>> From the Library of Congress in Washington, D.C. [ Pause ] >> Abby Yochelson: Welcome to another in the series of LC's Digital Future and You, a wonderful set of programs that's been going on for several years. I'm Abby Yochelson. I'm a reference librarian over in the main reading room in the Jefferson Building, and I'm the Humanities and Social Sciences Specialist in English and American Literature. So when Roberta Shaffer could not be here to introduce Dr. Rowe, I was asked to do so. And I'm very pleased to be able to do that. Dr. Rowe is a professor of English at Bryn Mawr College, and she also directs initiatives in digital teaching and scholarship with nearby Haverford and Swarthmore Colleges. She is a well-known Shakespeare scholar -- extensively published. You should see the list on this CV. It's really, really long. And a very popular teacher and speaker on subjects like Renaissance drama, teaching early modern English, and Shakespeare adaptation. She's also served on the Editorial Board of the Shakespeare Quarterly and is a trustee of the Shakespeare Association of America. But she also teaches and writes about media change from the Renaissance to today's digital era. And yes indeed, there was big-time media change back in the Renaissance. She has an emphasis on cultural adaptation, performance media, and the evolution or the history of human emotion. She's got books on New Age Shakespeare on Screen and Dead Hands: Fictions of Agency Renaissance to Modern. So I would describe Dr. Rowe as cutting edge in the field of digital humanities. But the New York Times described it much more eloquently, saying that she is one of a small vanguard of digitally-adept scholars rethinking how knowledge is understood and judged. That's way better than "cutting edge." So anybody else here remember the early days of the net? We called it the "net." Back when the complete Shakespeare was put up online by someone at MIT. It was in 1993. And we thought it was miraculous, this full-text searchable Shakespeare. It was revolutionary. It now looks unbelievably primitive twenty years later when you see what Dr. Rowe is working on. She is a founding partner in the Luminary Project and co-Director of the Luminary edition of Shakespeare's Tempest, and Macbeth will soon follow; is that right? Is Macbeth in the works? This is an app for iPads, and it makes Shakespeare so very 21st century that it can change the way we teach, and learn, and read. And so it's a wonderful example. So at this point the temptation is great to whip out that line from the Tempest, "O brave new world," but I'm going to do something from Caliban's speech instead: "...and then in dreaming, the clouds, methought, would open and show riches ready to drop upon me..." So I give you Dr. Rowe to drop some riches on us. [ Applause ] >> Katherine Rowe: Thank you, Abby, for such a gracious introduction. I'm really thrilled to be here today. I want to start by thanking an array of people that made this possible. When you work with technology -- and today I'm going to be working between different technology platforms -- it always takes a big team. So many thanks to Judith, Beth, Angela, Karen, Matt, and Roger for making this so far quite seamless. We'll see what it's like when I start to play with the tech in front of you. As Abby said, I began my life as a Shakespeare scholar, and I have been intensely interested for my whole career in the history of the book and the history of performance technologies. And over the last five years I've been drawn by my teaching at Bryn Mawr, as well as by my research, into working with new media forms of the book and performance. I'm in that category of digital humanists who's invested in studying media change itself. My particular interest over the last several years has been changing modes of scholarly communication, research, and practice, particularly classroom practice -- how academics organize our daily work as teachers and scholars, how we communicate, how we publish new knowledge and build audiences. Today I'm going to give you two examples -- specific examples -- of what I mean by "changing modes of work and scholarly communication" from my own research and publications. Each example should take about twenty minutes I'm hoping. And with the media change, that gets us to about 45 with ample time for Q&A and more demo if you're interested in that. My goal is to give you a starting place for a conversation with you about the changing landscape of digital research and what it means for research libraries like the Library of Congress and for the public that learns, and teaches, and studies here. My title comes from a wonderful interview with Jeffrey Schnapp on the humanities digitized in Harvard's alumni magazine last year. "When you move from a universe where the rules with respect to a scholarly essay or monograph have been fully codified," Schnapp says, "to a universe of experimentation in which the rules have yet to be written, characterized by shifting tool kits and skillsets, in which genres of scholarship are undergoing constant redefinition, you become by necessity a knowledge designer." I hope this idea will resonate as I talk about what has changed in my own work over the past five years. Now, there's two headlines I want to give you here. The first headline is that I and the scholars around me in a number of fields are increasingly making the resources that we use for research and teaching. I'm going to give you humanities examples of this, but I'm seeing the phenomenon across divisions of academia and in the sciences in particular, from undergraduates up through the faculty level. Second is that this movement into sustained knowledge making with digital corpora and tools requires new collaborators for us. My work has always been relatively collaborative. I co-authored a monograph and a collection of essays. I've written, co-authored stand-alone essays. But unlike those traditional research projects, the projects that involve digital text and methods include not just other scholars, but also students, library staff, information services staff, and software developers -- involve them necessarily. In other words, I can't do that work without them in a way that I can largely with my traditional scholarship. I'm going to give you two examples -- one for each of these changes -- so the change to the increasing making of the resources that we use to do our research and the change in collaborative modes. I'm going to be going quickly so that there's enough time to get through both of them, but then I'm happy to talk at more length about either of these headlines. Let me start with an example of new kinds of research. This is a research collaboration in the planning stages. We're moving to the second phase of grant planning. A project that's led by the Folger Shakespeare Library around the corner. The Folger Digital Folio of Renaissance Drama, which we're calling "F21" for short. This is a multi-year initiative, currently in the planning stage, as I said. Its goal is to create interoperable digital editions of some 500 plays by Shakespeare's contemporaries and subsequent writers that were written or performed from within thirty years of his birth to about 1700. So this is a text-based project. It aims to build a consistently transcribed and interoperable digital collection of plays that will be the basis for future machine-assisted research. Once established, it should sustain both automated and human tagging in the service of scholarly inquiry over a longer period. When I say "tagging," for how many of you is that a familiar term, "digital tagging"? Okay. Let me just say in simple terms to tag something is to put it in a bucket, to say, "I can now search for it" -- I'm interested in a stage direction or I'm interested in the gender of a character. I create a tag system, I build into the metadata, tag associated with those categories, and I can then search it. But I have to identify those categories in all the body of materials that I want to search. So some of that work can be done in an automated way, and much of it has to be done by intelligent human beings -- at least right now. The Folger's the lead institution in our recently concluded planning grant from the Mellon Foundation, but the planning process has included partner libraries, such as the Bodleian and consultants from an array of major digital archives. A key component of F21 is the inclusion of undergraduates as digital