This is the first part of a three post series (part 2 and part 3) that will be posted here in November about the thesis I did at Atex Polopoly. I did the thesis in the fields of human-computer interaction (HCI) and computer supported collaborative work (CSCW). In this post and two more posts I will shortly present my thesis, my experiences from in the wild and what to think about when designing the Polopoly Admin GUI. This post is about how I conducted my thesis and my conclusions.

The thesis was about content collaboration, which is when editors collaborate about producing and editing content, for instance, a teaser or an article. The purpose was to investigate how this type of collaboration was supported in web content management systems. The overall question was if the user experience of web content management systems would be better if it was more like a community where users worked together to produce and publish content.

A field study was conducted at seven web editorial staffs. Three different types of editorial staffs were participating in the study: distributed editorial staffs (a university and two municipalities), time-oriented editorial staffs (two larger media providers) and news-oriented editorial staffs (two newspapers). At all editorial staffs, collaboration was central in the process of producing content. These staffs did not only use Polopoly, but also Escenic, EpiServer and SharePoint.

People collaborated by talking to each other, sending e-mails and instant messages and, sometimes, even using other types of collaborative software. But they rarely used the web content management system directly to collaborate. One example of collaboration was that one editorial staff used workflows to manage content, which in this case was an approval process to where at least two users worked together.

Another example was that one web content management system had support for sending content to a “first page inbox”. Even if some journalists sent completed teasers to the first page editor, he or she did never check the inbox. The primary source, and the only they had time to follow continuously, was the first page editors e-mail inbox. Therefore, all journalists sent an e-mail to tell the first page editor to check the “first page inbox”. In the next three paragraphs, three examples are presented to support the three main conclusions of the thesis.

The first conclusion was that web content management systems would improve content collaboration by better integrating support of external collaborative software. Many editorial staffs had found alternative usages of features in the web content management to collaborate about content. For example, many editors used the ID of a content to share the content with other editors. That was accomplished by sending the ID in an e-mail or an instant message.

The ID is a string of numbers that the editors had to copy from, for example, an e-mail and paste into the web content management system. One editor said that it was like “sharing phone-numbers with each other”. By using URLs it would be possible to minimize the number of steps necessary to share a content to another user. For example, it would be possible to easily create an e-mail directly from the editor’s interface with the content URL in the body of the e-mail. When the URL is clicked, the content is opened in the editor’s user interface. It is much better to support the mental model of internet in the web content management system, then the mental model of phone numbers.

The second conclusion was to better support awareness of content activity and, that is, to make editors more aware of what other editors and content producers are doing. Even if all web content management systems had support to follow recent changes, they lacked efficient support to filter them. For example, editors waiting for content to be delivered wanted to observe if another editor had started working with the content, for example a teaser. It was not possible to follow only one user or to observe the activity of a group of users. Compare it to how you follow friends on different social networks.

Last, but not least, the editorial staffs had different requirements about content collaboration. They worked with different types of content and, therefore, had different requirements. Additionally, the organizations had different structures. Subsequently, collaborative features had to be customizable. For example, to improve the awareness of content activity different dashboards designs were tested. Each editorial staffs, or even editor, had their own favorite features in the dashboard. The dashboards designs were only very simple prototypes and it does not prove how the editors would use the dashboards in the reality.

Better collaboration support in web content management systems is likely to facilitate the management of articles and other content. The conclusion was threefold. Better integration with external collaborative systems, better support for awareness of content activity and making collaborative features customizable to support different types of editorial staffs are the most important guidelines.

To finally conclude, some organizations may benefit from using a web content management system where the user experience is more like a community. If you want to learn more about how I conducted my thesis and more examples and solutions, read my thesis. In my next post that will be published in one week I will write about my experiences from the wild.

Max Walter is a soon- to-graduate computer science student at the Royal Institute of Technology that did his master thesis at Atex Polopoly. He is also working as a usability consultant at Metamatrix. During his studies he has written computer books for beginners and articles in Sweden’s largest computer magazine for advanced users, Datormagazin. Read more about him at LinkedIn, follow him on Twitter or read his online business card at http://2mw.se.

The photo was taken by bwalsh.

Taggar: atex, cms, enterprise, interaction design, ixd, kth, polopoly, polopolyforum, thesis, usability, user experience, ux, WCMS