Reflection on Luther, K., Caine, K., Ziegler, K., & Bruckman, A. (2010, November). Why it works (when it works): Success factors in online creative collaboration. In Proceedings of the 16th ACM international conference on Supporting group work (pp. 1-10). ACM.
Summary:
The paper presents a study of online collaboration to create animated movies (called “collabs”) which were collected from Newsgrounds, a site for hosting animated movies and games. The paper aims to understand what factors influence the success of collabs. From interviews, the paper posits that three factors: (1) planning and structure for collabs, (2) reputation and experience of the leaders, and (3) communication and dedication of the leader and the members are important factors. To verify these, the paper presents a quantitative analysis which supports all three claims. Finally, the paper contrasts the findings with existing literature on Open Source Software (OSS) projects.
My Reflection:
I liked the paper. I also appreciate it asking us to refrain from generalization and calling for further research to understand online collaboration. I felt some of the findings were intuitive, something that whoever has worked in online collaboration project have all felt, and seeing it quantified on a paper felt nice. Many times, especially in HCI we take what people say (or say what they had observed) as the finding and this paper goes further to compare what people said with what they actually did. Thus I found the method refreshing to read.
Thinking about the method, the first thing that stood out for me was the low number of collabs. The paper says that there were 127,328 threads and only 2,797 of them were considered as collabs (as they had the word “collabs” in the title). That is around 2.03%! This raises a concern of whether the practice of collaboration is low in Newgrounds or that selecting collabs based on the use of the word in the title severely restricted the selection. It would have been nice to somehow know how many collaborations happen and then we would know what percent of those adhere to a supposed community norm of using the “collab” in the message title.
Another thing that stood out was the finding that leaders of successful collabs have previously completed more solo projects than leaders of failed collabs. Given that the community definition of “successful collabs” is just to be completed, I wonder if this finding raises further concerns in generalizing it for complex, collaborative tasks. These leaders of successful projects have experiences of completing projects on their own so even if other people don’t contribute a lot, they could take responsibility and bring it to completion. To ensure that this may not be the case, we should probably look at the quality of the contribution or even analyze something as simple as the number of collaborators listed as co-authors.
This brings me to conclude where I want to try to merge the papers on coordination from last week to the collaboration papers that we read this week. I consider collaboration as the umbrella term encompassing the alignment of interests among participating member (cooperation) and alignment of actions (coordination). In this paper, we see cooperation in the form of high-volume communication among members in successful collabs and coordination could be seen in the importance of planning and structure. Reflecting on these papers and my own work, I have started to think that collaboration has to have something more which makes it worthwhile/meaningful — something like umami which accentuates the taste of the thing in which it is added without it having its own taste.