DLF Keynote address
I’ve been hanging around at the Digital Library Federation’s Fall Forum in Baltimore. Most of my notes are jotted on low-tech paper, but I did type in my notes on John Unsworth’s keynote address, “Cyberinfrastructure for the Humanities and Social Sciences” (slides here) Here’re my raw, unedited notes:
Basic vocabulary of “cyberinfrastructure” can be problematic
Term comes from Atkins report (“Revolutionizing Science and Engineering Through Cyberinfrastructure”), Blue-ribbon panel funded by NSF
- NSF expects recommendations when giving funds
- ACLS panel more like an amicus brief, and less concerned with how to disburse actual funds
“Cyberinfrastructure” described in layers:
- layer below cyberinfrastructure comprises hardware, etc..
- layer above cyberinfrastructure comprises “software programs, services, information, knowledge and social practices” in specific local contexts and practices
- cyberinfrastructure itself lies between these layers, of enabling “hardware, algorithms, software, communications, personnel”
- most important is that this is a shared layer across practices and sites, but isn’t “electronic plumbing” shared by everything
We know what cyberinfrastructure for the sciences looks like [slide]
• Few of these components translate to the humanities and social sciences; only technologies for distance collaboration and software toolkits are applicable
What might these shared cyberinfrastructure for humanities and social sciences look like?
• Training and professional development for researchers; long-term commitment for an institutional support base, understanding of such training as “professional development” in the traditional sense
• can’t support hum/SocSci researchers in same way as science/engineering [?]
• standards and communities of (best) practices; applicable across disciplines and contexts
• well-curated federated collections of digital objects; hum/SocSci objects are distinct from science/engineering ones
• digital preservation
• data repositories/depositories for the social sciences [how are these different from “collections”?]
• analytical tools “appropriate to hum/SocSci goals and materials”
• “tools, legislation and institutions” that address both copyright and privacy issues
Universities more inclined to invest in layer under cyberinfrastructure (“wires in walls”, “hardware and software”) rather than people and “rich academic content”
• stems from appeal of one-time investment vs. long-term commitments
Emphasis on tools that will assist humans in their interpretation-work. (Michael Jensen quote)
Much of the object of humanistic inquiry is “non-computable” – difficult to quantify, etc.
Increasing linkages b/w data result in increased risk of harm to subjects of social-science research
• looming problem of tradeoffs for subjects (and researchers) b/w privacy and convenience
Primary constraint for humanities isn’t privacy, but copyright
• [**copyright slide; sheer volume of content being generated annually]
• copyright restrictions limit humanities research access to objects of study, just as privacy concerns limits access for social scientists
• “the ink on vellum problem isn’t expanding; the pixels on screen problem is”
Questions of audience: focused on scholars, or larger public?
• crisis of scholarly publishing is a crisis of audience