Michael Best
Department of English
University of Victoria
Victoria, B.C. V8W 3W1
Canada.

Scholarly Editions of Shakespeare for the Internet

1. The Internet as Medium

Vast potential but precious little content; thousands of pages pointing to other pages with pointers to other pages. This is a common -- and generally accurate -- complaint about the Internet. In a recent assignment for students taking a computer-mediated course on Shakespeare by Individual Studies, I asked them to list things they wanted to find on the Internet but couldn't. The list was long: where are student tools like introductory essays, plot summaries, glossaries, cheat sheets; where are the scholarly articles, reliable annotated texts, sources for performance criticism, reviews of Shakespeare plays on video? Where are the musical settings of the songs in the plays? Information about Elizabethan costume? [You can browse some of the student pages from this link]

Some content has arrived on the Internet, especially on renaissance figures other than Shakespeare. There is the Spenser Web, the Piers Plowman Electronic Archive, the Dante Project, and there are an increasing number of electronic texts.[1] But there is very little on Shakespeare, and no central scholarly archive of any kind. In this paper I will discuss some of the challenges in designing such a site and some of the critical and theoretical questions that need to be answered in the process. The site itself will make its initial appearance this summer, and will initially focus on providing scholarly editions of the plays.

I begin by looking at the nature of the Internet as a medium, suggesting some conclusions that arise from the kind of use Internet texts would stimulate; then I look at one important and rather technical matter, the kind and extent of "mark-up" that will be embedded in the texts.

1.1. Books and CD ROMs

The advent of the medium of CD ROM has led to some discussion of the nature of electronic texts in general, and of the way they diverge from conventional printed materials.[2] But there is one way in which books and CD ROMs are similar, and which differentiates both media from an Internet site: each is self-contained. There are two major consequences that follow from this fact. In a book and on a CD, copyrighted material can be reprinted, subject to the negotiation of acceptable fees; and any revision of the material involves a complete reprinting (or "reburning") of the edition.

At this point the book and the CD begin to diverge. The designer of the book of necessity has complete control over the appearance of the text, but while the designer of the CD has the option of controlling display, to do so is to reduce the power of the electronic medium, since the computer makes it possible to allow the user control over such things as window and font size. The addition of these variables makes the process of designing for the CD significantly different from the book.

When it comes to the kinds of materials that can support the text, the CD diverges even further from the book. The book can provide graphics, but the CD offers an opportunity for extensive text, graphics, sound and video -- and the technology is improving rapidly both in the amount of data that can be stored and the speed at which the data can be accessed for display by the computer. And of course the CD provides a far more elegant and efficient means of hypertext linking than the book.

[Return to top]

1.2. The Internet

It will probably be clear already from this brief summary that an Internet edition will diverge from the traditional printed edition at a much earlier level, since the site is neither self-contained nor fixed. This difference creates a number of opportunities -- as well as problems.

