products and processes. She is currently director of MICRO (Microelectronics Innovation and Computer Research Opportunities), the first industry-university cooperative research program in the University of California.

  Click here to see Debra Richardson's personal web page

   
  

• Laurie Williams, Assistant Professor at North Carolina State University

  Debunking the Geek Stereotype with Software Engineering Education

  Keynote address, Tuesday April 19th, 9:05 (click to see it in the schedule)

  Abstract: Alas, the stereotypical software engineer is depicted as spending long hours working alone in a cubicle filled with empty pizza boxes and soda cans. This "work alone" stereotype can dissuade talented individuals from considering a career in the information technology industry. As educators, we often reinforce this "geek" stereotype early in the curriculum by giving students lengthy assignments and forcing them to work alone — collaborating is cheating! However in industry, software engineers actually spend a large part of their day collaborating with teammates — the "work alone" stereotype is largely unfounded. Research results indicate that through providing students with more collaborative experiences, we could retain more students without compromising their individual learning and these students would be better prepared to be collaborative team members.

  

  Bio: Laurie Williams is an Assistant Professor at North Carolina State University. She received her undergraduate degree in Industrial Engineering from Lehigh University. She also received an MBA from Duke University and a Ph.D. in Computer Science from the University of Utah. Prior to returning to academia to obtain her Ph.D., she worked in industry, for IBM, for nine years. Dr. Williams research interests include empirical studies of agile software development including the pair programming and test-driven development practices, software reliability, software testing, and software security.

  Click here to see Laurie Williams' personal web page

   
  

• Bran Selic, IBM Rational Software and head of the UML 2.0 committee.

  What I Wish I Had Learned in School: Reflections on 30+ Years as a Software Developer

  Keynote address, Wednesday April 20th, 9:10 (click to see it in the schedule)

  Abstract: Computer science and software engineering curricula are, by now, long established in most institutions of higher learning. They have evolved from a few utilitarian courses that taught essential programming skills to fully-fledged academic four- and even five-year programs to match the growing body of experience and knowledge in the field. Yet, it seems that some quite fundamental gaps persist in the education of software experts rendering many of them inadequately prepared for industrial software development. These problems, which are both technical and cultural, can be addressed only if there is a clearer understanding of the nature of software technology. Knowing how software differs from traditional technologies will indicate where innovative approaches to teaching are necessary, whereas knowing how it is similar to those technologies will help us understand where classical time-proven methods can be applied. This talk identifies some of the key problem areas in current curricula and describes suggestions for dealing with them - from the perspective of a long-term practitioner of industrial software development.

  

  Bio: Bran Selic is an IBM Distinguished Engineer at IBM Rational and an adjunct professor at Carleton University in Ottawa, Canada. He has over 30 years of experience in designing and implementing large-scale industrial software systems. Bran pioneered the application of model-driven development methods in real-time applications. He is chair of the OMG team responsible for the UML 2.0 standard.

Workshop Details

W1: Best Practices for Teaching Object-oriented Framework Development

Object-oriented framework development is one of the most challenging software development tasks. Teaching framework development is even more challenging since there is no unified framework development methodology.

This workshop aims to share the knowledge and experience of different organizations and individuals in both practical and theoretical aspects of teaching framework development. The main goal is to identify, discuss and promote best practices to properly engineer and teach object-oriented frameworks.

The following are some of the possible topics:

• Methodologies for framework development
• Creative approaches for teaching framework development
• Approaches to evaluate framework development methodologies
• Teaching framework development using design patterns
• Tools for teaching framework development

W2: Software Engineering Education (SEE) Research and Publication: Issues, Challenges and Directions

We will review the researchers guide prepared for this conference. Particpants will be asked to suggest improvements to it and to comment on how it has affected the quality and rigor of the research presented at the conference.

The objective is to improve the guide in preparation for next year's conference.

W3: Infusing Software Architecture into Software Engineering Curricula

Call for papers: Please go to http://www.unf.edu/~asanchez/cseet05-w3/ for the call for position papers. Those interested in participating are asked to send a position paper by April 13th.

Abstract: Software systems of significant size, complexity, and lifetime are critically dependent on the match between their architecture and the quality attributes of greatest importance to the system stakeholders. Teaching software architecture, however, is a challenging task in an academic setting: bringing architectural issues to the fore may require problems of a size that conflict with the limitations imposed by quarters or semesters.

