text encoding

A Library Intern’s Maiden Voyage through Digital Publication in the Antarctic

This post was authored by Suzanne Raybuck, Intern with the Office of Digital Research and Scholarship in the Fall of 2019. Suzanne recounts her experience working with Special Collections Materials and creating a digital publication interface to display it online. The final version is not yet live, but this post contains previews of the interface.

A sturdy, metal oil lantern about a foot tall with grey metal and discolored glass protecting the wick.

A Hercules emergency oil lantern from Operation Deep Freeze.

When I originally was brought on as the Digital Publication Intern for the Office of Digital Research and Scholarship, I had virtually no concept of what I would be doing in my new internship position. But, very early on I knew that I wanted to work with the Robert E. Hancock Jr. Collection at FSU Special Collections. The Hancock Jr. Collection is a collection that “contains materials regarding military operations in the Antarctic, primarily focusing on the Operation Deep Freeze II mission.” Based on that description, it’s a safe conclusion to assume it contains lots of important and scholarly documents and artifacts. However, it also contains various memorabilia from Robert E. Hancock Jr.’s time in Antarctica (including many, many, tiny penguin figurines, a drawing of Mickey Mouse shaking a penguin’s hand, model navy destroyer ships, lumps of coal, emergency lanterns, and military rations). This wonderful collection of artifacts is endlessly fascinating because it provides a series of vignettes of life at the South Pole in the form of really fun random objects.

A highly detailed model of a navy wind-class icebreaker ship. Below the water line the ship is painted red, while above it is a steel gray. The deck is littered with various apparatus such as guns, lifeboats, radar equipment, and cranes.

A model of a wind-class ice breaker ship USCGC Southwind which participated in Operation Deep Freeze.

I found this collection while searching through Special Collections for a fun series of documents to use as guinea pigs for a new publication system we were testing. Essentially, I needed a bunch of documents in similar formats that we could transform into digital objects and then use to test out different publication tools. After spending maybe an hour with a variety of fun models and pictures, I found the Operation Deep Freeze Newsletters nestled into a box of other periodicals from Antarctica. The Newsletters were published by the army to send to the families of servicemen who were in Antarctica to let them know the news from the various bases. The Newsletters were mostly written by incredibly bored servicemen just trying to pass the time in their freezing posts. This boredom resulted in the inaugural newsletter detailing the long and involved process of how a band of grizzled soldiers tried to hatch live chicks from commercial eggs for the upcoming Easter Holiday. I had definitely found my guinea pig documents.

A very basically formatted newsletter titled OPERATION DEEP FREEZE NEWSLETTER. The text is somewhat faded and the first article details how several servicemen from Little America Station selected six eggs for incubation in a dental incubator to try to hatch live chicks.

The original front page of Volume 1 Issue 1 of the Operation Deep Freeze Newsletter.

After finding the newsletters, I was tasked by our Digital Humanities Librarian, Sarah Stanley, with first encoding these newsletters in a data-rich .xml format called the Text Encoding Initiative, or TEI, and then figuring out how to publish them online. To accomplish this, we had to take into consideration three key factors: maintaining the format of the newsletters, good display functionality (e.g. tables of contents, hyperlinks, page view/scroll view), and how easy it would be to use. With these in mind, I started trying out different publication methods such as eXist-db’s TEI Publisher, which proved to be a challenging introduction into digital publishing.

eXist-db is an XML database tool that can be used to build web applications. We used the TEI Publisher package to create a digital collection that would use our TEI data format and present it in a clean and simple interface. The process of generating an application was intricate and required lots of specialized knowledge of both TEI files and their accompanying customization files. Additionally, we had no idea how the digital edition would look before we generated an application and viewed it, so if some small part of the display of the edition was off, we would have to delete the app, minimally adjust our code and generate a new app from the very beginning. Once we did get a finalized version generated, the overall look and feel of the page was exactly what we had hoped: very clean and easy to read. However, because we were using a program to generate the app for us, we had a very limited capacity to tweak the website interface and design or add our own custom parts to the whole thing. Ultimately, the the functionality of eXist-db did not quite meet our needs, and we tried to find a solution that would let us get a bit more hands-on with our edition.

