A few weeks ago I had the opportunity to address the Duke University Library Advisory Board, a distinguished band of champions committed to helping to make and keep our Libraries great. They do not mess around; their schedule for the day started at 8 (that’s 0800) and ended at 12 (that’s 2400)! I followed Don Waters, from the A. W. Mellon Foundation, who gave a great paper on “Research Libraries and Scholarly Communications in the 21st Century;” his point of departure was a New Yorker piece written by Daniel Mendelsohn: “God’s Librarians: The Vatican Enters the Twenty-First Century.”
I thought I’d be the first to make the joke that if the Vatican is home to God’s librarian then surely Duke must be home to the Devil’s librarian! But my timing was off, and Deborah was sitting to my right. Oops. I meant of course, this Devil, the Blue one.
Great, an inauspicious start; so, now I had to crawl back from the edge. I gave it a go with the following:
You might not know it but one of the world’s best custom handmade titanium bicycle builders lives just down the road in Chapel Hill, a guy called Jim Kish. I am a bit of a road biking nut; so I had him build me a bike. If you’ve ever ridden a completely custom bike you know what a life altering experience it is. The bike becomes an extension of your body. It is awesome, just awesome. I even got to go watch him build it, measure, miter, weld the tubes. And man, if you’ve ever seen someone build you your own custom bike, you know you’ve never seen anything so … dull!(*) Jim teaches bike building and tells his students, who are all gung-ho to learn this exciting skill, that most of what you do is measure, file, remeasure, polish, clean. In other words, this completely awesome machine is the result of painstaking, disciplined, thoughtful, creative, composition of very many less awesome things.
In a community so charged to swing for the fence, think outside the box, break down barriers, bear the next big idea—and all of this good—we must also remember that awesome, much of the time, is built the way my bike was. The Egyptian pyramids. Epic poetry. The solar calendar. Facebook. Ant colonies. The library.
I am extraordinarily fortunate to be able to work in an environment, Duke in general and the Libraries in particular, that is as awesome as my bike is. I want to spend few minutes this afternoon explaining what I mean by that.
I am part of a three-person digital classics R&D group (with Ryan Baumann and Hugh Cayless) within the Libraries called the Duke Collaboratory for Classics Computing, or DC3, named and modeled after the plane, which was built quickly without a prototype, called a loose collection of parts flying in formation, and whose longevity was in large part owing to a very simple and flexible design–that by the way is the extent of my knowledge about airplanes; so, if I’ve gotten any of this wrong, please feel free not to let me know.
Our mission is the careful design, development, and composition of very simple (which does not mean trivial or easy to build) standards, services, and tools that support a wide range of scholarly activities aimed in the first instance at classicists, but developed with a view to applicability well beyond the contours of that discipline.
With grants from the Andrew W. Mellon Foundation, whose generous support, along with the Libraries and College of Arts & Sciences, also helped get the DC3 off the ground, we spent much of the last decade building a web-facing environment that allows papyrologists, that is scholars who study the documents and literature preserved on the ‘paper’ of the ancient world, to create peer-reviewed, version-controlled, collaborative, open and public scholarly editions of these 2000-year old text-bearing artifacts. That work is hard. Very hard. These scraps are beaten up, ripped, faded, usually divorced from any known archaeological context, written in non-standard orthography, often undated, and hardest of all, penned for an audience of insiders, actors who not only knew Greek better than we ever will, but also knew automatically all of the surrounding facts and circumstances, that we need PhDs and years of experience to reconstruct. Creating a full-blown scientific, scholarly edition of one of these things is just not a task you give to the uninitiated. It is too hard, has too many moving parts.
But we bring all of those parts together in a common environment; image, text, translation, commentary, scholarly and curatorial metadata; and we allow and encourage even rookies–especially rookies–to propose ideas about any of these parts, no matter how small; to log in and suggest, “You know what, Herbert Youtie read such and such in line three of this papyrus, but I am looking at the digital image and I see something different; here’s what is says and here’s why that matters.” Youtie, by the way is the Michael Jordan of 20th-century papyrology. And the student who made that comment was real, a second year graduate student, and he was right. He may not have been ready–at the time–to create full-blown editions that compete with those of the best, but he certainly could contribute something of value. And he could because we have broken the monolithic scholarly gesture called edition-making into its constituent parts: transcription, emendation, translation, commentary, authentication and attribution, peer-review, and so on. And we’ve allowed anyone to access the scholarly artifact at any level of complexity. We honor and respect even the smallest scholarly contribution, and we support it with thoughtful code and processes. Now, if someone wants to build us a bike, create a complete and exhaustive edition, we take it; but if someone wants to file a single a tube, we’ll take that too; we can work with that.
