Connexions: a Social and Successful Anomaly among Learning Object Repositories Xavier Ochoa Centro de Tecnolog´ ıas de Informaci´ on, Escuela Superior Polit´ ecnica del Litoral, Guayaquil, Ecuador Email: xavier@cti.espol.edu.ec Abstract— Connexions is a Learning Object Repository that has gained notoriety as a successful example of collaborative creation of learning materials. In a previous quantitative study about the characteristics of learning material reposi- tories, Connexions presented some anomalies that prevented it to be classified together with other Learning Object Repositories. This paper, working with updated data, finds that those anomalies are not errors of the previous study, but that the anomalies have increased and are more strongly expressed. Moreover, those anomalies, namely the exponen- tial growth in the number of contributors and content, seem to be the reason behind the success of Connexions. It is concluded in this work that Connexions strategy of “release early, release often”, together with the openness of its core community, that welcome and support newcomers and sporadic contributors, offer a reasonable explanation for the exponential growth trends found in Connection, that can be classified now as a Social Learning Object Repository. Index Terms— Learning Object Repositories, Social Net- works, Connexions, Social LOR I. I NTRODUCTION Connexions is a Learning Object Repository (LOR) created at Rice University in 1999 as an alternative way to create, share, maintain, use and reuse content [1], [2]. The primary goals of Connexions are to interconnect learning content across disciplines, courses and curricula and to create a collaborative environment where learning communities could share knowledge that would lead to the creation and improvement of materials [3]. One of the major differences between Connexions and other, more traditional LORs, is that the process of content development could be opened for collaboration between interested individuals worldwide. In a previous work of the author of this paper [4], several systems for learning content publication were quantitatively examined in order to find common char- acteristics and behavioral patterns. The main conclusion of this previous paper is that there are different kinds of these systems where the characteristics are clearly dif- ferent between Learning Object Repositories and Refera- tories (LOR), Open Courseware initiatives and Learning Management Systems (OCW and LMS) and Institutional Repositories for research papers (IR). While most of the systems, due to their shared characteristics, aligned quite clearly in one of those categories, Connexions presented some anomalies. Some of Connexions characteristics put it, as expected, inside the LOR group. However, one very important variable, the growth of the contributor base presented an exponential growth, while in all the other LORs, this growth was linear. The difference was so clear that it could not be assigned to biases in in the measurements or in the calculations. The only logical conclusion, implied in the discussion section, is that Connexions, due to its unique measured behavior, is a new type of repository. Given that the main difference of Connexions with more traditional LORs is the social interaction for the creation of materials, the name Social LOR is proposed for this new category. The first objective of this work is to test if the anomalies detected in [4] are still measurable and if new anomalies could be found. To conduct this study an updated set of data is collected from Connexions and other representative LORs. This data is taken two years after the collection made for the original study. Several characteristics will be measured and analyzed using the same procedures used in the original work. The second objective is to provide a in-depth quanti- tative analysis of the particular characteristics of Con- nexions and its community that could provide some explanation for the observed anomalies and could justify the creation of a new category of repository where Con- nexions would belong. First, the processes of publication and reuse will be analyzed and the obtained variables correlated. After, the social network that emerge from the collaboration to create materials will be analyzed and its characteristics also contrasted with the previous results. This paper is structured as follows: Section II provides some background for this study, presenting related work made to understand Connexions and other Social LORs. Section III explains how the data was collected and prepared for the analyses. Section IV compares Connex- ions with more traditional LORs in order to establish if the anomalies are still present in the current version of the repository. Section V analyzes the particular char- acteristics of the production and consumption process in Connexions and provide reasonable explanation for two of the observed anomalies. Section VI analyzes the social network of Connexion contributors, the result of the analyze offer an explanation for the third observed anomaly. The paper finishes with a general discussion about the findings of the analyses and further research that should be done to obtain a better understanding of Social LORs. JOURNAL OF EMERGING TECHNOLOGIES IN WEB INTELLIGENCE, VOL. 2, NO. 1, FEBRUARY 2010 11 © 2010 ACADEMY PUBLISHER doi:10.4304/jetwi.2.1.11-22