 Large long-term humanities projects leave in their wake a body of research that may become inaccessible or lost when publication is complete. Following publication of the final volume of the letters of Samuel Beckett came the big question. How can 30 years of essential primary research be preserved for the future of scholarship? Metadata provided a methodology while respecting literary rights it could describe and index the letters, integrate decades of interviews and research, and model humanities methods for new and public inquiry. The letters of Samuel Beckett were published by Cambridge University Press as a selected edition in four volumes from 2009 to 2016. We knew from the outset that we would not be able to publish every one of the letters because there are more than 16,000 of them that we had consulted and transcribed to prepare the selected edition. Indeed the published letters amount only about 25% of the letters consulted. Yet the letters form a unique corpus of writing itself a major literary effort. They describe the narrative arc of Samuel Beckett's work including writing and translating, as well as directing productions for the theater, radio, film and television. The letters reveal the author's self-perceptions. The young writer faces discouragement and insecurity. The mature writer expresses the fear that he has written himself out. The letters also reflect cultural and social changes of 60 years, 1929 to 1989. In Ireland, France, England and Germany where he lived and worked. As his work became widely known it touched even broader international audiences. Ours has been a long research journey. It began in 1985 when Samuel Beckett asked Martha Faisenfeld to edit his letters and approved my involvement as assistant editor. When we first met him in Paris at a small table over even smaller cups of coffee, he said, you will get round and see these people, won't you? Yes indeed we would and we did, for this was not a rhetorical question. Although we have done intense archival research to prepare the addition of the letters of Samuel Beckett, we took Samuel Beckett's advice and found even richer avenues of memory in the people we met through the project. While it was Beckett's wish we not published the letters until after his death, this was hardly a restriction because many years of work lay ahead of us. At the beginning we had only a few glimpses of the eventual magnitude of the project. Even though Ours was a selected addition of letters chosen from over 16,000 letters consulted, we did have to transcribe and often annotate each letter that we consulted. Sometimes Beckett's handwriting is merely a scrawl. So transcriptions are required, if only to be sure that we are all looking at the same common text. As we edited the letters of Samuel Beckett, we did not chase down an answer for every question that the letters posed, but we did gather a good deal of information that brought three dimensions to the relationships events by Beckett's letters. We also filled a great many file cabinets. So now nearly 40 years from 1985 and four volumes later, our offices in Atlanta are filled with research that undergirds both what has been published and that supports the transcriptions of all the letters consulted. This brings us to the issue, what to do with it all. The Beckett Project has the most complete collection of his letters anywhere in photocopy and transcription. However, we only own our editorial work. Letters are unusual in that they have two owners, the first, the copyright holder, either the Beckett estate or the writer of a document, and two, the recipient or owner of the letter itself. So to publish letters, permission is required from both owners. The residual copies of documents and mountains of research materials gathered and kept in the Beckett offices actually belong to others and cannot be used without permission. This constitutes what is called a gray archive. So the key question we asked ourselves was this, within the ethical and legal principles that adhere, what can be done to preserve such materials and their context so that they can be useful to future generations of readers and scholars? Well, we quickly dismissed a funeral pyre in the parking lot. We also reluctantly but realistically resisted embarking on another 35-year journey to publish, quote, it all, quote. Even had the estate been willing and the audience ready, there are human limits. While we understood that most archives collect originals and not copies, we also realized that boxes of uncatalogued materials, accessed out of context, would not be very helpful to future research. The digital world offered some good options. Our first concern was to preserve the record of Beckett letters consulted and transcribed. We decided to create a location register of Samuel Beckett's letters. This was the first step toward preservation of the cumulative knowledge of the project. The editorial project of the Beckett letters maintained document control with an internal database, which we have repurposed and regularized to facilitate searching. The result is an open access database that describes each letter in public archive bibliographically with data that is reusable. The Gladys Delmas Foundation has supported this effort. We