DH09 Tuesday, session 2: Supporting the Digital Humanities: Putting the Jigsaw Together
[OK, not putting information copied from slides in quotes: no time. Thank you, panelists, for concise wording in your slides! If you want specific attribution, let me know.]
The big questions to be addressed by the panelists, as Martin Wynne proposes in his introductory remarks:
1. What specific problems have you identified, and how are you seeking to address them?
2. What services, if any, will you provide?
3. How might you link with other related initiatives?
4. What are the further elements of the jigsaw puzzle which are needed to create a coordinated and more complete research infrastructure?
First up, Steven Krauwer from the University of Utrecht. “CLARIN — Some Pieces of the Jigsaw Puzzle.” Krauwer is a linguist, and coming at CLARIN from that perspective.
CLARIN stands for Common Language Resources and Technology Infrastructure. Basic idea = European federation of digital archives with language data and resources (text, speech, multimodal, gesture…) aimed at research communities in humanities and social sciences. All languages are equally important. Aims to cover all EU and associated countries.
Estimated cost 2008-2018: 200mil Euro; for now they have 5mil. 200mil sounds like a lot, but less than a million per language per year. But a lot of fundraising work to do.
At least five ways to fail. There are no unsolvable technical problems, no lack of enthusiasm from linguists, and no lack of resources or technologies. On the other hand:
1. Standards. Common standards are crucial. So, threat 1: lack of agreement on standards. CLARIN’s remedies: single standards not *always* necessary (support more than one, provide mappings between them); best and safe practices are very important; and they’re involving the community at large to ensure broad support.
2. Sustainability. CLARIN started as a bottom up initiative. Threat 2: lack of financial support. CLARIN’s remedies, and hopes: EU support (already evident; they’re on the roadmap of essential infrastructures for consideration by national governments); they’re talking to national funding agencies; and the investment per country/language is actually not excessive (see per-language per-year cost above).
3. Ignoring content evolution. Must consider not only written language but other linguistic modalities. Threat 3: CLARIN misses the multimodal boat, will fall into obsolescence before completion. Remedies: they’re reaching out to other communities (speech, frex); they remain open, standing invitation to institutions to join; and they’re teaming up with initiatives with a broader scope than only text.
4. Language coverage. Threat 4: “minor” languages will effectively be excluded because of poor resources and technology coverage. Remedies: BLARK (Basic Language Resource Kit) may have to bring all languages to a minimal level of coverage; they encourage porting of tech and expertise, to minimize effort redundancy; and they’re working on transnational actions to fill the gaps.
5. Take-up. Threat 5: Failure to reach and convince target audiences. No buy-in. Remedies: teaming up with other infrastructure initiatives targeting related audiences; focus on training, education, awareness [vz: YES].
Services they want to offer: 24/7 availability; persistent unique identifiers; single access/authentication framework; single licensing framework; virtual collections.
Links with other initiatives: they’re here! Even before existing as a formal entity. One key ingredient for creation of coordinated and more complete global etc: STANDARDS. Oh yes.
Next up: Sheila Anderson, impersonated by Seth Denbo, also from Kings College London. DARIAH is Digital Research Infrastructure for the Arts and Humanities. Harold Short and Willard McCarty, years ago (last revised 2002), made a conceptual map of the digital humanities: Methodological Commons. [will add link later]
Scholarly Primitives, per John Unsworth: Discovering, Annotating, Comparing, Referring, Sampling, Illustrating, Representing. There are objects, and then there are activities.
DARIAH isn’t solving specific problems, but exists in the liminality between research and infrastructure. Building on existing communities to answer (and ask!) new questions. They take the researcher as a starting point. They have to sustain activities that are at the heart of research, so they’re studying research practices. Not only mapping out the methods applied by individuals, but concepts and perceptions behind them. Based on interviews, they’ll be developing process models for digital and non-digital humanities research. This work will inform conception of an ongoing collaboration of DARIAH primitives: Digitize,, Curate (Model?), Publish (Preserve); Discover, Access/Deliver, Share; Connect, Collaborate, (Re)Use.
