This blog post was originally published by me in NYPL’s SpecialCollections.txt channel on July 6, 2017.
If you work in a cultural heritage institution, especially one that handles archival materials, it is helpful—even necessary—to have a system in place that tracks and monitors objects. This concept sounds deceptively simple, yet it is not. This became especially apparent when the NYPL’s Special Collections Department was created in 2015, bringing together several units charged with the description, management, and preservation of the Library’s special collections. At a Library-wide scale, tracking, monitoring and performing assessments of objects becomes complicated without a system that sufficiently understands all objects’ relationships with themselves (more about this later) and the spaces they inhabit over the short- and long-term.
In an effort to address this issue, key staff of Special Collections came together to form the Object Data Model Group. Over a series of months we thought conceptually about what we wanted an ideal system for object management to accomplish.
Physical Objects and Their Relationships
One of the first questions we asked ourselves was: What is a special collections object? Answering this was an important first step, because we wanted to have a basic conception of what we wanted our ideal system to track. What we determined was that objects can be divided up into two general groups: the first being something that contains something, but has no intrinsic historical or archival value in of itself, like a box or folder; the second being a collection object, like a letter, video, print, book, photograph, or map. Sounds simple enough: however, things started to get complicated when we began to map out their relationships to one another. For example, our conservation department will often repair items from NYPL’s special collections in preparation for exhibition, digitization, or to halt degradation. To a conservator, an object usually starts off as a single map or an album of photographs, depending on the project.
Let’s say that a conservator receives a photo album whose backing is acidic and therefore harmful to the photographs adhered to it. They decide that the best course of action is to separate the photographs from the offending material. So, something that starts off as a single photo album may, at the end of treatment, turn into 20 single photographs. What this demonstrates is that, in terms of a record for an object, you cannot assume that it will always refer to one and only one thing: depending on its needs over time, it can potentially transform into more things via common library procedures. Our ideal system, then, would be able to create nested relationships between objects in the event of a procedure, like a conservation treatment.
When you start to think about how an object relates to its digital surrogates—which we also conceive of as a type of object—things start to get even more tricky.
For example, a technician who works at NYPL’s Preservation of Audio and Moving Image (PAMI) lab may conceive of an object as a single reel-to-reel tape. However, upon transferring the reel-to-reel into a broadcast-quality digital audio file, they may discover that the tape, labelled “Recorded Stuff”, contains a variety of recordings the creator made at various points of their life in a variety of recording speeds, both slow and fast, and on a variety of topics: an outdoor concert on one part of a tape, a recording of an interview off their television set on another. The PAMI technician performing the transfer may then have to create more than one digital file, since they will have to stop the transfer upon the change in tape speed, recalibrate their transfer machines, and then start another transfer session. This obviously differs from, say, a scan of a paper document: generally speaking, a photograph will produce 1 digital surrogate, or have a 1:1 relationship. Of course, there are always exceptions to this: a Digital Imaging Unit lab tech recently told me about the process of photographing a 12th-century Japanese scroll, which required a series of photographs to be taken. Either way, it is safe to assume that, with digitization procedures, a single object can yield many digital surrogates.
The Right System
In the midst of all Special Collections operations (mainly: inventorying, processing, description, transferring, conserving, and assessing/monitoring), items are being physically moved from and to a variety of places, both physical and virtual. Objects are likely to be moved from at least once from one part of the library’s network of research libraries throughout Manhattan, to the Library Services Center in Long Island City, Queens for processing and then to a long-term storage facility in Princeton. Digital preservation masters are placed in repositories, and service copies are stored in the cloud. In addition to this, items may get shipped to a vendor to be digitized, be loaned out to institutions for exhibition, or [insert movement here]. And, most importantly, they are being requested for research purposes by our patrons and visiting scholars, and will be called up and moved to be serviced out in a reading room or other research area.
The question that those of us meeting in Special Collections have been trying to answer is whether or not there is a system that exists that sees objects not as static items with a lowest common denominator object (a box) as the be-all, end-all of the object, but as dynamic objects. By dynamic, we mean objects that can have relationships not just with other objects, but with themselves. In addition to this, they system should also be able to establish relationships between objects and the spaces they occupy over the long- and short- term. We believe that if these relationships can be expressed in a single system, we may be better poised to better understand our special collections as we digitize, treat, monitor and steward them throughout their long lifespans.
In our investigations, we have found that some existing systems, especially those built to track museum objects, have some, but not all of the desired functionality we would require to track objects within the special collections/archives universe. We are very curious to hear about how other archives and special collections, especially those dealing with collections on a scale comparable to NYPL’s, track and monitor objects. What systems do you use? Do they accomplish what you need to understand your collections, or do you wish they did more? Some of NYPL Special Collections staff will be travelling to the Society of American Archivists conference in Portland, OR this July, so we hope to also pose these questions in-person. We would appreciate any and all feedback!