Lars Bendix

The Unified Extensional Versioning Model

Versioning of components in a system is a well-researched field where various adequate techniques have already been established. In this paper, we look at how versioning can be extended to cover also the structural aspects of a system. There exist two basic techniques for versioning - intentional and extensional - and we propose a unified extensional versioning model for versioning of both components and structure in the same way. The unified model is described in detail and three different policies that can be implemented on top of the general model are exemplified/illustrated by three prototype tools constructed by the authors. The model is analysed with respect to the number of versions and configurations it generates and has to manage. Finally, the unified extensional model is compared to more traditional intentional models on some important parameters. The conclusions are that the unified model is indeed viable. It not only provides the functionality offered by the intentional model with respect to flexibility during development and management of combinatoric complexity, but also offers a framework for management of configurations that enables systems to provide much more advanced support than is commonly available.

A study of configuration management in open source software projects

Projects where developers are geographically distributed and with high personnel turnover are usually considered to be hard to manage. Any organisation that successfully handles such projects merits closer analysis so that lessons can be learned and good practice disseminated. Open source software projects represent such a case. One important factor is good configuration management practices. In this paper, the authors examine the configuration management process for some open source software projects and analyse how process, tool support, and people aspects of configuration management contribute to this success. Finally, we discuss best practices and how lessons learned from open source software can be transferred to more traditional ways of developing software.

CoEd - A Tool for Versioning of Hierarchical Documents

In this paper, we report on our experiences from developing and using a tool to provide optimal version control support during co-operative development of hierarchical documents. The main problem in co-operative work is to maintain an overview of how the document is evolving, and at the same time to maintain a complete version history. We extend traditional version control in the sense that we provide version control on both the entire structures of the document and its constituent parts. This enables us to establish natural baselines of the document that have actually existed, and furthermore our system is capable of manipulating directly the structure of the document while retaining complete version history. CoEd consists of four browsers, which show the configuration, version, hierarchy and text aspects of the document. We describe the analysis and considerations that led us to the actual design and implementation of the two prototypes of CoEd. Furthermore, we detail the experiences we have gained from actually using these prototypes, compare our work with that of others, and generalise our results to the field of software development.

VTML for Fine-Grained Change Tracking in Editing Structured Documents

The task of creating documents collaboratively is complex and it requires sophisticated tools. Structured documents provide a semi-organised writing environment where collaboration may assume more controlled forms than with other document types. CoEd is a writing environment that provides integrated structure support, content overview and version management for complex and hierarchical documents (e.g. technical documentation). The present implementation has, however, limitations in the efficient management of disk usage and in providing more sophisticated collaboration functionality. This led us to consider the VTML change tracking language as a backend for improving the performance and feature set of CoEd. This paper explores the advantages of using a sophisticated change-tracking language in a versioning system for collaborative writing.

Towards a suite of software configuration management metrics

Software Configuration Management (SCM) is an important support activity in software development. However, its transparent nature as a service that makes life easier for others and as an insurance against disasters, often makes it difficult to justify investments in tools and processes that apparently do not have any direct return. We have made a first step towards establishing a model for showing the return on investment in SCM, making the costs and benefits explicit. In this paper, we also sketch how we plan to take the next important step and establish a set of metrics that can be used to manage and tune the SCM processes and tools.

Fully Supported Recursive Workspaces

Version control deals with support for making changes. The way traditional systems handle this problem is centered around repositories of immutable versions. However, little or no support is given for controlling change outside of said repositories. We develop a conceptual model which places workspaces on an equal footing of repositories and thus we extend traditional version control support to apply to workspaces too. We show how this model was implemented and report on some of the experiences from its actual use.