現今軟體開發專案普遍有時程延宕、預算及人力不足、軟體品質低落、無法應付客戶頻繁的需求變動等問題。為思考相關解決之道，本研究參考極致編程(Extreme Programming，簡稱XP)軟體開發方法，於校園內選擇特定專案，配合XP的四個核心價值觀與十二項實行原則實際進行軟體開發。於專案導入XP後，透過對開發過程的觀察、記錄、資訊蒐集，及針對開發團隊所作的深入訪談，本研究綜合分析與評估XP於校園內軟體開發專案之適用性，並發掘可能的問題。 　　研究中發現XP在校園內不易達到完美實踐之目標，適用性不高，主要受限於時空環境難以配合、團隊專案經驗不足、缺乏激勵誘因、缺乏團隊約束力等因素；不過施行XP也改善了團隊內外部的溝通與協調機制，促進團隊成員之間的知識與經驗分享，亦有益於系統架構之完整性。 　　本研究排除了完全無法實施的要項之後，依團隊實際情況提出若干改善及彈性應變之建議做法，以期將XP進一步應用於業界個案及後續學術研究。

Agile methods change the software processes. Agile processes such as Scrum, ExtremeProgramming (XP), Open Unified Process (OpenUP) etc. have techniques that improve softwarequality. No doubt that the purpose of these techniques is to inject quality assurance into theproject under development. This thesis presents quality assurance techniques in Open UnifiedProcess (OpenUP) along with comparative study to extreme programming (XP) for agilesoftware development. OpenUP is an agile and unified process that contains the minimal set ofpractices that help teams to be more effective in developing software. It assists to achieve qualityby an iterative and incremental approach with artifacts, checklists, guidelines, disciplines androles. On the other side XP emphasizes on values such as communication, feedback, respect,and courage. In addition, XP prescribes a collection of techniques, which aim to improvesoftware quality. Both these processes have the same purpose, to develop software that meets the stakeholder’sneeds and expectations, however they uses different approaches to achieve their goals. Thisthesis compares both processes in four different points of view, by comparing their qualitytechniques, focus in time, and cost of usage and social perspective. We have proposed an extrarole of the quality coordinator (QC) in OpenUP/XP. QC can support and coordinate project inall quality assurance activities. The objective of an extra role is to use the knowledge of QC toachieve highest possible product quality in software development process.Keywords: Agile Development, Quality assurance (QA), Open unified process (OpenUP),extreme programming (XP), Quality coordinator (QC)

Detta arbete är en reflektion över ett programvaruutvecklingsprojekt, WAIS, och den arbetsmetodik som användes för detta projekt. Arbetsmetodiken för utvecklings-projektet var Extreme Programming (XP) och vi ifrågasätter om detta val var lämpligt för ett projekt som WAIS. Projektet genomfördes vid Blekinge Tekniska Högskola under våren 2004. Utvecklingsprojektet WAIS var en del av ett större projekt, AIS 42, som stöds av VINNOVA. Huvudprojektets bakgrund och syfte beskrivs för att läsaren skall förstå den roll projektet hade. Studenternas uppgift har varit att utveckla en prototyp för ett system som skall sprida information relaterad till logistik och transporter. Slutanvändarna till systemet är många och de tillhör olika organisationer som är spridda både fysiskt och verksamhetsmässigt. Arbetsmetodiken XP faller in under paraplybegreppet Agile manifesto vilkas principer beskrivs i detta arbete. Paralleller dras till arbetsprocessen och dess resultat som projektgruppen har utfört. Vi reflekterar över projektarbetet och arbetsmetodiken utifrån det perspektiv som växt fram under våra år på MDA-programmet. Synsättet bygger bland annat på particpatory design och människa-datorinteraktion där användarmedverkan har en central roll vid utvecklingsprocesser. Vår slutsats visar att utvecklingsprojektet gått miste om flera av de fördelar som arbetsmetodiken XP kan ge. De orsaker som identifierades var: tidsaspekter, tillgång till aktörer och avsaknaden av en tydlig gemensam målbild. / This thesis is a reflection over a software engineering project, WAIS, and the work methodology that has been used within the project. Extreme Programming (XP) was the work methodology used during the process of development. We question if the choice of work methodology is suitable for this kind of project. The project was running by student during spring 2004 at Blekinge Institute of technology (BTH). The project WAIS was a part of a larger project named AIS 42 that runs with funds from VINNOVA. The main projects background and purpose are described in thesis so that the reader can understand the students’ part in the projects as a whole. The students’ main task was to develop a prototype for a system that disseminates information about logistics and transports. The system has many different end users and they come from various organizations that are spread out both physical and organizational. XP is treated as a methodology with in Agile manifesto, which principles are being explained in the thesis. Parallels are being drawn between the work process and the result from the work done. We reflect over the process and the work methodology from a perspective that has been brought forth from our years of education at BTH. The approach is built up upon theories from participatory design and human computer interaction where the user has an essential part of the process of development. The conclusion shows that the project has missed some of the advantages that XP brings. The reasons that we identified were; timeaspects, access to users and the lack of a joint objectiv.