1. The open structure of the Internet allows the text to include hypertext links to sites external to the edition, a facility that will result in powerful research tools as the number of locations offering scholarly content grows.
2. The fact that the site is not a discretely published physical object opens the attractive possibility of continuous revision as improvements can be made in the published version. We have all experienced the frustration of the appearance of an important book or article as our own work is in press.
3. Perhaps the strongest claim for the significance of Internet editions is their accessibility. Books and CDs need to be bought or borrowed from the Library; the Internet editions would be "published" in metaphorical rather than physical terms, and will be accessible to anyone with a computer and a modem.
4. The very accessibility of the editions, however, creates a potentially limiting problem: that of copyright. There is the possibility of creating a site that would allow access or downloading of copyright material only after the payment of a fee through a secured channel, but this solution seems to me to run counter to the expectations of most scholars and students, and it would certainly restrict access to those using browsers capable of transmitting encrypted financial transactions. The problem of copyright and the Internet editions can be seen as both a limitation and a strength: all data on the site will have to be freely available for educational and non-commercial applications. Thus the range of material will be less than can be put on a CD, but the material that is there will be accessible for guilt-free use by scholars and teachers.
5. The structure of the Internet also means that the designers of the editions have far less control over the appearance of the text, since these are displayed by various browsers on various machines with varying user preferences for font sizes and types. One consequence of the uncertainty of appearance is that designers of Web pages are increasingly using graphics, since these are more controlled than text. Another is the continuing battle in the HTML wars between conceptual and visual tags (<cite> </cite> versus <i> </i>).[3] I have also noticed an increasing acceptance of the use of tags that are designed to be conceptual being used for purposes of graphic display: in particular the tag <blockquote> is being used to force a margin of white (or whatever) space between the edge of the window and the text, often for a complete page with no block quotes in it whatsoever.
6. Like the CD, the Internet offers the opportunity for extensive supporting text, graphics, sound and video clips, though the usefulness of much of this kind of material is limited at present because of the slow rate of data transfer; the prospect of wider bandwidth transmission is in the offing, but a well-designed Internet edition should recognise that the speed is unlikely to catch up to the technology of the CD ROM, and that the higher speeds will be attained only by a fraction of users in the forseeable future.
7. In creating hypertext links, the designer of a CD can build into the software flexible options for multiple windows as well as sequential choices, because the software is included as a part of the publication. But WWW browsers are by their nature (again, in their present stage of evolution) significantly more limited, and for the most part are limited to sequential hypertext (or. . . or . . . or, rather than and. . . and . . . and).[4] Even a pioneer work like the CD ROM Voyager Macbeth provides more ways of viewing hypertext material than is currently possible on the Internet.[5]
8. In a printed edition, there is a tradition of providing an extensive Introduction that covers a range of scholarly issues: the probable date of the play, the sources Shakespeare may have used, the nature of the textual choices that the editor has made; and there will usually be both a critical discussion of the play and a survey of the history of its performance. All this material should similarly be available for the user of the Internet editions, but the electronic medium introduces one important difference. The supporting discussions need not be sequentially hooked to the play as an introduction, and the different topics need not be all contributed by the same scholar. The user will see an introductory menu that offers access to separate essays, each hypertextually linked to the text where it is referred to. The editor of the text of the play will write the essays on the textual principles of the edition, and probably the discussions of date and sources, but other materials may be contributed by other scholars with expertise in the area. Tools that would be especially useful to the advanced scholar would be a thorough overview of criticism on the play and an extended discussion of its stage history. Since these will be free-standing essays, they can be more detailed than is possible within the confines of a printed edition.

[Return to top]

1.3. The Audience for the Editions

An important question about any edition is to decide what audience it is being developed for. Books target specific audiences: high school students who need basic explanations of the situation and language (Coles Notes and the like); students taking college introductory courses (Signet, Bevington, Riverside); scholars and advanced students (Arden, New Cambridge, Oxford). CDs will still probably be aimed at either student or scholar, though the medium makes possible a wider potential audience according to the quantity of material available and the capacity of software to offer different interfaces for different needs.

An Internet edition should likewise be aimed at a particular audience if for no other reason than to make the job of the editor possible. Current editions on the Net seem to be designed for a general browser interested in a traditional text rather than annotation. The editions I am proposing will be aimed at scholars and advanced students, at roughly the same level of interest and expertise as the original Arden editions. But Internet users should not be defined purely in terms of their level of expertise in Shakespeare studies; they must also be considered in terms of the sophistication of their equipment and the kind of connection they have to the Internet. For the immediate future -- and possibly for much longer -- we must assume that the site will be accessed by those with equipment that varies from plain text browsers and relatively slow modems to those with a fast ethernet connection and a browser capable of displaying advanced graphics, sound, and video. This variety of access leads to at least two conclusions concerning the nature of the material that should be made available.

