Designers of electronic spaces have chosen comfortably familiar terminologies to make a complex medium comprehensible. It is a sign of the degree to which the previously foreign medium has become a commonplace part of our interaction with it that the terms are rapidly becoming so transparent as to acquire the status of dead metaphors. Few of us will now consciously think of a book as we download a Web “page,” or of a supermarket when we obediently add an object (which is actually an item of electronic data) to our “basket” or “shopping cart.” This paper will suggest alternative metaphors as a means of delineating more clearly the aims of the Internet Shakespeare Editions: the electronic space as a stage for the performance of text, and the process of collaboration that produces the text as something akin to the processes involved in the production of a play. In A Midsummer Night’s Dream, as Peter Quince organizes the “hard-handed men of Athens” for the play that will be presented before the Duke, he chooses a “marvelous convenient place” (3.1.2-3) in the forest outside Athens for their rehearsal. The convenience Peter Quince admires is the result of the plasticity of the forest space: the “green plot” becomes a stage, a “hawthorn brake” their tiring house (the space offstage); the electronic medium at present is in a similar state of plasticity as techniques and conventions of display are malleable and have not yet hardened into tradition. Humanists have, I believe, both an opportunity and an obligation to help in the exploration of the new medium as a plastic space, suitable for the publication of data of special interest to those in our discipline.
Unlike the members of Peter Quince’s cast, who understand that they must work together to create the play they are to perform, Humanities scholars are not accustomed to thinking in terms of collaboration. Very few publications are co-authored, and the perception of most is that they work on a project in relative isolation, interacting only with libraries and with primary sources. Yet this process is nonetheless a collaboration with other authors, even if those authors are represented not by warm bodies, but by footnotes in their text and books on their shelves. The collaboration is often lively, if one-sided, and tends to be marked by (moderately polite) antagonism, as fellow scholars, alive or dead, are more likely to be cited if their work is the object of disagreement or modification. Once the article or book is completed, the scholar’s level of collaborative activity is almost over, though there may be some interaction with peer reviewers. The actual publication of the work is the responsibility of the publisher, who will see to formatting, copy-editing, and designing the work; while there is thus some level of interaction in the process of publication, it is fair to say that this has decreased in recent years as publishers have been forced to tighten their budgets.
At this stage in the development of the new medium, publishing on the Web can be a very different activity from the processes of traditional print articles and monographs. While the process of submitting an article to a journal like Early Modern Literary Studies (EMLS) is much the same as for a print journal, there exists a more fundamental level of Web publication that is very much in the development stage; here there is a unique opportunity for scholars to be involved in the basic design and display of their works. Much of the energy in designing Web sites has thus far been controlled by commercial interests, from the shady to the respectable, so that the shopping cart and the Google advertisement have become principal design features of major sites. The alternative commercial model is the site that is available through subscription only: in our field, for example, such sites as ProQuest/Chadwyck-Healey’s Literature On Line (LION), or Thompson-Gale’s Shakespeare Collection. In neither case, however, is the collection conceived primarily as a forum for the publication of new scholarly work; LION provides a searchable archive of texts, few of which have been peer-reviewed, and the Shakespeare Collection consists largely of work that has already appeared in print.
The new medium deserves better. Back in 1996, when I first started work on the Internet Shakespeare Editions (ISE), I argued that a book was not a screen, and that the electronic space presented both opportunities unavailable in print and limitations compared to the technology of the book (“A Book”). The limitations are beginning to disappear -- as computers become more compact and as the technology of electronic paper begins to mature -- but the opportunities have only begun to be exploited. Such is the complexity of the electronic space as a medium for publication, however, that those involved must interact with programmers and graphic designers as well as the already considerable network of contributors, copyeditors, and peer reviewers. If Humanities scholars are to have a voice in the construction and development of publication on the Web, they will have to develop techniques of collaboration and an appreciation for scholarly work created less in isolation than is traditional in the field.
The theoretical foundation for collaboration as fundamental to textual practices in the Humanities is well established, most notably in the work of Jerome J. McGann and D.F. McKenzie. McGann and McKenzie have argued persuasively for what they call the “social” text—one which is the result of various kinds of deliberate and accidental collaborations between the author(s) and those who produce the published work. In the case of Shakespeare’s texts, for example, Shakespeare’s original work (whatever it was) must be seen as necessarily at several removes from the final printed text as it was filtered through, and co-created by, theatre practices, scribal copy, printer’s markup, compositorial interpretation and, inevitably, compositorial error. In some ways the collaboration extended back in time to those earlier printers who had developed a commonly accepted print format—or interface—for dramatic materials: how such elements as speech prefixes or stage directions were to be represented on the page, for example. As modern scholars work on new editions, their work subsumes a long tradition of textual scholarship: the editor consults earlier editions, adopting and rejecting emendations of various kinds, and creating a record of what is in effect a further collaboration in the dense collation that forms part of the edition.