This workshop will bring together faculty who are infusing architectural concepts into computing courses for an exchange of ideas as to successful and promising approaches. The scope will span the range from those with a single undergraduate software engineering course to dedicated software engineering programs. To help ensure a successful workshop, participants will be asked to submit a short, one-page position paper outlining their involvement in software architecture and their experiences in teaching this material. Topics addressed in these position papers include (but are not restricted to):

• Weaving architectural concepts throughout a curriculum.

• Incorporating software architecture in a one or two course sequence on software engineering in computer science or engineering.

• Software architecture approaches appropriate for undergraduate and graduate degree programs in software engineering.

• Classroom or laboratory exercises to improve student comprehension of architectural quality attributes.

• Classroom or laboratory exercises to improve student ability to analyze architecture fitness for use with respect to the quality attributes required.

• Activities that increase student ability to synthesize potential solutions and to assess the explicit and implicit tradeoffs embodied in such solutions.

We plan to invite a member of the SEI's Product Line Practice Initiative to participate in the workshop as a way to facilitate interaction in this area between the SEI and the academic community.

W4: Using Active Learning Strategies and Web Support to Engage Engineering Students in Large Classes

Sponsored by the Centre for University Teaching, University of Ottawa

Outline As class sizes continue to grow, the goal of keeping the students engaged and taking ownership of their learning can be very challenging. Integrating active learning strategies and web support into instructional design and course management can go a long way towards meeting that challenge.

In this interactive workshop, the presenter will first share her experiences from teaching a large class of engineering students, where she has been implementing the active collaborative learning model championed by a well-known engineering educator, Richard Felder. Successful strategies used in the course to engage students will be identified, with several activities modeled with the session participants. The participants will have an opportunity to 'tour' the BlackBoard website supporting Dr. Zywno's current course in Process Control, and see how multimedia can be used to support learning. Video clips of group activities and short videotaped interviews will bring to the audience the students' point of view on educational technology and on active, collaborative learning. The participants will next have an opportunity to brainstorm strategies to meet the needs of different types of learners in an engineering classroom, discuss perceived barriers to a wider adoption of active, collaborative learning strategies, and how to overcome them.

Bio Dr. Malgorzata (Gosha) Zywno is a Professor of Electrical and Computer Engineering at Ryerson University in Toronto, with a cross appointment in the Learning and Teaching Office. Dr. Zywno has taught at Ryerson since 1982.

Dr. Zywno is passionate about teaching, and her focus on students and their learning is complemented by her research interests, which include active learning in a technology-rich environment, faculty development, and recruitment and retention strategies for women in engineering. She has taught courses and conducted workshops at universities in Canada, France, Germany and Scotland. Dr. Zywno is the recipient of several teaching excellence awards, including 3M Teaching Fellowship (2002), International Network for Engineering Education and Research Achievement Award (2002), and Technology Innovation Award (2004). Dr. Zywno's research and publications on the issues of technology-mediated teaching, active learning and learning styles have also won several awards, including the Best Conference Paper (2002) from the American Society for Engineering Education.

Tutorial Details

T1: Integrating Software Engineering Process in an Undergraduate Curriculum

Summary: This tutorial is intended to assist faculty members and administrators who are designing or modifying undergraduate software engineering curricula, and who wish to learn about alternative approaches to incorporating software engineering process. Curricular recommendations developed by the Joint Task Force on Computing Curricula, formed by the IEEE Computer Society and the Association for Computing Machinery, provide a context for this discussion. The applicability of the Personal Software Process (PSP) and the Team Software Process (TSP), developed by the Software Engineering Institute, is also considered. Experienced software engineering educators may benefit from the presentations, hands-on exercises, and group discussions, but the tutorial is designed primarily for educators with less experience in software engineering curriculum development.

Introduction: Software engineering process is an important component of the professional practice of software engineering (SE), and thus is also a critical element of any educational program intended to prepare students for work in the discipline. Recognizing this need, software engineering educators have reported on a variety of approaches to incorporating SE process elements into academic programs.

With the growth of academic software engineering programs, and the accreditation of the first undergraduate software engineering programs in the United States, many more educators are now grappling with identifying desired student outcomes, designing or modifying curricula, and assessing the results. Fortunately, the recent work of the Joint Task Force on Computing Curricula, sponsored by the IEEE Computer Society and the Association for Computing Machinery, has produced a document that outlines a body of software engineering knowledge (known as SEEK) and provides guidance on defining and implementing curricula that can be used to teach fundamental SE knowledge and skills.