1. The site should make data available in a variety of different formats suitable for different browsers, from those with simple ftp transfer to those with full hypertext and graphics capacity.
2. Since the user is likely to be charged connect time for any work on the Internet, the site must offer fast and efficient tools for the "hit and run" user: someone who wants to look up a particular passage, check the meanings of particular words, or access recent performance criticism on a particular play -- and then download the result of the search for further use on the home computer. Thus the user may be interested in anything from simple ASCII text to fully marked up material that can be browsed locally. In the second case, scholars might want to download the text of a whole play and its annotations, then use it with their browsers on their own hard drives: the Internet edition at this point becomes more like a CD ROM edition, with the choice of materials up to the user rather than the publisher. Since graphics, sound, and video take a lot of time to view or download, there should be small, compressed "thumbnail" versions available for sampling.

[Return to top]

2. The Texts

At some point the general editor of an Internet site must leave the relatively safe role of webmaster in order to provide consistent guidelines for the individual editors who are creating the texts that will be put on the site. And here I venture on an area of recent scholarly activity that has sometimes seemed like a fully fledged "flame" war: the question of the particular kind of modern text that should be published on the site.

To preface my discussion on this point, I should outline the range of texts I envision being made available on the Internet site. At the centre of the edition will be a modern text. From it will branch annotations, both explanatory and textual; from these annotations in turn will branch links either to further supplementary materials (sources, criticism, sound, video, graphics), or to material held on other sites (references to other on-line resources). In addition, the modern text will be linked to an extensive collation, and to both graphic and electronic versions of the original quarto and Folio texts. Thus the user will have immediate access to the texts that lie behind the modern text, and to a full discussion of the textual choices the editor has made.

If I may attempt to simplify a complex debate, the question is whether editors should assume that behind the printed texts there lies a single source, Shakespeare's unique foul papers, and it is the object of the editor to recover as far a possible what Shakespeare actually wrote (or, more arguably, meant); or whether they should assume the likelihood that the texts preserve different states of a dynamically revised text, the result of a collaboration between Shakespeare, his fellow actors, the scribe, the compositor, the proof-reader, and so on? Of course these propositions are not always mutually exclusive.[6]

There is a prior question that must be asked before the nature of the modern text can be decided, a question that has little to do with the substance of the flaming debate. We need to return to the question of the actual way the editions will be used, given the nature of the Internet as medium.

I have already pointed out that the nature of an Internet edition means that one major function of the modern text will be to provide a starting point for exploration: of annotation, of the primary texts (electronic or graphic representations), and of other supporting materials. In effect, the text is both a message and a medium: words to be read and pointers to click. Unlike a traditional text in book form, the modern text here will be both a gateway to the originals and a base text from which users can, if they wish, construct their own texts. Thus the aim of the editor should be less to try to reconstruct a text which approximates that elusive original Shakespearean handwriting or intention than to construct a text that provides a tool for exploration and reconstruction according to the varying needs of the users -- which may vary from creating performance texts to verifying the punctuation of the Folio.

In a book the printed text is the only one the reader will have access to (except for the rather clumsy solution of parallel text editions), whereas the electronic edition can offer both original texts and multiple versions of the modern text. There are a number of conclusions that follow.

1. Since the originals are just a click away, the editor is freed from the need to remind the reader that the text has been tampered with in the process of modernization, and can thus with a clearer conscience choose modern forms and spellings. In effect this means that the principles of modernizing spelling in the Internet editions will be closer to those adopted by the new Oxford Shakespeare than by the carefully pedantic Riverside.
2. On the other hand, the proximity of the originals may be expected to promote a reasonable degree of modesty in the editors when it comes to emendation since their modifications to the original will be palpable, no longer hidden in the runic code of collation that only reviewers, fellow editors, and particularly conscientious students in their obligatory course on bibliography can decipher.
3. As a tool for creating variant versions to suit the specific needs of scholars and actors, the initial texts should aim to produce the "poem" rather than the "play," to use a useful distinction revived recently by Anthony Hammond.[7] The tradition of theatrical texts, from the "bad/acting" quartos to the present day has been to look tolerantly on major revision for the sake of the immediate needs of performance; logically, the base text should be the poem before it has been turned into one of the many plays possible.
4. A corollary, however, of this principle is that one function of the Internet editions should be to make available variant versions of the base text as it has been used to generate acting versions, from the "bad/acting" quartos to modern prompt books. Thus in the case of Romeo and Juliet one modern version could rigorously present the "poem" from Q2, another could stress the additional information about early performance provided by Q1: an initial menu would present the user with a choice of the various texts, modern and original. For King Lear the user should eventually be offered a choice of a traditional conflated text (the most generalized tool, and thus the first to be offered), a modernized Folio, and a modernized Quarto. All these in turn would be linked to electronic and graphic representations of the original Folio and Quarto.