Given that McGann’s work has been hugely influential in the field, one would think that modern textual scholars would both welcome and be familiar with the processes of collaboration, but it is still in large measure the case that those preparing texts for the ISE adopt the familiar pattern of most Humanities scholars, working on their own, meeting occasionally at conferences, but otherwise self-directed. In a project for which the conventions of publication have been well established, there is no particular reason for scholars to do otherwise than to collaborate with the books on their desks, but in the still new electronic medium, where there are as yet no strictly defined principles for the presentation of the text, collaboration between scholar and interface designer is vital if the potential of the medium is to be realized; the time is ripe for the creation of a more comprehensively social text than ever before, one that intimately includes the team of programmers and graphic designers who create the vehicle for the text. Collaboration on the scale required by electronic publication is not common in the Humanities, and involves something of a learning curve from those of us steeped in the tradition of solo scholarship. The ISE is working on several intersecting projects to integrate the adaptability and convenience of the electronic medium: hypertext editions of the plays that involve multiple levels of complexity in annotation and navigation; databases of images of the original texts, and of the plays in performance; and software that makes immediately available sophisticated ways of re-viewing the text through automatically-generated linking, concordancing, and analysis. In different ways, each project both records performances of the plays and creates new performances as text and image move across the computer screen.
The collaborative network that leads to the edited texts follows familiar scholarly lines. Eric Rasmussen is the General Textual Editor for the series, and individual plays are edited along largely traditional lines by scholars approved by an Editorial Board. Several plays are being edited collaboratively between two scholars, one of whom focuses on the text, the other on the commentary. Expertise in the electronic medium varies from those who, having mastered an early version of word-processing, are reluctant to upgrade their skills any more than is absolutely necessary, to those with a sophisticated understanding of electronic texts and the Web. Interestingly, an attempt to persuade the growing team of editors to share ideas and experiences through a Yahoo! discussion group has produced little response.
In the creation of the database of Shakespeare in Performance, a rather different network is in the process of being created. The database already contains comprehensive information about Shakespeare films, thanks to the generosity of Kenneth Rothwell, who donated as a starting point his important work, Shakespeare on Screen: An International Filmography and Videography (1990). In addition to updating this work, the ISE is in the process of accessing and digitizing artifacts from theatre and festival productions of Shakespeare in Canada and the United States; it is our intention to expand geographically as the database matures. The collaboration here requires communication with the extensive theatre community in North America, in large measure through the good offices of the Shakespeare Theatre Association of America (STAA). The actual digitizing is done by research assistants at the University of Victoria, under the supervision of the ISE. Overseeing the academic quality of the productions chosen for inclusion is a Performance Editor, Paul Prescott of the University of Warwick.
None of this would be possible without collaboration between the creators of content and teams of programmers and graphic designers to realize the vision of both the edited texts and the database of Shakespeare in Performance. The challenges are considerable. The culture of scholarship in the Humanities resists close consultation, and textual editors are not normally acquainted with more than a very basic understanding of principles of markup. Programmers are similarly unaware of the complexities that underlie an edition of an Early Modern work, and they will often find it difficult to communicate effectively with non-experts. The relationship between scholars and programmers can be a frustrating one. Few scholars are able to keep up with the pace of technological change, and few programmers understand fully how their work can most fully benefit scholarship. At the meeting of the Canadian Symposium on Textual Analysis (CaSTA), McMaster University, Jerome McGann delivered one of the keynote lectures (“Culture and Technology”); in the question period following, he recommended that in order to achieve their goals in developing online resources, Humanities scholars should find the funds to hire professional programmers for specific tasks. Any other route, he argued, led to frustration, compromise, and continuing delay. The problem, for most of us, is that programmers come at a premium, and research grants do not. Appropriately enough, granting agencies in the Humanities tend to favor proposals that involve using students rather than professionals.
Though the ISE has been generously supported by the Social Sciences and Humanities Research Council of Canada (SHRCC), our funding has certainly not stretched to provide the kind of professional work recommended by McGann; SHRCC rightly insists that the bulk of its support goes to providing students with experience and an income. For this reason, we have employed students, both undergraduate and graduate, in the areas of both Humanities and Computer Science. The challenge has been to find a model for working with the facts of life where students are concerned: their commitment will be part-time, and they will move on. Our experience in the last two years has been both delightful and instructive as we have worked with an outstanding group of young programmers and student researchers. I add the term “student researchers” because we have integrated one or more students from Humanities disciplines (principally English) into the development team, providing the programmers with local and continuous feedback on the way that data is used and understood in the Humanities disciplines the website seeks to service, and in the process equipping our Humanities students with extensive knowledge of the principles of Humanities Computing.
