The Twilight Reference Zone
Picture if you will an American public library. Any library will do. A smartly dressed, clean-shaven man in his mid-40s approaches the reference desk.
“Can I help you find something?” asks the librarian.
“Yes,” says the man, “I’m looking for a copy of Bats and Bones, a Frannie Shoemaker Campground Mystery.”
“Let me see if I can find that for you,” says the young woman behind the desk as she checks FirstSearch.
“Just one thing, please,” asks the man as he waits. “I’m looking for a specific edition.”
Scrolling past titles, the librarian says, “Oh?” without looking up.
“Yes,” says the man. “I’m looking for the edition that will be published in the year 2102.”
[Cue creepy music.]
This library has entered…the Twilight Reference Zone.
The story you have just read is true. Dramatized…but true. It’s from a real ILL request for that very Campground Mystery, to be published nearly 90 years in the future. How can this be? You be the judge. Did the requestor:
- Know something about the future of kids’ mysteries that we don’t?
- Arrive at the library in a TARDIS?
- Make a typo when entering the request?
I’m going to choose to believe “option 2,” since it’s the most fun.
At OCLC, we have records for ILL requests for items that will be published as far into the future as 2915 and as far into the past as 429. So maybe Dr. Who is just a big fan of interlibrary loan.
Or it’s a typo.
But it’s a typo that points out two, time-related aspects of ILL that don’t get as much attention as they should: the future importance of transactional data and what we can learn about the past from the publication dates of (non-time-travelling) ILL requests.
All data is future metadata
We’re used to thinking about transactional data as a kind of throw-away byproduct of a process. We may aggregate some of it (number of check-outs, total circulation, aggregated requests by subject). But each individual activity may contain information that isn’t tracked.
Until it is. And since digital storage space is now inexpensive and fast, we tend to capture a lot of data that can become metadata for better understanding our systems and users. An easy example of this is thinking about ILL as an indicator of collection strengths (loans) or weaknesses (requests). Not a particular item—that’s pretty easy to spot if you find yourself requesting the same resource over-and-over—but of a certain subject, industry, author or…time period. And if the transactional data in an ILL request has an error, that’s going to skew the aggregated results that you’re using for evaluation.
This isn’t meant as a guilt trip about typos. It’s a reminder that your transactional data has great value to your future analyses. The ILL request you fulfill today is a data point for a librarian of the future.
Looking back at looking back
In our investigation into “questionable” past and future ILL requests, we found that prior to 1905, the most requested ILL publication year was 1770 with 724 requests. 1776, by comparison, had only 494 requests.
What was so special about 1770? Beethoven and William Clark (of Lewis and Clark fame) were born. The Boston Massacre occurred. But those wouldn’t have triggered published editions from that year. What are the most popular requests published in 1770?
- Archaeologia, or, Miscellaneous tracts relating to antiquity (Periodical): 336 requests
- The lady’s magazine: or entertaining companion for the fair sex, appropriated solely to their use and amusement (Periodical): 26 requests
- The Royal Danish American gazette (Newspaper): 11 requests
- The Bath chronicle (Newspaper): 9 requests
- Rise glory rise (Musical score): 9 requests
- Encyclopédie, ou, Dictionnaire universel raisonné des connoissances humaines (Print book): 8 requests
- The Nova Scotia gazette and the weekly chronicle (Newspaper): 7 requests
- Methode pour apprendre facilement la musique vocale et intrumentale, ou tous les principes sont developés avec beaucoup de clarté, et cent leçons dans le gout nouveau à une et à deux parties, ce qui enseigne en très peu de tems a solfier sur toutes les clefs, toutes les mesures et tous les tons (Print book): 7 requests
- The Massachusetts spy (Newspaper): 6 requests
- Instructions succintes sur les accouchemens : en faveur des sages-femmes des provinces. Faites par ordre du ministere (Print): 6 requests
That first item is more than ten times more popular for ILL than the next one.
Why is that? I don’t know. It’s held by 473 libraries, so it’s not that rare. But it’s an interesting data point for someone doing research on Great British periodicals, I’d think.
The important thing for us to remember is that measuring the use of library collections may sometimes be as interesting as the materials themselves.
So whether you’re helping populate transactional data that will inspire and inform future librarians, or delving into the use of materials from the past, remember that ILL can be more than a source of materials for today’s use. It can be the fuel for librarians’ time-travelling expeditions.
Question… if you could make an ILL request for an item from the future, what would it be?