The slightly surprising result of this analysis is that the medium of the Internet allows the editors less to douse the flames of controversy than to use its heat to generate multiple tools for users of the editions. The general editorial principle becomes the need to ensure the scholarly integrity and clarity of critical focus of each version of the play within its own terms, rather than to provide a single and necessarily limiting philosophy in order to ensure consistency.

Thus one important effect of the openness of the Internet as a medium is that the texts themselves will be collaborative in a way that is unusual for modern editions. The presence of multiple texts and supporting materials will mean that it is more likely that scholars will contribute parts rather than wholes: the modern text and notes, the electronic versions of quarto and folio, graphic. sound, and video archives. And the "texts" (including the attached notes and support materials) will be dynamic. Alterations and accretions can be accepted by a process of approval, the equivalent of reading articles for publication. There will also be a free discussion group attached to the site, focusing specifically on issues of the text and its annotations -- another important way in which Internet editions will be a completely different medium from books and CD ROMs.

[Return to top]

3. Mark-up Languages

A challenging necessity in making texts of any kind available in the Internet is that they need to be "tagged" -- marked up in what is in effect a computer language so that the user's browser will understand and display the text effectively. The editor of an Internet Shakespeare must make some difficult decisions both on the kinds of mark-up to be undertaken and the level of detail that the mark-up will record.

[Readers familiar with the distinction between HTML and SGML can skip this paragraph.] Standard web browsers at the moment recognize HyperText Markup Language (HTML). HTML is primarily designed to provide the browser with information on the visual display of text and graphics, and the presence of links to other materials. With the addition of various program extensions, a user can send a query to the site and receive an answer; typically this happens when users are given the tools to search a database. Standard Generalized Markup Language (SGML) is a more inclusive and powerful language that permits sophisticated textual analysis and multiple means of interpreting the text, since much additional information about it is included.

Neither of these mark-up languages is ideal for Shakespeare editions. HTML is still very much a language in development, and there are a number of features in text layout that are difficult to achieve elegantly in its present form. An obvious example is the difficulty of providing an attractive way of aligning line numbers, either on the left or right of the window, without resorting to "workarounds" -- technical tricks -- that will have unpredictable results in different browsers.

[Return to top]

3.1. Why SGML?

HTML is the native language of the World Wide Web, so there is no question that the Internet Shakespeare should provide texts in this format. But scholars are likely to want more than a text to look at. The combination of a reliable electronic text and a computer is a remarkable opportunity for those who wish to analyse the text in various ways; if the main purpose of HTML is to display text and make hypertext links between sections of text, the main use of SGML is to provide the scholar with the power to search, analyse, and index a text in various ways. The problem is that SGML, for all its power, is a bit like using a sledgehammer to crack a joke.

The specific guidelines within SGML developed by the Text Encoding Initiative[8] have become something of a standard in the encoding of literary texts, but there have recently been some cogent criticisms of them. Ian Lancashire has written of the difficulties of adapting SGML for Renaissance electronic texts; in particular he challenges some of the basic assumptions made by those who developed the TEI guidelines in stressing the importance of logical structure at the expense of the physical representation of the text.[9] Users of the Shakespeare Internet editions who consult an electronic version of the Folio will want to know how the text looked, and may wish to analyse accidentals, spelling variations (including ligatures and the long "s"), catchwords, and many other characteristics of the text that are the result of the physical reality of the press and the processes of setting type rather than logical structures. A modern text will use italics for logical reasons (a word is being emphasized, a title is being cited), but renaissance printing conventions were less consistent, with the result that the only reasonable way to mark italics is with the physical type style "italic."