Our student programmers today bring with them an extensive understanding—and often an enthusiastic acceptance—of techniques of collaboration pioneered in the Open Source movement. In his seminal essay, “The Cathedral and the Bazaar,” E.S. Raymond seeks, as I do in this chapter, to find metaphors to guide his thinking, in this case the differences between the creation of software by traditional means and the processes that have evolved in the Open Source movement. Raymond chooses metaphors from the worlds of architecture and commerce. Of the traditional model of software development, he writes, “I believed that the most important software. . . needed to be built like cathedrals, carefully crafted by individual wizards or small bands of mages working in splendid isolation, with no beta to be released before its time” (para 3); he contrasts this with the seemingly more chaotic activity in Open Source, characterized by the Linux operating system: “. . . the Linux community seemed to resemble a great babbling bazaar of differing agendas and approaches. . . out of which a coherent and stable system could seemingly emerge only by a succession of miracles” (para 3). To a purist, the metaphors here may seem rather mixed—an architectural structure of great beauty and complexity is compared with a commercial interchange that has no apparent structural outcome—but the distinction he is making is essentially between a “top-down” managerial structure, and one that is much less hierarchical. Cathedrals were seldom created in quite the way Raymond suggests, since most took generations to build, and many incorporate a wide variety of building styles; in many ways his metaphor would be more accurate if he chose as his model the meticulous architectural planning that is needed for the construction of a skyscraper rather than a cathedral. Raymond also extends his metaphors to include the biological along with the architectural and commercial:
The Linux world behaves in many respects like a free market or an ecology, a collection of selfish agents attempting to maximize utility which in the process produces a self-correcting spontaneous order more elaborate and efficient than any amount of central planning could have achieved (ar01s11.html).
There are some interesting assumptions here as biological systems are cheerfully equated with economic structures, but the distinction is insightful, and has been influential in the business of software design. Later in the same essay, Raymond returns to the metaphor of architecture, as he writes of an example that offers hope for those working with a moving student population:
The development of the GNU Emacs editor [an advanced text editor used by programmers] is an extreme and instructive example; it has absorbed the efforts of hundreds of contributors over 15 years into a unified architectural vision, despite high turnover and the fact that only one person (its author) has been continuously active during all that time (Raymond, ar01s12.html).
I shall return to the image of architecture later. The important thing to note here is the usefulness of being aware of the kind of approach that has been so effective in Open Source programming: a more decentralized structure that nonetheless provides for efficient interchange of code and ideas among programmers; indeed, Raymond argues that the software resulting from Open Source techniques is superior, and more bug-free, than that produced by centrally managed systems.
There is no doubt that fully professional programmers, presented with the equivalent of architectural drawings on the skyscraper model, would produce the desired results, possibly on time, and possibly on budget. There are, however, some real advantages to working with a younger, less restricted team. Both the excitement and frustration of the current rate of development in electronic communication make it unlikely that a scholar focused on the creation of content will be able to keep up; for this reason there will inevitably be technical capabilities that the academic directors of the site are simply unaware of. Students love technology, and, if given some leeway, will find ways of making the site more creative in its programming, and more forward-looking in its architecture; in addition, they will be keenly aware of the sites on the Web that are most “cool” in their implementation of newer technologies, and will be able to recommend them to their less trendy supervisors. Several features of the new ISE site are the result of suggestions from the student team. It is also true, of course, that coolness needs to be tempered by the need for simplicity and usability.
The interaction of a team as multiple and as diverse as this requires both organization and the use of appropriate technology. There is no possibility for the full application of Open Source methods, since the team is necessarily limited in number by available funding, and since the Web applications under development are of interest more to the users of the content than to programmers. But some Open Source techniques work very well indeed in a group of three or four programmers. Each member of the team has responsibility for a specific area: the database design for Shakespeare in Performance; the database for viewing facsimiles of the early editions; the XML textbase for the plays and poems themselves, together with their complex layering of commentary and annotation; the database of links we provide to other resources on the Web of interest to Shakespeareans; the section of the site that provides a background to Shakespeare’s life and the context of his work, structured as another XML database; and an Essay Viewer that automates the addition of critical and other materials in linked essays.
At the core of the success of this team is “egoless programming” (Gerald Weinberg qtd. Raymond, ar01s11.html). Team members have created both a Wiki and an issue-tracker (Trac) that records their progress, lists of items to do, and shared resources; all members of the team share code as needed, and there is a wider discussion list that reaches all those interested in the development of the site. During the summer break, the student researchers shared a common laboratory space with the programmers, where they were able to work together; during the school year, however, students naturally have conflicting schedules, and as a result one student programmer is now doing graduate work in a different city. The team uses not only the issue tracker, the discussion group, and email, but also instant messaging of various kinds to keep in touch. As the programmers’ work progresses, they become intimately involved with the design of the interface that will present the data they work with to the user of the website. This topic goes beyond the scope of this chapter; the range of skills needed in the development of an attractive, consistent, and usable interface is beyond even the best of students, so at this point we have to invoke volunteer work from those skilled in the area and to reach into the professional arena for the polishing of graphic design, browser compatibility, and usability.