curators and as student investigators. Undergraduates will take on key tasks of digital transcript correction and tagging and claim some pride of ownership in the creation of apparatus and research projects. We imagine three components to the project, managed centrally from the Folger. So the first is a digitization initiative. The Folger and the Bodleian would provide the digital surrogates of the plays for the purposes of transcript correction, which here means all non-Shakespearean plays of this period; approximately 30,000 page images. We'll be creating a lightweight working environment. MITH -- the Maryland Institute for Technology in the Humanities -- will create a working environment allowing transcribers to toggle between those facsimile page images, corrected transcriptions where they already exist, and the student transcription. Folger staff would oversee linguistic annotation, data storage, and other infrastructure issues. I'm going very quickly here, but I can go deeper if you'd like. The third component is a pilot lab with a research agenda governing downstream decisions about what to tag and why. A key assumption of machine-assisted research in Renaissance drama is its corpus value, the value of expanding the dramaturgical and dramatic context for Renaissance English drama to include the full scope of plays contained in early English books online. That's a radical expansion of what's in fact an extraordinarily small set of text, including Shakespeare's, that scholars in the field regularly study. The lab would establish umbrella research questions within which an array of future scholarly projects might be pursued. By tagging dramaturgical structures, for example, in and out -- when somebody goes through a door, off the present stage, or back out of a door -- we know at those in, exit, and entrance instructions that there are either two or three doors on stage. How are those doors being used? What fictional spaces are being mapped by the gestures in and out? It's an incredibly interesting question for a Renaissance scholar because it goes to dramatic action on stage; it goes to how characters -- players, that is -- establish the scene, the fictional scene for the audience and the audience imagination; and it gets to questions of how theatre architecture makes that imaginative work possible. We are hoping in this way to widen our range of models of Renaissance theatrical practice, to deepen our understanding of the performance conventions that shaped Shakespeare's plays and his contemporaries. To pursue such agendas the lab will convene a community of practice to design and implement some fairly complex workflows. We imagine senior post-doctoral fellows at the library as project managers, supervising undergraduate teams in transcription, TEI markup, and downstream tagging in the service of research. Those undergraduates would advance into this work from courses at pilot institutions. With Folger staff, the principal investigators are investigating a common introductory curriculum that we can trial over the next few years. We're aiming for small, low-risk activities that might sit within a survey of Shakespeare or Renaissance drama. For example, transcribe and tag a page or a scene according to our metadata tagging schema. Such short assignments will introduce students to ways of thinking about the print history of play texts. What gets onto the page of a play is a wonderfully interesting question for me as a historian of the book, someone that's interested in the history of the book, about text proximity and linguistic similarity, concepts taken up in broader debates in my field about the nature of genre change, dramaturgy, and literary influence. Now, a key ambition here is to pilot a model of collaborate humanities research that integrates the labor of scholars, library staff, graduate students, and undergraduates and that could be scaled up and extended via traditional academic structures: Our regular work in our classrooms, what we do when we go to conferences with each other. In other words, from the infrastructure point of view, from the workflow point of view. We want this to sit inside what we're already doing to the extent that we can. Students interested in advanced work would join the Folger lab via summer boot camp for training in digital curation in research development. Now, here's where I could get very literary nerdy and talking about the kinds of research and teaching workflow questions we've been grappling with: How can you anchor a really low-risked curation task, correcting or categorizing words, tagging lines to speakers, and so on in resonant research questions? I want to say that in the sciences I think that bridging task, which requires the imagination of a teacher and scholar, to bridge a routinized management task of some kind -- let's say staining slides in a diabetes lab -- to bridge from there to why this is meaningful work, what makes this compelling intellectually? We have been less good at doing that in certain fields in the humanities. Some fields are good at it. The classics, for example, has been very good at connecting transcription and translation, which they do regularly, with larger learning questions and with intellectual practices that they care about and are conveying, even to students who will never become classicists. In literary studies we have been less imaginative about that. And I think we need to become more so. So the point that I want to make about this is that these are research and teaching questions that I can't easily answer alone. I need the expertise of a librarian who's thought deeply about curation in both traditional and digital formats to design assignments that will work and that will build those bridges intellectually, and to craft a curricular path for my undergraduates that is going to implement those learning goals. The production of such a corpus, as we're imagining it in F21, opens new institutional paths for learners in turn as they progress from an intro course, through a boot camp, to perhaps a senior research project back at their home campus. That means a new structural role for the Folger as a collaborator in the enterprise of teaching undergraduates and it fostering undergraduate research, something that undergraduate institutions across the spectrum are very interested in right now. What's at stake in this project in the even larger sense is what it means to make a digital corpus accessible and usable and for whom. Access to and use of knowledge are the traditional missions of the university, the college, the research library. But to make our digital corpora play text really accessible and usable as digital objects, rather than as just electronic surrogates of print objects; in other words, images of a page that happen to be digital -- to do that requires robust data curation and print resources -- sorry, robust data curation that print resources have not required or invited yet to the same extent. This is a field-wide knowledge design and sustainable curation challenge for Renaissance studies. And as a field we need the Folger to convene to us to address it. There's no other entity right now in Renaissance studies positioned to do that. So that's the third really significant shift it seems to me or at least extension of the mission of a research library is to convene fields into action where it's needed around large-scale heritage curation projects. As a teacher, there's a difference set of interests, long-standing interests at stake here as well. What it means to educate students to be citizens and professionals in a networked world. Students involved in data curation get their heads and their hands under the hood of a digital corpus in a way that demystifies digital media. And that demystification, I think, lasts and is going to be crucial for sophisticated citizenship now. My co-PI, Martin Mueller, of Northwestern has proved this quite powerfully in earlier trials of collaborative data curation with undergraduates. So I'm going to close this section of my talk by sharing his results from one final course assignment in an advanced Renaissance drama seminar in which Martin's undergraduates prepared, corrected, and linguistically tagged digital editions of eight Renaissance English plays. Martin has been running assignments like this for three years, and he's shown that undergraduates with moderate training can produce resources that scholars will want to use. But what's most