Another significant resource for education in software process is the work done at the Software Engineering Institute. The current version of the Capability Maturity Model (known as the CMMI) provides an overall framework for improving software processes, while the Personal Software Process (PSP) and the Team Software Process (TSP) provide specific techniques for individual software engineers and teams to ensure product quality and improve process management. A number of reports describe efforts to adopt or adapt the PSP and TSP in software engineering education.

The purpose of this tutorial, then, is to assist faculty members and other interested software engineering professionals in integrating software process into a new or existing curriculum. While seasoned software engineering educators may benefit from the presentations and discussions, the tutorial is targeted primarily to an audience with less comprehensive knowledge and experience in defining, implementing, and assessing a software engineering curriculum.

This three-hour tutorial will include presentations and hands-on group exercises. The following is the proposed outline:

• Introduction
• SE2004 curriculum guidelines
  • Overview: SEEK (Software Engineering Education Knowledge); Guidelines for curriculum design and delivery; Courses and course sequences; Curriculum patterns.
  • SEEK knowledge areas related to software process
  • SE2004 courses related to software process
  • Group exercise and discussion
• PSP and TSP
  • Planning
  • Quality
  • Team work
  • Process improvement
• Approaches to process integration
  • Dedicated process courses
  • Process content distributed across courses
  • Large-scale lab course sequences
  • Capstone project courses
  • Case studies
  • Group exercise and discussion
• Implementation issues
  • Process elements and student maturity level
  • Faculty resources and preparation
  • Process support infrastructure
  • Consistency across courses and instructors

About the Presenters:

Mark J. Sebern is a Professor in the Electrical Engineering and Computer Science Department at the Milwaukee School of Engineering (MSOE), and program director for MSOE¹s undergraduate software engineering program, one of the first four SE programs to be accredited in the United States. He is also a visiting scientist in the Software Engineering Process Management group at the Software Engineering Institute and an ABET software engineering program evaluator.

Thomas B. Hilburn is a Professor of Software Engineering at Embry-Riddle Aeronautical University. He is an IEEE Certified Software Developer and an editor for the ACM/IEEE-CS Computing Curriculum-Software Engineering project. Tom has been active in efforts to integrate software engineering into academic computing programs. He is also a visiting scientist at the Software Engineering Institute, where he works in developing activities and materials for promoting the use of individual and team software processes.

T2: Problem-based Design Studios for Undergraduate SE Education

Summary: Both the increasing knowledge needed to practise as a professional, and the accelerating rate of change within the discipline suggest that traditional learning models may not address the requirements of learners. Problem-based learning (PBL) and design studios (DS) are two approaches that focus on learners developing characteristics of lifelong learning. This tutorial explores a Problem-based Design Studio (PbDS) model of learning. The goal is to enable participants to gain some understanding of the model so as to evaluate its' applicability in their teaching/learning context.

Introduction: In traditional models applied to professional education students first study basic science, then the relevant applied science, so that learning may be viewed as a progression to expertise through task analysis, strategy selection, try-out and repetition. The formal roles of lectures, tutorials and laboratory classes are intended for knowledge transfer using essentially uni-directional modes of teaching. In this model, the purpose of the practical work students are presented with is to apply knowledge learned earlier in the curriculum to real-life problems: the students deal with know-how problems that can be solved by knowledge acquired in their lectures. Students become experienced in the use of disciplines and theories considered necessary/relevant through the practical work that supports this knowledge.

However, with the on-going increase in knowledge in the discipline, and the accelerating rate at which this increase is occurring, students cannot learn all the material required to practice as professionals in their disciplines.

Problem-based Learning (PBL) and Design Studios (DS) are two approaches that focus on centering the learning environment on the student. Student responsibility and independence help to develop characteristics of lifelong learners - motivation, self-evaluation, time management and the skills to access information. Together PBL and DS provide mechanisms and processes for the teacher to build a learning environment that encourages a community of learners to interact to define and solve problems, and to garner skills that enable them to become self-directed learners.