Thus far in this discussion, as I look for ways of understanding and promoting collaborative structures in the development of Humanities electronic projects, I have explored the use of metaphors of architecture (cathedral, skyscraper), commerce (bazaar), and ecology (…). It is clear that, given the levels of funding that we are likely to achieve, we will be unable to build many skyscrapers, so the top-down model rejected by Raymond (who is clearly thinking especially of Microsoft) is impractical. On the other hand, the structures of Open Source software are unlikely to function in the relatively narrow, content-specific areas that are of interest to Humanists. Whatever ideological subtexts are hinted at in the various metaphors invoked by Raymond (see K. Best), all embed assumptions concerning management and management styles. If the Open Source movement suggests some useful techniques for enhancing collaboration among programmers, it is important to acknowledge that successful Open Source projects have at their centre a personality of some strength and understanding in the orchestration of collaboration. My metaphor in this last sentence is a first step towards examining some metaphors close to the heart of the Arts and Humanities might in the hope of finding some further insight.
Elsewhere I have suggested that the computer screen can become a kind of stage where Shakespeare’s text performs and transforms in ways that illuminate its variousness (“Standing in Rich Place”). The metaphor of collaboration as preparation for a performance is attractive: an orchestra, or a company of actors, rehearses, and in doing so creates an intensely collaborative work of art. A stage performance especially requires the cooperation of a wide variety of artists and administrators as the director integrates the artistic vision of the play; as I have suggested above, the manager of a major website involved in the publication of Humanities documents is in a similar kind of position, integrating academic, technical, and design teams into a coherent whole. Unsurprisingly, perhaps, there are a wide variety of techniques and styles when it comes to the business of directing a play. Two classic examples will illustrate my point. Laurence Olivier might represent a very top-down (skyscraper) approach:
The actor must be disciplined. He must be so trained that he automatically carries out the director’s orders. I expect my actors to do exactly what I tell them to do and to do it quickly, so I can see my own mistakes immediately if I have gone wrong. I believe the director must know the play so well that he grasps every important moment of every scene. He knows—and he alone—when the action should rise and where it should fall. (qtd. in Cole and Chinoy 412)
whereas Joan Littlewood adopts a much more organic, or bazaar-like method:
I do not believe in the supremacy of the director, designer, actor or even of the writer. It is through collaboration that this knockabout art of theatre survives and kicks (Littlewood 133, qtd. in Schafer 15).
It is reasonable to suppose that the same variety of approach is available to those managing projects in Humanities Computing. Perhaps the best balance might come from the source of my title for this paper: Peter Quince, director of “Pyramus and Thisbe” in A Midsummer Night’s Dream, combines suggestions from his cast—how to represent Wall, for example, and how not to fright the ladies with a Lion—with firmness in dealing with the outsize ego of Bottom, who wants to play all parts. He encourages Snug the Joiner, who is shy, while ready to improvise additional lines for the play to make all clear.
At this point, it is helpful to ask a very basic question: does our use of the electronic text actually create anything genuinely new in the experience of the reader? In Sonnet 59, Shakespeare explored a traditional debate between past times and the present, asking whether the modern world does anything better than the ancient:
If there be nothing new, but that which is Hath been before, how are our brains beguiled, Which, laboring for invention, bear amiss The second burthen of a former child! (Sonnets)
Those of us working on Early Modern texts in the electronic medium are familiar with a continuing paradox: at the same time as we insist on the newness of the medium we determinedly argue for the vital importance of the physical print object we are transmuting into varied electronic formats. And it may be that sometimes, as we eloquently argue for the originality of the work we do in our next grant application or conference paper, we find ourselves suppressing the thought that we may be laboring to give birth once again to a child already born—one of the possible readings of the rather obscure lines in the sonnet that forms my epigraph.
The ISE e-text, as it is being re-conceived, can be thought of as a stage for the display of the complexities of the texts as we receive them from books published in Shakespeare’s time. Not only can users view and compare multiple transcriptions and facsimiles, but they can choose to highlight different sets of variants, changing the display dynamically as they do so. Thus the interface performs the nature of the text with a script provided by the programmer, as interpreted by the user. Two structures are crucial in constructing this stage: a hypertext environment that satisfies the needs of users, while reflecting the underlying patterns of the text; and an interface that makes it possible for users to generate queries that will allow the databases and analytical tools to perform to their full capability.
