A Shakespearean in the Electronic Study
 
                       Hardy M. Cook
                  Bowie State University
                      Bowie, Maryland
 
     I clearly recall the last paper I wrote on a typewriter: I
remember especially the sheer physical effort of typing those
thirty-eight pages from my hand-written, cut-and-pasted draft and
the Wite-Out all over my fingers and typewriter.  I would type
for forty-five minutes to an hour at a time until I became ex-
hausted; then I would visit computer stores, muttering under my
breath, "Never again."  Indeed, my writing methods have changed
significantly since 1983 when I purchased my first personal com-
puter.  As a result of my interest in computing, I am currently
directing a project at my institution to explore ways computers
can be integrated into our composition courses to assist instruc-
tors with their teaching and record keeping and students with
their writing.  During the 1989-1990 academic year, we worked on
a data base to track a student's progress in basic skills through
the composition sequence.  For several years, we have included
word processing and other computer applications in our writing
courses and have ordered authoring software to fill gaps in ex-
isting software and to construct exercises specifically for our
students.  In academic year 1990-1991, we will be setting up a
computer-supported writing facility--a network of twenty comput-
ers that will include not only word processing but also prewrit-
ing and revising software.  In this paper, however, I am con-
cerned with ways that I have incorporated computer applications
into my work as a Shakespearean.
     I purchased my first computer primarily for word processing.
I relished having a spelling checker and later an electronic
thesaurus.  I welcomed the ease with which I could revise my
drafts.  I enjoyed sitting back and letting my printer effort-
lessly churn out corrected copies without the dread of having to
retype a page because I wanted to make major or minor revisions.
     This computer came bundled with WordStar, a spreadsheet, and
a data base.  The first software I bought to supplement these was
ThinkTank, an outlining program.  I recently upgraded to the more
elaborate GrandView and still begin most my writing projects and
oral presentations by working with this outliner.  In the past, I
have used outlining software to keep a record of the complete
credits of The BBC TV Shakespeare.  In this case, treating the
outlining program as a data base enabled me easily to search all
the parts an actor played in the series, to navigate through the
complete list by collapsing and expanding the credits of individ-
ual productions, and to export the credits to an ASCII text file
for inclusion in my WordStar-generated documents.
     In addition, I find editing and proofreading software help-
ful.  I began with the primitive Punctuation & Style.  This pro-
gram includes Cleanup, which searches for mechanical problems,
and Phrase, which examines diction and usage.  I eventually re-
placed Punctuation & Style with Grammatik because I found it more
comprehensive and easier to use.  Recently, I have discovered
Editor, a program soon to be marketed by the Modern Language
Association.  The philosophy of Editor differs markedly from that
of Grammatik and of its major competitor RightWriter.  These two
programs are conceived primarily for the business writer and are
designed to promote rapid changes and to favor the plain style of
writing.  Editor provides no quick fixes: it "can find more than
fifteen thousand common writing problems in more than forty cate-
gories" and is intended to encourage hard copy engagement with
one's draft, "offering some of the expert attention that is char-
acteristic of a good teacher or editor marking up a manuscript
for improvement or publication."
     One begins Editor by running a utility program that numbers
the lines in the text, either by creating another file on disk or
by sending a copy of the file with line numbers to the printer.
One next invokes the Usage program to generate a list of sugges-
tions keyed to the numbered lines, again with the option of
creating a file on disk or sending the output to a printer.  The
Usage program's dictionaries analyze diction, punctuation, me-
chanics, wordiness, trite expressions, cliches, slang, jargon,
colloquialisms, awkward expressions, and commonly misused words.
The writer then goes through the list making appropriate changes
on the line-numbered draft before transferring those changes to
the original draft.  The authors of Editor firmly believe editing
is done more accurately on paper than on a computer screen, and
their program encourages that approach 1 [". . . the Editor programs
are not 'interactive' with the user's text--that is, they do not
allow immediate, on-screen corrections--because studies indicate
that a computer screen is not the best environment for thoughtful
revision of one's work.  In keeping with recent discoveries about
writing and word processing (Collier, 1983; Haas, 1988), we want
writers to use printed copies of their texts for markup and
revision."  (Elaine and John Thiesmeyer, Editor manual, page 4.)],
although one may, if one wishes, revise the files on screen with
a word processor that supports split-screen editing.
