Hello dear readers!
This post is for anyone who wants to learn a little about the way we here at Special Collections and University Archives digitize our analog materials for online use and the platform where some of these can be accessed. The IAWA digital library platform (a selection of IAWA collections) gathers images that have been systematically digitized and described by University Libraries’ staff over the last decade. I was motivated to write this post now partly because new content is available on the platform and, while this is an opportunity to highlight these collections, I also want it to serve as a reminder that the platform is a growing, actively-maintained resource. But, as is the case with any resource, it may occasionally require some user guidance. First, here are a few of those newly available collections to explore:
- The Elsa Leviseur Collection: https://iawa.lib.vt.edu/collection/4d44p20b
- The Vera Jansone Drawings: https://iawa.lib.vt.edu/collection/v001qm0w
- The Women’s Development Corporation Records: https://iawa.lib.vt.edu/collection/rk831m8n
Second, and one of the many caveats I’ll give in this post, is that the “user’s guide” of the title may be slightly misleading. I will not be providing a step-by-step tutorial or walkthrough of the platform, but will be setting out some loosely organized tips, thoughts on digitization, and some limitations for the curious researcher to be aware of.
How use? Some tips on information foraging
Because I’m an archivist obsessed with context (and I’m familiar with the site’s limitations), I will generally recommend browsing over searching. On the home page of the platform there is a prominent search bar, and just beneath that there are two icons: one for browsing by collection and one for browsing by item. A lack of context can make browsing by item somewhat confusing for the layperson or casual researcher, while drilling down through a single collection can be a good way to get your feet under you while you explore. What makes search hit-or-miss is that there are inevitably inconsistencies in metadata (depending on who created it) both in the way subject terms are applied and overall completeness (more on that below); occasionally, something simple that was missed in quality control, like a misspelling, can mean that an item fitting a user’s search criteria will not appear in their search results. Moreover, users will typically use queries that mirror natural or casual language use rather than formal subject headings or tags.
I will also note that the terms “collection” and “item” are used in an unintuitive way on the platform, which may reflect my own bias as someone who’s familiar with the physical arrangement of the manuscript collections represented on the site. From an archival perspective, “collection” is more intuitive because it corresponds with and is grouped under the creator (this is how archival records are organized, according to the principle of provenance), whereas an “item” on the platform refers to a grouping of physical objects (a real world folder, for example). Whereas in the real world environment I would consider an item to refer to a single drawing, sheet of paper, etc., an item on the platform is an aggregation that may represent anywhere between one object and two hundred (technically, there is no upper limit). The images below show what it looks like to move from the collection-level page to selecting and viewing an item.
These images are meant to show what it looks like to browse and click through the platform. Here I’m using the Elsa Leviseur collection to show how one might navigate in a top-down manner (from collection down to item). The collection page (screenshot 1) contains identifying information, like the manuscript number, and a biographical sketch of its creator. Scrolling down the page reveals the “items” in the collection (screenshot 2 above). As noted above, these items refer to aggregations of varying size (usually corresponding to physical folders). Screenshot 3 then shows what it looks like to select one of these items. The image viewer is populated with the first image in the “item,” here a greeting card, which can be manipulated using the zoom and reset functions in the bottom righthand corner. Below the image itself is a carousel showing thumbnails of the other objects in the item (in this case the entire item is comprised of 21 images). On the righthand side of the page there’s a title and item-level description, which serves as a good illustration of metadata limitations for collection-scaled digitization and search functionality. Metadata, often used interchangeably with “description” in archives, cannot be done at the level of individual objects because there is too much material, and description is therefore mostly done at the aggregate level. The problem here is a lack of granularity that doesn’t allow the user to search with specificity. The image in Screenshot 3 is obviously a collaged greeting card, a piece of information nowhere listed in the metadata. Other information visually available (but not searchable) is the text in the image itself. Although it may not be immediately apparent to a casual user, “Fleischmann” was Elsa Leviseur’s married name, and it can be easily inferred that this is a family greeting card with pictures of her children. From this small example, one can easily imagine how much subject matter expertise and time it would take to give a full account of every object in every collection. Because every image is just one among many, the goal is to give objects a title that accurately (and briefly) describes varied content so that a given user can make educated guesses and inferences. In other words, some precision is sacrificed to accommodate more content. The researcher still needs to use their judgment when finding and interpreting these materials, and actively engage in the discovery process.