striking about his results is the payoff that the students ascribe to their participation in the scholarly transcription and digital curation project. The first is their powerful sense of being useful, of doing something real and lasting. "I felt like a digital new age monk or something, having contact with the original texts or the digital copies of them and putting them in a database that would be widely accessible and searchable. It was cool to be part of a project so grand." Second, their deepening understanding of the verbal constitution of texts, a kind of radical close reading, another change in mindset that this task has led me to is that a text is built of words and words alone. It's not that this is a startling revelation, but it's easy to let that fact slip from consideration. And we forget that each word has its own important place in a text, and its frequency, associations, and placement are all loaded with meaning. There's a deep and clear connection to literary values in these two points, I hope you can see. Finally, Martin's students recognized in a way that I find really compelling that all texts, whether in codex or online form, are built things -- designed for purpose along specific interests and biases that a sophisticated reader needs to understand before consuming. I can see you smiling. Any teacher in the audience will really love this. "I chose to do the data curation project because I thought it would take less time than a paper. [Laughter] That was absolutely not true. However, it did engage my brain in a way in which it hasn't been engaged in a long time, and that is in the critical analysis problem solving sort of way." If you haven't been involved in digital tagging, it sounds absolutely like drudge work and the opposite of critical analysis problem solving work. And what the students here are showing is that that's not the case, counter-intuitively. I'd say that the biggest thing that doing data curation with Analex has offered me is some perspective. Ironically, it's programs like this that lead us to feel like computers perform most tasks, when I'm now really struck by how much manpower goes into getting such an application right. The recognition that our knowledge resources are designed for some purposes and not others is a crucial kind of sophistication for anyone living and working in the networked world -- crucial, I think. For these students, it comes from having inhabited the role of scholar producer engaged in making a corpus available and accessible to research. I've been describing a deeply humanities-based kind of scholarly production, but I want to underscore that the current interest among academics in making the resources that we use for research crosses divisions. And we're seeing the same effects of participant interest at Bryn Mawr and I'm sure at many of the other institutions that you're affiliated with in this room. We're seeing students outside the computer science department, for example, not only using digital tools, but creating their own digital tools from the ground up. Here's two quick examples. A geology student who designed and implemented a software tool to analyze data from an open ocean buoy using mathematica. Chemistry students who can write code to implement the quantum mechanical models that most students learn in theory only, a standard assignment in a class of a colleague of mine who's been building her own software models for molecular interactions for about a decade. So what I've been describing in Martin's work with Analex is part of a phenomenon that's much wider than the humanities, though it has particular challenges for humanists as we begin to harness the energy of students who are interested in making the tools that they want to be using. A participant's interest in and an understanding of the designed nature of digital resources are to my mind the crucial aspects of what it means to be an educated citizen now. And to be a teacher or scholar, I think, we have to extend that interest in making tools into the arena of the tools that we use regularly in the classroom. So I'm going to shift to my second example now of the scholar as knowledge designer and talk about a reading app I developed about 18 months ago with a colleague at Notre Dame. My partner is Elliott Visconsi, also a Renaissance scholar. My partners are also the team of experts; scholars; secondary teachers; actors; directors; editors; designers; and software developers that helped us build this app, which is now in the app store live and is the prototype for five more apps that we will be releasing in November in partnership with the Folger Shakespeare Library and with Simon & Schuster. What I'm going to show you today is this first release, Shakespeare's The Tempest for iPad. It was launched by Luminary Digital Media, a small company that Elliott and I founded to publish iPad editions that match what a literary scholar and a teacher needs an e-book to be. Our initial motivation for doing this work was a keen sense of frustration about our field's inability to solve the audience problem of the humanities. We don't think we have a crisis of the humanities, but we do think we have a challenge of building audience, connecting with audience in the humanities. The second motivation was our sense of a growing breach between scholar-teachers and conventional publishing. We were moved to experiment with alternative ways to reach readers potentially interested in the difficult and satisfying text that we study and teach, to lower the difficulty of their first encounters, and to find new genres for public scholarship. The pedagogical context in which we work as Renaissance scholars also gave us a keen sense of the inadequacy of the current generation of e-books for classroom use. So let me give you some background. The software framework was developed, as I said at Notre Dame, at the Center for Research Computing, which used this project to grow and train an iOS development team under an internal grant for scholarly innovation projects. We launched the first publication, The Tempest, in April last year, then sought and won modest angel investment that has brought us to our current point with a framework that can stamp similar apps for other texts and we hope for other teams of scholars. The app's in 28 countries and a score of high school and college classrooms last year, and the response from students has been lively and enthusiastic. I'll talk a little bit more about that later on in this presentation. We launched in iOS, that is, on the iPad because that gives a fairly large initial distribution system, and it's a simpler development environment because there's paradoxically because it's so controlled by Apple; whereas Android developers are generating new code for new form factors fairly regularly, it's harder to manage that development process. But we're hoping to expand to Android devices by the end of this coming year and to be platform-agnostic as quickly as possible. What makes this possible and sustainable from a technical standpoint is that what sits behind the app is a database. All right? So not a facsimile of a page. The text itself -- the play text that I'm going to show you -- exists as a database and can be displayed as a page or a screen in many different modes, but it's not being reconstructed from a page, tagged, that is, and rebuilt as a database as in the prior project that I just told you about, F21. The play and the college scholarly commentaries that are paired with it exist in a way that's expressible in multiple electronic text formats and tagged in order to be searched or manipulated. Elliott and I came to our collaboration and the idea of building an app through similar paths. We're both early modernists who pursued an intensive period of mid-career retraining with the support of a New Directions Fellowship from the Andrew W. Mellon Foundation. His was in constitutional law and free speech; mine in media studies. We're both scholars who work by habit, pleasure, and conviction at the edge of our own mastery. I think that's a little unusual for academics, to like to be right on that teetery edge of what you really know. We both work with texts of a similar kind: Large heritage corpora edited in successive generations for sale by academic and commercial publishers. And we were encountering similar phenomena in our classrooms at very different kinds of institutions -- me at a small liberal arts college; Elliott at