>> From the Library of Congress in Washington D.C. ^M00:00:04 [ Silence ] ^M00:00:18 >> Mark Sweeney: Well good evening, ladies and gentleman. I'm Mark Sweeney, the Associate Librarian for Library Services, and on behalf of Dr. Billington, the Librarian of Congress, it's my privilege to welcome you to the 8th Annual Kislak Lecture. This series of lectures, which showcases the work of the foremost scholars, working in the subject of the early Americas, is one of the many great benefits to the library and a nation that has resulted from the library's relationship with one person: Jay I. Kislak who is with us this evening. Mr. Kislak. ^M00:00:55 [ Applause ] ^M00:01:03 As I'm sure, most of you are aware, Mr. Kislak is one of the library's greatest benefactors. As a result of his vision, leadership, and generosity, the library now owns the Jay I. Kislak Collection, an unparalleled assemblage, he created that comprises more than 3,000 rare books, manuscripts, artifacts, through which enumerable facets of the early Americas can be explored and understood as never before. The Kislak Collection provides the basis for the library's premier exhibition, "Exploring the Early Americas," which draws millions of intellectually curious people to the library, from all around the world. Although a decade has passed since the material from the collection first came to the library, Jay Kislak's generosity has been unabated. I'm excited to tell you about his most recent gift which we received a few days ago. It is a manuscript drawn around the 1516 by the astronomer and mapmaker Johannes Schoner. It was part of the portfolio that contained the Great 1507 World Map by Martin Waldseemuller. And also contained his masterpiece, the 1516 Carta Marina. It is simply wonderful that thanks to Mr. Kislak's generosity, is one again complete and is in the library's collection and this available to the world. I was going to ask you to join in thanks, in the acknowledgment of Mr. Kislak, but we've already done that. So, now it's my pleasure, introduce my colleague, Dr. John Hessler, the curator of the Kislak collection at the Library of Congress, who will introduce tonight's distinguished speakers. Mr. Hessler. Dr. Hessler. ^M00:02:55 [ Applause ] ^M00:02:59 >> Dr. John Hessler: Well I want to thank everyone for coming. It's going to be an interesting evening tonight. We're taking a little turn from what our normal Kislak lecture would be, which would be kind of focused on the history of the early Americas and archaeology, to something that is a little bit more timely and a little bit more technical. Now, a couple of weeks ago, I was asked to give a talk at a very imperialistic search engine that will remain nameless. And it was a summit convened by them on what they termed the Computational Humanities. And after my talk was finished, the talk had the catchy title of "The Stochastic Markov Processes, Medieval Manuscript Transmission, and the Algebraic Igon Value problem," which is extremely memorable. I was asked three questions by a rather senior member of the manuscript studies community. A rather senior scholar. A conservative scholar. This was one of those conferences where you look out into the audience and if you've given enough talks at these cross-disciplinary, digital humanities conferences, you see that the field is divided into two groups. On one side, you've got people dressed in black, with little Apples glowing in front of them, and the other side of the aisle is the tweed and Hermes scarf senior academics. And so, this piece of question came from the senior academic side. And the first question he asked is, "How can I, as a senior academic, someone who's been studying manuscripts for so long, understand what it is you are talking about? How is it that I can come to some terms with the things that are going on in the digital and computational humanities?" The second question is he asked was, "How fast is this field moving? Could you have done the research that you have just presented two or three years ago? And how do we keep up with that research?" And the last question he asked was a little bit more poignant towards the humanities themselves. "How is it that this huge influx of scientific imaging, of computation, of digital tools, is changing the way the humanities structure? Is it helping the humanities? Is it really, really increasing our knowledge, or is it radically changing it into some that it wasn't or isn't?" So tonight we're going to explore some of those issues. And our lecturer tonight, before we get into the round table, Mike Toth is going to explore the first one. He's going to explain a little bit about hyperspectral imaging, how it's done, and more importantly, how it's managed. This is one of the great problems in the digital humanities, how to manage this stuff? How do we manage this data? How do we keep track of it and what is it useful for? And so Mike has been on the ground and one of the innovators in this field for a very long time. He has worked all over the world, on everything from The Dead Sea scroll papyra [phonetic] to work at the Library of Congress on our top treasures, to the deserts of Mt. Sinai. And so without further ado, I am going to give you our eighth Kislak lecture, Mike Toth. ^M00:06:18 [ Applause ] ^M00:06:25 >> Michael Toth: Thank you John. Thank you Dr. Billington, Mr. Kislak, for being here. Yes, okay. And I especially want to thank here at the library, I served in the government for 28 years. And I really want to thank John and Fenella, and of the other career federal employees here. I know Jennifer from NEH. It's a tough time to work for the federal government when your training, your travel, everything is under so much scrutiny. And we do appreciate all the work you're doing, not just for the humanities but for our country. And so thank you very much. We appreciate that. So moving ahead, what I wanted to talk to you tonight, we talk about forensics, but in reality, we're talking -- we've got to come to the core which is how are we studying manuscripts? This comes back to what John was saying in terms of you have the technical types and you have the scholars. How do we bring manuscript studies, which is based in the earlier times, looking at the manuscript itself -- the object itself. And how we -- how do we bring that to our current state of technology? To our changing state of technology? Whether it's in the lab with the ability to study the object with the latest technology, as Dr. Fenella France, Megan Hill [assumed spelling] and the team down in the preservation lab, are doing with hyperspectral imaging here at the Library of Congress, or it's the sharing of information such as the Walter's Art Museum, or the ability to get information just from in a mobile device, or in Flicker or whatever it may be, how do we bring those together? And how do we apply digitization, and not just the advanced digitization in terms of hyperspectral imaging or spectral imaging, but digitization in terms of making ones and zeroes that are meaningful, out of these precious objects that represent our cultural heritage? How do we use it for research support, whether it's for conservation, for scholars, for scientists, or for cartographers? Here in the Library of Congress, the spectral imaging system, was used to study the [inaudible] on the Carta Marina. And as you can see on the right, using the infrared with some of the other wavelengths, to be able to reveal what is not seen by the naked eye. So how do we support research, whether it's objects from these continents or from Europe or elsewhere? And how do we provide digital access? Will Noel will be on our panel, and he's working up at the University of Pennsylvania, on a project they call O-PENN [phonetic], which is to make freely available, make O-PENN LI [phonetic] available, all the manuscripts images from their collection, and to provide it as a standard form and format that others can use for sharing this information. So how do we provide digital access? How do we share it globally? ^M00:10:00 And sharing takes on many forms. There are projects such as we've worked on, Dr. David Livingston's diaries, there's of course the Archimedes Palimpsest where we sort of --- where we initially pioneered some of these techniques, as well as now as I said, Flicker, Pinterest, those are the flavors of the year or the decade in terms of social sharing. But what will come next? And how do we provide data, provide information, in a form that will support, not only what is used now, but will be used in the future, and be malleable if you will, for future social media and sharing? So what we've done, we talked about forensics but really the roots are in astrophysics, into astronomical studies, medical studies. We euphemistically call earth resource studies or national security studies. Applying that science and technology to the support of researchers, wherever they may be. The two examples I have here are a team of researchers in Paris, working on an early Syriac manuscript, and two researchers in the Vatican also working on a Syriac and also some Greek manuscripts. So how do we take this technology and use it to support researchers? And of course, in the title of this talk, we talk about forensics and taking the technology that is used for forensics. More importantly, unfortunately what we face is not just what they're using in forensics, but Hollywood's depiction of forensics. And the abilities that are proclaimed for us, using