My recommendation to browse by collection notwithstanding, there is an obvious benefit to browsing the platform’s content at the item-level: it allows you to look at examples of the same kind of content across collections. There are a number of filters that can be applied to return a subset of items. For example, if you expand the options under “Medium” on the lefthand navigation menu, you can isolate the items tagged with “charcoal” or “photographic prints.” The gallery of images below are screenshots showing the ways these facets can be selected (note: filters will not automatically deselect when you apply a new one so you will have to manually toggle filters on and off as you go). While imperfect, these filters go some way toward allowing users to remix and reuse materials in ways that would be onerous, if not impossible, in a physical reading room.
There are other ways to use the platform, of course. We cannot possibly anticipate every type of use, but my hope is that this primer will answer a few questions about why things are organized the way they are and why some metadata tends toward vagueness rather than specificity. My final caveat for this section is that the digital library platform, while a useful tool, is not the place to search across all IAWA collections (see below).
The broader context of digitized content
At the beginning of this post, I emphasized the word “selection” because the digital library platform is not a comprehensive or complete representation of the IAWA, but in fact a rather small subset of incompletely digitized collections that vary dramatically in size and content. All this is to say that there’s often a misconception that archives can and should “put everything online.” Most archives simply don’t have the institutional capacity, meaning staffing, to support this work at scale. In the case of the digital library platform, the IAWA instance (there is a general instance for University Libraries and an IAWA-specific instance, which we’re discussing here) has largely been seeded by work on a CLIR (Council on Library and Information Resources) “digitizing hidden collections” grant-funded project that lasted from 2017 to 2019. During this period, several team members were hired (myself included) who were entirely dedicated to this project, which included image capture (scanning), metadata creation, and macro-level project management.
Understandably, there’s some user confusion that tends to happen around online representation of analog content. I sometimes field questions from patrons who mistakenly believe that when they search the digital library platform, where a subset of images live, they are searching across all IAWA collections. Typically archival institutions provide access to their materials in multiple ways, but the primary method is by describing collections in “finding aids,” a term of art meaning a resource that contains contextual and structural information about a collection. The context generally consists of biography, provenance, and descriptions of the scope and content of a collection; the structural information usually consists of a box/folder inventory where folder titles are indexed to a particular container (a box, for example). “Finding aid” usually refers to one of these specific inventories (I call it a “table of contents” for primary source material), but it can also refer to a more general guide like the one linked here: https://guides.lib.vt.edu/iawa/collections. This list of collections (ordered alphabetically by creator name) is probably where I would point someone starting their research; another place I would point them is the platform ARVAS (Archival Resources of the Virginias), an aggregator that can be used to search for and across finding aids. There are many ways to find and learn about our collections, even if at first glance it’s frustrating that information lives on different platforms that don’t talk to each other.
Even when we’re just talking about IAWA materials, let alone ALL Special Collections and University Archives materials, what’s available on the library platform is a fraction of a fraction of what is digitized, which in turn represents a fraction of all analog IAWA collections (not to mention born-digital records). Digitization in the context of everyday archival work is mostly done in a piecemeal ad hoc manner according to patrons’ requests. Some of these patrons are architectural historians, while some are homeowners interested in original drawings for their home, and others are family members who want to feel a sense of connection with the work of their grandmothers, mothers, aunts, or sisters. The question of how this case-by-case workflow differs from the one for ingesting collections into a platform is often the difference in the amount of context needed by a specific (and interested) researcher and a general audience coming to collections with no frame of reference. More specifically, what I mean is that it takes time to create quality metadata that supports understanding, reuse, and physical collocation for general users that remains useful long into the future. It’s not so much that creating metadata is not a good use of time, but since it could go on infinitely, it’s a question of priorities and resource allocation. For this reason, the amount of programmatic (systematic as opposed to ad hoc) digitization that takes place in archives varies from institution to institution. Even so, I tend to view digital reproductions as just another discovery pathway for the archives. They cannot (and should not) “replace” the original objects nor will they be sufficient for some forms of research, especially in design pedagogy where the full visual impact and materiality of the original object can be critical.
Thanks for reading!
























