     When I want to study the differences between two versions of
a paper, I execute CompareRite, which produces a redlined docu-
ment that shows all variations between the two copies being com-
pared.  With this program, one controls the appearance of addi-
tions and deletions in the redlined text--the printout or screen
format--and the comparison style--the way CompareRite organizes
those changes.
     In my work, I also rely on Bibliography Generator.  I have
found that this inexpensive bibliographic program adequately
meets my present needs.  With it I can set up a variety of bibli-
ographies and have them automatically formatted in the MLA style.
I can make a selected bibliography from a master bibliography or
combine two or more bibliographies into a larger one.  I can
print bibliographies from the program itself, or I can export a
bibliography to an ASCII text file and edit it with WordStar.
Bibliography Generator can append annotations to entries in a
bibliography, but it does not have the more sophisticated feature
of Pro-Cite, which allows multilevel sorting, multiple automatic
formats, and complex searching.
     When I am writing, I frequently take advantage of WordStar's
Windows function.  With it, I can view two documents at once,
easily switching between them.  I also can copy from one file to
another or simultaneously edit two parts of the same document.
However, when I want to look at a file on another subdirectory or
when I know what text I would like to examine but I do not know
the file that contains that text or even where that file is lo-
cated on my hard disk, I run Gofer, a memory-resident program.
Without leaving the document I am working on, I invoke Gofer.  At
the entry screen, I can type key words or phrases into as many as
eight fields, using Boolean Logic (and, or, not, nearby) search
criteria.  After I have selected the drive, path, and files to
search and have indicated whether the results should be viewed on
screen, written to a disk file, or sent to a printer, Gofer be-
gins searching.  I usually view the results on screen; then I can
decide whether I want to mark the text to be exported to the
document I am preparing or to a file or printer.  Besides my
papers, I keep many notes I have taken from my reading on my hard
disk.  Thus, with Gofer, I effortlessly can include in my writ-
ings direct quotations or summaries from my sources without hav-
ing to retype them.
     Recently I have discovered an alternative to typing out
notes from my reading; now I scan selections from journals and
books and convert them to text files, using a scanner and text
recognition software.  I chose to buy an inexpensive hand-held
scanner, ScanMan Plus, even though I have seen advertisements for
full-page scanners in the $500 range.  I originally thought that
I would have more flexibility with a hand-held scanner, believing
that I could scan books and journals directly without having to
photocopy them first as one must do with a full-page scanner.  I
must report, however, that I have not yet developed techniques
that enable me acceptably to scan directly from books and jour-
nals.  As my description of the procedure below will indicate, I
am also slightly disappointed in the software I selected to con-
vert the scanned images to text files, Read-It O.C.R.
     Before using optical character recognition (OCR) software,
one has to run the scanner's software, scan the text, and save
the text in the TIFF format, one of many digital formats for
graphics files.  One scanner parameter of importance to the ease
and accuracy with which OCR software operates involves resolu-
tion.  ScanMan Plus scans at 100, 200, 300, and 400 dots per inch
(dpi).  Generally, one selects the scanner's highest resolution
for OCR applications.  Unfortunately, the higher the resolution
the smaller the total surface area one can scan with a hand-held
scanner.  I scan at 300 dpi for maximum text input (350 to 400
words per scan) and at 400 dpi for shorter text scans.  For OCR
software to work most efficiently, lines of text must be scanned
evenly: skewed lines cannot be recognized.  Before scanning, one
must set the width and length of the scan.  With texts printed in
columns, like PMLA, I set the width to 3.25" and scan vertically
(the length will vary depending on how much text one wishes to
include).  With texts that are in a single column, like Shake-
speare Quarterly, I set the length to 5" and scan horizontally
(here, the width will vary).  Before saving scanned text as a
graphics image, one must use the scanner's software to cleanup
the image as much as possible.  Such cleanup might include eras-
ing unwanted text or rotating horizontally scanned text.