forensic technology, which may or may not be based on fact. In this case, it was based on fact, where Dr. Bill Christens-Barry had developed a prototype system for the Archimedes Palimpsest. Very much a prototype. There's a cigar box here with these wave guides coming out of it, the fiber optics, to illuminate the manuscript. And this was borrowed by the National Treasure book of Secrets, with a bunch of plastic and they were able to recreate it for the movie. But in actuality, we run up against so many things that we see in the movies that people then have higher expectations than what technology can really do. I love -- I'm not going to go through the whole list, but I like the ninth item on the list, which cites that whenever they get a match, it always comes up in big, red letters, "99% match," and there's a bonus, you get a driver's license photo with it too. So, this is what we're up against is we have these expectations of, "Oh, the technology can do this. I saw it on CSI." Or "I saw it in National Treasure." And that is not necessarily the case. There is a lot of sweat, a lot of labor, a lot of intellectual effort that needs to go into it, which is not to say forensic scientists don't do the same. They're running up against the same issues. So, when we look at the technology, and I'm not going to go get into the technology or [inaudible] or [inaudible] anything like that. We've got to look at the engineering starting with the technology, but we also have to look at in terms of systems. Now with the spectral imaging that we use, that we -- that was actually pioneered on the Archimedes Palimpsest process, we used narrow band illumination, that we take these very bright white lights, all the spectra -- the broad spectra lights, and instead of using a single white light and filtering it down, we use these LEDs which are -- which break the light into its components or emit the light and their components. And this was done by Dr. Bill Christens-Barry. You can tell he's an engineer because he has white socks and sandals on. I've told that when he's in the room, so he's aware that I make that point. But he developed these prototype panels where we go from the ultra violet, the 365 nanometers, and we work through the visible and into the infrared at about 700 nanometers. And by taking a sequence of images in each of these narrow band waves of light, we can then provide a product. And this is the system, the first generation system, or second generation system - I'm sorry - which is now in the Library of Congress in the lab here, in the subbasement, past all those old card catalogs there, where they have the 39 megapixel camera, which is a very large camera, these narrow band light sources, and the object, in this case the Gettysburg Address. And with that, you can gain large amounts of information. The LED illumination is the key part of this, using these lights, which are very low heat which is important for conservation and preservation. Using these lights to illuminate the object. We're now on our third generation system, which is more modular and portable. So, and part of this -- we talk about the lights and say, "The lights are so important." Well what's important is really to have an integrated system, to integrate the lights with the camera, whatever camera you may have. We're now using a Phase I camera which is fairly ubiquitous in cultural heritage. You have them in prints and photography here and elsewhere. But to integrate the camera, with the lights, any filters, but it's more than that. We focus on the technology. I mean, John was just talking about the technology. I've talked about the technology. Forensics, we talk about the technology, we talk about the sources of it. But the technology is but one part of this system. We always have to think in terms of the people. Not only the people operating the system. What skills do they have? What skills do they need? But the people who are going to be using the products of these systems. All too often, we think in terms of, "Okay, we're going to get these -- you know, this data out here. We're going to get these images out there." But we never really think in terms of that other -- as John referred to it, "That other side of the aisle. The people in the tweed coats and the academics." And it may be academics. It might be users on Twitter. It might be -- it could be preservation scientists. It could be a range of people who need to use this data. So we need to think in terms of not only the technology, but the people, and then the processes. How the data's going to flow? How we bring the metadata, the cataloging information which is the real value at the library [inaudible], that wealth of cataloging information that goes with the data. How do we incorporate that into the metadata? So bringing not only the technology to bear, but insuring we have the processes in place so the value added to the institution can be brought to bear to the object we're studying, with the people, with the skills. And it's only by doing that, that we can really claim we have an integrated system, that we have a system full stock, much less an integrated system. We always think, "Oh okay. I've got the latest camera. I've got the latest lights. I've got this object. I'm going to get these pretty pictures." That is not the do all and end all. You've got to think about how that is going to be a value to the end user there. And there's many different imaging capabilities. We tend to here again, think in terms of what is the ultimate capability? Spectral imaging, hyperspectral imaging, x-ray fluorescence, whatever it may be, whatever energy level. But, we can do a lot with regular digitization. We can do a lot with spectral imaging. We can do it from manuscripts, books, maps, manuscripts that are bound in other things. So, there's a range of technology we can use, and part of the challenge is figuring out what is most appropriate? And I always think of it, or I try to depict in terms of a pyramid. And since we usually deal with two dimensional objects, this is a two dimensional pyramid. So, but you think about the money that's invested in the -- in prints and photography for example, and your digitization suite there. And they do large amounts of digitization, very efficiently, and provide large amounts of data. You put a lot of money into it up front, but you get large throughput and you still get -- you get a very good bang for your buck there, in terms of quantity. And you get quality in terms of very good digital images. And that supports a broad base of users, whether it's scholars, whether it's for preservation, whether it's for global access, whatever it may be. And then you've got to look at, "Okay, what can we not get with that?" And then we use spectral imaging. And we'll use just kind of standard production imaging where we use the -- whether it's the LEDs or whatever system you have, and you do standard processing and you get a good amount of information. That's what we did with the Archimedes Palimpsest, for most of it. We're going to redo a few and step it up a level, and do some experimental processing. But you have to look at what is most appropriate for the challenge facing you for what the users need? ^M00:20:10 So you've got the standard imaging. You've got standard spectral and experimental. And then and only then, do you really pull out the stops. If you've got something that's very high priority, "Okay, we need to get it in front of nuclear [inaudible]." Or, "We need to bring an x-ray fluorescence machine in." To try to get that valuable piece of information. And so to scope it up and to say, "Okay, when is it appropriate to use the appropriate technology?" So, you've got to think in terms of the technology, but you've also got to think in terms of the system that's needed to support that technology. So you've got high quality, color images, which provide a lot of information for a broad range of users. You've got the spectral images. Those that are collected. Those what we'll call raw images from the ultraviolet, through the visible, into the infrared, each of those individual images. You've got then color that could be provided from those spectral images. And the processed images. In this case, looking at the under text. If you look at the over text, and then digitally we can reveal the under text here. This is from a St. Catherine's monastery. A palimpsest. This is a Greek [inaudible] Greek II. This happens to be one of the librarian's favorite works which is a gospel of Paul. You can see on the right, the Arabic text. This is a Christian Arabic text. And on the left, the Greek. And then the underlying Greek text there. Just as an example, of the type of work that can be done with processing and with spectral imaging. The same with Galen's Palimpsest on an early medical text. Much more difficult but to reveal the under text on that. But you have to be thinking in terms of your product because remember we said the user is so important. We've got to get the information to the user in a form that the user can apply into their studies, their research, whatever that may be. Now in the case of for example, the Archimedes Palimpsest, we just put all the images out there, the raw and processed, for people to use as they felt was most appropriate. And in some cases then, people could develop what we call the GUI, the graphic user interface, or some application by which people could then access that, could manipulate those images. But throughout