Preservation in Digital Libraries

Rieger
I am glad to see digital preservation issues being addressed so thoroughly. Though it’s more of a minor point, I’m particularly intrigued by the concept of shared print storage facilities – it strongly suggests that digital is becoming the default, with print just the backup/alternative. It also has interesting implications for the way competing institutions might find themselves collaborating. I think one of the crucial points raised in this report (though maybe not directly related to digital preservation) is the call for libraries to develop more “scaleable and flexible infrastructures” and to rethink their collection development policies. Digital resources are truly changing the landscape of traditional resource management practices. The most important issue raised, however, was the fact that so many standards and best practices are so far out of date. That’s a field that requires a lot of development very quickly, especially with the advent of these large-scale ventures.

Jones and Beagrie
Although digitization has been touted by some as the solution to preservation of print resources, it has become clear that digital objects require just as much if not more effort and funds to maintain. Paper can sit on shelves untouched for decades and still fulfill its intended function; it would be very difficult, at this point, to access a digital object from forty years ago. This handbook emphasizes the fragility of digital media, and the advance planning and quick action necessary to keep it intact and functional. It also identifies the key stakeholders for long-term preservation efforts. Unfortunately the section on cost is probably all too accurate when it says, essentially, that there is no way to estimate it.

Littman
This is a useful reflection from someone who clearly has much experience in the digital resource field, and has faced many of the problems that digital libraries have and will continue to face. The four categories of failures he describes -- media, hardware, software, human -- are ever-present risks in any digital library, and it is important to keep those various risks in mind when designing and managing a DL. Littman suggests that human error was the most serious threat to his own project, which is a lesson that any DL team should consider very seriously.

Muddiest Point (Access in Digital Libraries II)

Point 1: OAI-PMH seems to rely on consensus amongst all participants as to which metadata standard to use (i.e. Dublin Core). However, I imagine it is difficult to convince people to provide DC metadata if they are using a different standard for their own purposes -- it's just more work. What I wonder is whether it is easier to try to get people to agree on one standard, or whether it is possible to develop a new protocol that is capable of recognizing and harvesting multiple metadata schemas.

Point 2: If Z39.50 is not designed for full-document retrieval, is there a different standard in the works that addresses that growing need? Is there any institution that has successfully implemented a different system?