WikiTeams

WikiTeams - The PJAIT team presents the results of the project.
The future of expert community-based programming.

Aristotle already defined humans as "social animals." One of the underlying skills for this characterization is the ability to collaborate and work in a team, now considered crucial for the development of innovative projects in the knowledge economy. The "WikiTeams" project led by the PJAIT team led by Prof. Adam Wierzbicki investigated collaboration in virtual, emergent teams, so characteristic of the modern Internet, facilitating this goal.

The research conducted was on the organization of work and management or coordination of virtual teams of Wikipedia authors, or developers on the GitHub platform dedicated to building open-source software (open-source software). Both Wikipedia and GitHub are communities of creators, based on knowledge and meritocracy, whose modus operandi is imitated by companies in the new economy (as described by Dan Tapscott in his famous book "Wikinomics") and often referred to by the common adjective "Wiki."

Specialization through experience

More than 80% of developers who devote their time to open source software development aim to gain experience and improve their programming skills. They mostly choose the GitHub platform, where more than 30 million projects have been created, with more than 10 million developers working on them.

The proportion of 3 projects per 1 programmer shows the scarcity of work occurring on this platform, where most projects are single-person. Few projects get huge popularity, and new programmers tend to choose the most popular and active projects. This leads to an inefficient division of labor, and is not to the benefit of programmers, whose activity in a very popular project will be much less noticeable.

The research in the WikiTeams project has made it possible to develop a method for recommending GitHub projects to programmers who want to work on this platform, says Professor Adam Wierzbicki, project leader. This method is more effective than individual selection by programmers, while also promoting the programmer's gaining experience in the programming languages he or she prefers.

The method developed in the research is based on the similarity between a programmer's programming experience and the technologies required for an active project. The research showed that the proposed method has the potential to significantly increase the efficiency of work in the GitHub developer community, the largest platform that brings developers together around projects.

Open Programming on the Market

From the user's point of view, open source software projects on the GitHub platform resemble projects developed by external, commercial, software companies. A business owner who is considering choosing open source software that meets his needs has an alternative: he can contract an external company (or his own employees) to produce software with similar functionality, or buy such software if it already exists on the market.

One of the criteria for choosing between open source or proprietary software is the quality of support provided in fixing bugs and adding new software features. The WikiTeams project has developed methods to evaluate the support provided by the OpenSource development team on the GitHub platform. - highlights the PJAIT team responsible for the project.

The models developed make it possible to predict how quickly a GitHub development team will fix bugs or add new functionality to software. Moreover, the study found that the quality of support provided by a GitHub development team depends on the organization and management of the team. Teams that have fewer permanent members and more centralized work provide better support than large and decentralized teams.

In addition, the quality of support is heavily influenced by the use of task management functionality provided by the GitHub platform.

The research in the WikiTeams project also allowed the creation of a social network model for the Wikipedia author community. Authors who frequently edit Wikipedia and discuss articles on its pages form a network of acquaintances whose structure influences social phenomena and processes on Wikipedia.

One example of the influence of familiarity is the selection of administrators on Wikipedia. The network of acquaintances also influences the quality of articles created on Wikipedia. In the Wiki model, then, we can place our hopes on the substantive development of Internet functioning and the exchange of knowledge and experience.