Approaching SGML from the point of view of an encoder working on a modern text, Michael Neuman shows that the encoder continually encounters problems of interpretation as the codes become more detailed, yet it is precisely the detail of the tagging that makes the text useful for analysis.[10] It might, for example, be useful to tag all Shakespeare's references to mythological figures so that they can be indexed and examined. Lear's "No, do thy worst, blind Cupid; I'll not love" (4.6.139) is easy, but is Duke Vincentio referring to the same blind god when he says "Believe not that the dribbling dart of love/Can pierce a complete bosom" (Measure for Measure 1.3.3)? And is Orlando writing of upper or lower-case love: "He that sweetest rose will find/Must find love's prick and Rosalind" (As You Like It 3.2.102-3)?

To these shortcomings of SGML/TEI I would add a further one that becomes pressing in the specific realm of electronic texts. Paradoxically, while the TEI Guidelines avoid tags that describe the physical nature of the book, they assume a general model of a conceptual book that is guided by the traditional printed medium: <front>, <body>, <back>. This sequential and hierarchical ordering of text is largely foreign to the electronic medium, where sequence becomes multi-linear rather than linear, and where there are often multiple and overlapping hierarchies. I have already suggested above that material traditionally presented in the Introduction will be accessible in parallel rather than sequential files; similarly, SGML/TEI does not manage electronic footnotes well, since any note can become the centre of a new web of interconnections to the original texts, to Shakespeare's sources, and to critical or performance materials rather than adjuncts to the "main" text. Lancashire remarks accurately that "SGML and TEI make anachronistic assumptions about text that fly in the face of the cumulative scholarship of the humanities."[11]

Why then should the texts be marked up in SGML as well as HTML? The answer is that (faute de mieux) SGML provides the scholar with indexing and searching tools that no other method of ordering the text can. In addition, if in the future more flexible or readable schemes of text mark-up become available, a text already in SGML will readily be converted by relatively simple programming. The general solution is to prepare texts that are lightly encoded, ready for scholars who are looking for more detailed information to add their own tags. To take an example from earlier in this paper, they can then add their interpretation of what constitutes a reference to a mythical figure and index the result.

In keeping with the general principle of providing the scholar with tools, the site will make available the kinds of relatively simple text-processing programs that will allow users to modify the SGML file. A simple example would be a program that translated the SGML text into a file that could be loaded into a word processor, complete with basic formatting. [12]

[Return to top]

3.2. The Folio and Quarto Texts

The original Folio and quarto texts will be transcribed literally, with no emendation or correction. The editors will add only such necessary navigational material as modern act and scene divisions, and line numbers anchored to the modern text. Links within and from these texts will be minimal, referring back periodically to the modern text and cross-linked at intervals to any other original texts.

1. HTML. Since HTML browsers at the moment do not display special characters like the long "s," the HTML versions of the quartos and the Folio will aim to produce uncluttered, readable copy, suitable for a quick check on original punctuation, spelling, and layout. These texts will be useful for scholars who wish to look at original readings for critical purposes, and will be valuable teaching tools for students at the undergraduate level, since the texts will be far easier to read than the originals.
2. SGML. The aim in tagging the original quartos and the Folio texts in SGML will be to produce texts available for further analysis; again the emphasis will be on providing basic tools for further use by individual scholars. The tagging scheme will stress the physical characteristics of the original text, and will follow a simplified version of the guidelines devised by Ian Lancashire for his Renaissance Electronic Texts, and suggested in his paper on "The Public Domain Shakespeare."[13]
3. Ultimately no physical attribute is free from the interpretative intervention of the editor, but where possible interpretation will be kept to a minimum. Thus, for example, the tagging scheme will identify compositorial effects such as justified text and hanging words, but will not attempt to indicate which parts of a text may have been set by different compositors, since this is a matter of inference from the physical characteristics of the text. Scholars interested in bibliographical and linguistic questions will thus be able to use these texts as starting points for their own schemes of "deeper" tagging -- and will be invited to make their versions available to other scholars.