have worked very closely with archives around the world to confirm descriptive information for the letters. This collaboration with archives on location register has already improved the documentation in archival collections and has even led to the discovery of new letters. The location register describes letters individually at item level, not just as part of a collection. Identification of materials by collection, of course, offers the advantage of an immediate context, but it provides only one context for a relationship or a partnership between two people, whereas in real life there are great many. Moreover, letters to a single recipient may be scattered among archives and even parts of letters have shown up in different repositories. For example, enclosures are separated from a letter with which they were originally sent. One of the goals, one of the first goals of the location register of Samuel Beckett's letters is to get all the pieces back together. Ownership of Samuel Beckett's letters still is shifting from private to public archives. Letters continue to be discovered and consulted. Over 500 new letters were found within six months after the editors closed the book on volume 4. The Beckett letters project is well poised to continue the collection process and merge new materials with letters already consulted and transcribed. The location register will also guide future scholarship by identifying widely dispersed and largely unpublished materials, thus establishing an essential foundation for new scholarship in the literature and culture of the 20th century. In short, it will preserve the legacy of the letters and open access to them for all. The interactive index of Samuel Beckett's letters in public archive provide the text of the letters, but it makes the metadata of his letters more accessible to a wider audience. It builds on the purpose and principles of the location register and adds context. It provides a searchable digital index of each letter. The data is curated and enhanced with chronologies, profiles, major recipients, filmed interviews, and advanced search capacity. The goal is to invite new inquiry. It will not replace the work in archives, but rather guide it and help students and scholars focus their research. The database provides access to rich, linked, and nuanced data from the contents of each Beckett letter. All letters have been tagged to create a searchable database sortable in the following ways. Beckett's reading, writing, translating, and directing. As mentioned, publications, productions, places, and organizations. Beckett's attendance at sporting events, exhibitions, plays, concerts, whether in person, on radio, or on TV. References in the letters to artworks, music, and world events. This metadata will allow users to discover patterns in interrelationships and thus will give future students and scholars a myriad of starting points for their own research in archives. Even when Samuel Beckett is not the director or the translator of record, we know when a letter shows him writing or directing or translating and link this to the production or the work and his publication. The participle is important here because letters record the process. While there are major Beckett archival resources of manuscript drafts and directing notes, it's the letters that record the struggle and experiment that happens while writing. To his directors, Beckett expresses the need to see a play on stage in order to know if it works. In letters, Beckett shares his doubts and disappointments as well as his discoveries as a writer. Who will use this information? Students, scholars, directors, actors, readers, and persons with an interest in the fields of art, music, film, theater, publishing, and even sports in popular entertainment. In short, the letters of Samuel Beckett provide a particular survey of 60 years, 1929 to 1989, with broad appeal, a period of significant cultural, social, and political change. The Beckett archives are exceptional and deep. They offer traces of the living record of performances through photos, reviews, recordings, interviews, and anecdotes by the actors and directors. The ephemeral nature of theater is that each performance is different every night. Even productions that Beckett directed varied markedly. Yet it is fortunate that Beckett did direct and or participate in several productions of the same work, not for the record of how a perfect performance should look, but for the experience that each performance finds its own measure and its own audience. Nuance is central and brings us back to the archives as a continuum, one that is constantly reflexive. Although the interactive index of the letters of Samuel Beckett is built empirically, it is intended to stimulate and not replace reading of the letters in context. It is a curated database, not a data dump. It is not intended to merely mine the data of the letters, but to open fields of inquiry organically with the letters as a starting point. Chronologies of Beckett's writing and public events, profiles of individuals and personal interviews, film snippets and photographs. All of these help to contextualize the letters. The letters are linked to archives of corollary collections and to two other international