Information Technology is enabling large-scale, distributed collaboration across many different kinds of boundaries. Researchers have used the label new organizational forms to describe such collaborations and suggested that they are better able to meet the demands of flexibility, speed and adaptability that characterize the knowledge economy. Yet, our understanding of the organization of such collaborative forms is limited. In this dissertation, I study distributed knowledge collaboration in the context of a unique setting - a large, distributed, professional legal association, where practice involves knowledge that is complex, highly contextualized and failures have extremely consequential results. The first essay focuses on knowledge sharing at the individual level. Differing approaches have been developed for the study of knowledge sharing - I distinguish between approaches that focus on knowledge transfer and those that highlight the need to transform knowledge to be effective. The former emphasizes availability of and access to knowledge sources while the latter argues that knowledge is difficult to share since it is `localized and embedded in practice.' In this study, I empirically examine the notion that, in the presence of novelty, knowledge sharing involves not simply the transfer of information but rather the transformation of knowledge and understanding. I proposed a theoretical model and tested it by gathering 160 survey responses from individuals who answered questions about two specific cases they encountered - one routine and one novel. The results largely support the key arguments presented here. The second essay examines, at the organizational level, the practices used to mitigate the challenges of distributed collaboration. For example, since larger geographic dispersion may result in pockets of local expertise, how is such knowledge shared with the community? What practices are used to mobilize members for collective action? I undertook a field study using a grounded theory approach and a practice lens to investigate the every day activities that are used to coordinate knowledge work. I found evidence for two distinct sets of practices - one with an internal focus and the other with an external focus. I describe these in detail and suggest that the way in which distributed communities balance the two is essential for their continued viability.