The Archive Relational Model is a new approach for recording and presenting images and information about people, places, events, organizations, and objects in the real world. It provides archivists with a structured, but easy to use method for presenting their collections online in a way that engages users by making it fun and easy to learn about local history.
The model was developed by the Southwest Harbor Public Library in Maine following a lengthy search for software that would showcase the Library’s large collection of historic photographs, documents, maps, and research material. The search turned up solutions that display this kind of information, but none that reveal the real-world interconnections among people, places, and objects that make our lives rich today and make history so fascinating. Unable to find an existing solution, the Library opted to use Omeka to create the Digital Archive Software and share it with others.
The basis of the Archive Relational Model is that each item in the Archive serves as a stand-in for something in the real world such as a man or a boat. The key to understanding the model is to think of the item for the man as the man himself, and to think of the item for the boat as the boat itself.
Now consider another kind of item in the Archive, a photograph of the man on his boat. Obviously the photograph itself is not the man and it is not the boat. It’s a piece of paper with chemicals that formed an image showing the man and boat during a moment in time.
Now let’s reinforce the concept of an item being a stand-in. Clearly, a database cannot store the flesh and blood of a man, the wood of a boat, or a photograph’s paper and chemicals. It can however store information about the man’s life, the boat’s history, a digital scan of the photograph, and information about the photo such as when and where it was taken. This information, called metadata, tells us what the item is a stand-in for and provides a place to record how one item is related to another.
In the real world, the man owned the boat, and the photograph depicts the man and the boat. The words “owned” and “depicts” are relationships. The Archive Relational Model lets us record these very same relationships between the items in the database that stand-in for their real world counterparts.
Note that the phrasing of a relationship depends on your perspective. The man owned the boat, but from the perspective of the boat, it was owned by the man. The owned by relationship is simply the inverse of the owned relationship. The model and the software allow archivists to establish relationships between two items from the perspective of whichever item they happen to be working with.
Most relationships are established between what are called Articles in the model. An explanation of this key concept is deferred until later following further discussion of relationships.
The Importance of a Relational Model
Relationships are what make the world go round. A man and a boat by themselves are not very interesting until you learn that the man who owned the boat tended a lighthouse located on an island where the man and the woman he was married to were the parents of eight children who were students at a one room school house that has since been restored by the historical society!
An archive that does not reveal the relationships among its items, serves only as a repository of information that users must search to locate items of interest. It’s like fingering through a library card catalog to locate a single book. In contrast, using the Digital Archive is like seeing a book on the shelf along with the related books that surround it. By looking at the shelf, you discover other books that you didn’t even know about. The Archive Relational Model enables this discovery process in the Digital Archive.
The discovery process is fun, but more importantly, it is an essential feature that an organization must provide if it wants the public to fully appreciate the content of the collection. However, to provide users with this benefit, archivists must establish relationships between items in the collection. Fortunately, it’s easy to do and can be performed little by little over time as resources allow, and users begin to benefit immediately.
The Archive Relational Model both encourages and enforces good practices for entering item metadata and establishing relationships between items. The user interface for adding relationships is quick and easy to use and it prevents an archivist from inadvertently setting inappropriate or non-nonsensical relationships such as a man married to a boat. This enforcement of referential integrity is possible because of relationship types and relationship rules.
When an organization first adopts the Archive Relational Model, it defines the kinds of relationships that make sense for items in the collection. The organization also specifies rules that must be followed when an archivist adds a specific relationship between two items. An example of a rule is that only a person can be married to a person. Another rule is that a photograph can only depict an item that is a stand-in for a person, place or thing. For example, a photograph cannot be married to a person, because as discussed earlier, the photograph is not the person, but only an image of the person at one moment in time.
In the Digital Archive, an item that serves as the stand-in for a person (or boat, house, business etc.) is called an Article. The concept of an article is central to the Archive Relational Model. The next section describes the meaning and use of articles and explains why they are so important.
The term Article refers to an Omeka item where the type is set to either “Article, Text” or “Article, Document”. An Article serves as a stand-in for a real world entity such as a person, place or thing. “Article, Text” means that information about the entity is in the item’s metadata. “Article, Document” means that additional information is contained in a PDF document attached to the item.
Like a Wikipedia article, an Article in the Archive Relational Model contains encyclopedic information about its corresponding real world entity. However, unlike Wikipedia, an Article in the Archive Relational Model may contain very little information, but serve as a placeholder that “glues” other articles together as will be explained in the next section.
The purpose and importance of placeholder articles can be explained by an example. Suppose you have two Articles about people, one for Robert and another for Theodore. Archivists learn of a woman named Carol, but know nothing about her except that she was Robert’s daughter, so they create an “Article, Text” item for her that contains only her name. They also establish an is child of relationship to Robert’s Article. Later, while researching Theodore, they learn that Carol was his mother and so they establish a parent of relationship from Carol to Theodore.
Now, when you view the Article for Robert, you see his daughter Carol and his grandson Theodore. Conversely, if you look at Theodore’s Article, you see his mother Carol and his grandfather Robert. Though Carol’s Article is by no means encyclopedic, it glues together grandfather Robert and grandson Theodore, and it provides a placeholder to accept additional information if archivists learn more about Carol in the future.
The use of placeholders to glue other Articles together is what makes the Archive Relational Model so powerful and so different from other models. Relationships are like the lines that connect the dots in those puzzles you did as a kid. Until you drew the lines, all you saw was dots and numbers, but once you connected the dots, you saw a picture! By simply creating a “dot” for Carol and adding relationships, two seemingly unrelated men became part of a three-generation family tree.
The Archive Relational Model in the Digital Archive
The diagram below depicts how the Southwest Harbor Public Library uses the Archive Relational Model in the Digital Archive. Another organization would apply the model differently depending on the nature of its collection, but Articles should be the heart of any application of the model.
Click the diagram to see a larger version
The ovals in the diagram are the Digital Archive’s primary item types and the arrows are the relationship types allowed between those item types. A list next to each oval identifies secondary types, for example Book is a specific type of Publication.
The long list on the left side of the diagram names relationship types that are only allowed between two Articles. As you can see, just a handful of other relationships like depicts, about, and refers to, are used to relate an Article to a Document, Gallery, Image, Map, or Publication. That’s because the list on the left contains the real world relationships between people, places and things, whereas the relationships that appear on the arrows are more clerical in nature. In all cases, a specific relationship between two items is only allowed if the relationship’s rules permit it.
Notice that relationships are allowed between two Articles (the loop labeled “relationship” from the Article oval back to that oval) or between an Article and another item type. Non-Article item types are generally not related to each other. So for example, a photograph cannot depict another photograph. The two exceptions are an annotated map derived from a base map and the fact that a gallery contains items of other types.
For additional information about the Archive Relational Model, please contact George Soules by sending email to email@example.com.