that project, we didn't worry about the GUI. We didn't worry about, "Oh, what's the latest application, the latest technology," because that's going to be obsolete. That's going to be obsolete in probably a year and a half. So we focused on making sure there was good, standardized data out there that you, if you knew the standard for those data, you could then apply your application to that. You could do that in 2008 when we released all the data, or you could do it now in you know, almost ten years later, well not quite. Eight years later. So, we focus on the data itself. Very simple data. To you, you know, it looks like a very, straightforward structure and it is. The goal is, just put it out there in simple form to insure that it's accessible by as many technologies, as many people as possible. And then if you drill down into these data, you can get all the different images there, in different forms. But a key aspect to this is metadata. And this is where -- this is a tremendous strength of libraries, Library of Congress, and other institutions because you have rich metadata, usually done in a MARC record which is the library standard. And you need to insure that that accompanies, that that rides with these images. Unfortunately, metadata is something we give short [inaudible]. We think, "Okay, we've got pretty images," and this is represented in Will Noel's books, "The Archimedes Codex," where in the copy for Doug Emery [assumed spelling], who was our data manager on that project, is now working up at University of Pennsylvania coding various applications there. And Will says to Doug - to Doug Emery - whose central contribution to this project goes unrecorded in this book, "Sorry, metadata doesn't sell." Thank you very much. So, and this is truly representative of the problem we face. We talked about the system. We tend to focus on the technology. We tend to focus on the product of that technology, the pretty pictures. But a big part of it is including all the other information for archaeologists, the provenance, the source, where did that object come from, and all the information about the object itself. I won't go into detail, but a key part of metadata and I've cited before and we may discuss in the panel, are standards. We use double [inaudible]. Our library uses MARC. But what standards do we apply to the metadata? And making sure those are available so that people understand how that data is being provided. And including it. For those of you who are into XML, this is maybe your favorite slide here, but we'll move on. So I want to talk, put this in the context, of two programs. One is David Livingston's diaries. His 1871 diaries. When he was in Africa, he was kind of trapped in Africa, he was in ill health. He's been robbed. He had really, very little, and he started writing his diaries in berry juice on old London standard newspapers, which made it rather difficult to read, especially as the juice ate through it and -- or dissolved through it, and you're obscured by various texts there. So these are in the David Livingston Center in Blantyre. And I used to live in Southern Africa and when I was approached on this project, I thought, "Oh Blantyre. I'd love to get back to Southern Africa and visit Malawi," until I realized that this is the birthplace of David Livingston in Scotland, after which Malawi named its capital. But nonetheless, beautiful -- if you're ever in Glasgow area, I recommend you visit. So they have these holdings, these newspapers, that are the diaries of David Livingston, and also in the National Library of Scotland. So we have handwriting on the printed text here. As you can see, it's rather difficult to read. And if you look at a close-up, you also have seepage from the other side there. But then there's different spectral responses we're able to capitalize on. These numbers here refer, in the upper right corner, refer to the different spectral bands. And by capitalizing on that, and I won't go into the technology. Happy to chat later about it. You can see, this is that same newspaper page, but you're not seeing the print of the newspaper. You're only seeing David Livingstone's handwriting there. We were fortunate that these are such different inks. We had to make a little mental flip because we were always talking in terms of the Archimedes Palimpsest and trying to reveal the under text, and here we're trying to reveal the over text. So we kept slipping up and referring to under text and over text, in a way that was backwards, but anyway. So we were able to make good progress with that, suppressing the printed text. For those of you from geography and maps, he also had maps and drawings where we were able to enhance it. On this map, you can see vertically, especially on the right hand side, you can see a bit of the newspaper. This drawing was on top of a newspaper, but we were able to bring the drawing to the fore. And when I say we, it's not a matter of just taking those pretty pictures. There was a lot of processing there. And in actually, we were able to run around the clock because we would do the imaging in Scotland during the British working day, pass the data onto our data manager on the east coast of the United States, five hours behind us, go off and drink our ale and watch the World Cup, while our imaging scientist in Hawaii, 12 hours off, would do the processing, and then we'd get his feedback the next morning to then optimize our process. So, literally spanning the globe and spanning the clock to get these products. Pseudo color, sometimes where we'd have ink on top of ink, we'd give it an artificial color which helped the scholars, which is a key point here. We have the scholar with us, but we found on all of our programs, you have to be able to work with the scholar, more on the processing side. The imaging has becomes pretty much straightforward. But on that processing piece, remember we talked about the people? And it's, "How do we provide information that's useful to the scholar?" I have a little colorblind test on my phone that I use because you know, 10% of the men may not see the colors. And if you can't see red-green, no worries, we'll make it a different color for you. But you have to keep that in mind. ^M00:30:02 So, the end product of this was transcriptions -- or is, because the project continues with some of his other diaries, under an NEH funded grant. And NEH I should point out -- I'm sorry. I think I left the logo off here Jennifer. NEH did fund this project, I should point out. But to encode the transcriptions so they're linked with the images. So you can go in and line by line, go through an image and go through and see where that appears on the text or on the image itself. And you can link the different diaries and different -- I'm sorry. Not different diaries. Different renderings to the diary, whether it was his published journal, his original diary, or a later publication. But all of this is dependent -- these all look -- you know, it's nice to be able to compare these and to search through these and to compare the image with the text, but it's really dependent on the data. And getting those data out there, here again, in standardized form. You may note, this looks a little familiar. Yes, it's just a simple, flat, stream of files here. We did this for Archimedes. We did it for Livingstone. It can be used anywhere around the globe. You can build whatever application you want. It's in standardized form. It's -- or the transcriptions are encoded in text encoding initiative. The images are all tips. So with all these standards, it offers a utility of these data that can be used by, whether it's the scholars on Livingstone, or scholars who are looking at cartography of Africa, or the early European explorers knowledge of Africa, whatever it may be. So, it comes down to those data there. Now I want to turn to another project. And since we've gone with the forensics role here, and the movie side of things, we'll call this, "The Case of the Missing Leaves," or as we refer to it, "The Galen -- Syriac Galen Palimpsest Early Medical Work by Galen," which was translated into Syriac. And in 2009, this was actually purchased by the owner of the Archimedes Palimpsest, because he thought we needed something interesting to look at, having done the Archimedes. And we very much appreciate that. And so we first did it as a test with the bound manuscript. Found spectral imaging offered good results. And then Abigail Quandt at the Walter's Art Museum, dis-bound it and made it available so we could see down in the gutter region, down where the choirs [phonetic] come together there. So we imaged that and we put all the data out there. It's 300 gig. It's all on the web. You go to www.Digitalgalen.net and you can get any of those images. So, independently of this, we have had a team of scholars who've been studying Galen. They're actually looking at herbals in ancient medicine. And then came across this and started studying it. And we think this is great because it wasn't something that was funded by anyone. It was these scholars who were all working together to try -- who made use of these data. And it's out there. They're using it. They're studying it. Now we've helped them out. I was over -- in fact there's that one picture in Paris there. And I don't know if you can see on the left side of this, there's some circling. The screen doesn't show you. So there's some circling of the image and then there's notes, "Well what about this? What