Since taking over management of the Digital Media Group at Cornell University Library in 2005, a/v has felt like my own personal white elephant in the room. I knew that at some point (always “very soon”) I was going to have to deal with it, but didn’t have the staffing, resources, or know-how to really begin to tackle it. I sensed the overwhelming complexity of formats; the problems of nascent delivery platforms; and the lack of a viable preservation system (not to mention standards) to handle the files in perpetuity.
So instead I punted, outsourcing when necessary, but also developing very small “boutique” collections of a/v content on which to cut my teeth. (One particularly fun and successful project involved partnering with the University of Bremen to digitize, describe, translate, and caption 22 films by the German filmmaker Alexander Kluge: http://muller-kluge.library.cornell.edu/en/index.php.) While I learned a huge amount in the process of developing this type of collection, it in no way approached the range of activities needed to deal with a/v in a systematic way across the Library’s vast holdings.
In pondering audio/video more expansively, I’ve always broken it into three primary components: digitization, delivery, and preservation (of course, metadata is integral to all three). I felt like, until we had a viable option for at least 2 out of 3 of these, it would be folly to pursue it seriously. Then in 2010 Cornell University contracted with Kaltura to support delivery of a/v assets campus-wide, and Cornell Library completed Phase 1 of its (long-overdue) Fedora based archival repository. When, in 2011, the Goldsen Archive became the official repository for the Experimental Television Center’s 40-year history of artist produced tapes (a phenomenal and historically significant collection), I knew it was time to begin to tackle the digitization piece. We applied for internal funding from the College of Arts & Sciences, and were awarded a small grant to digitize a portion of the Center’s VHS tapes.
By design I knew I wanted to begin simply. Starting with a single collection and a seemingly more accessible format like VHS seemed doable. Despite these good intentions, it’s been significantly more complicated than I anticipated. The classic trope, “the more I learn the more I realize how little I know,” definitely rings true in this case. That said, we now have a functional space and workflow for digitizing VHS content and will expand to cassette tape soon. We contracted with Chris Lacinak at AudioVisual Preservation Solutions (AVPS) in New York to help us establish our lab-space and workflow (highly recommended- those guys are great), and hired an internal person with subject expertise in video art to begin digitization. We knew we wanted preservation and access quality versions, so we decided to benchmark at an uncompressed 10 bit format, which upped our storage requirements significantly (and also negated the possibility of using our current archival repository, which isn’t set up to ingest such large files). All in all, it’s been a rewarding endeavor, but if I had to do it all over again, here are a few things I wish I’d known in the beginning.
- It was naïve of me to think that I could get away without hiring an official a/v technician. While we have made progress in digitization, it’s been a very steep learning curve, much of which could have been mitigated by having a knowledgeable person on staff. It was equally important to have the subject-expert involved, but trying to cut corners by merging the positions was unadvisable. We have, however, just posted for an a/v specialist, and should have the position filled in January.
- Starting with an albeit very compelling collection of experimental video art dating from the 1970’s — well, I probably could have thought this one through a bit more. I can see it as either incredibly demanding or remarkably forgiving, depending on the day and my mood. In short, it’s really hard to determine whether that tracking error or color shift was by design (think Nam June Paik) or a problem in the tape, further complicating the capture and qc process.
- Setting up a viable lab-space for a/v is much more demanding than for still images. One huge factor I didn’t consider: the potential electrical frequency interference from the small mechanical room adjacent to our lab. We had to find another location within the Library to set this up (never an easy prospect with the shortage of space typical of academic institutions). We are still in the process of finalizing appropriate furniture layout for the space.
- Overall, the a/v workflow is simply way more complex, from capture to qc to preservation — every step of the way is fraught with significant technical demands and challenges. Proceed with caution.
While I’m glad to have started this process, it’s still very daunting to contemplate what lies ahead. We have vast holdings of a/v content at very high risk, and no real way of dealing with it. For this reason, a colleague at the Lab of Ornithology and I have initiated a campus wide group to try to develop a comprehensive plan to deal with the University’s a/v assets, modeling our approach on the successful work done by Indiana University. While there is no guarantee of success, I’m looking forward to learning more about what I don’t know (and probably won’t ever know), but hopefully getting the funding to hire someone who does. Stay tuned.
PS — On the delivery front, this is a whole different story. Let’s just say nothing is ever “out of the box.” We are in the final stages of implementing Kaltura for the Library, after making a number of regrettable compromises in the process. Write me if you want the gory details.
// Danielle Mericle
Digital Media Group, DCAPS
Cornell University Library