Our capacity here results from a thoughtful collection of a number of simple but sophisticated services. Basically, we are mitering tubes, measuring twice, laying down tiny welds. But, meanwhile that grad student has formed a geeky club of other grad students whose members use our platform to try to improve on the editions of the world’s best papyrologists. That’s not dull. A team of ambitious undergraduates at Brigham Young University will soon launch a project to translate, in our system, the single largest archive of papyrological documents (about 1800 of them). That is not dull! One of our most active contributors is a retired Dutch high school teacher who spends his days adding print-published texts to the system. Another is a retired papyrologist, who continues to make tremendous contributions from home. That is not dull. Three of our Editorial Board members are graduate students (find me a scholarly journal 25% of whose lead editors are students!). That is not dull.
And because we reside in the Libraries we’re also able to participate in a fantastic experiment to open up some of Duke’s Byzantine Greek MSS to the kinds of fine-grained digital crowd input that we currently support for our papyri. And part of our test bed will be a travel journal, held in the Rubenstein, written by Viscountess Emily Anne Beaufort Smythe Strangford during her travels in Greece, Turkey, Syria, and the Ancient Near East in 1859/60 (her father-in-law is the very Strangford for whom the British Museum‘s Strangford collection is named; see e.g. the Strangford Apollo). One of the project champions of this collaboration among Special Collections, Digital Projects and Production, and the DC3 has been calling this team the Avengers! I love that; it reminds us that we can do big things when we bring our different, complementary strengths to a venture. (Plus, Will said I can be Thor if want). I want to make a couple points about this. First, only in a library would such a diverse team bootstrap a toolset designed for papyrology to work with Byzantine MSS and 19th century travel diaries at the same time! Second, I was recently at a conference on scholarly communication where a panel leader asked whether libraries can afford to innovate in the Digital Humanities space. But when I think about this particular project I think that libraries cannot afford not to; that if we innovate by starting small, nimble, flexible, we can build our way up or over to adjacent applications; that this bold strategic R&D gesture of embedding traditional academic interests inside the library, will help us not only to “afford to innovate” but also to break a bit out of the dominant paradigm of libraries serving scholarly developments and a bit more into the role of driving them.
I’ll move from here to our next big project, and then close with a brief comparison of the library and my bike.
The DC3 is now aiming to tackle the world of ancient Greek and Latin epigraphy, public documents carved in stone and erected in the city-states of antiquity for anyone to see and read. We have laws and decrees, gravestones, letters from emperors and kings, inventories of dedications to the gods, records of property confiscated as a result of legal judgment, the list could go on. We know well over a million of these, from Britain to Afghanistan. Maybe 700,000 are published in some digital form in one of about 8 major resources–several other smaller ones; thousands are translated into a modern language; tens of thousands of pages of scholarly articles and books have been written about these things; photographs or paper impressions (called squeezes) exist in research archives, or on the web; these texts mention persons and places; they are cited in innumerable lexica and encyclopedias; they are ‘eyewitness’ evidence of important legal, social, linguistic changes that would otherwise be more or less unknown to us. They are our single best window into the lived public life of real people from the full footprint of the ancient world. And epigraphy as a discipline started in the 1820s or so. So, there is a lot of work done already.
Digital too. But the majority of it has been conducted by epigraphists working with little technological background, without reference to accepted encoding standards, without a view to interoperation across the landscape of multiple related resources–basically, without the benefit of collaborating with world-class programmers and a world-class library. There is currently no stable, vetted, and transparent way to assert that the text in one database has been translated in another resource, and in yet another has a corresponding image that shows that the text should be reconstructed in some other fashion, which runs contrary to scholarly views expressed in article that can be found in another. It is a giant mess of disjointed data. A beautiful mess. Imagine a library in which the catalog records were written by 30 different teams, without reference to any published standard and without communication with each other, on 9 different incompatible software platforms; and then imagine that one night elves snuck in and rearranged all the books randomly. Imagine an internet with giant holes in it, so that you can post to Facebook, but the only way to get that content into your blog is to send a hand-written note through the US Postal service to a transcription firm in Munich. That is the state of digital epigraphy. We mean to provide the suite of services that brings the network together; to create an environment for generating, reviewing, publishing, and archiving scholarly contributions that run across this vast and varied landscape of epigraphic content.
We’ve only just started. But for example, we are mapping provenance data for about a half million of these artifacts to their geo-coordinates. We’ll soon have the ingredients for a smart phone app that knows where you are and can show you texts and images of inscriptions that were found within some number of miles of your location, plus JSTOR articles related them. We’ll be able to browse texts not by lists of publications, but via geographical interface. Our killer app, if we succeed, will allow you to capture an image of an inscription on your phone, ping a service that ‘reads’ the inscription, matches it to published texts, translations, secondary literature, other images, then deposits the image in an institutional repository. This will of course be a great boon to scholarship but it is also essential from a preservation and access perspective. The majority of inscriptions have been published without accompanying photographic control. The legibility of these stones, which often sit out in the elements, is not improving with age. There are parts of the world where inscriptions come out of the ground and may vanish before they can be edited. But how many tourists, students, locals, or classicists walk past these objects every year; some even taking photographs, and even posting them to flickr or other sites? Once the infrastructure is in place and the app built, we’ll have a simple mechanism that allows these individuals to use the photos they’ve taken anyway and contribute meaningfully to the progress of knowledge. We don’t need to train a legion of epigraphists. We just need to break epigraphic science down into small parts that anyone can handle, and create a fabric of simple services that allow people to contribute what they can.