about that?" And then our image scientist would go back and try to provide more information to the scholars. So, the -- and the under text, which is not in purple. It depends what colors the scholars want, but you can see some of the under text there. That one you can see a little better, is all available and that's what they're studying. So one of the scholars, Gregory Kessle [phonetic], a Russian scholar who's teaching in Germany, was here at Dumbarton Oaks on a grant doing study, and he's a Syriac scholar whose been trying to catalog Syriac works. And since he was here in Washington, I said, "Hey, why don't I bring you by to meet the owner of the Galen Palimpsest, see his library, and see the actual Palimpsest itself?" So I brought him in the library, pulled the palimpsest. It's rebound. Pulled it off the shelf. Put it out for him as you can see in this picture. And he holds it and he says, "I feel like I've held this before. I've seen this before. And there's something tangible about the object, not just looking at the screen." And he pulls up his computer and he'd been at Harvard about a week or two before, up at the Houghton Library and he looked at this one folio in Syriac, that he knew his palimpsest. He said, "This leaf came from this book." And sure enough, we went up to Harvard and we set up the gear there in the Houghton Library, thanks to Bill Stoneman [assumed spelling] and the team up there, and we imaged Houghton MS Syriac 172. And that is one of the missing leaves from the Galen Palimpsest. And you can see, since it's a -- the under text runs left to right here. They're sideways, since it's a single folio. So, and with that, he was on a hunt. And so started scouring catalogs. He didn't -- actually he happened to have seen this object, but he started scouring catalogs, paper catalogs. And this comes back to the role of the library and the role of the cataloging information because in a catalog, which Father Justin, the librarian at St. Catherine's monastery [inaudible] is pointing to, he found four lines by -- this is by Sebastian Brock. It's all the Syriac holdings in St. Catherine's monastery. He found four lines in there, referring to an under text in Syriac. And he said, "I think this is Galen." So as part of our work in St. Catherine's monastery, we threw in, "Syriac New Find 65." Sure enough, it's a Galen text. And then with that he did some more research and he looked at a catalog in the Vatican library. And he said, "I think you've got a couple there." So we went trucking off to the Vatican. And I have to say, having worked now with the Greek Orthodox Church, the Roman Catholic Church, the U.S. government, I think I have some expertise in three of the greatest bureaucracies in the world. But in any event, so we went there and we looked at this. And then we imaged it. And we were able to find additional -- actually three-folio. There's one bi-folio and two other folios. And this is one of the processed images there. So, one of the issues though is open access. And I have to say that the monastery of St. Catherine's, Father Justin the librarian, was very good and he said, "Well what--?" I said, "Can we put this in digital Galen? We've got a conference of these scholars on herbals and ancient medicine in Paris and they would like to be able to share this." He said, "What do you want me to say?" I said, "Well, how about 'Creative Commons Noncommercial License for Attribution?'" So he sent me, "Creative Commons Noncommercial License for Attribution." Harvard did the same. Of course the owner makes all of the -- his holdings available for commercial use or whatever you want. Vatican is a little more of a challenge. I didn't work the initial negotiations there, so we'll see what we can do about making that more broadly available. But this is -- I can share this with you because you are the educated public or scholars, right? Yes. But other than that, we're still working on making -- to allow hosting on the digital Galen. And next month, we will be imaging the last of the missing leaves and that's at the BNF, the Bibliotheque nationale de France. There's one leaf there, Syriac 382. Another bureaucracy but we think we've got that one resolved too. So, with that, we will have all the -- well, we won't. We'll have almost all the data freely available because as I said, we have the Harvard and the [inaudible] leaves now available on the www.Digitalgalen.net. And we will then have the Bibliotheque nationale de France, that was one of our prerequisites for going in there was we have to be able to host these data. And then we're still working on the Vatican. So, and with that, all these data -- you can go there now and get what we have and hopefully we'll have the French one there. Building on this, building on these data files, up at U-Penn, a woman named Doc Porter, has been working on applications to support cultural heritage. And as we noted, these are leaves from a book. And there's choirs, and there's a choir annotation which looks like a bunch of concentric views. So she set up this collation tool. It's how it's collated. This collation tool, in which she used the Galen data. And so you can piece together where these leaves fit into the collation. She's used some publicly available images of the upper text of the Vatican leaf in this example here. We also have the Harvard leaf and the [inaudible] leaf. ^M00:40:23 So, by being able to use these data, she could then make it -- use it in her tool, since it's standardized data, which is a tool that's useful for anything else that follows the same standard. And I should point out, this follows the other standards and not the other way around. So, the data has more uses -- we didn't think about this. And there are many other applications for this. And it's not for us to prejudge what it will be used for, for us to mediate how that data will be used, but to make it available in a form that others can use it. So with this, we've been pushing open data. We being, one doctor Dr. Noel, and the team whose been working on the Galen Palimpsest and others, which was recognized by the White House for pioneering efforts with making data available, which brings up an issue. If it's all available, how long will it be available and how do we preserve those data? How do we insure these ones and zeroes -- I was just down in John Hessler's office and he's got, you know, these big, magnetic tapes. And you know, how do you read important information from those magnetic tapes? How do we read from floppies? How do we read from the cloud? How do we read TIFFs henceforth? So our view is, get those data out there, make it available and have people copy it. We know for example the Islamic manuscripts that the Walters has hosted online, have been sucked down in the Emirates, and they're being hosted there. That the Galen we know is being hosted around the world. So not by conscious decision but people just wanting it, wanting to use it and taking it, which helps the preservation there. And preservation's changing. And I know in this audience, in this institution, I don't need to say that. But it -- you know, it's a big change from you know, locking the books down, making sure they don't walk out, as to, okay, these servers, how do we make these data available? How do we back it up? I should point out, this server farm is not in St. Catherine's monastery. It's dry, it's dusty, there's no connectivity, the power goes out, but I'm just trying to compare. I had someone say, "Well that's [inaudible]." That there's you know, how do we, misuse the term a paradigm shift, but I think this is appropriate here? How do you shift from keeping those objects on a shelf, to making this data as Will Noel says, as making those data available? And I always -- given some of the folks I've worked with, as stewards we talked about the role of civil servants here. We talked about the challenges the leaders have here. And we look at trying to preserve these data. And three people I've worked with, Father Justin who I mentioned in the [inaudible], Michael Suarez who's at -- in Charlottesville with the University of Virginia Books, and Columbus Stewart. I ran into him again an Islamic conference, and he gives camera systems to people in the Levant, to communities in the Levant, to digitize their texts. And he was saying, "Yes, some of those have been destroyed. And some of these people are going to great efforts to try to preserve those objects they can." But we have all the digital copies here. And I look and here's three men of the cloth who have kind of a broader view if you will, and they're -- you know, what's their perspective on preservation and how do we take that broader view in terms of preservation? So as we look at where are we going and what are we doing, we have to think in terms of what is going to be available for the next generation and how do we support that with systems that include the technologies, the people, and the processes, that will provide this data for generations to come? So thank you very much. I appreciate the opportunity. ^M00:44:46 [ Applause ] ^M00:44:54 >> Dr. John Hessler: Thank you, Mike. And I'd like to invite our round table participants to come up, and I'll just do a brief introduction once they arrive at the table. So, and the round table's going to be kind of a loose conversation on some of the issues that we just talked about. And then we're going to take some obvious questions from the audience that will kind of spur it on