a large research university. An increasing number of students are using tablet devices, and I can see the difference between my freshman and my seniors in the number of tablets that come into the classroom. The change is very, very rapid. And the research that we're looking at suggests that the younger students get, the more likely their families are to own a tablet, the more likely they are to want to use it for homework, for pleasure, and so on. These tablets are increasingly the primary device of undergraduates, partly because they're cheaper than laptops and partly because they are catching up in their functionality with the functionality of laptops. Students are regularly asking for e-texts and seeking sophistication in how they use those e-texts. And we didn't have a resource for them in our classrooms that was adequate from either an editorial standpoint or from a functionality standpoint. Both of us also see in the rapid pace of media change now opportunities for highlighting the value of long-standing reading practices. And both of us teach comparative courses -- his on the history of the First Amendment; mine on the history of reading and writing technologies. We read promiscuously on multiple genres and platforms. I am a binge reader, I like to say. And there's nothing better than an e-book at 11:30 at night when you need a new novel. And we were finding ourselves, like our students, more promiscuous rather than more settled with the advent of tablet readers. We were both frustrated with our traditional publishers. We both sought ways to begin to build the e-books that we wanted with publishers who we worked with for a long time and really respect and recognize the difficulty that they face with large companies in moving their business practices to a different platform. For the first six months of our intense collaboration our relationship developed entirely via electronic media. We met once in June, and then we worked via phone, Skype, email, text, and Words with Friends. When you're trying to get to know somebody and you're a literary scholar, your ability to play with words is a great asset. Our design principles grew out of shared frustrations, as I've said, with key aspects of the current state of incunable readers. First, their unsuitability to core humanist practices, the readerly writing and writerly reading that requires handling a text that constitutes textuality in any medium for a literary scholar. One of the chief powers of electronic print is its manipulability. Traditional publishers, commercial and academic, have been slow to move their EPUB technologies and DRM policies in this direction. Second, the economics disincentives that they face coincide with a model of knowledge that emphasizes solo cognition in reading. What Ian Bogost would call the procedural rhetoric of commercial e-readers is captured in that iconic one in a tree slide branch -- right, of Kindle, the reader under the tree -- a liberal arts brochure update of Jane Eyre in the window seat. "Made by a means of a computer model," Bogost explains, "a procedural rhetoric makes a claim about how something works by modeling its processes in the process native environment of the computer, rather than using description, writing, or depiction images." What I'm going to show you going forward is how we've been thinking about the procedural rhetoric of a reading environment. Even the most exciting of the new high-end app books model reading as multimedia processes of delivery and consumption, rather than of curation, co-curation, and social engagement. In a [inaudible] at Flow TV in last December, Mara Mills observed that reading is quote "a sphere of activity like play." What that means is that in our most intensely social modes we read as we play to collect companions who intensify the experience. So our design logic for the luminary apps is social reading. Our goal is to bring the procedural rhetoric of the app to a place where it models the social constitution of knowledge via reading that thrives in our classrooms. This means that we view texts, including multimedia formats, both as gateways to social experience and exchange and as social experiences in their own right. Even Jane Eyre reads to find different company, companionship in a writer, and in the writers that that writer reads. This logic dictated an array of annotation and social note-taking tools with more planned as we iterate design. We wrote terms of service and privacy guidelines governing this co-created content that we would be happy to sign ourselves. So I'm going to show you what that means concretely by switching to my iPad. How many of you have tablets? Use them regularly for reading? Okay. So what I'm about to show you is going to feel familiar to that crowd, but I'll go slowly for those who don't. Our first title, Shakespeare's The Tempest for iPad, is available in the app store. It took about 11 months to develop -- pretty intense. Development at Notre Dame, as I said, was on an internal grant. We've had terrific support from Bryn Mawr as well for my time. We've taught with it in multiple classes. So I'm going to give examples as I go about how the students are using this environment -- how they read, how they take notes. The first goal that we had for this app is that the initial reading experience is going to be a quiet one. This is a customizable reading text. It can be as complex as you want. So I can open up a little gloss there to find But there are no media links out, unlike reading on a laptop. You can find them when you need them, but the tablet form factor keeps you inside your text. And students remarked that it's far less distracting -- that's the way they describe it -- than reading on the web in a web-based environment. There's no Facebook pings here. It's also lying flat. You can't see that here. But the difference between my laptop, which has even a short sort of bar between us at a seminar table or in a lecture hall and a nice, flat text like a book is really significant one socially. All the usual resources are here for reading Shakespeare, so I've shown you glosses, there are scene summaries as well, a cast of characters. For example, these come up as pop-ups. And there's bookmarks. Let's see, where will I go? I'll pop right here that you pop in yourself. And there's also some very significant improvements. My favorite is this dynamic line counter. This is the one that scholars working with this text like best. You can pop a class immediately to the same act, and scene, and line number without having to shuffle through the pages. And that wait time of just a few seconds is enough often to lose an initial reader who is really having trouble orienting himself or herself in the architecture of a print play as opposed to a novel. Even better from a scholar's point of view: You never have to count line numbers because that line counter is dynamic. This is the thing that thrills the academics I show it to. You just slide from our little left-hand pointer to whatever line you're at, and it will tell you where you are. So when I'm writing about The Tempest now this is my default text because I never have to count line numbers. You can highlight. You see I've done that here with the standard navigational tools that iOS supplies. You can take a note. This is the note that I took for a class that I taught this spring. This is in a really interesting moment in the play, a moment where our perspective shifts from watching a shipwreck at sea to suddenly being with those on -- we thought -- abandoned island, watching that shipwreck as well. So we've done what in film studies would be called "180-degree shift." And I'm really interested in that shift in perspective. We're about to get a lesson from the two characters introduced here in sympathetic spectation. Miranda's going to teach us how to watch sympathetically, and her father, Prospero, is going to control that watching. So this is my note for what I wanted to do with the scene in class. What I want to emphasize about this text is that the margins are both dynamic and porous. I can send that note out via email or via Facebook to the class group if I want to. I could take notes of my own, running notes. I can collect them and send them off to an essay that I want to write. So unlike a conventional EPUB format, the margins here are even more porous than they are in a print text, and they make social the possibility of exchanging your responses with each