One constant thread in the energetic discussions that have revolutionized editorial theory in recent decades has been an emphasis on the vital importance of the physical artifact that recorded the text we work with. In their different ways, scholars like Jerome McGann, Margreta de Grazia, and Randall McLeod have, with clarity and wit, shown how both the production of a text and our understanding of it are profoundly influenced by the physical medium. At the same time we argue that the hypertext edition—and its extension, the database-driven edition—are revolutionary in their re-imagining of the process of reading and interacting with the texts we edit. A common aim in many of the electronic texts recently created or under construction has been therefore to record as much information as possible from the original artifact, while exploiting the additional capabilities of the electronic medium. There are two ways in which the ISE is moving towards integration of graphic with text: working with digitized facsimiles of the original quarto and folio texts, and generating a database of Shakespeare in performance.
The starting point is, of course, the electronic transcription of the texts. From the earliest application of SGML (Standard Generalized Markup Language) to the recording of Early Modern texts, there has been an ongoing debate about the limitations of most tagging schemes in representing both the conceptual and physical structures of a text. One of the pioneers in this field has been Ian Lancashire, who in 1994 created the encoding guidelines for his Renaissance Electronic Texts, in a format that permitted the markup of multiple hierarchies where needed. The SGML for ISE texts follows Lancashire’s guidelines closely; more recently, however, we have developed a system for converting these texts to standard XML, capable of being searched and transformed by readily available Open Source software.  It is a relatively simple task to set up a database of facsimiles where one searches on the electronic transcription in order to locate specific information in the linked image; the fun comes in setting up the linking structures and a reasonably comprehensible interface. On the ISE site it is now possible to view and compare multiple facsimiles of the plays, and to see them side by side with a transcription of the text. Because the publishing history of the plays is complex, the interface for accessing the facsimiles must, of necessity, reflect that complexity. What we have realized in the development of the initial interface for viewing facsimiles is that it reflects the collaborative process of editing the plays, discussed earlier in this chapter. In addition, the process of refining the interface has become something of a model of the iterative and exploratory nature of the development of an interface that is unique to the electronic medium, and which must navigate a complexly inter-related dataset. Both the programmer and the scholar learn from the process. The list of physical books for which we have digital images is substantial and growing: we currently have access to Folios 1 to 4 from the State Library of New South Wales, and a number of early Quartos from the British Library. The choices will grow: thanks to the generosity of the British Library we will be able to add images of all significant early Quartos; we are in the process of adding a second copy of the First Folio, from Brandeis University, a second copy of the Second Folio, in the possession of Eric Rasmussen, and we expect to be able to add later editions at least from Rowe to Malone.
Some challenges in solving problems of searching and accessing the texts are illustrative, again, of the long tradition of editing and publishing the plays. It is no simple matter, for example, to search the images by the common academic benchmark for citation (act, scene, and line) since these concepts are unstable and fluid in the published texts, ancient or modern, that the visitor to the site may be referring to. The upshot of this fluidity is that we must perforce anchor our conceptual divisions of act and scene to one or another print-based, physical representation of the play. For historical reasons, we have chosen the influential Globe/Cambridge edition of 1863-6 as our guide. The same ambiguity of act, scene and line numbers means that we use the Through Line Number (TLN)—numbers generated from the Norton facsimile of the First Folio(s) edited by Charlton Hinman—as the one constant that we retain through all editions and books. We have added the concept of a Through Page Number to deal with the various complexities of unnumbered, misnumbered and variously numbered pages in the different editions—in the First Folio, for example, pagination starts again as we move from Comedies to Histories, and again to Tragedies, with the further anomaly that Troilus and Cressida, apparently added at the last minute, has no pagination at all.
The fact that we have accurate transcriptions of the original texts means that we can enable the visitor to the site to search by word or phrase, limiting the search at will to a specific play or poem. From the list of “hits,” the visitor can then branch to view the appropriate page in the various facsimiles. One common problem in a search of this kind is that of the old-spelling itself, which is both variable, and difficult for any but an advanced student or scholar to navigate. For this reason, we have instituted a basic “fuzzy” search that checks for common variants in Early Modern spelling and searches for them as well. The screens for viewing the images are necessarily complex, and we anticipate learning from further usability studies to refine them.