a little bit. So, you know Mike by now, and you know a little bit about Will, being he appeared so much in the talk, Mike's talk. But Will Noel is Director of the Kislak Center at the University of Pennsylvania. He has been on the absolute forefront of digital preservation and -- of medieval manuscripts and basically the tools and the amplification of texts. Mostly medieval but now that he's at the University of Pennsylvania, he's taking a much broader view of those things because they've got a lot of Revolutionary War materials in Philadelphia and things like that. Dr. Fenella France is the head of our preservation research and testing, and I guess at this moment is the acting director of all of the conservation activities here at the library. Fenella has been also on the forefront of using this technology, has made some really interesting breakthroughs, especially with some things related to Thomas Jefferson and the Declaration of Independence, which I think we'll get her to talk a little bit about. Obviously, we know Mike. And Chet Van Duzer, has been here many times. He is a former Kislak Fellow. He's a historian of cartography and he has used hyperspectral imaging on one of the most important early maps called the Martellus Map in Yale. And so we'll a little bit about that. So, my first question I guess, we can open this up to, and I think maybe Will can start talking a little bit about this was really the -- one of Mike's interesting points is the interaction between the scholar and the technologist. And how does this take place? How do we think through making these tools fit the way scholars, the more traditional scholars who have perhaps the language skills, the paleography skills, begin to look at this material? So maybe you want to say a little bit about your medieval manuscript programs and we'll start from there? >> William Noel: Yes, thank you John. I remember when we started out on the Archimedes Palimpsest project and we were doing some experimental imaging and the images -- it produced some images that they were incredibly pleased with. And we're talking serious images, right, from the [inaudible] to technology. All the [inaudible] or whatever. And they spent a lot of time processing these images. And we put them in front of the scholars to ask for their judgement and they said, "Well, they're out of focus, aren't they? They're blurry. Who are you people?" And there's a great deal of misunderstanding between what a paleographer can think an image is and what an image it produces. I mean in that case, the images were very highly processed images because the spectral characteristics of the inks were very similar. And so they had to process them a great deal. And when you process them a great deal, you produce a lot of artifacts and the artifacts don't matter if you're imaging a cocoa field in Southern America, but they do matter if you're trying to decipher 10th Century Greek cursive. And another problem for example is that we were imaging in 300 dpi because 300 dpi was what the imagers thought the scholars would need. But if you ask a scholar to decipher a difficult text, he'll want to blow up a single letter to the size of a computer screen. And we just sort of had to radically rethink this. And when you think about these complicated imaging operations and these processes, you know, and this is what Mike was so brilliant about over the Archimedes project, is that it's truly an iterative process between the scholarship and the imagery and the conservation. And if you don't have a feedback loop between these people, you're not going to get a good result. And everybody learns from it. And what you really need is a big hospitality budget because-- >> Dr. John Hessler: You have to put a lot of people up in hotels. That's right. That's right. >> William Noel: Yes, and get them talking. >> Dr. John Hessler: And buy them drinks too. Right, right. >> William Noel: Buy them drinks. Yes. >> Dr. John Hessler: Well that's why we have Mr. Kislak. We [laughter]. Well, [inaudible] okay. Thanks. And Fenella, you worked on a project here. Most of my experience with this material has been -- I first met Mike in a very, very dark room, wearing very orange glasses, when we were working on the Waldseemuller map. The imaging at that time which was almost a decade ago, was new to me. And since then, I have become one of the biggest proponents of this, I think. I think it's one of the most important breakthroughs that we have and one of the most important tools. Fenella France here has also been at the forefront of using this on many of the documents here at the library. And would you tell us a little bit about your Jefferson experience? ^M00:50:31 >> Fenella France: Well we have the honor of -- well, I had the honor and the humility to be able to work with some of the most amazing documents at the library here. And one of them was the original rough draft of the Declaration of Independence, written by Thomas Jefferson. He had made various changes on the document, and generally it was just a very neat cross-out, and he wrote the change above it. And as I was going through some of the different wave bands as Mike talked about, I realized there was one place on Page 3, where there was a deliberate erasure, and it looked like he'd tried to cover it up. So, the challenge is [inaudible] that sometimes when you make a change on a document, the inks are very similar. And so their spectral response is very much the same. With a lot of processing and this was during the time of Snomegeddon, so I went out and I shoveled a few inches and I'd come back inside and I would do a bit more processing. And finally found that the section where he'd written "Fellow citizens," he'd originally written "Fellow subjects." I thought that was a fairly significant change so of course you know, we -- running up to the sixth floor to say, you know, "Does anyone know about this?" And particularly because everywhere else, he'd just made this neat cross out. So we did find that a researcher from Princeton had been going through Jefferson's papers. Jefferson noted that he was copying from the Virginia Constitution that was written two months earlier and he said in his notes, "So, I realized as I copied that this was the wrong word for our new country. So I expunged it, never to be seen again." So I guess we are de-expunging things in the subbasement at the Library of Congress. But the wonderful thing is that that gives us this insight into the original thoughts of our founding fathers. They though that they suddenly went, "No, this is wrong. Let me just hide this." But no. We can see and we can actually start to understand more about it. But as everyone's been saying, I think this process of working closely with the curators and they're just so knowledgeable about their materials, it's been such a wonderful experience for me. I'm a mere scientist. You know, I can't know everything about cartography, about manuscripts, about Hebrew. And looking out at some of the colleagues I've worked with, I learned so much from that, but together, it's that together application of new technology and their innate knowledge that makes this such a strong process. >> >> Dr. John Hessler: Okay, so much for Question 1, "How do we integrate this with the scholarship and how do we make the traditional scholars understand what it is we're working with here?" Well the obvious answer is, "We work closely with them." And as this technology gets better and better, we will reveal more and more. The second question which was, "How fast is this technology changing?" was basically something that was thought through -- when it was asked to me, "Could I have done it two years ago? Could I have done it five years ago?" And the question was sort of misplaced because the technology, as Mike said, is one thing. The other problem sometimes is actually getting permission from curators to actually work on these materials. There's a very important map called the Martellus Map which is hanging on the wall at Yale which Van Duzer had unfortunately noticed and became a dogged fan of "Let us get this off the wall, take it out of a frame, and hyperspectral image it." Chet of course is not on the staff of the Yale Beinecke Library, but had the [inaudible] of actually being able to do it and has some pretty radical findings. So even though the technology was the same, some of the patience that has to take place in order to actually get the stuff done, is another story. So Chet, why don't you tell us a little bit about what your experience with the [inaudible] Martellus was? >> Chet Van Duzer: Yes, it was quite an experience getting the permission to do the project. I had a fellowship at the John Carter Brown Library to study the map in 2011. And I had been working with the Beinecke to try to get images of the map. I wanted multispectral images. And they were not -- at that point, were not able to make multispectral images. They were able to make ultraviolet, infrared, and very high resolution natural light images. And I was able to use those to study the map and write a book about it. I was able to decipher, let's say 25% of the text on the map, which was very useful and generated a lot of insights about the map. And at the point that I was finishing up that book, I happened to talk to a woman at the NEH who said that they fund projects like making multispectral images of the Yale Martellus Map, and I applied for such a grant and obtained it, fortunately. And it was