Characteristics of PBL: In contrast to the stronger emphasis on teacher-direction and the coverage of academic content found in most traditional models, Problem-based Learning incorporates many of the practices that are now considered the desiderata of good teaching: it is student-centered, fosters intrinsic motivation, promotes both deep and active learning, taps into students¹ existing knowledge, encourages reflection on the teaching/learning process, develops collegial learning skills, and can support student self-assessment and peer-assessment. In a PBL environment students are self-directed, independent and interdependent learners motivated to solve a problem. Evaluation of this learning model confirms that having authentic (ie feasible in the real-world practice of the discipline) problems assists students in understanding and later deploying their new knowledge. Skills acquired in this way are transferable to professional practice. When undertaken in the group environment advocated, students also develop generic interpersonal skills to draw on after their formal education is completed.

Design Studios Studio-styled learning models have evolved from the master-student relationship of the classical apprenticeship. They are commonly used in a wide variety of professional education disciplines including:

• architecture (and other creative arts) ­ where groups of students work with the architect ³master² across all aspects of a design task. This style of education is widely used and well recognised in architectural education as being highly desirable, if not essential, as a means of achieving the essential learning outcomes for the graduate

• clinical professions, including medicine, dentistry and veterinary science ­ where exposure to the processes of observation, diagnosis and treatment in a face-to-face environment is taken as being an essential part of the education process. We can also include a number of the para-clinical professional in this category, for example psychology, chiropractic science and physiotherapy

• the profession of teaching itself is generally recognised as requiring exposure to real-life work situations as a part of the training process.

The engineers from Aalborg

• were assessed to be stronger in problem-solving, communications, co operation and general technical knowledge

• perceived a convincing agreement between the composition of the knowledge and experience used in the project-oriented education and in the professional engineering practice

• after three years of employment still derived their applied professional knowledge from their project work rather than from taught courses, colleagues or postgraduate courses.

Objectives include:

• exploring issues around PbDS in some detail, both from
• a student perspective and
• implications for the teacher
• designing a small problem on the SE discipline
• evaluating issues concerned with implementing PbDS.

The tutorial itself will model a PbDS in action. The implication of this is that, while some presentations are included, they will usually take the form of summarising the discussion undertaken. The bulk of the tutorial will be based around activities and resource exploration in the context of a PBL process. Resources and a workbook will be provided to participants.

Jocelyn Armarego: Senior Lecturer - Software Engineering:

I have been involved in the application of Design Studio learning and PBL in Software Engineering at Murdoch University. I initially applied the Studio model to Software Design courses. However, evaluation of these highlighted the need for a process to anchor Studio learning, and led to the integration of PBL with the Studio (PbDS). This model was first applied to Requirements Engineering learning. Subsequently I have been involved in moving all courses in the Software Engineering program to PBLbased Design Studios. I have been actively involved in the developing a programme of staff induction in PBL and Studio learning. As of 2005 all Engineering programs will apply PbDS at years 3 and 4.

Sally Clarke: Senior Lecturer - Medical Education:

I have worked in the area of teaching and learning in higher education in several Australian Universities. As Evaluation Officer in the Graduate School of Medicine at the University of Queensland, I conducted a number of evaluations of the medical curricula before and after the change to Problem­Based Learning (PBL). Following that I have been involved in introducing PBL in Information Technology. At Queensland University of Technology I was an active member of the team from the Faculty of Information Technology implementing PBL in intermediate level programming. At Murdoch University, I worked with Jocelyn Armarego introducing PBL in software engineering. I also facilitated a workshop for Engineering staff at Murdoch on Design Studios and PBL in August 2004.

Panel Session Details

P1: Professional Engineering and Software Engineering

Abstract: The purpose of this panel is to inform attendees about the current state of the accreditation of software engineering programs and licensing for software engineers in Canada and the US. We also intend to stimulate discussion of the more controversial aspects of this topic.

Panel outline: The motivation for this panel comes from the following observations:

• Most people who graduate from either computer science or software engineering programs, or even computer engineering programs, develop software in their careers: i.e. they do the same general task.

• The public is now exposed to the term 'software engineer', meaning generically anyone who develops software, in news reports and films.

• The term 'Computer Scientist' has less and less meaning for the public; and in fact, very few graduates of computer science programs actually do anything that involves the 'scientific method'; they mostly do design, just like engineers.

• Accreditation agencies (following different models in Canada and the US) accredit most software engineering and computer science programs.

• In Canada, most universities that offer computer science or computer engineering degrees are now offering software engineering degrees.