other. We're modeling a reading experience here; our aim is to model a reading experience so it's like that in the classroom. So we also have a rich array of commentaries -- oops, sorry, come back -- from the actors who've provided full text audio for this play, from the editor, from a series of wonderfully different scholars who have their different takes on the play. This is the marvelous Harry Burger, one of the lions of our field. Prospero is like someone who lost his job in Manhattan and was forced to raise his family on Long Island where he lived in humiliation for twelve years without a kitchen sink until the Big Apple gave him another break. When the play starts, he's still rankled, still humiliated by his layoff and exile. Our goal here was to bring multiple voices to bear on the text. We did not ask these scholars to agree with each other. We didn't tell them where to cluster. And when you look at this line of commentary and hop around in it, what you find is that the places that the play generates interpretive opportunities and disagreement rather than as with a conventional edition going to the introduction to hear how those interpretive problems are named and resolved. Those commentaries are tagged to the play, and so as you slide through the play you'll see how they cluster. And you can be carried to them as you go. You see that? I'm adjusting it slightly. And it will tell you if there are multiple ones at any given moment. I'm particularly pleased in our recent review in the TLS, Times Literary Supplement, that this commentary function was called out for what I think it is, which is an archaic reading practice as well as a cutting edge one. The reviewer said that we were reviving the defunct -- that's probably not the word he used -- but reviving the genre of the commentary in a way that made it lively and accessible. And he saw as the future of e-texts of this kind. That was partly our goal. And we've built in, as Renaissance scholars, an array of longstanding humanist practices, tying them into what you can now do in an e-text. So the text itself is also manipulable. I can collect passages, as you see here. These are passages for a class that I thought that are examples of the characters who are sharing conflicts or amorous affection over that central figure of Miranda. This is a thematically collected set of passages. And this is the set of passages that show us how many different characters are framed in one way or another as slaves or as indentured to each other. And what the differences are between the vocabulary of slavery that clusters around Caliban, the monster on the island, and that cluster around Ferdinand, the prince and lover. You can see those differences in an immediate and really visceral way. We've also hard coded what I would think of here as collections of the text. We've hard coded the roles with cue lines. Let me turn on the cue lines. Come back. Miranda. [ Pause ] iPads don't like very much to be projected. So let's see if it works. We were having trouble earlier. There's Miranda. I don't know why the King of Naples is still there. Go away. Okay. This is Miranda with her cue lines. And as you see as you scan through this world, this is what a text version of a play text would have look liked in the Renaissance -- the earliest written versions of the plays would have been roles with cue lines of the other parts. When you collect the text here, you notice something really striking about her. Who's she talking to? Right. Basically one person. She has a few exchanges with Ferdinand, Caliban prompts her in one occasion -- I'll find it -- but other than that, her dominant interlocutor is her father, Prospero. And you can see the way a part is constituted -- a character is constituted intersubjectively, socially -- this way when you can look at the part with its cue lines quite immediately in a way that's really not visible in the print compiled version of the play. You might be able to hear that on stage as you notice those kinds of interactions, but it's not something that the reading experience generally gives. This space on the left here -- the commentary space -- is open to a teacher or class to publish out to. Let me see if this version has. There's my multilingual Shakespeare class. Hang on, let's go back. This is a Folger master class. So these are collections of commentaries published into this space to private teams or to public teams in the case of the Folger master class that allows you to use it as an opportunity for the students to create rich curatorial content themselves, to be experts. This was an advanced class that was preparing a study of Caliban's part for production in Abu Dhabi this past spring. And so we were reading the translations of The Tempest in multiple other languages, and they were commenting on those and redacting key elements of those translations for each other and the scholarship and criticism around them. So I should add that this is unicode development, so that we have a Chinese character set here. And we can do any other kind of character set that we want. So that we imagine this work as multilingual as well as multivocal. The play is polyvocal in another sense: There's a full professional audio performance. Let me go back to the main text here. Actually, let's go back to this spot. Again, where Prospero and Miranda come in. There's a full audio performance of the play by actors from the London stage. >> And more, Miranda, but how is it that this lives in thy mind? >> Katherine Rowe: And it scrolls with the play as it goes. And then we've created multiple audio takes. I'm looking for them. I have a note here somewhere. Right here. So let me play you -- the goal here, again, is to think about what it means to have a play that's speaking in multiple voices rather than a single voice. So we asked Dale Rapley, who plays Prospero, to give us his version of Prospero, which I'm going to play first, and then to give us a very different one. So here's Dale's version. >> Never meddle with my thoughts. >> Lie there, my art. Wipe thou thine eyes; have comfort. The direful spectacle of the wreck, which touch'd the very virtue of compassion in thee, I have with such provision in mine art. So safely ordered that there is no soul-- no, not so much perdition as an hair betid to any creature in the vessel which thou heard'st cry, which thou sawest sink. Sit down; for thou must now know farther. >> Katherine Rowe: All right. So that's Prospero as the loving, nurturing father who's concerned for her anxiety. Here's our Prospero, and I want you to listen to the pronouns in particular, the way Dale delivers them. >> 'Tis time I should inform thee farther. Lend thy hand, and pluck my magic garment from me. So lie there, my art. Wipe thou thine eyes; have comfort. The direful spectacle of the wreck, which touch'd the very virtue of compassion in thee, I have with such provision in mine art so safely ordered that there is no soul -- no, not so much perdition as an hair betid to any creature in the vessel which thou heard'st cry, which thou saw'st sink. Sit down; for thou must now know farther. >> Katherine Rowe: So that's Prospero as a tyrant. Yes? And the same set of pronouns have really been flipped. In his first version, Dale leans on the yous, that second person address. In his second version he leans on the first person -- I, mine, and my art. And the difference in tone is very vividly demonstrative of how plastic this language is, how richly interpretable. In about 15 seconds a student who's coming to the play for the first time can hear that instability and that opportunity. And one can then go on and, say, use the iPad's recording capacity to give us your own version of a different speech here. Read it differently yourself. The goal here is to move the students from comprehension, making the play text more accessible, to interpretation relatively quickly even when they're working on their own. We've done some informal measures of the way students interact with this app, and it's been really satisfying. When you're a Shakespeare scholar, you can get a sense of mastery from some key cues. For example, how confused or comfortable the students are by the character names. All the Sebastians, and Solarias, and Salanios, the Antonios, and Alonsos are very difficult to keep apart. And students who are uncomfortable with who the characters are won't say their names. And comparing this version of The Tempest to the Folger's