repositories for the rest of us: DigCCurr 2007: Closing Plenary

2007/04/20: DigCCurr 2007: Closing Plenary

Cal Lee and Cliff Lynch shared the results of the survey and lessons from individual feedback.

Cal Lee went first. He gave a big caveat that the data is as close to research as "The Situation Room" is to journalism.

Survey Questions

What do you see as the biggest digital curation challenges in your institution?
*A few high level categories: need to change/influence beliefs/perceptions of people outside; skilled IT staff that understand the issues; expectation management (don't promise the impossible without expanded staff and/or time line); organizational commitment; insufficient buy-in from the top;
*lack of IT support either internal or outsourced
*essential technological components are lacking
*money or funding
*ownership of the problem-space (other disciplines besides LIS/AS
*how to identify roles and responsibilities
*planning, quality control
*skills of digital curators-- lack of wide spread competencies
*volume of data, long term preservation and access, metadata
*define what digital curation is, what it takes to do it

Discussion: very difficult to disambiguate the challenges listed above since they all touch upon each other.

2. What are the most important topics to cover in digital curation education?
*High level conceptual orientation, being aware of information/archival theory, OAIS model, risk management
*Functions and tasks like cost modeling, systems analysis, cataloging, web design
*Artifacts
*Standards
*Current landscape: economic models, major players, trends, users and services

3. What do you look for if you were hiring a digital curator for your institution
*Communication skills were mentioned far more frequently than any other skill
*Programming
*Leadership
*Project management
*Metadata
*Service orientation, history of profession

4. What are the skills required that are now currently lacking?
*technology and IT
*programming
*server administration
*knowing alternative technologies (not just pick one but how to evaluate between options
* a good BS radar
One comment in the survey about distributing these skills over many people so one resignation doesn't hold up production
*people with fundamental respect for research process

5. Other comments?
*Allow for retooling for professional development for current LIS/Archivists but also for CS and systems people to remediate in the LIS stuff
*Metadata
*comment on being comfortable with digital curation but less so with digital curators
*don't underestimate the importance of management skills
*strengthen still useful skills from LIS/AS
*think about who you want to recruit -- do they need the discipline degree?
*eventually this will be "the" curriculum, we should teach it within the context of everything else
*seems to be more suited to a specialization on top of other degrees

Cal comment on "vacancy in the professions" this is distributed among a bunch of different professionals. Who will be responsible?

Cliff Lynch - giving a mix of his opinion, reaction, based on many conversations with people here and in the past year. Some study of what was and wasn't covered in the program at this symposium.

Suggest that language does matter. We moved from the phrase "data curation" which comes out of developments in the sciences such as Chris Greer's long life data report to the "truly frightening" term digital curation--which we may want to consider getting rid of. We start from the perspective of archives and records management which are traditionally marginalized although recognized as important.

Better to recognize that we're not the only ones being impacted by large scale computing and networking. Everything has changed radically in the last couple of years. CL just came from NSF/JISC workshop -- it started about digital repositories, then morphed into talking about "data driven science" and then finally became about the entirely new ways of doing science/research. The new types of roles emerging.

"Research facilitators" start looking around S&E facilities of large grant receiving institutions -- they are finding ways to squirrel these people into their organizations (i.e. "staff technologist")

Humanities "critical editions" used to be important -- are they still?

There is a set of activities around data curation that we need to define, specifically regarding management and preservation of scholarly data. One set of activities for long term memory organizations. One set of activities for the creation of the data. Just how much specific scholarly expertise do you need? Critical question.

Difference in use of word "curatorial" in bio-informatics/biology data sets. It's more of a critical editorial role.

Digital curation vs. data curation. Curate. Recognize that once upon a time that libraries had curators who built and managed collections then this role got sliced and diced into all the various types of librarians you've got now (bibliographers, catalogers, etc.). The return to curation as term reflects the changes that need to happen in the way we think about acquisitions in libraries.

Other comments: stunning how different the participants view curation. we all have different opinions of what skills somebody with a certificate in digital curation would have.

We need people who can sort through social, organizational, economic issues around sharing, destruction, anonymous-ization of information resources across time. Not just a narrow view of records management. Broader view needed of social policy and impact in order to make the necessary case for stewardship.

We have a lot of case law that gives primacy to individual ownership rather than social commons.

Economics - besides business and cost models for individual organizations, it is a social good and needs that type of funding rather than a bunch of organizations recharging each other in a circle.

Risk management - very important but difficult to do as it's hard to quantify value of irreplaceable objects. CL says we haven't yet discussed an acceptable loss rate. Keeping all bits for perpetuity is not doable from engineering standpoint. Must get explicit about that.

If your bits don't make it to next week all discussion is mute. We're in an environment where our information is exposed to all kinds of complex threats all the time.

How do you teach the next generation how to do something we don't know what to do? Go back to original principles underlying sound curation as place to start. Foundational principles are a good place to frame analysis.

Labels: DigCCurr 2007