• In the professional engineering community it is a given that those who do any form of engineering need to graduate from an accredited engineering program (or prove equivalent competence through exams). It is also considered by many in this community either desirable or essential that they be licensed to protect the public. On the other hand, most people developing software are not licensed and do not, in North America, come from accredited engineering programs.

• Many people argue that licensing of everyone developing software, who the public is generically coming to call 'software engineers' is not necessary, or in fact would be infeasible or harmful. Some argue that even accreditation is not necessary. The ACM famously pulled out of the development of SWEBOK over a concern that licensing is a bad idea.

• Some say that software engineering is just another branch of engineering, whereas others say that it is sufficiently distinct from all the other defined branches of engineering that it should be treated distinctly.

• Provincial and territorial Engineering Acts in Canada provide exclusive scope of practice over engineering practice and right-to-title legislation for Engineers. Therefore, by law those calling themselves 'engineers' and those taking responsibility for engineering work, must be licensed.

• Given the relative immaturity of the field, it is argued by some that moves toward licensing will limit the development of innovative practices -- since engineers are expected to follow established practice.

• Some say that only those engineers working on safety-critical systems, or those selling services directly to the public, should be licensed. But if the education of all software engineers should be broadly similar, should not all at least be 'licensable', regardless of whether their degree is called 'Computer Science' or 'Software Engineering'?

• In the British model, both computer science and software engineering graduates are eligible to be Chartered Engineers.

Panel 2: Software Assurance Education

Abstract: Software engineering processes for building safe and secure software have existed for a long time. However, these processes‹particularly for secure software engineering‹have not been widely taught within colleges and universities leading to a shortage of graduates skilled in these areas. This panel will discuss the increasing need for colleges and universities to produce graduates that are skilled in building safe and secure software. Panelists will share their experiences teaching courses in these areas and future directions for curricula.

In each of the last two years, the CERT Coordination Center has reported roughly 4000 vulnerabilities, demonstrating that software security is a significant problem. Exploitation of these vulnerabilities continues to grow exponentially, with well over a hundred thousand incidents reported to the CERT CC in 2003. The problems with software are not just in operating systems and web servers; a large portion of the new vulnerabilities now being reported are within applications.

For more than twenty years, much has been known about safe and secure software development that builds on the necessary foundation of well organized, managed and disciplined software engineering processes. Distributed systems and the Internet have complicated systems design, but methods for producing dependable software remain largely unchanged. Understanding how to develop secure software is a necessity since software produced by developers often interacts with users via the Internet.

Among the questions the panel will consider are:

• Is there a difference in the methods used to produce safe or secure systems? How are the methods the same or different? Can they be effectively taught in the same course?

• What types of class projects that will help students understand how to build safe or secure software?

• What are the roles of formal methods in producing secure and safe software?

• Should changes be made in beginning courses to introduce the proper mindset and techniques as well as avoiding students developing ³bad² habits?

• What lessons can be learned from the Safety Argument and documentation processes that can be used in developing secure software?

• Are design vulnerabilities or code vulnerabilities more fundamental?

• What are the roles of threat and hazard analyses in system requirements determination?

• Why are security and safety emergent system properties not simply specialized security functionality?

Related meetings

• Meeting of the Software Engineering Program Leaders Association (SEPLA): This comprises all people who are leaders of Software Engineering academic programs (they may be called 'chairs', 'department heads', 'program heads', or whatever). The meeting is planned for Sunday April 17th. Further details will be posted here later.

  SEPLA has met annually at CSEE&T for several years. The meetings provide an opportunity for participants to learn from each other's experiences and make contacts.

  
  

• Working Group on Software Engineering Education and Training: This group has met for many years at CSEE&T and on other occasions to discuss and advance the field.

Birds-of-a-feather sessions

• Monday, 11-12:30: PSP Certification: The Software Engineering Institute (SEI) is developing a certification program for software developers and practitioners who use the Personal Software Process (PSP). The certification will be based on the PSP Body of Knowledge (BOK) document currently in development. This document will outline the core knowledge and skill set required by software engineering professionals who use the PSP methodology. The PSP BOK is currently available for review and feedback at http://www.sei.cmu.edu/tsp/psp/bok/index.html. A PSP BOK "birds of a feather" discussion session is scheduled for CSSE&T 2005, at which SEI staff will present the document and solicit feedback from the software engineering community.

  The SEI Professional Certificate and Certifications programs are a means for transitioning technology best practices and standards to the software engineering community through products, services, and support.