print edition in a class that I just taught last fall, the students who were working with this say the names really freely. They do in part because they hear the names being spoken, so they're confident that they have them right, but also because they attach the names to characters when they hear inflection in play in the way that I just showed you. That continues on into their writing as well. There's a higher instance of quotation. You can easily grab passages in this text and select them and email them for yourself using the standard function there and email them out. And students are freer with quotation here when they can simply grab something and punch it into a collection of quotes that they'll then use with an essay. So those are the informal kinds of measures that I regularly see. They also really like, as I said, being able to hop altogether to the same location in the play very easily. So what's next for us? We're about to announce -- we have just announced an agreement with the Folger and Simon & Schuster, and we're going to be heading into a second round of fundraising so we can pivot it into a platform-agnostic development path. Our most recent sprint has been to be able to stamp apps out to do that ourselves. There's an authoring backend that anybody can use who's got the app to write those commentaries, for example, or to put a new text in. And we're doing QA with that right now. We're testing it out for the next set of plays. And a long-term aim of project is to find a way via platforms like this to relocate ourselves and our colleagues outside of the traditional publishing system that offered us narrowly defined roles as receivers or deliverers of fixed scholarly content in established genres. Our goal has been to explore new genres, to revive old genres like the commentary, and to build an apparatus that would be to some degree customizable by users. We've created that web-based authoring environment in the hope that other teachers, scholars, book groups reading together would be self-publishing commentaries for themselves, tagged to the passages of this play and others that they find so compelling. I've spent twenty years of teaching and studying the history of the book. And I recognize a phase of what the book historians and librarians would call "incunables." That's the phase that we're in right now, the swaddling cloths of the e-book. Incunable means that the cradle dressings or the swaddling cloths of the print book. This is a phase of digital incunables that we're in, and it's my strong sense that if teachers and scholars don't lead the development of the resources that we want, we aren't going to get what we need. It wasn't what I expected to do when I trained as a Renaissance scholar to be an e-book entrepreneur, but I've become by necessity, as Jeffrey Schnapp says, a knowledge designer. That's his phrase: By necessity a knowledge designer. I really don't feel I have a choice because I can't wait for the e-books that I want to be produced in the time that it's going to take my students to learn to be sophisticated readers. Now I'm going to need the rest of you, librarians, those who use the library, to help me figure out how to make sure that the new tools of reading and writing that we need in our classrooms and outside them have a sustainable infrastructure in which they can circulate more widely. Thank you. [Applause] I think we have time for questions. I'll try to keep this screen awake here as we go. All right. The question was: Can I compare the experience of creating an app with creating a textbook? And I can't directly; I can compare it more with creating a monograph. And the differences are both authorial -- in other words, I'm sharing very broadly the authoring work of this app -- and formal in the sense that the form that I'm writing for, creating for, is plastic. We're figuring out if you want to read an essay in an app, what's the format that's really going to work on a tablet? We're working right now with an idea of a swipe essay with interspersed illustrations. We're working with the Folger's amazing digital collection of images for Shakespeare. Gorgeous. And the retina display of the iPad is so beautiful for showing off that collection. So there's an opportunity for displaying collections much more widely here and for building them into essays. So we're taking essays that were entirely in print in the Folger Shakespeare editions and beginning to rethink them as swipe essays that might have be richly illustrated. How can we do that intelligently in a way that doesn't just clutter the page, allows the flow of the essay to work? And is this swipe version the right format? So the biggest difference, I think, in terms of my authoring work is that I have to start -- I can't assume a genre. I'm often having to feel my way into a new genre. And the writers who are working with us, the scholars describe that, also, as both a pleasure and the challenge of writing for this format. It's a known -- what you do when you make an addition -- is known labor; this for them is unknown labor. It turns out to be pretty easy. It takes a scholar a couple of days to write a strong commentary stream, not unlike writing a lecture, right? So there are cognate kinds of intellectual activity. But you sort of have to move sideways. You figure out: What is the cognate kind of activity that I'm doing here that this is like, even if what I'm doing appears to be print? Does that answer your question? I guess the other thing I would say is working with software developers is a blast. And I had no idea that that would be the case. One chiefly, it's been a pleasure to work at their pace. It was hard to learn to move up to that speed. If I don't deliver it to them by Monday, it's not going to make it into the next build. So either I deliver or I don't. And that's not the way this publishing cycle works in academia. [Laughter] The other thing that's been really exciting about them is to learn that I'm a good collaborator for them, that my training as a literary scholar -- which is in very precise description of the way readers read, and writers write, and the way texts work on the page -- that very precise verbal description is something software developers care about enormously because the more precise you are in describing the kinds of activities, practical and intellectual that a user is doing, the swifter and more accurate their development is. And I didn't realize that I was training for twenty years to be a good partner for a software developer. In the back. >> How did you select the commentators? >> Katherine Rowe: How did we select the commentators? We brought together scholars of different generations -- teachers, the actors -- there's a wonderful commentary stream by Jennifer Kidd, who plays both Ariel and Miranda, talking about playing both of those parts and about the cross-casting dynamics interestingly in this play. We wanted diversity, disciplinary diversity, diversity of expertise in our commentators. And we wanted people that we knew would not necessarily agree, would come from different angles into the play to show how expansive it can be and open to interpretation and to different kinds of readers. And then we went for people whose work we really love, we find exciting, who we know are interested in writing for broad audiences, not just other academics. And people that we are happy to hang out with because when you're working in a new genre, they have to be very tolerant of you, and you have to spend a lot of time with them. So. Anybody doing a software development project knows that you really want to do it with people that you're happy to have a beer with. That's a great question: What's the relationship of this work to EPUB 3 and the iBooks reader? It's divergent. EPUB 3 has the great advantage of being purposable across form factors and platforms, right? So it will display in lots and lots of different media but the disadvantage of not giving us a richly manipulable text. So for us, we wanted a text we could get our hands into, that is to say, the reader could get her hands into and manipulate very, very freely in the ways that I've shown you here: Grab passages and commonplace with them; rearrange text in various different ways; annotate; share annotations; and so on. And EPUB 3 didn't allow us to do that full array of manipulations that we wanted. And to iBooks Author I think is probably your question. iBooks Author is terrific. I'm