projects, the Beckett digital manuscript project based in Antwerp and the staging Beckett project based at Reading, England, as well as two wider world events. We will explore the platform together after Sarah Palmer, the project coordinator at Emory University Center for Digital Scholarship, discusses how we developed the database. Secluded in a wooded area north of Emory's main campus lies an administrative building named the Loose Center, which houses the extraordinary archive of the Beckett letters. The sheer volume of paper presented challenges for a data making endeavor. Thousands of transcribed letters were stored on volume level Microsoft Word files. Assembling data at the letter level required cutting and pasting content out into individual records and integrating with metadata from an access database. This process illuminates the difference between workflows or composing books and gathering data. Volume level collections of information make sense for producing a printed collection, but data is atomic in nature and requires different processes to realize its full potential. The Emory Center for Digital Scholarship established a workflow for building this dataset. Like many projects seeking to get off the ground quickly, we selected tools and platforms that were low in cost and easy to use. This led us to initially tag XML versions of the letters using Oxygen XML Editor, a powerful tool that allows tagging to be conducted without writing code. We also stored data in Google Sheets, which were easy to edit and share between users. Despite these affordances, we struggled to query our data and ensure its consistency. We soon realized we needed to migrate the information to a centralized database. Thanks to some innovative software engineering, we developed a more comprehensive system for managing the data. This diagram illustrates the components of the system, which are colored by the team that maintains them. In blue are the data sources from the Loose Center, including the transcriptions, an Excel file of letter metadata affectionately titled Big Sam, and multiple Google Sheets with data related to the tagged entities. And red are the technologies maintained by the Emory Center for Digital Scholarship, including a tagging platform coded in Ember that sits atop a Ruby on Rails application, a PostgreSQL relational database, and an application programming interface that publishes the data for machine use. Finally, in yellow is the evolving public website that contextualizes the data for end users. This system has proved in a measurably more convenient and reliable way to maintain and present data. Of course, every system has its weaknesses, and in this case, the Google Sheets were that pain point. It was difficult to enter and harvest italics into the database, and duplicates were all too easy to create. In the past year, we have updated the system to replace the Google Sheets with a custom design tagger with an interface for data entry. We used a Ruby gem called Administrate, so that content creators could add and edit data in an interface directly connected to the database. The process of working together on this system has revealed some aspects of data-driven workflows. It became abundantly clear that humans and computers are good at different things. Our early challenges with managing and identifying numbers across platforms emphasize the fact that humans struggle with the mundane while computers excel at repetition. Conversely, development of a website that presents the data with proper context underscores the reality that nuance is not something computers present as readily as the human mind. Data making requires translating between humanistic inquiry and technological outcomes. In sum, we have learned that the maintenance of data that speaks to multiple complex relationships requires careful curatorial and technical decisions. We have amassed a considerable amount of information about thousands of letters and this volume requires thinking about how to present it as fertile soil for exploration and not simply a data dump. Everything from how entities are presented on the site to the development of search filters to adding contextualizing videos, narratives, and graphics serves to further a promise of rich metadata and that is to create new avenues for human knowledge. The purpose of this open access data and open source software is to democratize the materials of both archive and scholarship. Forming links between sources of knowledge is not only an intricate editorial and technical endeavor but can also be a form of public service. As such, this project aspires to be a model for other long-term research projects to adopt. Now we will explore the in-progress platform to discuss the kinds of research it supports. Among the features accessible to users on the opening screen are a tally of letters related to the search term that have been published in the letters of Samuel Beckett and a tally of letters written in English, French, German, and Italian. Also available from the opening screen is a list of repositories that provide public access to originals or copies of originals. This will make it easier for users to plan research visits to or consultation with archives. The archives are listed