quite surprising to me that they Yale -- the Martellus Map at Yale is one of the Beinecke's prized possessions. And speaking diplomatically, they didn't seem very enthusiastic about the project to image the map. Things went very slowly in trying to arrange all the details of the imaging project. Once I was there and I was there with Gregory Hayworth [phonetic], Roger Easton [phonetic], Ken Boydston [phonetic], and Michael Phelps [phonetic], once we were there and once they saw what we were doing, everything went very well, but getting to that point was very difficult and I can imagine Mike at the BNF will -- these things can be very, very time consuming and very frustrating. But one thing I want to say about that project is, when the object is a good candidate for multispectral imaging, the scholars' perspective is that the object goes, in this case the Yale Martellus Map, it goes from being an un-studyable object to one that's studyable in all its aspects. The map had been acquired by Yale as a gift in the early 60s, and there was a great excitement when it was acquired. And there were a few articles published. And then for almost 60 years, nothing happened, precisely because all of the text on this large world map, had faded to the point of complete illegibility and actually invisibility in many cases. And multispectral imaging was the only thing really that could bring this text out. I'd seen the ultra violet images which were surprisingly un-useful with that map. The infrared images were not very helpful either. But the multispectral images, it's night and day. This wonderful, important map that no one could do anything with is suddenly available to scholars. >> Dr. John Hessler: And just so you know, this is a giant, wall map, considered one of the sources for Waldseemuller's you know, early work. So it's an extremely important piece of cartography. All of us are looking forward to Chet's reading of what's on it and any breakthroughs that we might anticipate. I guess the best thing now is to kind of open the floor up a little bit for questions, if people have any. Are there any questions? Yes? ^M00:58:27 [ Inaudible audience question ] ^M00:59:06 Yes, well you hope for peer review. >> Michael Toth: I'll take a [inaudible] at this and then Will may have some thoughts here. As scientists or as technicians or those who are collecting the information, it is our obligation to provide it as objectively as possible. And I think one has to look at, to insure the team doesn't have any vested interest, if you will. If you're working with a religious text, you don't want some really religious people working on the collection. I'm not saying the study, but the collection of the information. If you're working on you know, some cartographic or geographic, you want to make sure it's just data. Unfortunately to me, it's become that sometimes I'm in an institution, and I go, "Oh wow." I have to think, "Oh wow. Wait. Remember, this is this because--." You know, you're trying to get this done. You've got limited time. You're in a dark room. You've got these goggles on. And so that objectivity I believe is important. ^M01:00:06 Second, and this is where Will can speak more to, especially with regard to [inaudible] and Nigel Wilson [assumed spelling] on the Archimedes Palimpsest, where we had a balance with the scholars, one of whom knew Archimedes very well and one of who knew the Greek very well. Do you want to talk about how they played off each other and--? >> William Noel: Yes [inaudible]. So, if you throw enough money -- problem, you can get a pretty spectacular image. But if you've got sort of 177 folios of diabolical manuscripts image, that can get very, very expensive. So what you want to do as -- when you're running a program like this is not create the best image you possibly can, but to actually create the worst image you can possibly get away with. And you want the scholars to do as much work as they possibly can. And that's what scholars do. I mean scholarship is subjective, right? So scholars use their minds and they look at marks on pages and they form an opinion. And, what you really want to do is to get a consensus of more than one scholar. And in the case of the Archimedes Palimpsest, we had two very different characters and two very different approaches to the material. One was Nigel Wilson who's [inaudible] when it comes to the transmission of the Greek classics through the Middle Ages, but he doesn't really care about mathematics at all. He cares about [foreign name], right? So he was pretty bored with the actual math. And then there was [foreign name] who is ancient mathematics. And he was all about the math. And so they had to agree. And in the end, they agreed on all the readings. And with the other texts in the manuscript [inaudible] from the ancient world, they were essentially crowd sourced. And if there were disagreements amongst the scholars as to what they saw, they were noted. But there weren't many disagreements in the end of the day. You have enough eyes around the table and you can establish a consensus and that's what we did. But scholarship is subjective. That's what it is. Collecting a data isn't subjective but the parameters that you initially set, they're choices. They're financial choices. They're wavelength choices. They're resolution choices. Their best guesses. And they can be reevaluated. >> Fenella France: I'd like to just add one point which is we've worked very hard, particularly here at the library and I'm so grateful for the library for embracing this technology early on. It's standardizing the processes so whatever we're doing, can be reproducible. And that takes a lot of that question out of it that we can easily track exactly what we have done with those images, and you haven't introduced any new components that are not really there. So I've found that all of the scholars that we've worked with, have been very open to whatever has been found, even it if wasn't perhaps what they wanted to. But it's just that knowledge of hidden information that's being pulled out, that adds such a lot of content knowledge to our collections and layers that on the original materials. >> Dr. John Hessler: Yes, I just want to add also that this is -- you know, the way we portray it, the way we show it in these images as if it sort of just pops out, this is incredibly, incredibly, difficult, mind numbing work. If you've ever been on one of these imaging projects, you know, just setting the camera up, just getting everything running and being rectified so the images fall on top of one another, is just -- you're in a dark, hot room with you know, just day after day after day, banging your head against the wall. And then you've got the images. And now you've got to process them and you've got literally hundreds of algorithms to choose from. And what is it that you're going to -- what is it you want to show? Most of the time, the stuff you're working with is somewhere between you know, dirt and new. I mean it's not a brand new thing. It's not a controlled experiment. So you are really knocking your head against the wall. And sometimes it's, "Oh my God. Holy shit, look what we found." And sometimes it's like, "Oh my God, we've been working on this for two months and we have -- we can't see anything." So it's not all glory. A lot of times it's just failure and failure is fine also because you learn a lot from it. So. >> Michael Toth: That doesn't even include two revolutions, bombings and-- >> Dr. John Hessler: Right, that -- exactly. Exactly. Mark? >> [Inaudible] geography map division, looking at the Kislak Collection. And we were specifically, we were looking at pottery, figurines, three dimensional objects. So what are the -- how can this technology be used for that type of material? And what is it we're likely to learn from it? >> Fenella France: We actually have already done some imaging on the Kislak flasks. And we're hoping to get a student and a summer intern to do some of the processing. We have actually been able to look at the hieroglyphics on the surface of that, and particularly as well as not just pulling out hidden text, but we can identify without taking a sample, what the actual pigments that have been used to create that. We can see all of the treatment changes in that where it may have been correct in the past. And depending on what part of the spectrum that shows up in, we can actually then tell you whether it's a more recent treatment or whether it's historic. So, it's really very adaptable. We've looked at textiles. We've looked at documents, books. We can image through them and do transmitted and the 3D is, in terms of the objects, is actually really exciting. >> Dr. John Hessler: Yes, as far as the objects go, one of the things we're interested in, in the Kislak so-called poison bottles but Maya flasks, there's a great deal of them that are exactly the same. They look very much the same. I think John Carlson [assumed spelling] is here, one of the great experts on these flasks who's sitting in the back. But there's a group of these flasks that are very much identical. They have a [inaudible] two gods and then they have a series of glyphs in the middle that haven't been deciphered. And one of the things that we're working on is, we think these might have been produced in a factory setting. And so what we're looking is to see how identical some of the actual, very, very intricate forms on these are to each other. And using hyperspectral imaging, you get