3.3. The Modern Texts

1. HTML.The aim will be to produce clean, readable copy, following a conservative subset of HTML in its most recent generally accepted form. The HTML text will be the base text for navigation, linked to annotations, the collation, and the original quartos and Folio.
2. SGML. The purpose of the modern SGML text will be to make it available both to more sophisticated browsers as these become available, and to provide a base text for linguistic and stylistic analysis. Thus the tagging will more fully recognise structural and conceptual elements than the tagging for the original quartos and Folio. There will accordingly be more interpretative activity involved. Some elements to be tagged would include literary divisions (act scene line);[14] supporting text (speech headings, stage directions); editorial decisions (departures from the copy text, additions and deletions); literary types (verse or prose).

4. Conclusion

There is much that I have not discussed in this paper, some of which will involve breaking new ground. Especially in the new areas of graphics, sound, and video, we will have to find solutions to the problems of linking, indexing, and searching different media. But the challenge of solving these problems can only enrich the scholarly resources of those working in the area of Shakespeare studies.

Quite suddenly, the arena of scholarly Shakespeare editions has become competitive. Oxford single editions, the New Cambridge Shakespeare, and the new round of Arden editions are appealing to overlapping audiences in the world of print; there are a number of CD ROM editions under development, and one paper in this seminar will be discussing the forthcoming New Variorum editions. Not surprisingly, in this climate, there is a similarly sudden interest in creating editions that are different -- innovative in some form. If in no other way, the Internet editions of Shakespeare will be innovative in the way they use a medium that is making different kinds of texts available to an increasing audience. It is not the intention -- or the capacity -- of the proposed Internet editions of Shakespeare to compete on the same playing field as books and CD ROMs. The book will never lose its appeal as a closed, portable, and attractive medium for Shakespeare's texts. CD ROMs have not yet proven themselves capable of attracting and holding a paying public, but the ability of the medium to present copyright material, to format it predictably to the screen, and to provide a secure and speedy transfer of data will give these editions capacities the Internet cannot.

The Internet editions will, however, fill a real gap. In the field of Shakespeare studies there will be some content at last for the pointers to point to.

[Return to top]

Endnotes

[1] A sampling of the kinds of resources that are currently available will include these sites: Classical Greek and Roman texts (in English translation), provided by The Tech at http://www.perseus.tufts.edu/. Edmund Spenser Home Page at http://darkwing.uoregon.edu/~rbear/ EMLS (Electronic Journal, refereed) at http://unixg.ubc.ca:7001/0/e-sources/emls/emlshome.html. The Oxford Text Archive at http://info.ox.ac.uk/~archive/ota.html. Project Aldus at http://www.jhu.edu/~english/aldus/aldus.main.html. Renaissance Dante in Print at http://www.nd.edu/~italnet/Dante/. Renaissance Electronic Texts (RET, Toronto; a growing source of scholarly e-texts) at http://library.utoronto.ca/www/utel/ret/ret.html. Sidney's Defence of Poesie (Richard Bear) at http://www-vms.uoregon.edu/~rbear/defence.html. The Dante Project, gopher://gopher.dartmouth.edu/11/AnonFTP/pub/Dante. The Piers Plowman Electronic Archive (Hoyt N. Duggan) at http://jefferson.village.virginia.edu/piers/archive.goals.html. The Spenser Web (uses copyright material from Johns Hopkins) at http://www.artsci.wustl.edu/~jntolva/shepcal.html. Thomas Middleton at http://dayhoff.med.virginia.edu/~ecc4g/middhome.html. [Back]

[Return to top]

[2] See my article on this subject, EMLS 1.2 (September 1995) at http://unixg.ubc.ca:7001/0/e-sources/emls/01-2/bestbook.html. [Back]

