I had an interesting chat with a researcher on Saturday night; it seems my library hasn’t been doing its job in the researcher’s opinion. The issue is that the library doesn’t catalogue its special collections appropriately . Now I actually work with these self-same special collections, and I suspect that the researcher has a good point — our simple lack of complete registration is fatal for unique documents. We do register boxes of content, but with no specific declaration for each item in the boxes. In other cases, the general collection is registered at an upper level, but boxes are unregistered.
Each time a document is digitized, we do create metadata for this, but this is item-by-item, and doesn’t give the overview a researcher needs; it’s just-in-time from the library’s perspective, not from that of the researcher.
All the same, I can finally say that we have a system to ensure that we have unique IDs, so we can theoretically register each and every document with an HTTP-URI and create a resource description for each URI. However, rolling this out for the entire collections is as yet a pipe dream — there is no data to convert, just original cataloguing to be done. This really puts the scale of the task into perspective, and scuppers much of the likelihood of the work being done.
The weird thing is “how did they know”? They had come into contact with the materials in different ways, but mostly, they suspected that we had more stuff.
So, the researcher has a clear message: the job of the library is to catalogue things well before they are needed. It really is common sense. Find a manager, tell them.