Data Structure and Software Engineering

An Architecture to Support

Learning, Awareness, and

Transparency in Social

Software Engineering

Wolfgang Reinhardt and Sascha Rinne

ABSTRACT

Classical tools for supporting software engineering teams (collaborative de-

velopment environment, CDE) are designed to support one team during the

development of a product. Often the required data sources or experts reside

outside of the internal project team and thus not provided by these CDEs.

is paper describes an approach for a community-embedded CDE (CCDE),

which is capable of handling multiple projects of several organizations, pro-

viding inter-project knowledge sharing and developer awareness. e present-

ed approach uses the mashup pattern to integrate multiple data sources in or-

der to provide software teams with an exactingly development environment.

10 Data Structure and Software Engineering: Challenges and Improvements

Keywords: Learning Systems, Knowledge Management, Cooperative

Development Environments, Learning Communities

Introduction

Traditional clichés about software developers lose their validity more and more.

Times, when programmers sat in dark cellars and tried to solve all problems on

their own are over once and for all. In the meantime software engineering has

become a very knowledge-intensive [5] and communicative process (not only but

also triggered by agile methods for software development) where the actors heav-

ily exchange data (see Google-Code), connect with like-minded (see Google Sum-

mer of Code), blog about experiences in their own weblogs, provide code snippets

free of charge (see Django-Snippets) or help novices with words and deeds in large

mailing lists. is social software engineering—the creation of software and relat-

ed artifacts within a social network—gained a lot of attention in recent software

engineering research [1,17]. Besides the improvements of integrated development

environments (IDE, e.g. Eclipse) or procedure models (e.g. eXtreme Program-

ming [3]) research is addressing improvements of the daily working and learning

environments of the developers. e main function of collaborative development

environments (CDE) [2] is to support the whole development process of a team

of software developers from start to nish. is includes version control of code

artifacts as well as process documentation, coordination of tasks or support for

division of labour.

CDEs usually are set up for one specic project; the possibilities for inter-

project-collaboration within an organization with multiple software projects are

very limited because the single CDEs are not able to exchange data.

Furthermore many developers are using data pools (bulletin boards, developer

communities, mailing lists and a lot more) outside the organization in order to

solve a specic problem. Furthermore existing CDEs lack in providing a transpar-

ent view on the progress of a project, awareness of developers’ competencies and

support for individual informal learning processes.

is paper describes an approach for a community-embedded CDE (CCDE),

which is capable of handling multiple projects of several organizations, provid-

ing inter-project knowledge sharing and developer awareness. e presented ap-

proach uses the mashup pattern to integrate multiple data sources in order to

provide software teams with an exactingly development environment. Further-

more we present requirements for a community of developers and sketch a rst

prototypical architecture for such a CCDE.