A second way that the ISE will integrate image and text presents rather different challenges. But as I wrote that sentence I realized that I was implying a familiar, but false, distinction between text and image: just as the physical book is a vital component in our understanding of the words printed on it, the concept of the “text” of a play can—and arguably should—be extended to include its performances. The history of criticism of Shakespeare through performance is a long and honorable one, and there has been a recent flourishing of interest in theorizing the practice in the work of William Worthen, Peter Holland, and Michael Bristol, among others. The fundamental problem is that performance is evanescent and mutable, each night or matinee producing a different reading of the text. As a result, a popular sub-genre of Shakespearean performance criticism focuses on Shakespeare on film, since film can be replayed, and on DVD analyzed if necessary frame by frame in a satisfyingly fixed format. Yet it will be clear that the medium of film requires a more radical transformation and adaptation of Shakespeare’s words than all but the most wildly experimental stage production. Other discussion of Shakespeare in performance depends on the happenstance of the writer’s visits to the theatre, or possession of a budget that permits travel to theatre archives.
One result of this inevitable limitation is that performance criticism that reaches beyond film tends to focus on what I might call “canonical” productions: the Royal Shakespeare Company, the British National Theatre, and if the reach extends beyond the UK, Stratford, Ontario, and productions in New York. Even Ashland, Oregon gets few mentions. And of course students and scholars in cities that are far from theatre archives or major production companies have limited access to the rich range of materials that illustrates the interaction between Shakespeare’s text and recent cultures. It is probably fair to say that a significant number—even a majority—of today’s students have minimal experience of the theatre. But the astonishing fact is that Shakespeare’s plays are performed more now than ever before, especially in North America. The Shakespeare Theatre Association of America (not to be confused with the academic SAA), has an active membership of over 75 companies that focus primarily on productions of Shakespeare—and this group does not include the many companies that will choose a Shakespeare play once every two or three years. In Canada there are professional festivals in Vancouver, Calgary, Edmonton, Saskatoon, Toronto, Ottawa, Stratford (of course), Halifax, and St. John’s. Shakespeare is performed from time to time in translation in Québec, and there are also a number of amateur or semi-professional festivals in smaller centers, like Victoria and Prince Albert, Saskatchewan.
Working collaboratively with these companies, the ISE has developed a database that will record and preserve some, at least, of this almost overwhelming wealth of materials. Our focus is on the material evidence of the activities that go on backstage: prompt books, adaptations of the text, set and costume designs, production photographs, directors’ notes, and so on. Collectively these companies provide a remarkable resource for research in a number of directions. The database is a storehouse of resources for performance critics and students, directors and actors can look for ideas for their own work, and there is an opportunity to explore the phenomenon of the continuing success of Shakespeare’s plays as a kind of high culture reaching a popular audience by companies that depend on the box-office for survival -- very much as Shakespeare’s own company needed the pennies and sixpences from audiences at the Globe.
Ironically, given my earlier comments, initial work on the actual database began with Shakespeare and film. The premier scholar in the field, Kenneth Rothwell, whose catalogue Shakespeare on Screen: An International Filmography and Videography has for many years been the primary reference work in the area, donated his work to the ISE; his book appeared in 1990, and has been updated by Tanya Gough of Poor Yorick, a specialist bookstore/video store in Stratford Ontario. The process of rescuing the original electronic data from Rothwell’s book was an instructive exercise in the rapid obsolescence of technology: the book was originally prepared on 5 ¼ inch floppies on a PC using WordStar. As we find ways of inter-relating the materials we are digitizing, we are conscious of the importance of providing the kinds of metadata that will facilitate searches, since most materials will be binary objects rather than machine-readable text. Each object will have at least a standard Dublin Core entry attached, with further relevant information where possible. There is, as always, a trade-off to be made between searchability and complexity.
The potential for learning about both Shakespeare’s text and our own cultural history as we compare productions separated by space and time is, I believe, impressive and exciting: a further kind of collaboration with the past that goes far beyond examining books in the library. This second phase of the interconnection of text and image—or the extension of the concept of text to include image—moves well beyond the simple aim of showing an image from some antique book, in the words of the sonnet which I quoted earlier. Later in the same sonnet, the persona goes on to wonder, once again comparing earlier times with the present:
That I might see what the old world could say To this composèd wonder of your frame; Whether we are mended, or whether better they, Or whether revolution be the same.
I will not claim that the extended texts in the new medium are “better,” or even truly a “revolution,” but I do believe that the kind of edition/archive we are preparing will be both delightful and instructive as it performs on screen the multiplicities readers, actors, and directors of Shakespeare continue to reveal as the plays are re-created and re-searched.
