Related Projects

Full Title

Open Source Software


open source, software project characteristics, inter-disciplinary study


Open source is a phrase often used to describe a piece of software that comes complete with the source code and usually can be downloaded for free. Even though this description is pretty much correct, there is more to open source than just that.

In order to understand open source better, we conducted a year-long study as part of DIRC’s “Open Source Software” project activity. This project activity involved three sites (Newcastle, City and Lancaster) and several disciplines (computing science, sociology/ethnomethodology and statistics). The major achievements are the publication of a final report along with several papers, and an International Workshop on Open Source Software Development that we organised in February 2002.

The activities performed at Newcastle included literature surveys on the open source subject in general and on individual projects; interviewing several people involved with open source while playing different roles and/or in different projects; giving presentations to and getting feedback from external audiences; and holding discussions at various DIRC sites and with one of DIRC’s Senior Visiting Fellows, Michael Jackson. There was also participation in the first Workshop on Open Source Software Engineering (as part of ICSE 2001), with a position paper about software architecture issues on open source [7].

Lancaster University had been mainly involved in the investigation of open source groups; in particular, a psychological and sociological study of the Cocoon open source development project was conducted. Cocoon is a Java publishing framework for the creation of web content using the latest technologies such as XML.

City University’s effort was spent in clarifying issues of quantitative description and evaluation of dependability. This included researching examples that would support or serve as counter-examples for claims generally made about open source products which may be intuitively plausible but not generally true; and generating a list of research questions that may be worth investigating and would contribute towards identifying "openness" process factors which affect product dependability. Towards this last goal, a speculative model of the reliability growth of a product was built [2], which takes into account the multiplicity and diversity of usage profiles involved in fault finding by execution of the software.

One of the strands of work in this project activity focused on understanding the actual meaning of the term “open source”. We identified characteristics that are common among open source projects as well as those which vary from one open source project to another. By exposing these characteristics, we created a taxonomy against which people can compare the characteristics of any so-called open source project.

In the end, we found enormous variation among projects claiming to be “open source”. Therefore, it is imperative to ask people using the term which of the open source features they really mean. Actually, many features claimed of open source can potentially be –and are at times– used in “conventional” projects. Consequently, we recommended that DIRC should have a further project activity ( “Collaboration in Dependable Software”), would leave aside the over-used term “open source” and study group problem solving in the specific domain of software design and development.


DIRC's Open Source Software



[1] Cristina Gacek and Budi Arief, The Many Meanings Of Open Source, IEEE Software, Vol. 21, No. 1, pp. 34-40, January/February 2004

[2] Diana Bosio, Bev Littlewood, Lorenzo Strigini and M.J. Newby, Advantages of open source processes for reliability: clarifying the issues, Proceedings of the Open Source Software Development Workshop, University of Newcastle upon Tyne, pp. 30-46, February 2002

[3] Cristina Gacek, Tony Lawrie and Budi Arief, Interdisciplinary Insights on Open Source, Proceedings of the Open Source Software Development Workshop, University of Newcastle upon Tyne, pp. 68-82, February 2002

[4] Adrian Mackenzie, Phillipe Rouchy and Mark Rouncefield, Rebel Code? The open source 'code' of work, Proceedings of the Open Source Software Development Workshop, University of Newcastle upon Tyne, pp. 83-100, February 2002

[5] Tony Lawrie and Cliff Jones, Goal-Diversity in the Design of Dependable Computer-Based Systems, Proceedings of the Open Source Software Development Workshop, University of Newcastle upon Tyne, pp. 130-154, February 2002

[6] Budi Arief, Diana Bosio, Cristina Gacek, and Mark Rouncefield, Dependability Issues in Open Source Software - DIRC Project Activity 5 Final Report, Technical Report CS-TR-760, School of Computing Science, University of Newcastle upon Tyne, February 2002

[7] Budi Arief, Cristina Gacek and Tony Lawrie, Software Architectures and Open Source Software - Where can Research Leverage the Most?, Position Paper included in the Proceeding of 1st Workshop on Open Source Software Engineering, Toronto, Canada, pp. 3-5, 15th May 2001


Budi Arief (Newcastle)


Page Maintainer: Credits      Project Members only Last Modified: 10 August, 2005