English Editors: Joint issue with N OVÁTICA 2 Presentation: Software Engineering. A Dream Coming True? – Luis Fernández-Sanz (original) (raw)

Free Software Philosophy and Open Source

Handbook of Research on Open Source …, 2007

Product or company names used in this set are for identifi cation purposes only. Inclusion of the names of the products or companies does not indicate a claim of ownership by IGI Global of the trademark or registered trademark. Library of Congress Cataloging-in-Publication Data Handbook of research on open source software : technological, economic and social perspectives / Kirk St.Amant and Brian Still, editors.

Open Source Software Development

This article examines and reviews what is known so far about free/open source software development (FOSSD). FOSSD is not the same as software engineering as that is portrayed in common textbooks. Instead, it is a complementary approach to address the many challenges that arise in the development of complex software systems that are often built outside of a traditional corporate software development environment. This article highlights some of the basic understandings for how FOSSD works based on empirical studies of FOSSD projects, processes, and work practices in different communities. This includes identification of different types of informal online artifacts that facilitate and constrain FOSSD projects. This article also identifies what different studies examine as well as the approaches used to sample and systematically study FOSSD. Next, opportunities for constructive crossover studies of software engineering and FOSSD help reveal new directions for further research study. Finally, the last section presents limitations and conclusions regarding studies of FOSSD.

The Practice of Free and Open Source Software Processes

2008

Abstract: Free and Open Source Software (F/oss) is widespread today. F/oss is primarily a movement that permits access to software source code, and, en- compasses a software development paradigm that harnesses the efforts of a dis- tributed community. F/oss is not simply a computer science phenomenon, but also a social, economic and legal one. The goal of this survey paper

Open Source Software Development*Open Source Software Development

Open source software development (OSSD) is a community-oriented, network-centric approach to building complex software systems. OSSD projects are typically organized as virtual enterprises that lack an explicit managerial regime to control and coordinate decentralized project work. However, a growing number of OSSD projects are developing, delivering, and supporting large-scale software systems that are displacing proprietary software alternatives. Recent empirical studies of OSSD projects reveal that OSS developers often self-organize into organizational forms we characterize as evolving socio-technical interaction networks (STINs). These STINs emerge in ways that effectively control semi-autonomous OSS developers and coordinate project activities to produce reliable and adaptive software systems. In this paper, we examine how practices and processes enable and govern decentralized organizations like OSSD projects when coalesced and configured as contingent, socio-technical interaction networks. In so doing, we draw on results from two ongoing case studies of leadership and governance activities and elements in a small and a large OSSD project.

Dependencies, networks, and priorities in an open source project

Handbook of research on …, 2007

Open Source Software: Lessons from and for Software Engineering

Computer, 2011

Despite initial suggestions to the contrary, open source software projects exhibit many of the fundamental tenets of software engineering. Likewise, the existence of category-killer apps suggests that conventional software engineering can draw some lessons from OSS. Open source software can elicit strongly contrasting reactions. Advocates claim that OSS is high-quality software produced on a rapid time scale and for free or at very low cost by extremely talented developers. At the same time, critics characterize OSS as variable-quality software that has little or no documentation, is unpredictable as to stability or reliability, and rests on an uncertain legal foundation-the result of a chaotic development process that is completely alien to software engineering's fundamental tenets and conventional wisdom. Research suggests a more balanced view. On one hand, OSS is not the "silver bullet" championed by its most vocal partisans. On the other hand, it does not radically diverge from traditional software engineering practice as its severest detractors claim, and, as evidenced by some notable successes, OSS offers many tangible benefits. OSS AS A SILVER BULLET Twenty-five years ago, IBM software engineer Fred Brooks famously contended that "there is no single development, in either technology or management technique, which by itself promises even one order of magnitude improvement within a decade in productivity, in reliability, in simplicity." 1 However, many claim that OSS is indeed such a silver bullet. Defenders argue that OSS, beyond its obvious cost advantages, is of very high quality. Contributors to OSS projects are in the top 5 percent of developers worldwide in terms of ability, and are self-selected and thus highly motivated. Furthermore, the testing pool is global, and peer review is truly independent. Another key advantage cited is the rapid development time of projects. The OSS community has taken odds with Brooks' law-namely, that "adding manpower to a late software product makes it later," 2 a conclusion based on his experience managing development of the IBM OS/360-by endorsing Linus's law: "given enough eyeballs, every bug is shallow." 3 There are many examples of OSS products of exceptional quality and reliability across a range of application domains-indeed, "category killers" such as the Linux kernel and Apache webserver perform so well that there is no market for an alternative. NOT SO FAST Critics concede that a staggering mélange of OSS products is readily available for free download, but they claim it is virtually impossible to predict the usability, stability, and reliability of these products. The uneven quality is not helped by a lack of documentation and the reliance on support and upgrades from a voluntary community who must be convinced to accept changes to suit specific circumstances. These flaws are exacerbated by a complex licensing situation in which even the lawyers cannot definitively resolve IP rights issues. 4 Furthermore, OSS arises from a development process that seems to flout traditional best practices. For example, typically there is no real formal design process, no risk assessment or measurable goals, often no direct monetary incentives for developers or organizations, informal coordination and control, and much redundancy as tasks are duplicated in parallel initiatives. All of this is anathema to conventional software engineering. Other analyses of OSS say that 30 years of prior software engineering research cannot be discounted so easily. The claims in relation to the quality of OSS products and of community feedback are particularly questionable when exposed to scrutiny. Quality A study by Ioannis Stamelos and colleagues assessed quality issues in the SuSE Linux 6.0 release. 5 Using the Logiscope code analysis tool, they examined more than 600,000 lines of code across 100 modules and found that only 50 percent were acceptable. Of the remainder, 31 percent required comments, 9 percent required further inspection, 4 percent required further testing, and 6 percent needed to be completely rewritten. These results are quite average in the software industry: only half of all modules meet generally accepted standards. In a similar vein, Srdjan Rusovan, Mark Lawford, and David Parnas studied the implementation of the Address Resolution Protocol in the Linux TCP/IP implementation and identified numerous software quality problems. 6

English Editors: Joint issue with N OVÁTICA 2 Presentation: Software Engineering. A Dream Coming True? – Luis Fernández-Sanz (original) (raw)

Related papers