If the process of managing collaboration for publication on the Web is a performance that requires some tact, and ideally some balance between the extremes of styles found in directors of Shakespeare plays, the resulting website is unlikely to be either a cathedral/skyscraper or bazaar. It will be more humble than the magnificence of the skyscraper or cathedral, and it will be more organized than a bazaar. Nonetheless, architecture provides a useful foundational metaphor, particularly since it is the nature of Web publication to be always “under construction” in one form or another. A more appropriate building might be something close to a house in an older suburb; one that has been added to and modified over time, as the wiring is updated, and additional rooms added at the back for expanding families and increased expectation of spacious living. In an academic context, we might take a university library (the metaphor that underlies the structure of the ISE site) where, again, modifications required by earthquake code, computer wiring, and an ever-expanding collection lead ideally to a multi-purpose building. If we choose to retain in our metaphor the ecclesiastical flavour of Raymond’s image of the cathedral, no more appropriate example could be chosen than the Holy Trinity Church at Stratford-upon-Avon: the whole structure expresses a unity of balance and design, despite the fact that parts of the church come from many different periods; the nave alone includes sections built from 1280 to the beginning of the Perpendicular style two hundred years later.
As we adapt to the demands of the new medium, Humanities scholars face a fundamental challenge in adapting the discipline to the demands of collaboration. As well as ensuring that collaborative work is adequately recognized in the academy (see Siemens and Best), we need to forge techniques of managing collaboration in a manner that accords with the best of the Humanities. It is for this reason that I choose to explore metaphors that describe and illuminate both collaborative and management styles. The electronic space of a modern server is a marvelous convenient place for the construction of an elegant and capacious virtual library that celebrates Shakespeare. It remains for us to manage the development of the electronic space to create a performance on screen worthy of that most demanding of audiences, the global world of the Web which encompasses a range from the modern equivalents of Duke Theseus to Francis Flute, the bellows mender.
Best, K . "Beating Them at their own Game: The Cultural Politics of the Open Software Movement and the Gift Economy." International Journal of Cultural Studies 6.4 (2003): 449-70. Print.
Best, Michael . “The Text of Performance and the Performance of Text in the Electronic Edition.” Computing in the Humanities 2002 36:3 269-282. Print.
—— . “From Book to Screen: A Window on Renaissance Electronic Texts.” Early Modern Literary Studies 1.2 (1995): 4.1-27 . Web. <URL: http://purl.oclc.org/emls/01-2/bestbook.html>.
—— . “Standing in Rich Place: Electrifying the Multiple-Text Edition: Or, Every Text is Multiple.” College Literature Special Issue 36.1 (Winter 2008) 26-39. Print.
Bristol, Michael D . and Kathleen McLuskie. Shakespeare and Modern Theatre: The Performance of Modernity. London: Routledge, 2001. Print.
Cole, Toby and Helen Krich Chinoy, eds . Directors on Directing: A Source Book of the Modern Theater. Rev. ed. Indianapolis: Bobbs-Merrill, 1963. Print.
de Grazia, Margreta . "The Essential Shakespeare and the Material Book." Textual Practice 2.1 (1988): 69-86. Print.
—— and Peter Stallybrass . "The Materiality of the Shakespearean Text." Shakespeare Quarterly 44.3 (1993): 255-83. Print.
Dublin Core . Metadata Initiative. 1995-2002. Web. <http://dublincore.org/>.
Erlande-Brandenburg, Alain . The Cathedral: The Social and Architectural Dynamics of Construction. New York: Cambridge UP, 1994. Print.
Hinman, Charlton, ed . The Norton Facsimile of the First Folio of Shakespeare. 2nd ed. New York: W.W. Norton, 1996. Print.
Holland, Peter . English Shakespeares: Shakespeare on the English Stage in the 1990s. Cambridge, New York: Cambridge UP, 1997. Print.
Huot, Wendy . 2005. “Wendy's Panel Presentation.” Web. <http://mustard.tapor.uvic.ca/~isedev/isewiki/doku.php?id=wendy_panel>.
Internet Movie Database . 2003. Web. <http://www.imdb.com/> January 2006.
Internet Shakespeare Editions (ISE). Coordinating Editor, Michael Best. 1996-2006. Web. <http://ise.uvic.ca/>.
Lancashire, Ian, ed . Renaissance Electronic Texts. Toronto: University of Toronto Library, 1997-2005. Web. <http://www.library.utoronto.ca/utel/ret/ret.html>.
—— . RET [Renaissance Electronic Texts] Encoding Guidelines. 1994, 1997. RET Supplementary Studies 1.2. Web. <http://www.library.utoronto.ca/utel/ret/guidelines/guidelines0.html>. February 2006.
Literature On Line (LION). Chadwyck-Healey (ProQuest Information and Learning Company), 1996-2001. Web. <http://lion.chadwyck.com/>. Access by subscription only.
Littlewood, Joan . “Goodbye Note from Joan.” In Charles Marowitz, Tom Milne, Owen Hale, eds., The Encore Reader: A Chronicle of the New Drama. London: Methuen, 1965. 132-4. Print.
McConnell, Steve . Software Project Survival Guide. Redmond: Microsoft Press, 1998. Print.
McGann, Jerome J. A Critique of Modern Textual Criticism. Chicago: U of Chicago P, 1983. Print.
