Bamboo Technology Project: Building Cyberinfrastructure for the Arts and Humanities

Bamboo Technology Project: Building Cyberinfrastructure for the Arts and Humanities
Cole, Timothy, University of Illinois, Urbana-Champaign, t-cole3@illinois.edu
Fraistat, Neil, Maryland Institute for Technology in the Humanities (MITH), fraistat@umd.edu
Greenbaum, David, University of California at Berkeley, dag@berkeley.edu
Lester, Dave, Maryland Institute for Technology in the Humanities (MITH), dlester@umd.edu
Millon, Emma, Maryland Institute for Technology in the Humanities (MITH), emillon@umd.edu
Introduction
This poster will provide an overview of the Bamboo Technology Project and demonstrate some of its prototypes. We are currently midway through an 18-month implementation phase that builds on the Bamboo Planning Project – a process that engaged approximately 600 faculty, librarians, and technologists from 115 institutions between Spring 2008 and Autumn 2010 – both of which have been funded by the Andrew W. Mellon Foundation.

The Bamboo Technology Project (BTP) is a multi-institutional, interdisciplinary effort that brings together humanities scholars, librarians, and information technologists to tackle the question: “How can we advance arts and humanities research through the development of shared technology services?” Comprised of an international partnership of 10 universities,The partner institutions are Australian National University; Indiana University; Northwestern University; Tufts University; University of California, Berkeley; University of Chicago; University of Illinois, Urbana-Champaign; University of Maryland; University of Oxford; and University of Wisconsin, Madison the project is devoted to building applications and shared infrastructure for humanities research.

Structured Work Areas
During the Bamboo Planning Project, we found that scholar participants repeatedly expressed their desire for technology that would allow them to annotate, to collaborate, to gather materials, to organize information, to share materials, to store and preserve materials, and to use social media. Building upon workshop discussions, we identified a more general set of categories to describe central scholarly practices in the humanities that could benefit from technology. Prominent among them were the practices of aggregation, annotation, consideration, engagement, and interaction.Theme Groups may be found here Using these categories as guidelines, and drawing on the project’s leadership and participation from different partner institutions, we organized ourselves into four major interconnected areas:

Structured Work Areas
Research Environments
“Adapters” are being built that will enable scholars working within the Project Bamboo Research Environments to access and apply research tools to dispersed content collections such as materials in the HathiTrust and Perseus collections. Research is being conducted to enable OpenSocial gadgets to be leveraged within the Project Bamboo Research Environments and potentially other areas of the Bamboo ecosystem. OpenSocial is a promising approach to enable us to develop functionality once and apply it many times. We are also building a Tools and Services Registry through which scholars can discover other software tools that their fellow humanist scholars are using.
Corpora Space
We will carry out a community design process to plan for a future generation of corpora-centered virtual research environments to support the central and growing importance of data and of corpora curation for humanities’ scholarship. The results of this design process will be implemented during the second 18-month phase of the project.
Scholarly Web Services on a Services Platform
We will model existent digital humanities applications as web services and then deploy these or their proxies, and other web services, on robust platforms to help support many application developers in the humanities.
Collections Interoperability
We are selecting and recommending metadata standards for digital content that will allow for predictable use of important scholarly material across all Project Bamboo Work Spaces and with all Bamboo tools. We are building an initial suite of adapters and connectors that facilitate easy, cross-collection use of these “target” collections within the Bamboo environment: content from the inter-university HathiTrust repository; selected content from the Perseus Digital LIbrary; and 400-years of English texts from the Text Creation Partnership (TCP). The CI group is collaborating with other Bamboo working groups and external experts to identify the most appropriate standards to implement for both content access and structural interoperability.
Interoperability and Shared Infrastructure for the Humanities
The BTP is establishing shared infrastructure, in the form of web services, interoperable collections, and organizational partnerships to meet the technology needs of humanities researchers and institutions.

RESTful APIs
Project Bamboo is modeling services as resources, and will leverage RESTful APIs to manage these resources across the Bamboo ecosystem.
AuthN-AuthZ protocols
Project Bamboo will delegate authentication to established institutional infrastructure, such as Shibboleth, and to social media identity providers. Ongoing work with Internet2 committees such as Shib-dev and COmanage will leverage edu- space experience and adopt new and existing software to meet Bamboo's requirements.
Future Work
Our next eighteen-month phase will primarily be concerned with implementing the Corpora Space Road Map, which will focus in terms of content on the 450 years of print culture in English from 1473 until 1923, along with the texts from the Classical world upon which that print culture is based. As an initial set of collections, we will be focusing on Perseus, TCP, ECCO, EEBO, HathiTrust, Google Books, Oxford Text Archive, and AUSTLit. Although we will begin with these corpora, we envisage a principle of user-driven growth. In gathering this material, we will move from the earlier to the later, because earlier texts pose more difficult textual and linguistic issues. In designing protocols that will address these issues of early texts, we believe we will anticipate most of the difficulties of the later material.

To enable users to work with these corpora, we will provide tools from the following five classes: 1. Curation; 2. Corpus Search and Discovery; 3. Collation; 4. Annotation; 5. Analysis/Visualization. We aim to build a modular environment that will attract content and collections from providers beyond the Bamboo partnership, and that will encourage the broad-based development of tools to operate on the data.