by letter counts with the largest collections first. There is also a date range option so that users can selectively narrow their searches. When searching entities, it is possible to select a keyword, all, locate the repository, and find the total number of letters available and their language. Here we are searching for Robert Ponget at the Burns Library of Boston College. All of Beckett's letters to Ponget are written in French. Of these, 19 have been published in the letters of Samuel Beckett. When we select one letter, 28 September 1955, we can see a description of the letter as well as a listing of the mentions in the letter. If we select Waiting for Godot, we find a description of the first production in English, Peter Hall's 1955 production. This entry gives the name and city of the theater, a cast list, notes about the production, and links to other resources. By looking through all the letters that refer to this production, it is clear that references are not just from contemporary letters, but also those written much later. This alone extends the scope of inquiry. If the keyword searches Alan Schneider, who is Samuel Beckett's preferred American director, the user will find there are four archives that hold relevant materials. For all letters to Alan Schneider, the user can select all repositories. But there is also interest in the details. Selecting one letter from the Alan Schneider collection of the Burns Library of Boston College offers an example. Note the information given about Beckett's letter of 11 February 1966. The user can see that the letter was addressed from Paris, that it has no envelope, that it is an autographed handwritten letter signed, that it is one leaf with two sides, and that it was published in volume four of the letters of Samuel Beckett. Of interest in this letter is the mention of Ed Joe in Alan Schneider's production. And he, Joe, the German production directed by Samuel Beckett at Suddeutscher Rundfunk in Stuttgart. This was the first production where Beckett was the director of record, though he had poked his nose into many previous rehearsals of his work. Beckett's own feelings about this production are apparent by the number of letters written about the experience, some written much later. More detailed results can be found through entity searches. There is a drop-down menu for the entity types, which allows the user to narrow a research query to select the most relevant results for their purposes. One search might be to select a label or a keyword. By selecting any, a very wide search will be returned for any description that contains the word search. By selecting all, only searches that use the full phrase or name will be returned. The user can search for the label pros, and then select the entity writing to locate a list of Beckett's pros writing. When the website launches, this list will be sequenced by date. By using the label, Proust, and then selecting the entity of translating, a list will be supplied of the translations of Samuel Beckett's monograph, Proust, that are mentioned in his letters. Included is a proposal for a French translation, a proposition that Beckett rejected several times, not wishing his youthful remarks on this notable French writer to be available to the critical assessment of French readers. By selecting keyword Malloy, the first novel in the trilogy, and opening writing, and selecting the first result, the user will discover that this novel was begun in 1947, first published in 1951, and was mentioned in letters from 1948 to 1986. By selecting the English translation undertaken by the author Patrick Bowles, the user will find the archival and publication details for the translation. This inquiry will also reveal the letters in which this translating is mentioned. Learning then Beckett's own words, how much translation was a process of re-envisioning the text, and so listed here under writing. By selecting translating, there are five results. By selecting productions, the first listing is a rejected proposal for production. Selecting five yields a large number of letters that touch on Patrick McGee's readings from Malloy for the BBC. Beckett was so taken with McGee's voice that he would write craps last tape for him. From these, note the letter of 24 March 1958 to Patrick McGee. It provides publication information so that the user can consult the letter in print, as well as in the archives of Trinity College Dublin. By selecting person, it may be valuable to learn the names of eight persons with connections to Beckett's novel Malloy. Note it is possible to toggle from one entity to another and explore a single reference from several facets. Selecting the keyword Royal Court Theatre, and then the entity of attendance, we note that Beckett saw the production of John Osborn's play Luthor, performed by the Royal Court Theatre in Paris on 6 July 1961. By selecting person, one locates a list of individuals linked to the Royal Court, a useful guide to the several decades of key personnel named in Beckett's letters. Selecting Albert Finney, who acted in the 1973 production of Craps Last Tape. Here we note the link to the via for Albert Finney. Further research can be done by exploring the letters mentioning this production of Craps Last