Conference
Presented by:

Research Papers

Research papers reveal the aspirations and advance the framework of thought in the field. Come listen to the questions and conclusions from the members of our diverse community. Trade notes and discover where the thinkers are heading. Learn from those who are applying complex systems theory to organizations, reflective practitioners proposing new forms of agile development and dissidents who feel that "agile" is hot air.

Scheduled Research Papers

XP and components thereof (RP1)
Paper Speakers Doc
An Initial Exploration of the Relationship Between Pair Programming and Brooks’ Law Laurie Williams, Anuja Shukla, and Annie Antón
Paper - Adobe PDF (397K)
Exploring Extreme Programming in Context: An Industrial Case Study Lucas Layman, Laurie Williams, and Lynn Cunningham
Paper - Adobe PDF (229K)
The XP Customer Role in Practice: Three Studies Angela Martin, Robert Biddle, and James Noble
Paper - Adobe PDF (68K)
Communicating Communities (RP2)
Paper Speakers Doc
Agile Methods for Large Organizations – Building Communities of Practice Tuomo Kähkönen
Paper - Adobe PDF (61K)
Direct Verbal Communication as a Catalyst of Agile Knowledge Sharing Grigori Melnik and Frank Maurer
Paper - Adobe PDF (288K)

Research Paper Abstracts

An Initial Exploration of the Relationship Between Pair Programming and Brooks’ Law
Laurie Williams, North Carolina State University, williams@csc.ncsu.edu
Anuja Shukla, North Carolina State University, anuja@idealake.com
Annie Antón, North Carolina State University, anton@csc.ncsu.edu

Through his law, “adding manpower to a late software project makes it later,” Brooks asserts that the assimilation, training, and intercommunication costs of adding new team members outweigh the associated team productivity gain in the short term. Anecdotes suggest that adding manpower to a late project yields productivity gains to the team more quickly if the team employs the pair programming technique when compared to teams where new team members work alone. We utilize a system dynamics model which demonstrates support of these observations. Parameter values for the model were obtained via a small-scale, non-probabilistic, convenience survey. Our initial findings suggest that managers should incorporate the pair programming practice, particularly when growing their team.

Exploring Extreme Programming in Context: An Industrial Case Study
Lucas Layman, North Carolina State University, lmlayma2@unity.ncsu.edu
Laurie Williams, North Carolina State University, williams@csc.ncsu.edu
Lynn Cunningham, Clarke Collete, lynn.cunningham@clarke.edu

A longitudinal study evaluating the effects of adopting the Extreme Programming (XP) methodology was performed at Sabre Airline Solutions™. The Sabre team was a characteristically agile team in that they had no need to scale or re-scope XP for their project characteristics. The case study compares two releases of the same product. One release was completed just prior to the team’s adoption of the XP methodology and the other after approximately two years of XP use. Comparisons of the new release project results to the old release project results show an increase in design complexity, a 50% increase in productivity, a 65% improvement in pre-release quality, and a 30% improvement in post-release quality. These findings suggest that, over time, adopting the XP process can result in increased productivity and quality.

The XP Customer Role in Practice: Three Studies
Angela Martin, Victoria University of Wellington, angela@mcs.vuw.ac.nz
Robert Biddle, Carleton University, robert_biddle@carleton.ca
James Noble, Victoria University of Wellington, kjx@mcs.vuw.ac.nz

The Customer is the only non-developer role in eXtreme Programming (XP). The Customer's explicit responsibilities are to drive the project, providing project requirements (user stories) and quality control (acceptance testing): unfortunately the customer must also shoulder a number of implicit responsibilities including liason with external project stakeholders, especially project funders, clients, and end users, while maintaining the trust of both the development team and the wider business. In this paper, we report on a series of case studies of the Customer role in XP projects. We have found that Customers have a pressured and stressful role, leading to issues of sustainability.

Agile Methods for Large Organizations – Building Communities of Practice
Tuomo Kähkönen, Nokia, tuomo.kahkonen@nokia.com

Agile development practices respect tacit knowledge, make communication more effective, and thus foster knowledge creation process. However the current agile methods, like XP, are focused on practices that individual teams or projects need and use of the methods in organizations consisting of multiple co-operating teams is difficult. Community of Practice theory suggests that large agile organization should have various overlapping informal cross-team communities. This paper studies three agile methods, developed in Nokia, that use facilitated workshops to solve multi-team issues. Using Communities of Practices theory paper explains why these methods work in multi-team settings. The results of this paper suggest that workshop practices that gather people from different parts of organizations together to perform a specific well-defined task can be used effectively to solve issues that span over multiple teams and to build up Communities of Practice. This result suggests that the Community of Practice concept could provide basis for adapting agile methods for the needs of large organizations.

Direct Verbal Communication as a Catalyst of Agile Knowledge Sharing
Grigori Melnik, University of Calgary, melnik@cpsc.ucalgary.ca
Frank Maurer, University of Calgary, maurer@cpsc.ucalgary.ca

This paper discusses the role of conversation and social interactions as the key element of effective knowledge sharing in an agile process. It also presents the observations made during a repeated experiment on knowledge sharing conducted in various groups of professionals and students. The study suggests that the focus on the pure codified approach is the critical reason of Tayloristic team failure to effectively share knowledge among all stakeholders of a software project. Drawing on the knowledge-as-relationship perspective of knowledge sharing we theorize that verbal face-to-face interaction facilitates achieving higher velocity accomplishments by software development teams.

Selection Committee

  • Pekka Abrahamsson, VTT Technical Research Centre of Finland
  • Scott W. Ambler, Ronin International, Inc.
  • Ann Anderson, Oak Tree
  • Arlen Bankston, C.C. Pace Systems
  • Mike Beedle, e-Architects Inc.
  • Pascal Van Cauwenberghe, Nayima
  • Mike Cohn, Mountain Goat Software
  • Jens Coldewey, Coldewey Consulting
  • Lisa Crispin, Fast401k
  • Jutta Eckstein, Objects in Action
  • Klaus M. Hansen, University of Aarhus
  • Jim Highsmith, Cutter Consulting
  • Kenji Hiranabe, Eiwa Systems Management
  • Michael Hirsch, Zühlke Engineering AG
  • Ron Jeffries, Consultant
  • David Kane, SRA International
  • Mary Lynn Manns, University of North Carolina at Asheville
  • Michele Marchesi, University of Cagliari
  • Klaus Marquardt, Dräger Medical AG
  • Frank Mauer, University of Calgary
  • Pete McBreen, McBreen Consulting
  • Rick Mugridge, University of Auckland
  • James Newkirk, Microsoft
  • Jeff Patton, Tomax Corporation
  • Gary F. Pollice, Worcester Polytechnic Institute
  • Linda Rising, Consultant
  • Mary Beth Rosson, Penn State University
  • Ken Schwaber, Consultant
  • Helen Sharp, The Open University
  • David Stotts, University of North Carolina
  • Don Wells, Extremeprogramming.org
  • Laurie Williams, North Carolina State University