Tuesday, March 5, 2013
[ This is a guest post from Bonnie Lewis, a researcher at the University of Kentucky’s Center for Visualization and Virtual Environments. Bonnie has been working with the digital imagery of the biblical manuscripts from Lichfield Cathedral, captured in 2010 by a team led by Brent Seales, who was the director of the U.K. Viz Center. I was fortunate enough to get to be present and help with this digitization (the Center for Hellenic Studies donated some equipment from our work on the Homer Multitext). In this post, Bonnie describes the important work being done in Kentucky toward bringing these manuscripts to a scholarly community in useful ways. At Furman, we have been developing editing workflows that we hope will allow us to contribute to Bonnie’s work of capturing the full semantic complexity of the St. Chad Gospels, Lichfield’s Wycliffe New Testament, and other biblical manuscripts. The promise of a fully-integrated publication of these manuscripts is most clear from Bonnie’s description of her treatment of this challenge, a treatment that is both innovative and rigorous. — C. Blackwell ]
The Vis U program at the University of Kentucky has been working on several projects this summer. The one I am involved with is called InfoForest. This project is a system that works through Apps that can display data about ancient manuscripts (images, XML, and media) in a way that enhances the meaning of the documents and makes them available to a wider audience. What is unique about this system is that it has been designed to function on several platforms and to grow as our “Forest” of knowledge grows. It was built with the intention that it will be easily replicable.
Our project really began in 2010 when a research group from the Vis Center traveled to Lichfield Cathedral in England to image the Gospel book. They took Multispectral, 3D, and RGB images of each page. This was a part of a larger project called FoLIO. These images are where we primarily pulled the data for the Chad Gospels to put into the App.
My role in the project, as a History Major and Classics minor, was to assemble the data in a way that was organized and enhanced its meaning. I also saw my unofficial role in the project as trying to figure out what will be meaningful to a wide audience of users.
What that came down to in the project was that I wrote XML that corresponded to the images of the manuscript. People at Furman University had already done this for Matthew, so I finished out the rest of the manuscript, trying my best to mimic the XML structure they used for the first book. In the XML, I marked the line breaks, page breaks, verses and variations in words in the manuscript. My intent in this process was to create data about the manuscript that was informative organizationally (the Chad Gospels do not mark verses in the text) and easily searchable as many words are misspelled in the Latin and need to be able to be searched in their correct form as well as their incorrect. This process took the better part of 5 months.
I also organized the images of the Chad Gospels in a way that would make it easy for the server to respond to requests from the devices for a certain page of the manuscript. This meant that I, with the help of John Broadbent (another student working on the server side of the project), had to correctly size and name around 20,000 images. This was done through the use of scripts created by John that could rename and modify folders en mass. These scripts can be used on other data sets to resize and rename other images.
The result of this process was that we came up with a method to organize these sets of data in a way that was standardized and replicable for other data sets. There are still aspects of the project that will not be able to be as easily automatable as other parts were. For example, the transcription and mark up of the documents needs to be done, for the most part, by hand and is a time intensive process. However, in the XML, we were able to follow TEI standards in the mark up and were also able to establish a framework for writing the XML that can be replicated for other data sets.
Throughout the project, our goal was to put all of this information up on a server at the Vis Center. However, about two weeks before our deadline, the University shut down our server. Thankfully, with the help of Dr. Blackwell, we were allowed to put our information on the Furman Server. This whole project has been a collaboration of many minds and resources, but I love how the final weeks highlighted how dependent the project was on collaboration. The devices we are using to show the apps in Kentucky are pulling information from a server in Houston that was modified by Dr. Blackwell in South Carolina. There is still much work to be done, as we were only able to build a proto type this summer, but it is exciting to see it come together. One of the other students working on the project described what we are doing as giving old manuscripts, new life. I think that summarizes what we did and are doing perfectly. It is exciting to see new life breathed into these manuscripts and watch the information the hold begin to come alive to new audiences.
University of Kentucky