XP2003, May 25-29, 2003, Genova, Italy  An Experiment Working with RUP and XP Patricio Letelier, José H. Canós and Emilio A. Sánchez Department of Information Systems and Computation Valencia University of Technology

XP2003, May 25-29, 2003, Genova, Italy  Contents 1. Introduction 2. Project Settings 3. Some Results 4. Conclusions and Future Work

XP2003, May 25-29, 2003, Genova, Italy  1. Introduction An experiment with upper-level students in the subject “Laboratory of Information Systems” at the School of Computer Science Aim: compare RUP and XP regarding how easily they are learned and applied by students

XP2003, May 25-29, 2003, Genova, Italy  2. Project Settings 2 teams using RUP and 4 teams using XP Same case study: Develop the information system of a virtual store Laboratory with 20 PCs 2 hours sessions (x2, a week) 12 hours a week of workload for each student

XP2003, May 25-29, 2003, Genova, Italy  2. … Project Settings Previous knowledge of some tools and notations but they are beginners in process and teamwork experience Basic concepts of RUP and XP were taught by the instructors One instructor acted as client and another acted as coach Each team was composed of: 1 Project Manager, 1 Tester/Tracker, and from 4 to 6 Programmers

XP2003, May 25-29, 2003, Genova, Italy  2. … Project Settings 1st Iteration 2nd Iteration3rd Iteration InceptionElaboration 1st Iteration Construction 2st Iteration Construction XP RUP 1 week2 weeks3 weeks

XP2003, May 25-29, 2003, Genova, Italy  3. Some results XP Simplicity, few artifacts, less traceability and updating efforts First iteration: under estimation of user stories/task and some conflicts among team members Most of XP practices were easy to learn and apply. Only some troubles with “test first”, refactoring, coding standards and metaphor Our typical laboratory facilities were fine for teaching and working on XP

XP2003, May 25-29, 2003, Genova, Italy  3. … Some results RUP Project ran smoothly during the first two phases (inception and elaboration) but problems came after Precise definition of the different artifacts Artifact updating and requirements management need an important effort Customization of RUP to the proyect is not a simple task

XP2003, May 25-29, 2003, Genova, Italy  4. Conclusions and Future Work The success (quality and productivity) of the project can be reached independently of the process Teams using RUP needed more mentoring XP needs and generates stronger relationships among team members Open issues: impact on “software maintenance”, impact if some XP practices were discarded, etc. “Customers on-site” was not a big dificulty Next step: a more controlled (and maybe specific) experiment

XP2003, May 25-29, 2003, Genova, Italy  An Experiment Working with RUP and XP Patricio Letelier, José H. Canós and Emilio A. Sánchez Department of Information Systems and Computation Valencia University of Technology