very, very precisely rectified images. So we can begin layering these things on top of one another and looking at them very closely. Some of the images that we have gotten of the treatments and the preservation of these things, have been striking. So we have been working on some of the 3D stuff. So. >> Fenella France: And I just want to gently note that one part of the process is also the [inaudible]. A lot of the objects that you saw that Mike were showing were nice pieces of paper, kind of about this size or smaller. I very much enjoy and love working with John Hessler. He did used to challenge me a little though because he would come down to the lab and he'd say, "I've got this map." And I've learnt now, my first question is, "How big is it?" Because he'd turn up with something that's about five feet by four feet. And you're like, "Okay, this is going to take all day." So, but that adaptability has really helped us to look more closely and expand the breadth of what we can do and how we can use this technology. >> Dr. John Hessler: Yes? >> I'm one of the white socks guys. And sitting here I've -- it occurred to me that there's a bit of a [inaudible] problem here in which the people who are doing this great technical work and imaging in two dimensions and now three dimensions, are producing great databases and part of the problems are administering them and coming up with [inaudible] for them. And this is now well on the way to be accomplished. And then they're giving them over to the [inaudible] tweed guys who are applying centuries old techniques for coming up with answers to problems. And of course, you gave a speech at Google yourself and I'm sure that Google has -- the Google search engine itself could be adapted to many of the quote, "scholarly," may not want to call it scholarly if they do it, problems which are problems of correlation and association and coming up with general rules from the basis of detailed data. And this will surely happen. Is there any motion in this direction, probably being pushed by the people in the black turtle necks? >> Dr. John Hessler: Yes, I think a C.P. Snow problem is really where we -- we're beginning to fade. For those of you who don't know, C.P. Snow wrote a very famous book which talked about the two cultures and really it's the humanities and the sciences, and how they don't interact that much. I think one of the things these technologies are doing is beginning to bring those two people together, which kind of segues us into the third question which was really, kind of how is this technology affecting the way we study these things? Will mentioned crowd sourcing. One of the great things that has been going on with a lot of these images is crowd sourcing. I mean the papyrus project, the Egyptian papyrus project which has done -- teaches you basically in a few hours to read ancient Greek and in a few moments, you're then deciphering some of the papyra that have been sitting on the shelves in Oxford for the last 70 or 80 years. So crowd sourcing has become a really great way to use this data. So I think some of those boundaries are breaking down, but no matter what, no matter what you do, if you take a perfect image of beautiful Syriac manuscript, you still have to have someone who reads ancient Syriac. And unfortunately, that's probably the part of it that's least being pushed now. The idea of languages and things like that is something that we need and they need to do their work. ^M01:10:23 Just because we have an image, if we didn't have a person who was the expert in Greek mathematics, the Archimedes wouldn't have been read. So all of that stuff is important. And we're beginning to get hybrid people. People who sit on both sides. I mean Chet is one. Will is certainly one who basically stride both sides of this. So any comments from--? >> Chet Van Duzer: I personally think it's very exciting to be able to work with these technologies and learn about them. I like studying maps and I like reading the texts on them and determining the sources of the texts and the images, but this technology gives you an access that you don't have in other ways. And it's very exciting to learn about. >> William Noel: One of the things that I would say is that digital images are often called surrogates for originals. They're appalling bad surrogates for originals. Originals are such complicated objects that a single image will never capture all the stuff that you need. So for example, there are books written on bindings, right? Books written on individual bindings actually. And there are bindings' experts who love bindings and they love bookmarks. And the way that standard digital imaging works, there's no real record of the binding. So if you're going to record the binding, you have to do a completely different kind of imaging. Maybe you want to make a video, or something like that. But you're never going to replace -- you're never going to replace the original. The wonderful thing about data is that you can do all sorts of things with data that you can't do with originals, like you can cut them up, which is very handy. I mean, a lot of people think that people who study medieval books actually read them. And a lot of imaging is predicated on the notion that people actually read medieval books. You know, most of the time, people don't read medieval books. What they do is they compare the handwriting in one book to the handwriting in another book. Or the illustration in one book to the illustration in another book. And so what you really want to do is to bring these two books together, but the trouble is that one's in the Bibliotheque nationale and one's in the British Library, and then you're never going to get them together actually. But you've got a Syriac leaf in -- you've got Syriac leafs all over the world, and digitally you can pull them together. And that's really helpful and that's what digital imagery can do. And so these two worlds have a sort of problematic interesting relationship, one to the other. And it's not a case of the old world dying and the new world growing. There's a much more complicated, rich set rewarding operation going on which is incredibly exciting. >> Dr. John Hessler: Yes, we are definitely on the cusp of -- things are changing so rapidly, but they're extremely exciting. There's a new project at Stanford which is the bringing these medieval manuscripts together that are in various institutions and bringing them into one place, into one area. So you wouldn't have to go to five different links. You would just pick up the -- do the one link and you'd have the whole manuscript, even though the manuscript is in 20 different places. And so those sort of tools are just increasing the speed of scholarship. And I think that's really something. And you know, I have to -- I think all of us at the table agree about the original. Yesterday, I was actually giving a presentation on a new geospatial system that we're doing here at the library, and we had a power failure. And so I went onto the Twitter feed and I pulled down all the hits about power failure. And one of the Tweets came from the Georgetown Law Library. Their official Tweet was, "The internet is down. All digital resources have collapsed. But traditional books, still working fine." And so I think that's kind of what we're talking about here. There still has to be some of that. So, Jay, you have a question? >> Yes, it's sort of aligned with your last subject about the lights going out. I fit in with some of the people in the audience here who don't fit neither of the two categories that you talked about. The scholarly or the, I guess the wealthy, highfalutin that you discussed before. But we sort of fit in the middle. But I'd like to ask a question which was discussed earlier that maybe Will you, or Michael or Chet can answer since you investigated or looked into the Archimedes. Were you able to find out, when you stepped into the [inaudible], did [inaudible] really say eureka? >> Michael Toth: Will? [laughter] >> William Noel: Well, if I stepped into a bathtub, I wouldn't have said I'd found it. I would have said I'd probably lost it. So I don't think so. >> It's to show the relevance to the present of the kind of exploration that you're doing, I'd like to just mention that along with Jay Kislak, one of the greatest benefactors to the country in the Library of Congress has been a guy named David Packard -- David Woodley Packard. And I asked him why he built basically this enormous, up to date, audio-visual conservation center that we were able to build. What got him into it? And he said, "Well you know, I was trained as a classical archeologist" and he's currently building the first museum ever built for [inaudible]. But he said, "I was so concerned that we have so little of the deciphered fragments of the classical history, that I thought to myself, 'You know, I'm an American. My dad founded the Hewlett-Packard Company. And I've learned enough about technology to know how important it is to answer these questions." And so he said that, "I figured that the audio-visual creativity of the American people was so rich and so little understood and so poorly preserved, that nobody's going to be able to really decipher [inaudible]." So he said, "I thought 'Let's conserve. Let's get this stuff together. The audio-visual creativity of the American people over the last hundred years. And let's try to preserve it so that the future people who try to study ancient cultures and learn more