What ever happened to Project Bamboo?

This paper provides a description and analysis of the trajectory of Project Bamboo, a major humanities cyberinfrastructure initiative in the United States. While previous work has addressed the opportunities, consequences and form of humanities cyberinfrastructure (Svensson 2011; Blackwell and Crane 2009), and the development trajectory of cyberinfrastructure initiatives in the sciences (Ribes and Finholt 2009), very little has been written about Project Bamboo (2008-2012), and how it did and did not reflect visions put forth for humanities cyberinfrastructure. In addition, there has been little discussion of the precipitous drop in community enthusiasm and even awareness of Bamboo, despite its considerable amount of funding.
During its 2008-2010 planning phase, approximately 600 scholars, librarians, and IT professionals from 115 institutions participated in conversations about “enhanc[ing] arts and humanities research through the development of shared technology services”. While this broad engagement contributed to the development of a rich picture of the technology needs of humanists and librarians at a wide range of institutions, it resulted in a multiplicity of visions for what Project Bamboo should specifically do to address those needs. By the time the Bamboo phase one technology implementation proposal was funded in fall 2010, there was already a great deal of unclarity among the planning project participants about whether and how Bamboo was translating the enthusiasm for community building and technical interoperability that emerged in the planning project into concrete deliverables. Two years later, most of the digital humanities community had largely written off Project Bamboo for having failed to produce anything of value, or even make clear what it was attempting to deliver. This paper will elucidate the evolution of Project Bamboo’s goals and scope from its inception to its demise, the factors that contributed to its increasingly negative public perception, including an assessment of which form inherent problems for cyberinfrastructure initiatives vs. avoidable tactical missteps.
The Bamboo Planning Project received a $1.4 million dollar grant from the Andrew W. Mellon foundation in March 2008 (towards a total budget of $2.4 million), to carry out a series of five workshops that would lay the foundation for a follow-up proposal geared towards the technical development of infrastructure. The planning project proposal anticipated a total of 200 participants from 65 institutions (Bamboo 2008, 27). Instead, nearly twice that number of institutions applied to participate in the first set of workshops, and participation remained significantly higher than expected throughout the entirety of the planning project. The first workshop—held in four different locations, with a process refined after each iteration—encouraged teams of participants to discuss open-ended questions about the challenges facing the humanities and potentially ameliorating applications of technology. In many cases, the Bamboo workshop was the first time that scholars, librarians and IT professionals from the same institution had met one another, or discussed issues of common concern. These conversations had wide-reaching impact on local campuses, and were among the most well-regarded aspects of the Bamboo Planning Project.
Bamboo’s discussion-centric approach to workshops was intended to shape the enterprise IT-oriented vision of shared services and service lifecycles that was at the core of the project’s vision. Instead, it quickly became clear that many of the workshop participants—most emphatically, but not limited to, the humanities scholars—felt that the emphasis on tools and services was misplaced. Instead of focusing primarily on identifying services that could best be run centrally by a consortium, and how those services could be integrated with one another, subsequent workshops included tracks for participants interested in improving scholarly networking by connecting people, projects and tools; [1] and for participants interested in sharing faculty stories, exemplary practices, and curricula (both student-oriented, and curricula for improving scholars’ fluency with digital tools and resources). The long-term, visionary “Bamboo Program” that gained consensus among participants in June 2009 included a scholarly network; a tool and content guide; a collection of scholars’ narratives about the use of digital technologies across the arts, humanities, and interpretive social sciences; a repository for documenting workflows; and an “exchange” where individuals in need of assistance could connect with others who have the necessary expertise. These areas of work were included alongside the infrastructure components: a cloud-hosted service delivery appliance, the establishment of a service lifecycle, and the formalization of partnerships with tool developers and collection holders. [2] The scope was ambitious to the point of risky, but it addressed the stated needs and interests of almost all the participants who were involved throughout the planning project. Rather than simply voicing their support for the work in principle, most participants expressed an active desire for their institutions to be involved in building these aspects of Bamboo in subsequent phases.
Multiple factors contributed to the significant downscoping of Project Bamboo between June 2009 and the funding of the Bamboo Technology Project in September 2010, including the impact of the world financial crisis and a change in program leadership at the Mellon Foundation. The community-oriented aspects of the project—which fueled the interest of many planning project participants—were eliminated. This did not, however, result in a single, unified vision. One area of the project, led by librarians, focused on developing collection interoperability connectors for text repositories. Another area, led by technologists, worked on the development of a centrally-run Java-based services platform that would run core services for identity and access management, policy management, notification, and result caching, and could either run or proxy services that provided scholar-oriented functionality (e.g. morphological analysis, geoparsing, concordance). Groups of technologists also attempted to integrate these infrastructure services into “Work Space” web-based applications (Alfresco ECM and hubZero). A scholar-led group focused on the development of a “corpus tool set” application that could provide an integrated scholar-facing interface for tools that can work together to analyze textual corpora. Deviating from the planning project model that required scholars, librarians and technologists to collaborate, rather than focusing only on the areas of the project that were the most comfortable fit, led to a sense of disconnect approaching animosity between sub-groups. While progress was made in each of these areas, both tactical and strategic issues around internally- and externally-oriented communications made it difficult to convey these accomplishments intelligibly to the digital humanities community. The lack of a clearly articulated scholar or developer use case that would be facilitated by Bamboo’s technical developments exacerbated the impression that Bamboo was not developing anything of use.
As the Bamboo Planning Project illustrated, a high level of engagement from a broad community can invigorate a project, allowing it to strive for significant goals that have wide-reaching impact. However, this level of engagement comes with the risk of alienating a large swath of the community if exigencies force the project to scope down significantly, without the project fostering “spin-off” efforts to address those needs that are no longer in scope. A clearly-stated scope connected to use cases that resonate with the target community is essential. Infrastructure projects face a number of unique challenges around communication and demonstrating progress to a broad community, in large part because their technical deliverables do not provide functionality that itself directly furthers scholarship (e.g. via an innovative search algorithm or a novel text analysis interface). There are multiple approaches to ameliorating this, including working closely with digital humanities developers on pilot integrations or mockups for integration with scholar-facing tools, developing mockups, and providing documentation and a sandbox environment that developers can explore, share, and blog about. Regardless, long lapses in communication jeopardize a project comparably to significant delays in technical deliverables.
At the request of the Mellon Foundation, Project Bamboo began to shut down active development work and transition to a wrap-up phase in December 2012. By the end of March 2013, all “artifacts” from the project will be publicly available via links from projectbamboo.org.
References
Blackwell, C., and G. Crane (2009). Conclusion: Cyberinfrastructure, the Scaife Digital Library and Classics in a Digital Age. Digital Humanities Quarterly 3(1).
Project Bamboo. (2008). Bamboo Planning Project: An Arts and Humanities Community Planning Project to Develop Shared Technology Services for Research. January 28, 2008 public document revision. Retrieved athttp://www.quinndombrowski.com/sites/default/files/blog/bamboo_planning_project_proposal.pdf
Ribes, D., and T. A. Finholt (2009). The Long Now of Technology Infrastructure: Articulating Tensions in Development.Journal of the Association for Information Systems. 10(5). 375–398.
Svensson, P. (2011). From Optical Fiber to Conceptual Cyberinfrastructure. Digital Humanities Quarterly. 5(1).
Notes
1. http://quinndombrowski.com/projects/project-bamboo/wiki/w3-implementation-proposal
2. http://quinndombrowski.com/sites/default/files/bamboo/pbw4_discussion_draft.pdf