——. “Culture and Technology. The Way We Live Now; What is to be Done?” 2004. Web. October 2005. <http://tapor1.mcmaster.ca/~faceoftext/abstracts.htm#mcgann>.
——. “The Rationale of Hypertext.” Institute for Advanced Technology: University of Virginia. 1995. Web. 28 September 2001. <http://jefferson.village.virginia.edu/public/jjm2f/rationale.html>.
McKenzie, D. F . Bibliography and the Sociology of Texts. Cambridge: Cambridge UP, 1999. Print.
McLeod, Randall [Random Cloud]. “The Marriage of Good and Bad Quartos.” Shakespeare Quarterly 33.4 (1982): 421-431. Print.
Meier, Wolfgang . eXist. XML indexing and searching program. 2006. Web. <http://www.exist-db.org>.
Raymond, Eric Steven . “The Cathedral and the Bazaar.” Thyrsus Enterprises. Version 3.0, 2000. Web. <http://www.catb.org/~esr/writings/cathedral-bazaar/cathedral-bazaar/>.
Rothwell, Kenneth . Shakespeare on Screen: An International Filmography and Videography. New York: Neal Schuman, 1990. Print.
Schafer, Elizabeth . Ms- Directing Shakespeare: Women Direct Shakespeare. 1st ed. New York: St. Martin's, 2000. Print.
Shakespeare, William . "Sonnets." Web. March 2009. <http://isebeta.uvic.ca/Annex/Texts/Son/Q1/Work/Work>.
Shakespeare Collection , The. Thompson-Gale. 2005. Web. <http://www.galegroup.com/shakespeare/>.
Siemens, Raymond, Michael Best, et al . The Credibility of Electronic Publishing: A Report to the Humanities and Social Sciences Federation of Canada. 2001. Web. <http://web.mala.bc.ca/hssfc/Final/Credibility.htm>.
Siemens, Raymond . “Disparate Structures, Electronic and Otherwise: Conceptions of Textual Organisation in the Electronic Medium, with Reference to Editions of Shakespeare and the Internet.” in Michael Best, ed. The Internet Shakespeare: Opportunities in a New Medium. Early Modern Literary Studies 3.3 / Special Issue 2 (1998). Web. <http://purl.oclc.org/emls/03-3/siemshak.html>.
STAA [Shakespeare Theatre Association of America]. Shakespeare Editions, University of Victoria. 2002. Web. <http://www.uvic.ca/shakespeare/STAA/>.
Trac . Edgewall Software, 2003-2005. Web. <http://www.edgewall.com/trac/>.
Van Hardenberg, Peter . “Treespaces.” Contribution to SIG [Special Interest Group]:Overlap, entry 2.1.5. Web. <http://www.tei-c.org/wiki/index.php/SIG:Overlap>.
Weinberg, Gerald M . The Psychology Of Computer Programming. New York: Van Nostrand Reinhold, 1971. Print.
Worthen, William B. and Peter Holland, eds. Theorizing Practice: Redefining Theatre History . New York: Palgrave Macmillan, 2003. Print.
Worthen, William B. Shakespeare and the Force of Modern Performance . New York: Cambridge UP, 2003. Print.
 See Erlande-Brandenburg for a discussion of the complex economic, social, and architectural forces in play in the construction of many cathedrals.
 Most notably the automatic generation of a table of contents for each page, the dynamic generation of a home page for each play or poem that collects information about it from the various sections of the site and is automatically updated as new material is added, and the general design of the database of Shakespeare in Performance, which uses many of the highly relational structures of the Internet Movie Database.
 I would like to give full acknowledgement to the young scholar/programmers who have developed these sections of the site: Wendy Huot, who has developed the databases for performance and viewing facsimiles; Peter van Hardenberg, who has created the Text Viewer (including an original and elegant method of dealing with the thorny problem of overlapping hierarchies in XML); and Michael Joyce, who has had responsibility for the links database, Shakespeare’s Life and Times, and the Essay Viewer. More textual analysis software is currently being integrated by Matt Crocker.
 See McConnell, for an example of an introduction to the techniques of software development advocated by those involved with Microsoft.
 One of our developers, Peter van Hardenberg, has generated an elegant solution to this problem by adapting the accepted XML structure of namespaces, creating what are in effect “treespaces.” It is possible to flatten the overlapping treespaces rapidly and elegantly into a single well-formed XML document as needed. We believe that this is a solution that is less a workaround than a consistent and logical application of the principles of XML itself.
 We use the Open Source XML indexing software eXist, developed by Wolfgang Meier, <http://www.exist-db.org>.
 Our developer, Wendy Huot, presented a paper at the recent ACH conference in Victoria exploring the challenges in this process (“Wendy’s Panel Presentation”).