A screenshot of a webpage titled TEI Publisher with the content of the first Deep Freeze Newsletter in it. The interface has a search bar and small arrows to navigate between pages, and all the text is displayed in a old-timey typewriter font to mimic the original newsletter.

A screencap of the eXist-db interface we created, its very clean and easy to navigate but at least four iterations of apps went into getting this particular layout.

Another possible publication tool didn’t arrive until the next semester, when I was working on publishing a collection of poetry translations online. Sarah pointed me towards a Jekyll (static website generator) template for minimal editions called “ed”. After looking at the examples, the display was again very clean and easy to interact with, so we decided to give it a shot. After deploying some quick test sites, we found that it was incredibly easy to work with and consistently generated beautifully designed websites that intuitively displayed our editions. It also had a built-in search function and annotation, which we were looking for in our poetry project. The only problem was we had to translate our TEI format into markdown, which caused us to lose huge amounts of metadata and information about textual styling that would be useful to other researchers. We made a judgement call and decided to keep looking for something that would preserve our format while giving us all the functionality and display options that we found with ed.

A screenshot of a webpage titled ‘A Wilfred Owen Collection: Just testing this out’ with the text of the first Operation Deep Freeze Newsletter below the title. The interface is very simple and appealing with dark red accents and a collapsible menu bar.

A screenshot of our test site for ed. Most of the sample texts we used were poems from Wilfred Owen, hence the name. Here you can see that the layout is slightly different since ed automatically creates larger title text. Unfortunately, we had to change all our TEI files to markdown, which got rid of most of our metadata.

The final option we looked at was a JavaScript library called CETEIcean, which takes TEI files and translates it directly into HTML. With a single script added to any existing HTML page, we could take our TEI files and easily publish them. Again, we started making some test pages and playing with the code and quickly ran into a problem. Because CETEIcean is just a JavaScript library, it doesn’t automatically build websites for you like with existdb and ed. If we used CETEIcean, we would have to make every single page on our website from scratch, repeating tons of HTML and JavaScript along the way. Sarah was enthusiastic about using CETEIcean since it did arguably check all our boxes, but I wanted to find a more efficient way.

In the end, we settled on using a combination of CETEIcean and ed along with chunks of original code to create our own web application which we named Pilot: Publishing Interface for Literary Objects in TEI¹. We essentially used the quick and intuitive page generation from ed, the javascript transformation of TEI from CETEIcean and mixed it together all running on a node.js server. Because we made Pilot from scratch, we can include or add all the functionality we want such as annotation, interactivity, and variant readings of the base newsletters.

A very basic interface that has a navigation bar at the top labelled ‘Pilot’ with links to a homepage, collections page, about page and contact information. The contents of the Operation Deep Freeze newsletter are again reproduced here with very basic text styling and layout.

A screenshot of what the first draft of our Pilot interface looks like. This page was automatically generated by the server file after reading a folder of TEI files, transforming them to HTML, and finally running them through three templates to get the desired display.

Though this project was long and frustrating, it ended up teaching me one of the most important points of digital publishing: digital representation of texts adds to the work, rather than merely representing it. Digital publishing is at a unique intersection where we have to negotiate the appearance of the facsimile, the functionality the editors want, and the demands of a digital medium. With all of these competing agendas, it’s hard to remember that a digital edition is a creative opportunity. With the vast array of tools offered by the web, developers can take advantage of things like interactive elements, user input, and different types of media to create editions that can only exist in digital spaces. In a way, digital editions represent a new kind of edition that acts more like an archive; where researchers can explore a digital space to find artifacts that are curated through organization and interface.

We plan for our iteration of the Newsletters in Pilot to allow for full-text searching, public annotation, different readings, and interactive displays. With these new features, we hope that the Newsletters will be read and understood in entirely different ways than their paper counterparts, and allow readers to interact with such an engaging yet little known collection.

Notes

¹ As an homage to CETEIcean (a pun on “cetacean,” which means “of or relating to whales”), we decided to keep with the whale theme and name our project after the pilot whale.