Then build teaching and research structures around them; so, lead study abroad students to archaeological sites and museums and assign them treasure-hunt style punchlists of inscriptions to photograph. Or, better yet, spin up a MOOC (maybe a smallish one; call it SMOOC) and draw in students from around the world, even from the very regions where inscriptions are found. Or hold a contest in which the first team to produce so many hundreds or thousands of images of a certain character or distribution will win a prize. It is not hard to imagine a ‘collection development’ campaign that could muster many thousands of invaluable images for little money in just a few years. Once this works for Greek and Roman epigraphy, it ought to work for medieval manuscripts, for gravestones around the world, for archaeological remains, etc. This could revolutionize the way we do epigraphy, the way we do text criticism, the very way we think about “the collection” and the Library’s role in its constitution and care. What if we’d had such a collection development tool and network in place 12 years ago in Kabul, or 2 years ago in Libya or today in Syria. It would have been awesome. And we’d have gotten there through painstaking, disciplined, thoughtful, creative, composition of very many smaller contributions, a picture here, a few annotations there, some translations there.
What does a citizen with a cellphone standing in front of an inscription in the Syrian town of Homs have to do with an American University Library? It’s a fair question. Libraries have always been catalysts, I think, specialists in creating conditions that allow communities to turn information into knowledge. That information has for centuries been trucked in, in physical form. Of course in the modern research library that space is both physical and virtual; that information is both print and digital. Whether the source is a publisher or a cellphone-carrying student in Tunis, the Library’s role and responsibility to the data is potentially the same. The ‘collection’ is whatever the Library can get its arms around. Its meaning and significance is whatever sense the Library and the community that it nourishes are able to impart by curating the relationships of materials to each other. In this way, the challenges and disruptions posed by the digital revolution are not new. They ask the Library to do one of the things that the Library does best; in different ways and with different costs (not to be underestimated), but the intellectual work and its value are much the same.
Working in a place with such an ecumenical view of the inherent and abiding value of human knowledge is energizing and more than a little bracing. The academic side of the quad tends to reward innovation; you discover it, you publish it, you move on to the next thing. And that is fantastic. An academic builds something and another academic asks, “Great, what does this do and why should anyone care?” A librarian asks the same and then, “Will it be able to do the same 5 years from now? Or for a user who does not know English? For a user who is blind? Or for a user who is not an expert in this field? Or one who does not have access to the following for-fee services? And how can we host it?” These are not small or easy or merely administrative questions. Many of us don’t like to be asked them! But repeatability is an essential attribute of scholarly production, and access is a precondition of repeatability. When a librarian asks these questions she is in effect asking whether a scholarly project has an honest right to the title scholarly! Knowledge is not the next big thing, but the next big thing, in the stable context of its relation to all prior and future things. On university campuses, the Library is the steward of that ever changing context. Generation and stewardship of knowledge are not, in my opinion, separate but equal endeavors, the split domains of faculty scholars and librarian service providers. Rather, they are inextricably linked, each depending on and nourishing the other. And so for us, for the DC3, one of most exhilarating things about being embedded in the Libraries is the opportunity to do work at this interface between a faculty culture that encourages “Make it new”, and a Library culture that insists “Make it last!”
The DC3 feels at home in the Library, in part, I think, because we are attempting to build things whose awesomeness is an emergent property: the individual parts are crucial in and of themselves, individually beautiful; but their real value is seen when it all comes together, when a whole discipline, from students to emeritus professors and even outsiders, convenes around them to own its core data resources. Libraries are the same. Their distinct awesomeness is an emergent property; it is all of the individual moving parts coming together and functioning as a seamless whole, as an extension of the user; it is bringing millions of items under a single real or virtual roof, and cataloging them so that the user can move through and across them as they like (the web was not the first technology to deliver non-linear navigation through information; the library was; one book is a nice thing, a quarter million are a serious problem!); it is staffing the place with both domain and tech experts; it is creating an environment in which every user’s experience is in effect custom tailorable to his or her needs and objectives.
The Library, in other words, is and has always been awesome, in precisely the way that the DC3 aims to be, in precisely the way that my bike is.
(*) Jim: It wasn’t dull…this is just a ham-fisted rhetorical device. To me, it was almost as exciting as riding my ti beauty over 20 miles of half-frozen muddy gravel!