Video: “Visualization of Heritage Science and Humanities Data” by Fenella France (Library of Congress), an Ignite talk for the Openlab Workshop Unconference, December 1, 2015, in Crystal City, VA. Published on Jul 1, 2016

All Ignite talks



Fenella France, Library of Congress

Hi. I'm Fenella, and if you don't already know where I work, that's great, because I want this presentation to be about what I'm talking about not where I currently work.

Visualization of data. How can we bring our collections alive for any audience and make it engaging? We really need to make it more interactive and effective in the way we interact with the data that we have. We need to make that accessible and standardized.

So imagine you have a global map of your data. How does that look? How do we make it effective and link it back to the actual object or collection that we working with?

So imagine we're looking at the Gettysburg Address, and as you look at that, we can start to pull up all the different types of scientific data that tell us about the content: the handwriting, the particles, the processing. What does the ink look like, and the thumbprint, which might be Lincoln’s.

How do we connect the scientific and the digital humanities together? The hyperlinks between that, that help the curators think about the collections, but no it’s not connecting, it’s integrating. We’re integrating that data together. We’re making that social media component to bring the data alive.

What are some of the challenges? It’s transdisciplinary, not just multi. We are going across, we are not segregating our disciplines, we’re integrating them. We are linking the content back to the original object, and we’re trying to get away from the proprietary data components.

Just an example of large data when we talk about spectral imaging. We can get one to two gig just from one capture of an image. What does it mean? What does it look like, and how do we actually access those components together?

So what we actually see with that is what I affectionately call shooting in the dark. There’s different information in every one of these wavebands that we capture from the original object, and then we can take that further and take it a little bit to what do we see within each of those bands. You'll see things start to appear and disappear as we go through all the different bands here, and the way we process that date brings different parts of it alive.

Also we can take that a little bit further, and say how much data do we get from one single document? We can get upwards of 24 to 32 gigabytes.

There's a lot of data, but with public domain we want to make that accessible. We also want to show the different components of the hidden text that’s within our collections that we didn't know was there.

So taking it back to the challenges that we have. Well manufacturers don't really like making metadata available. We’ve got to push through and make them want to share it. We can do that. We're doing it, and researchers are restricted in how they access that if they don't have the same software. We have to push through that.

The one token slide. Don't be afraid. Underlying this visualization we had to have a robust database to support that and make it available, and easy to integrate data into it. So an example of bringing that alive.

The Waldseemüller 1507 World Map. The first map to refer to America. When we look at this with different types of imaging, we can look at the first reference to America there, and we can look at central sheets that originally had cartographic lines that had faded over history.

Well those don't exist anymore, but with this special imaging we can bring those back. We can show where they were, we can do a little bit more processing, even so and show where they start and finish. For a curator that's really important back in 1507. As a scientist I need the curator to tell me that it's important. We have to integrate.

We can then look at how do we process that differently to see what the original woodblock might have looked like. What tools did they have in 1507? We're pulling the curatorial and the scientific together, and making it come alive.

But wait a minute. 1507. That’s a long time ago. What did the world look like in 1507 compared to now? When we spin on what they had in 1507, they knew a lot about the bottom of South America there. What journeys were people sharing? How do we link that together? So let's bring that together again. When we’re layering this data in terms of all the different types of scientific information, and we are linking it back to what the curators know and understand about these collections. Making it alive, linking it, layering it.

Taking it one step further. What’s the geospatial component? What parts of our collections aren’t linked yet that we know about?

So here’s a 1513 Ptolemy Geografia. When we looked at this, and we went through some of the different papers that we used in it, we found a crown watermark. You know that might be kind of interesting, but what's even more interesting about that is that we could actually link the Ptolemy with the watermark from the Waldseemüller.

Well cutting a really long story short we have now put those two documents together and everyone who worked on them. They start in 1506 in St. Dié. A funder comes alive to print the Waldseemüller map. He dies. They regroup. They go to Strasbourg. They finally get better printing the Ptolemy Geografia. You didn't know that they could link those documents together.

So what do we have? We need to integrate the visualization through the object. We need to create a shared language, and we need to have a transdisciplinary linking breaking down the silos.

Thank you.