interactions cover Community Experience at

Matthias Müller-Prove in <interactions> 6 (Nov/Dec) 2007


Developers, scratching just their personal itch, are a well-known challenge for open source projects [5]. If the project’s target audience is not the typical software engineer and the application’s interface becomes too technical, intended users will turn away from the product. As a consequence, more-mature projects have established a culture that values—in addition to source code—contributions from other members such as quality testing, user documentation, globalization, and user experience. The cooperation of all teams with their specific expertise makes a usable and successful product.

Usability—or the lack thereof—in open source projects has been discussed during recent years and is also the subject of this special section of <interactions>. Usability is not a given. User research, usability testing, and user-centered design need to be actively integrated into the development processes to change the mind-set of engineers.

However, this article takes a different angle. It discusses the community perspective for open source projects with as an example. Such projects have more than one user interface: the application user interface and the Web user interface of the community sites. If the community portal is designed without a thorough community-centered approach, then the development community (including all participating teams) cannot cooperate properly, and new contributing members are not attracted.

OPENOFFICE.ORG COMMUNITY is the leading open source office suite, with about 85 million downloaded copies worldwide. It is available for all major platforms, and has been localized for almost 100 languages. The project was initiated by Sun Microsystems in October 2000 by open-sourcing StarOffice’s code base. The real size of the community today is hard to measure. However, there are 62,000 registered mailing-list subscribers, and 720 organizations signed the Joint Copyright Assignment to actively support the project [3, p.131].

Since its inception the community has established processes to handle requirements and requests for enhancements. Feature development follows a specification work flow. Furthermore, according to a product’s life cycle, new versions are planned to release in a predictable way for the user. Sun also contributes a team of dedicated user-experience engineers to the project, in order to support the development process and to improve the usefulness and usability of [1].


For some reason people join mailing lists, forums, social networks, and collaborative online communities. Sometimes it is just for the sake of being part of “the tribe,” but if the service turns out to be sustainable, it is often organized around so-called social objects. According to Jyri Engeström, sociologist and founder of, social objects are objects that people care about. People share bookmarks in social-bookmarking services. They share photos with friends in online photo communities, as well as music, “funny” films, contacts, etc. The list of Web sites in this area goes on and on. Each of these services has one main kind of social object, a common interest that community members share.

Wikipedia is the most prominent example with regards to worldwide collaborative systems. Everyone is invited to contribute and edit articles. They are discussed, disputed, sometimes completely rewritten, until they have reached an acceptable level of quality. On a cultural level, the Wikipedians have established a set of social rules about how to deal with issues that might jeopardize the desired standards.

Esther Dyson describes four basic principles that need to be observed for any community to prosper. Otherwise, the community risks collapsing. She says:

Open source projects can also be seen as a kind of social network, with the open source product as the connecting social object. Such communities follow the same principles outlined above. People can join by creating an account on the community website; they can then contribute to the project by filing bugs and feature requests or submitting source code to the project.

The next section briefly presents a newly founded group that is devoted to the user experience of


In January 2007 the User Experience Project was approved by the community council as an incubator project. The main objective of the group is to consolidate usability activities that were scattered all over the project in concept documents, specifications, the bug-tracking system, newsgroup discussions, private email conversations, etc. Another objective is to create a visible and active open source community of usability professionals and interaction designers [4].

A special focus lies in growing this new part of the community with Dyson’s principles in mind. A charter on the project home page explains the team’s identity. A membership list makes them distinct from a casual gathering, and a to-do list—hosted on the community wiki to stimulate participation—shows the current agenda for the team.

Compared with Sun’s user-experience team for StarOffice, the current size has increased by a factor of four. A survey conducted in July confirmed that the newly formed team is on the right track. With respect to flow of information and decision making, there are still some issues that need to be addressed. The visual appearance of the project pages has also been mentioned as an obstacle to creating an identity for the team.

Theses issues are not unique to the User Experience Project. They apply to the entire project at different scales.


The website is hosted by CollabNet, a collaborative platform for software development. The site is divided into more than 120 sublevel domains (remember the native language projects?). Furthermore, there is a community MediaWiki, numerous mailing lists, a task tracking system, IRC channels, several blogs, and of course the CVS system with the code base.

To sum up, the Web presence is a complex network of websites and databases that has a significant impact on the perceived image of the open source project. The lack of a coherent design and navigation structure is an especially severe usability issue for first-time visitors, which hinders them from quickly finding the information they need in order to understand the processes and to start contributing.

By redesigning the websites’ information architecture according to the needs of the community members on the one hand, and the users of on the other hand, it is presumed that the cooperation between the different development groups will be better supported and that the community can continue to grow and prosper.

An application that seamlessly fits into the user’s working environment supports him or her on the daily job. In order to build such a productivity suite, the open source community ought to be connected with a well-designed collaboration infrastructure. Both aspects have to be considered for a successful open source project.


  1. Benson, C., Müller-Prove, M., and Jiri Mzourek. “Professional Usability in Open Source Projects: GNOME,, NetBeans.” In: CHI 2004, Vienna. p. 1083-1084. ACM Press, New York, 2004. Extended Abstracts.
  2. Dyson, E. Release 2.1: A design for living in the digital age. New York: Broadway Books, 1998.
  3. Goldman, R., and R. P. Gabriel. Innovation Happens Elsewhere – Open Source as Business Strategy. San Francisco: Morgan Kaufman, 2005.
  4. Müller-Prove, M. “User Experience for” Interfaces, Summer 2007, no. 71.
  5. Pemberton, S. “Scratching someone else's itch: (why open source can't do usability).” <interactions> no. 11(January 2004) 72.