Open VRE Platforms, Generic Workflow Editors, and the Future of Humanities Scholarship

One of the most promising developments in digital humanities over the last several years has been the introduction of VREs: virtual research environments.  These are digital “environments” where scholars can conduct most, if not all, of their research, from the finding of resources to the publication of results.  VREs have found wide acceptance in the natural sciences but very little within the humanities.  In this paper I will present one major reason for this, the lack of access to appropriate analysis tools, suggest a promising solution, the open VRE environment, discuss the place of generic workflow editors within such environments, and demonstrate the open workflow environment Taverna, a system the Göttingen Centre for Digital Humanities (GCDH) is working to make compatible with open VRE environments.  The GCDH is part of the further development of VREs within the EU DARIAH Project (Digital Research Infrastructure for the Arts and Humanities).  DARIAH is developing a research infrastructure for the arts and humanities, upon which VREs can be built and used.  DARIAH will be also be represented by a poster presentation.

Humanities VREs have typically been built as closed systems, where tools must be specially designed for the VRE.  The current direction in VRE development, however, is to develop open VREs that can interact with the external tools that scholars already use.  But making these tools available is just the first step.  The next step is to allow scholars to bring tools together to build complex scholarly workflows.  This is where the workflow editor comes in.  So, say a scholar had 50 scanned pages that she wanted to OCR, combine into a single document, tokenize, and lemmatize.  And what if she wanted to do this for 50 different documents?  And then do similarity analysis on all of these documents?  With a workflow editor, instead of doing each step individually, she could bring these processes together into a workflow that she could then save and perform as often as she wanted.  And she could share this workflow with other scholars so that they could use it or, more importantly, critically evaluate it.  This latter is possible because the workflow saves the exact methods and their relationship to each other, allowing another scholar to evaluate and even change it, for example, selecting a different tokenization or lemmatization algorithm to produce a better similarity analysis.

The inclusion of a generic workflow editor within an open VRE environment does more than just simplify the work of the individual scholar.  It also allows, and even requires, a level of openness about the scholarly processes and decisions that stand behind research results that has seldom been expected or possible in the humanities.  And this is the primary advantage of the generic workflow editor: it opens every step of the scholarly process to critical examination, removing the veil of mystery that often clouds humanities research.  This may be a frightening prospect for the individual scholar.  But it can only be a gain for scholarship as a whole.