Much of what DARIAH wants to include in its infrastructure already exists in other communities; they want to bring those in, instead of building from scratch. Also, infrastructure is not only tech: it’s a digital ecosystem of people, information, methods, services. DARIAH wants to be: a network of centres; which provides access to and allows use of content; employ collective intelligence to build its intellectual networks; and is thinking about the architecture of participation.
Chad Kainz from U of Chicago talks about Project Bamboo. Bamboo is another part of the jigsaw. Joint planning project with funding from the Mellon Foundation, nearing the end of its planning phase. Over 130 institutions from around the world are engaged in it; it’s jointly led by UC Berkeley and U of Chicago. Everyone wrestling with the same problems: scale, sustainability, etc.
Typical project: 1/3 of the time spent on reserch and scholarship; 2/3 of the time spent sorting through technology — which turns out to have been done by others already! Can we turn this equation around and double the intellectual output of the project?
How can we advance arts and humanities reserach through the development of shared technology services? And the “we” is everyone involved, and everyone whose interests PB seeks to address.
Hard line between research and teaching is quickly dissolving; we need to support that too.
PB, Chad emphasizes, is currently a planning project. Goal: building up a consortium of universities, colleges, and other orgs. to work together to advance arts and humanities teaching and research through the development of shared tech services. As the planning progresses, you have to be willing to throw out your previous ideas and let stuff evolve as involvement results in new information. Major deliverable: ____ [help!]
Bamboo vision of cyberinfrastructure: understanding of practices and commonalities interwoven with shared technology services interwoven with organizational, partnership and social models. Bottom line: we have to all work together so that infrastructure is not irrelevant to large swaths of relevant research communities.
Bamboo is looking at a ten year program. Draft plan for first year is to work on three key areas: scholarly networking; Bamboo Atlas; Bamboo Services Platform. Also, develop plan for second year and connect with other projects.
Bamboo Atlas: Define a collection of services that will enable articulation, collection, review and discovery of shared services, related scholarly methods, and other digital tech that supports scholarship. Researcher starting on a new project: what tools have others, working on similar projects, have used? How have they gone about their research and development?
Bamboo Services platform: how do we create a hardware infrastructure that just doesn’t fail from the user’s perspective, and make it available to digital humanities?
Connecting beyond Bamboo: fit in rather than stand out. Expand tool interoperability. Crete common interfaces and open data models. Adopt common standards and specs. Collaborate, adopt, reuse existing technologies.
[vz: Wow. So, I've known about this project for a while, but hadn't explored in depth. Wonder how BU can become one of the involved institutions. I suspect it's a matter of emailing them. Digital humanities are informal like that.]
Finally, Neil Fraistat talks about MITH and the importance of digital humanities centers to innovation. We need more centers, and more collaboration and communication among centers and funders, and between both groups.
Neil references the ACLS Cyberinfrastructure Commission report from 2006, titled “Our Cultural Commonwealth.” John Unsworth says that digital humanities centers could transform the humanities if they collaborated, and were not silos of information and expertise. The Cyberinfrastructure Commission’s most important decision is to create such a network of centers and funders. Network must be international; broadly inclusive, including related fields that don’t use the words “digital humanities”; and free. This last is the most under pressure, these days. A center is defined as an entity larger than a single project.
The working group that Neil and Kay Walter co-chair has created centerNet. Neil describes centerNet’s work so far; information is available at their website. [vz: n.b. valuable resource!] They’re working to address the final report from the Tools for Data-Driven Scholarship: Past, Present, Future workshop that took place in 2008.
centerNet needs users to connect with each other and with tools! Multidisciplinarity is an empty word unless we make social, cultural, economic, technological and international changes to enact multidisciplines. It may be difficult for individual digital humanities centers [vz: not to mention institutions without such centers!] to engage with this cosmopolitics of digital humanities, and that’s what centerNet is looking to address.