sure it will continue to get better. Right now you have two choices: You can either only sell your text if you're going to sell it through the iBooks store, so you can't then repurpose it for Android; or you have to distribute it for free. Those are your two options. So we share some interests with iBooks Author, particularly in making it easy for users to publish. But it didn't turn out to be a useful environment for doing something quite as complex as this. >> [Inaudible] >> Katherine Rowe: iBooks Reader. I'm sorry. [ Inaudible ] >> Katherine Rowe: It's compatible with the app world. This isn't really an e-book. Right? This is an app. And do I have concerns about that? I think my answer to that goes back to the fact that we're in a phase of digital incunables, that we're going to see a lot of different formats experimented with and succeed or migrate sideways in interesting ways. And we've basically committed to participating in that mess -- exciting mess -- and trying to push it in a direction where the resources that the reader gets are the ones that we think readers need. >> [Inaudible] >> Katherine Rowe: Yeah. And but in fact, as I said, our goal is to be on Android next year and so to begin to expand into the multiple tablet platforms. But this is always going to be tablet-based, not web-based. A great advantage of this text is that you can read it and do most of the things that I just showed you without being hooked up to the web. You can sit underneath a tree. >> It's always going to be tied to your app reader. >> Katherine Rowe: It's always going to be tied to our app reader or somebody's app reader, right? Somebody's. Yes. Both of these questions are really getting at the same thing, which is: What's the sustainable path forward, right? Particularly to my very last question, which is: How do we create a sustainable environment for these resources to circulate? And we don't know the answer to that yet. And where are we going to see standardization across e-reading apps versus EPUB readers? I don't know the answer to that yet, although I'm reading fairly intensely every New York Times article that comes down the pike. But I think just to the specific question, this is an app. This is an environment, a framework for manipulating text in the way that readers want to in the classroom and teachers want to in the classroom. This is an emerging area of research. That's another arena in which I'm reading really, really actively. There are some quiet accessibility things here that I didn't show you. For example, there's no page turning. We're a scrolling play so that there's no moment in the middle of a speech where you have to drop even for a second -- increase the cognitive load -- while you wait for the next line. The lines that you're reading are always potentially at the center of the page. There are a host of small features like that. The synced audio and the highlighting bar is something that came from an experience in my family. There are a number of us who are very late readers, had reading assistance of various kinds, and reading challenges of various kinds. And a common strategy that was being used was to bring a bar across the page so that you screened out the other text. And so our highlighter's in there for that reason. So I think that the opportunities to understand how people learn to read, learn to become more sophisticated as readers -- those are two different things -- and what's happening cognitively are really live. And that's to my mind one of the most promising and interesting paths forward. It's not one we're doing in a sustained way yet, though; we're harvesting, as I think I've shown you, some small insights from it. >> [Inaudible] >> Katherine Rowe: We've been thinking a lot about that. So there's a couple of paths forward. Particularly let's say our student's a high school student, which would be the case for the Folger Shakespeare editions very commonly. The content stays attached to the Apple ID right now, or it will be to the Google ID, the Android ID. So the archiving of that content is subject to the practices of commercial platforms. This is a huge, I think, massively significant heritage curation challenge that we face right now across all of our media. Right? So my kids are using Facebook to lead their social lives, to collect their photos and share their photos. That assumes that there will be some future to that, that they will then -- if they want to keep those photos -- be able to come back to ten years from now the way that I can from all my high school photos. Not clear that that's the case. It's not a government agency charged with sustaining that curatorial environment; it's a commercial entity. So I think about that a lot. In this particular case the content is tied to the Apple ID and travels with the Apple ID. So, of course, I hope that the student is the purchaser and that that ID goes with the student into college, for example, and she can then go and look at her notes from high school and then ten years later continue to look at her notes from college. That's what I love about my print Shakespeare editions. Right now there's no provision for that electronically. And I think that we need to have one. So we're grappling with "Do we keep it on the server; do we export it in some format?" With my students here, I have their permission to show this dramaturgical stream. But it's always been assumed that it would be shared and published with the students in the class. This is not a public stream. They have their own part of it electronically on some paper, but they don't have everybody else's. So this gorgeous set of commentaries as a set doesn't exist anywhere except now on this app because most of my students don't own iPads. We have a flight that we share around at Bryn Mawr. That's not going to be retained in other apps besides on our server. So we're thinking a lot about -- and I'm talking with the students a lot about -- how would they want this to be retainable? Do we then publish out and keep it in some other place or print for them? Yes? So the question was: What are the opportunities for learners at lots of different levels to play a part in knowledge designing? And I think you're asking me in that case about the F21 project more than about the app, yes? >> Yeah, in general. >> Katherine Rowe: Yes. Yes. Well, I should say that I've been pushing hard and I hope succeeding in bringing more what we would call "amateur curator populations" into our pilot that we're seeking grant funding for now for F21. I would like to bring retired high school teachers, for example, into our pilot -- five of them. You know, five undergraduates, five retired high school teachers, five somebody else to see whether there is a population that's really interested in going to a boot camp and doing tedious transcription of Renaissance plays and whether the kinds of learning goals that the undergraduates were describing are more realizable through different kinds of populations of amateur curators. "Amateur" is a word that academics are really uncomfortable with, don't like viscerally. But I think that that is a word we need to be rethinking because it simply means you're not being paid for it. All the Olympians or most of the Olympians are amateurs, right? At the Olympic Games. And they're very, very good at what they do. So that distinction between amateur expert and amateur professional -- it's an unstable word, and it's paired in both directions, is one that I think we really need to be grappling with and rethinking. And I hope that if F21 succeeds, there will be a curatorial and workflow process that will allow someone who's really excited about this work to join it, to apply to join it, right? And to do it remotely. That's a downstream goal. We're not there yet with the pilot. Does that answer the question? >> Yeah, absolutely. >> Katherine Rowe: Yeah. Yes? >> I have two questions totally unrelated. >> Katherine Rowe: All right. >> The first is: Did you consider embedding video? That seems to be the one piece that's missing is to see a performance of this play or of the scenes. The other question is about academic support or recognition of this kind of work. You said your institution has been really supportive, but I think of a young academic going on the tenure track and wondering: Do they recognize this new form of work, or are they still looking for published paper monographs? >> Katherine Rowe: It's a really important question. And