     After one has a cleanly scanned image saved to a graphics
file, the OCR software can begin to process it.  The image pro-
duced by the scanner consists of a pattern of pixels (dots).  OCR
software compares the shapes of these dots to the shapes of the
letters of the various typefaces stored in type tables.  Because
Read-It O.C.R. is "trainable," it can be taught to recognize al-
most any typeface.  Although the program contains a few type
tables of popular fonts, one still must, because of the many
variables involved, teach the software to identify each new type-
face, that is, build a type table for each typeface.  To build a
type table, one first chooses an existing type table to adapt or
runs one of the learn modes to create the initial table, after
which one executes the recognize and learn command.  In this
mode, as it processes the graphics image into an ASCII text file,
Read-It O.C.R. will stop when it cannot find a good match for a
character; one then has manually to enter the correct character
from the keyboard.  This learning procedure is invoked until the
type table is so refined that it achieves a nearly 100% recogni-
tion accuracy.  The process of building and refining type tables
is a time consuming one, but once one has a well-refined table,
processing to an ASCII text file only takes a few minutes.  Since
I bought Read-It O.C.R., I have discovered another OCR program,
Omni-Page, that does not require building type tables; Omni-Page,
however, costs roughly three times as much as Read-It O.C.R.
     I also transfer text to my hard disk from my Radio Shack
Model 100 laptop computer through a null modem cable.  The Model
100 is the original laptop computer2 [The Model 100 was superseded
by the Model 102, a slightly smaller but in other ways virtually
identical computer.  In the past few years, however, the Model 102
has been eclipsed by MD-DOS compatible portable computers.].  It
weighs about four pounds, operates for considerably more than its
estimated twenty hours on one set of four AA alkaline batteries,
and can easily be carried in one's briefcase.  The Model 100 does
have many limitations: for example, its maximum RAM is 32K, and
its LCD screen only displays eight lines of forty characters.
To overcome some of the limitations of the built-in word processing
program, I added a ROM chip that includes Write ROM, a text processing
program.  This program gives me greater control over my text output
and even has a page preview function.
     I have employed the Model 100 in many ways.  When I was
writing on The BBC TV Shakespeare series, for example, I would
take notes on the productions with the Model 100, which creates
ASCII text files.  Normally, I would sit in my living room or in
the media room of the library with the Model 100.  As I watched a
production, I would pause the video recorder frequently to take
notes.  Although the 32K RAM capacity restricts the maximum size
of a text file, I always had sufficient capacity for my ten to
twelve page production notes.  When I finished the notes on a
production, I would connect the null modem cable from a serial
port (communications port) of my desktop computer to the serial
port of the Model 100; using the built-in modem in the Model 100
and the Write ROM program, I would send my text to the telecommu-
nication software, Mirror, on my hard disk.  Then, I simply edit-
ed my notes in WordStar.
     With a null modem cable, I did not need a modem with my
desktop computer to receive files from my Model 100, but when I
did purchase one for my computer, I discovered several uses for
it in my work.  With a modem and telecommunications software, I
can access information services like Compuserve.  I can send and
receive E-Mail at my university Internet address.  I have even
joined Humanist, an international electronic discussion group for
computing humanists.  I have not yet accessed the MLA Bibliogra-
phy on the DIALOG Information Service, but I regularly connect
with academic library catalogues through Internet.  If I want to
know, for example, the publication information for a book mentioned
in my reading, I am less than a minute away from an online catalog
and the answer to my query.
     Finally, I regularly run WordCruncher with the Riverside
Shakespeare.  When I am reading, I often check references and
frequency distribution claims3 [For example, the word "blood"
appears more times in King John than in any other of Shakespeare's
plays.]; I also look up quotations in context and search for echoes
of them in other of Shakespeare's works.  In addition, I export
quotations to ASCII files to include them in my writing or in
handouts or tests for my students.  I realize that I now am only
taking advantage of a small portion of this program's capabilities
and intend to pursue other applications of it.
     I began this paper by claiming that my writing methods have
changed significantly since I began computing.  At my paper's
end, I realize that I oversimplified the impact the personal
computer has had on the way I work in general.  Not only do I
have writing and editing tools that I could not have imagined a
dozen years ago, but I also have far greater and faster access to
information than scholars of a generation ago.  I eagerly await
the new technologies and applications on the horizon.  Now, I am
trying to figure when I can afford a CD-ROM drive, yet I know
that as soon as I have a CD-ROM drive I will begin considering
when I will be able to purchase the Oxford English Dictionary on
CD; then there is the MLA Bibliography, and so on, and so on.
 
Hardy M. Cook
Bowie State University
HMCOOK@BOE.TOWSON.EDU
 
 
___________________________________________________________________
The contents of this electronic file are copyright (c)1990
Hardy M. Cook, Bowie State University.  Quotation for scholarly
(non-commercial) purposes is permitted, but please contact the
author  to verify the material in question
and advise him of your intention.  Please do NOT distribute.