Dependability in Open Source Development

Open Source development involves making the source code of a system publicly available. This means that many people can propose changes and improvements to the software. Analyze the dependability issues surrounding the process of Open Source development.

Dependability Issues Surrounding the Process of Open Source Development
 

‘Open Source’ is a term used to describe software development projects (Arief et al. 2002). Projects that are significantly different and possess different characteristics are called open source projects (Lawrie et al. 2002). Some examples of projects that are open source are operating systems, web and mail servers, and developmental tools. These examples point towards the formation of a community that can create software that is claimed to be very dependable (Lawrie and Gacek 2002). 

Because Open Source development involves sharing the source code of a system, there have been issues regarding its dependability. Dependability is a relatively broad term which includes security, reliability, availability, and safety (Randell 2000). There have been several arguments about the dependability of Open Source Software development. Many suggest that Open Source is more protected because it provides its source code to all, including intruders, which is a challenge to the basic intuition (Bosio et al. 2002).

Systems where you can trust the services the system provides, with absolute justification, are known as ‘dependable systems’ (Reis et al. 2002). Neumann (2002) says that trust and trustworthiness are two different things. Trust may be present without any proof to justify the confidence in a specific system, while trustworthiness emphasizes the presence of assurance criteria that justifies the confidence in the system. A dependable computer system is one which possesses qualities like reliability, availability, and security (Lawrie & Jones 2002). The Open Source Software is vulnerable to attacks by sharing of altered versions of the software systems. This is a potential problem that raises the question of trustworthiness of the software system. 

A major issue for dependability in Open Source development concerns the necessity of research based evidence that would declare what attributes of Open Source Software and Non-Open Source Software can aid in assuring the dependability of the software products produced. Due to the public and open nature of the Open Source Development process, the privacy barriers of access for influence and involvement in the process are lowered (Lawrie and Gacek 2002).

There needs to be comparative research done to determine the benefits of introducing formal software engineering initiatives into Open Source projects in order to determine if programs like CHATS (Composable High Assurance Trusted Systems) have been successful in increasing trust in open source software products.
Instead it can be that the introduction of such software engineering methods, tools, and techniques may only give an impression of products being more dependable, rather than actually increasing dependability (Murphy and Mauhgan 2002).

Another consideration for dependability is the nature of the products that can be successfully developed in the process of open source development. Dependable system software like operating systems, developed by Open Source Software processes are seen as a prerequisite for further building and creating dependable and trustworthy systems (Neumann 2002). 

Therefore, the open source process may actually be the most effective development approach for completing a dependable system in IT infrastructures or in cases where high levels of dependability are required for initial system deployments, for e.g. safety critical systems (Bosio et al. 2002). 

Lawrie and Gacek (2002) establish that although Open Source Software products are generally limited to only developing system oriented software; these systems are essential to further build up dependable and trustworthy systems.  Due to the growing scope and complexity of software, its trustworthiness has become a major issue. Central to developing trustworthy software is software fault tolerance. Software that is trustworthy is always stable. 

References

• Arief, B., Bosio, D., Gacek, C. and Rouncefield, M. (2002). Reliability Issues in Open Foundation Software-DIRC Project Activity 5 Final Report. Technical Report CS-TR-760.

• Bosio, D., Newby, B., Strigini, L. and Littlewood, M.J. (2002): Advantages of Open Foundation Processes for Dependability:  Clarifying the issues. In Proceedings of the Open Source Software Development Workshop, Newcastle up on Tyne, UK, p. 30-46.

• Lawrie, T., Arief,B. and Gacek, C.  (2002):  Interdisciplinary Insights on Open Source. In Proceedings of the Open Source Soft- ware Development Workshop, Newcastle up on Tyne, UK, p. 68-82.

• Lawrie, C., T. and Gacek (2002). Issues of Reliability in Open Foundation Software Development. Software business Notes, 27 (3). P. 34-36.

• Lawrie, T. and C, Jones. (2002): Target-Diversity in the plan of trustworthy Computer-Based system. In Proceedings of the Open foundation Software Development Workshop, Newcastle up on Tyne, UK, p. 130-154.

• Murphy, R. and Mauhgan, D. (2002): Trusted Open Source Operating Systems Research and Development. In Proceedings of the Open Source Software Development Workshop, Newcastle up on Tyne, UK, p. 20-29.

• Neumann, P. (2002): Developing Open Source Systems: Principles for Composable Architectures. In Proceedings of the Open Source Software Development Workshop, Newcastle up on Tyne, UK, p. 2-19.

• Randell, B.  (2000): Turing Memorial Lecture: Facing up to faults. Computer Journal, 43(2), p.95-106.
• Reis, C., Pontin, R. and Fortes, M. (2002): An Overview of the Software Engineering Process and Tools in the Mozilla Project. In Proceedings of the Open Source Software Development Workshop, Newcastle up on Tyne, UK, p.  155-175.

Elad Shalom,
CTO at ITweetLive.com