[3] I discuss the argument about what constitutes "good" HTML in my article cited in the previous footnote. Any brief exploration of the World Wide Web will provide examples of both conservative and radical (not to say erratic) graphic layout of Web pages. [Back]

[4] See Jim Rosenberg, "Navigating Nowhere / Hypertext Infrawhere," 1995, at http://www.well.com/user/jer/NNHI.html. [Back]

[5] William Shakespeare, Macbeth, ed. A.R. Braunmuller (New York: Voyager, 1994). Introduction and commentary by David S. Rhodes; produced by Michael E. Cohen. Macintosh. [Back]

[6] Gary Taylor both documents the conflagration and fans the flames in "The Rhetorics of Reaction," in Crisis in Editing: Texts of the English Renaissance, ed. Randall M Leod [Randall McLeod] (AMS Press: New York, 1994) 19-60. [Back]

[7] "The Noisy Comma: Searching for the Signal in Renaissance Dramatic Texts," in Crisis in Editing, 203-250; see p. 205. [Back]

[8] In print media see Charles F. Goldfarb, The SGML Handbook, ed. Yuri Rubinsky (Oxford: Clarendon Press, 1990); Guidelines for Electronic Text Encoding and Interchange (TEI P3), ed. C.M. Sperberg-McQueen and Lou Burnard, 2 vols. (Chicago, Oxford: Text Encoding Initiative, 1994). In electonic media see "TEI Guidelines for Electronic Text Encoding and Interchange (P3)" at http://www.hti.umich.edu/docs/TEI/; "The official TEI Home Page" at http://www-tei.uic.edu/orgs/tei/; "A Gentle Introduction to SGML" at http://info.ox.ac.uk/~archive/teip3sg/; "CETH SGML sampler. A sample of SGML texts, encoded according to the TEI DTD, p3" at http://www.princeton.edu/~gjmurphy/sgml/index.html.

The Text Encoding Initiative has lightened up in its most recent release: a new Document Type Definition (DTD) "TEI Lite." the CETH SGML sampler explains: "TEI Lite is a subset of the full TEI DTD. Researchers who may have been put off initially by the TEI's elaborate use of parameter entities and driver files (necessary to initialize a full TEI setup) should return now [July 1995] to have a look at the much simpler TEI Lite." [Back]

[Return to top]

[9] Ian Lancashire. RET Encoding Guidelines. December 1994 at http://library.utoronto.ca/www/utel/ret/guidelines0.html. The home page for RET is found at http://library.utoronto.ca/www/utel/ret/ret.html. [Back]

[10]Michael Neuman, "You Can't Always Get What You Want." Paper delivered at the ACH/ALLC conference (Santa Barbara, July 1995); see the Conference Abstracts 84-6. [Back]

[11]"Early Books, RET Encoding Guidelines, and the Trouble with SGML." Ian Lancashire. Paper read at the Scriptorium conference in Calgary, November 1995. For an introductory discussion of the advantages of "lemmatizing" -- marking texts specifically for searching and developing concordances -- see the article by R.G. Siemens "Lemmatization and Parsing with TACT Preprocessing Programs," in Toronto Computing in the Humanities Working Papers. B.2: February 1996 Siemens also discusses some of the ways in which the process of lemmatizing the text is an act of interpretation as well as description, with the problems that this process involves. [Back]

[12]While the appearance of Web browsers capable of reading and displaying SGML files means that some of the more powerful features of the language may be translated into display, the continuing evolution of HTML, which is dedicated specifically to display, is likely to be the continuing language for initial browsing of the Internet Shakespeare Editions. [Back]

[13]Ian Lancashire, "The Public Domain Shakespeare." Paper delivered at the Modern Language Association (New York, 29 December 1992). Available at http://library.utoronto.ca/www/utel/ret/mla1292.html. [Back]

[14]Note that a scheme for line numbering in the electronic world of freely flowing text will need to be agreed upon; see the discussion of this issue in my article in EMLS cited in footnote 2. [Back]