Tape using the play's title as the keyword for an entity search. Beckett wrote in both English and French and translated his own works, occasionally with others, but most often, though reluctantly, by himself. He worked closely with the German translators Elmar and Erika Tophoven. His empathy with translators of his work into languages he did not know was expansive. To them, Jakob van Velde, Dutch, Christian Ludwigsen, Danish, Antony Libera, Polish, he generously and patiently responded to inquiries, knowing too well the impossible demands of translation. Here we can see Dutch translations of Samuel Beckett's works. Finally, a brief survey to explore the value of other entity searches. By entering the keyword strike and selecting public events, one finds an indication of major strikes that affected Beckett's letter writing, a postal strike for five months in 1979, and also daily living, an Air France strike in 1966, a bank for weeks in 1978. If one opens the entity person, there is mention of a hunger strike by Clara Colissimo, who had been denied the dramatic rights to perform happy days in Italy due to a prior contract with another actress. As with other examples, opening the entry for Clara Colissimo includes letters that mention this issue. Selecting his keyword, Algeria, and opening the entries for the entity public events, yields letters in which Beckett mentions the Algerian conflict that dominated the news and life, particularly from 1960 to 1962. We should not neglect Beckett's deep connection to music and heart. He enjoyed playing the piano, attended concerts, and had personal connections with musicians, and took an active interest in the career of his nephew, the flautist, Edward Beckett. With his wife, Suzanne, herself trained as a concert pianist, he listened to recordings and read the music composed in connection with his writing, ranging from contemporary composers Humphrey Searle, Charles Dodd, and Roger Reynolds, to the jazz artist Henry Crowder. His letters mention musical performers in the Dublin of his youth, Harry Lauders' music hall songs, and John Larch's performances of interval music at the Abbey Theatre, as well as highly regarded contemporary composers Morton Feldman and Earl Kemp. Among classical composers, his favorite without a doubt was Franz Schubert, whose composition, Nockt und Traum, influenced Beckett's radio play of the same name. His attendance included concerts and recordings. Thomas McGreevy was a significant influence on Samuel Beckett's interest in art, although their tastes diverged. He introduced Beckett to Jack B. Yates, whose painting and friendship was important to Beckett. The Italian painter Giorgione was a talisman for McGreevy, and Beckett does not fail to include his works in letters from Italy and Germany to his friend. Contemporary artists became friends. Gier and Bromvom Velde, the Dutch painters, were among them. The index to Beckett's letters provides insight and entry points for research and exploration of Beckett in art. Selecting the keyword Bromvom Velde, selecting the entity of works of art, one can note that items 3, 7, 8, and 9 were owned by Samuel Beckett. Looking further, selecting Sans Tietra and opening it notes the current ownership by the center Pompidou, a gift from Samuel Beckett in 1982. Selecting the link for C also on that page gives further detail about current ownership. The same is true of the last entry, which bears the same title. Here the images and descriptions, as well as the current owner and location, are key to knowing more about this art that Samuel Beckett discussed and or once had on his walls. The index of the letters of Samuel Beckett in public archives provides the data about and from his letters. Its data sets can be accessed through many avenues of inquiry. Searches can be saved by the user. Details of archival location and related materials provide places to begin fruitful research. Both the data and the platforms of the interactive index of the letters of Samuel Beckett are available for public access. Editions can readily be made as new material enters public archives, making this a dynamic resource for future scholarship and wide general interest. By making both the platforms and the data open access, the discovery of the intersections between the humanities and digital methodologies offers a model for other projects in the humanities, opening research to new questions and new evidence. Scholars and interested readers will find ample evidence of the cross currents of the 20th century in these letters and in corollary collections. As a digital postscript to the letters of Samuel Beckett, our intent is to make the methods and the accumulated research of scholars accessible to those who may want to pursue a specific interest, to spark curiosity, and to enhance the value of the humanities as it intersects with general culture. One need not be a specialist to be curious, to raise questions, to pursue reading with wide purpose and with depth. Indeed, those who create the programs of our humanities councils recognize the hunger for such discovery and look for ways to encourage and enhance this interest. We hope that our projects may provide suggestions for extending this reach to an even wider general public.