about them, will have the technology and the intellectual curiosity to help us better understand what needs to be preserved, of what people before us have done.'" So he gave the shortest acceptance speech when the Congress was kind of amazed to see him. And he said, he said, "Well don't," everyone was thanking him. He said, "Don't thank me." He said, "Thank the hardworking technicians at Hewlett-Packard that made it possible to earn enough money so I could help out at least a temporary problem." So I just think we ought to thank you all for doing the technology and speaking about it so eloquently about how institutions can work together and how the two sides of the humanistic and technical and scientific side are in fact, working together. So I think it was a wonderful presentation. I just wanted to thank you on behalf of this institution. And I think if I may speak presumptuously, for the American people, for what you are doing to help us better understand the full range and beauty of what people have created in the past. Thank you so much. ^M01:19:19 [ Applause ] ^M01:19:24 >> Dr. John Hessler: And we probably should end with that but we have time for one more question or two more questions. >> I'm just concerned with all the strife that's going on in the Sinai that what steps are being taken to preserve those treasures out there that are actually irreplaceable? >> Michael Toth: I'm going take it [inaudible]. I can't speak for the monastery or his eminency, archbishop and the monks, but I think a key factor here is their relationship with the local Jabaliya Bedouin tribe, a relationship that's gone on since 400. ^M01:20:12 And it comes down to that interpersonal relationship and to maybe not common goals in terms of preserving the manuscripts, but to preserving that institution which is important to the community, both the religious community around or into which they invest their lives, as well as the surrounding community that supports them. So I think beyond all the digitization, all the technology, physical barriers and those walls and everything, it comes down to the people. Will? >> William Noel: St. Catherine Sinai has preserved its manuscripts for -- many more hundred years than anybody else. And the reason it's been able to do that is because it's in the middle of nowhere. And actually, it's also quite hard to scale. Books have strange fates, you know? And you never know how they're going to survive. But if I was a book and I wanted to survive, I would find a place in a desert. And I wouldn't find the nation's capital because when you want to destroy a culture, you destroy its capital and you destroy its library. That's what you do. So over the long term, finding yourself in the middle of nowhere is really good thing. And St. Catherine's is in the middle of nowhere. >> Fenella France: And just following up on that, the library is working with the [inaudible]. They have a conference coming up on what happens when you lose a library. We are thinking very seriously, particularly with all that -- what's happening throughout the world, of what we can do as an institution, all of us, to actually stop things like that happening. >> Dr. John Hessler: And there's a question [inaudible] in the back there? Arthur? >> I have a question actually with a three part answer. You showed a [inaudible] Michael of the -- a [inaudible] from the Carta Marina. And then there was an image of what's under some covering. So I have a question, Fenella, what did you do to look through it? And to Chet, what does it say? And John, why is it important? >> Fenella France: Though I came probably -- essentially it came up very nicely in the ultraviolet. And it was on the back of the [inaudible]. Essentially it's [foreign name]. It's Latin. And it speaks to all of the changes that were made on the 12 sheets of the map. We did have someone go through and look at those, and they'd made changes to everything but, was it Sheet 6? I may have the numbering wrong [inaudible]. >> Dr. John Hessler: Yes, no, Sheet 6, yes. >> Fenella France: Which was the original parchment. I talked with people who were - you know - were drafters and they said that it's actually common for someone when they make changes to keep it with the original. Why they stuck it down so you couldn't read it, I don't understand, but they did keep all the changes with the original. >> Dr. John Hessler: Well I think that's a fairly concise -- one of the things it has shown is that the piece that Mark talked about that was just donated by Jay, is Schoner's drawing of Sheet 6 in the Carta Marina. It's the coast of Africa. That's what was originally bound into -- the Carta Marina is a large map. It's 12 sheets. This manuscript was originally bound into the volume with the other 11 sheets. The printed sheet did not exist in the volume. It was all printed -- 11 printed sheets and then this manuscript sheet. Later on, somehow, the printed sheet was folded, found, and then laid into the book. So there's actually 13 sheets of this map. It gives us some indication that piece because Schoner in his drawing, makes the correction that is supposed to take place to Sheet 6, when the printed copy doesn't. So the printed copy must have been a proof copy. So, that's the importance. It was kind of an interesting thing when we first saw it. There's a number of those sort of things on the Waldseemuller materials. There's on the 1507 map, there's a [inaudible] that is part of a palm reading manual on the other side. So, so. And I guess we can go one more? >> Yes. I have a loud voice so. The first thing I wanted to say is to Mr. Kislak for saving Miami Circle. I was down there when he did it. It was really important. And then, I wanted to find out from your perspective. I'm not a scholar. I'm not a scientist. So I don't the parameters. But what do you do with the imaging in order to stop people from knocking it off, like a Michael Kors bag? >> William Noel: Like a what? >> Dr. John Hessler: They can do it all they want. The stuff that the library has that we image and put online is perfectly free. People can use it for whatever purpose they want. It's out of copyright. And there are plenty of people who actually run small poster companies using the treasures that we've digitized and put up online. Other institutions have different ideas about their cultural property and it kind of goes from institution to institution. Mike can talk a little bit about the creative commons which is the way most of this is being put up. >> Michael Toth: The collections we've worked with have allowed free use under a creative commons, which was -- fortunately we were just trying to figure out how to get the Archimedes Palimpsest out there for free use and creative commons was developed. And this allows with attribution, you to reuse the images which is important. And we want people to do that. You can knock it off however you want because that means you're accessing it, you're using it, you may copy it somewhere else, that may help preserve it. So that -- we want that. >> William Noel: I love it when my images are knocked off. Do. It's great. You're welcome to them all. [ Inaudible audience response ] Love it. >> Michael Toth: Now some of them are creative commons noncommercial. David Livingstone's diaries for example. Archimedes Palimpsest, Galen, all the Walters and I believe perhaps Penn are commercial because there's also an issue with you know, if someone wants to host it on a commercial server or something. So you have more flexibility with a commercial, but there is also one license that is noncommercial, depending on -- and it depends on the host institution. We don't make the decision. We try to negotiate with them, as we're still working with the Vatican, to try to make these things available. >> William Noel: It's very important for the future of cultural heritage study that everybody advocates that images created are essentially free cultural works. That is to say that they're in the public domain or they're creative commons, share alike, something like that. Our discipline is being crippled by restrictions put on the use of images that only act as barriers to the free sharing of knowledge in the generation of research. >> Fenella France: And I think it comes back to the fact that I know we all feel that you cannot replace the original. But if we didn't have the original, we couldn't use these new technologies to bring out the information that's within them. So the more that researchers use them and spread that knowledge and become more interested in the original, it's better for all of us. >> William Noel: This is something that federal agencies know very well. So the NEH has been banging on the table about making data openly available. And you know, it's people who are in charge of these -- the physical assets that don't really understand the difference between a physical asset and a digital object. They're so used to looking after physical objects, where what you want to do is really make access very restricted to a very few people in order for the object to survive. But for digital assets, it's the object is going to survive. You happen to open it up so it's completely free to morph into whatever future lies ahead in a digital world. And it's a real fundamental problem. >> Dr. John Hessler: And I think we will end it there. And thank you all for coming. ^M01:28:48 [ Applause ] ^M01:28:51 >> This has been a presentation of the Library of Congress. Visit us at www.Loc.gov.