as I said, I have been very supported to do this. In fact, the first wonderful conversation with my provost about this was: So when you come up for review next what does this look like? And we agreed that I was doing an edition. It happened to also be the case that I was building the book for the edition and that those were two kinds of contributions. And we -- like many institutions -- are really actively grappling with the new forms of scholarly labor, labor and communication, labor and publication. I think that that's an ongoing process. Some organizations -- Four Cs, the MLA -- have been very early out in front with standards for evaluating work like that; others are following suit pretty steadily. And I think we're going to be seeing an organic shift into new methodologies that is so pervasive that it's not ignorable. In other words, I don't see myself really as a digital humanist. I'm a humanist. It happens that I need the word "digital" right now to name the transitional methods and technologies that I'm interested in because this is a phase of incunables. Our methods are in their swaddling cloths in certain ways as well as our tools. But I think that adjective is going to go away within about ten years, maybe fifteen. So that's the second question. And the first question? Say again. >> Video. >> Katherine Rowe: Video. Oh, yes. We do have video, and we have video capacity. Let me just pull it up here. What we have is this wonderful podcast from the Folger on bookmaking that shows you really quickly what the video interface would be like. Very straightforward, standard iOS video interface. This sits next to Caliban's wonderful speech injunction first to possess his books. Get Prospero's books and you'll control the magician. And we wanted to explain what it might mean to possess books in the Renaissance. In fact, we picked this play, The Tempest, in part because it's a play about magic books. So that's where we have the video embedded. It's actually an incredibly engaging and wonderful little three-minute videocast. We do not have full video for the very simple reason that we began with readers as our primary audience. And you read and listen through different perceptual paths that really reinforce each other, but you watch video and listen through the same path that you use when you read. And my experience of the gorgeous and exciting e-books that were out with video was that this form factor made those amazing actors something like talking heads and that I wasn't reading at the same time. That's the book as media delivery device rather than the book as a shared reading environment. We may turn out to be wrong about that. And I'd be very interested in partnering with a filmmaker, for example, who's got some ideas about how this form factor -- that little screen -- works. I know that my students are watching videos on their smartphones and reading books on their smartphones. So the practices are changing very, very rapidly. But that was our initial thinking. So the question was: Can I elaborate on the multilingual elements of the app; can users manipulate as freely in other languages as they can in English? Right now, no. Our goal is yes. In other words, I have envisioned -- we have envisioned -- these texts as having any content here and any content here. The commentary stream is the secondary material that's curating the content that sits here, the primary material and that these could be freely text, video, audio. My notion is that we could manipulate the language fairly freely in any of these cases. Probably the most important is that the reader annotator be able to take notes in his or her own character set and language. Yeah. We haven't assumed that we would translate the primary text as part of the project here. But we'd be interested in talking with publishers that have done that translation work. And so we're beginning to explore, for example, who in China has done the most publication of Chinese Shakespeare translations. But my idea was that the apparatus would be repurposable in any language -- so scene summaries, commentaries, annotation tools, and so on, even if what you're working with in the primary content here in the case of Shakespeare is English. Yeah. And that will probably be -- we'll hit those development milestones if all goes well sometime next year. So the question is: How long will I be at the Folger? Just this summer for a few weeks. But I work in Philadelphia, and I'm down here pretty regularly. The second part of the question was: Do we see this as a tool that could be used in massively open online courses? And specifically because this commentary stream is one that you can publish out to, to groups, we'd be very interested in working with the APIs of some of the large purveyors to tie their fora and their postings to have the opportunity for users to capture and pull in comments that they think are really illuminating and attach them to specific parts of the primary text. One of the things that's really exciting here is the synchronization of commentary and text, that those are together. You can see that there's five interesting commentaries on this one passage. So we're beginning to have those conversations. And I know the Folger's beginning to think downstream about what its MOOC mission will be. And so those are conversations in prospect, really exciting ones. Yes. This is some scholar working in advice literature and thinking about how exciting it would be to be able to tag by category the kinds of advice fathers to son, fathers to daughters, daughters to fathers in the 19th century. And how do you get started? How big a project is it? So I think this goes back to the earlier question about: At what stage in your career do you want to take on a project like this, and how do you stage that? It can happen at any stage. Pre-tenure, the advice that I'm giving scholars is to choose projects that really integrate well with traditional research modes of publication right now so that there's positive reinforcement for both kinds of work, that what you're doing in your traditional modes helps you build the research agenda that you can bring into the digital methodology. The data, the things you learn from the digital methodology can help advance your traditional mode and vice versa rather than separate, stand-alone projects. It's way more efficient that way. Places to get started would include DHSI -- the Digital Humanities Summer Institute -- which would be a wonderful way to spend a week getting to understand a project management process, thinking about metadata design because you'll make a metadata system; you don't just receive one. You say, "Here are the things that matter to count and here's why." And you ideally want that system to be peer-reviewed by other scholars who would agree "This is what matters to count and why" so that anything that you build onto that metadata system is going to be valuable to others. You've validated that from the beginning before you start building. I would look for -- at conferences in your field -- for the digital humanities panels, and go there and find out who's doing work like this. You might be interested in the Early Novels Database, for example, out of Penn and Swarthmore, which is as I describe it, it's a historical metadata project. They're taking all of the apparatus of early printed books that's usually left out of the scanning process -- not just publication information, but also, you know, the lists at the back of the commonplaces that are exemplified in the book, pages and pages of wise ideas that this book will teach you. Things like that. And they're cataloging that metadata and making it attachable, therefore, to the text of the novel itself. And that's the kind of material that you would find incredibly useful. And you might be able to partner with an initiative like that or contribute to it and then learn about the process so that you can do it more efficiently. So I think my answer is: Start with the learning before you build; find a community to conceive of and build projects with; collaborate wherever possible; the more value that we get out of the projects that we're building -- each individual project -- the better. It goes back to the question from the very back about the sustainability of individual platforms. Yes. Other questions? All right. Well, thank you so much for this very thoughtful conversation. [Applause] >> This has been a presentation of the Library of Congress.