Many schools have begun using microcomputers for classroom instruction, as opposed to simply providing them for individual work in a lab setting. Planning such instructional environments forces us to examine our beliefs about how students can improve as writers, how we like to work as teachers, how we use classroom time appropriately, and how we can best structure and manage learning activities.

Many labs have been set up by the folks at the computer center, who typically see the labs as places for students to drop by and do work for classes. And though it always turns out that the major use of campus microlabs is word processing, this fact is only admitted reluctantly and has little to do with planning. Usually there is no explicit consideration of how the space will be used--the question is, "How many computers will fit in this room?" So on our campus, for instance, we have an IBM PC lab in the old band room with about 65 networked IBM and cloned microcomputers tied to an assortment of dot matrix and laser printers. In the old choir practice room, about 30 Macintoshes and a few other assorted computers share the available space on two levels, on raised platforms and down on the floor. These are jumbled spaces, fine for dropping in and working on a paper, but not at all useful for group instruction.

What many of us have discovered, however, is that we want microcomputer classrooms for instruction. We want to use the machines for instructional intervention, opening up the writing processes of our students, so that we are not witnesses after the fact of writing, but collaborators during the process. Mary Peterson, of the University of New Hampshire, underscores the power of microcomputer classrooms:

Computers bring the process of writing right into the classroom; they're like a window on the work itself. This is the only writing class I've ever taught in which students have said, "Let's just write today." (M. Peterson, personal communication, Oct. 1988)

Teaching in a microcomputer classroom allows us to open this "window on the work itself." But as many of us have discovered, the design of the space has much to do with how comfortable we feel as teachers, how far we can encourage collaborative exchange, and the extent to which we can realize our instructional goals.

In this article, I would like to raise what I see as the major issues associated with designing ideal university microcomputer classrooms (which I will call labs out of convenience). I rely here not only on my experience, but also on personal correspondence--telephone calls, letters, and electronic mail--from colleagues, like Mary Peterson, on various campuses who have been gracious enough to respond to my requests for advice. Cynthia Selfe, coeditor of Computers and Composition, was the first person I called, and she was quick to remind me that the point is not having the best hardware or the ideal arrangement. The question is: "What are your goals? How do you believe writing should be taught?" (C. Selfe, personal communication, Oct. 1988) Without this initial instructional analysis, it is impossible to design a classroom. Selfe's reminder provides a basis for the issues discussed here: how to plan a lab, select space, arrange the space, select hardware and software, and establish funding. Excellent articles by Barker (1986), Schwartz (1987), and Selfe (1987) also provide information on setting up microcomputer classrooms. These articles, like the following discussion, are intended to be useful to English faculty who are planning microcomputer classrooms or redesigning existing instructional spaces.

While it has been tempting to establish spaces devoted to writing instruction and controlled by English departments, it seems much better to rely on established expertise, typically found in the staffs of university computer centers. These colleagues are accustomed to providing maintenance and support, and they often have budgets for computer supplies, at least for mainframe systems. With early involvement and committee-based planning, the services (and budgets) of these colleagues can be enlisted in planning and staffing a microlab. The computer center often has technicians who can fix disk drives and bright student "hackers" who can solve network problems or spot and cure "viruses." Workstudy funding can support student staff at low cost.

Working with other campus units always means arguing, explaining, and compromising. Lee McKenzie, of Weber State College, Ogden, Utah, spoke metaphorically of the marriage contract between English and other departments, pointing out the advantages of sharing space with other academic units, in her case, social sciences. Like any marriage, such unions have rough spots that need to be smoothed over, but the marriage is generally worth saving. (L. McKenzie, personal communication, 1988) Microcomputer classrooms and labs create communities (open marriages?) of students and faculty; they foster collaboration. The cross-disciplinary collaboration we seek through writing-across-the-curriculum efforts can be enhanced through public spaces where writers from various disciplines work in full view of each other.

Working together to establish a lab can help other campus units appreciate the English department's commitment to student literacy. When Bruce Appleby and I worked to set up labs at Southern Illinois University at Carbondale, we worked with Computing Affairs and with representatives from departments without large labs (defined as more than 20 machines). Thus, we brought computer resources to departments without traditionally large equipment budgets. We played up the campus-wide value of strong writing skills coupled with computer skills, and we found our strongest supporters in the engineering faculty, people who wanted to see both skills in their students.

I am now working with a cross-disciplinary committee at New Mexico State University to establish a Publishing Design Studio, a workshop space for advanced students in professional writing, graphic design, and theater. We are anxious to provide students with powerful machines for desktop publishing, advanced graphics, hypermedia, and computer-aided design. We have worked with our computer center to plan and seek funding, assuming all along that collaborative use of resources would increase our chances of achieving our goals, but also seeing real value in bringing together advanced students from several disciplines in a teaching/workshop setting.

Deciding where to put a microclassroom is an early, important planning issue. The initial decision about space can have a huge effect on the overall cost and can influence satisfaction with the new classroom.

At Southern Illinois University at Carbondale (SIUC), we decided to convert a row of classrooms. Remodeling was expensive--new walls, ceilings, lighting, improved wiring, an independent cooling system, window shades, alarms on the doors and windows, a central desk, and a repair shop all added up to more than twice the price of the machines. The lab is a showplace, though, and campus visitors are always trotted through. Floor-to-ceiling windows on the long side of the lab look out on the mature woods that give the campus a living heart. A few large plants suggest modem office space. It feels good to be in the lab, and that's important.

A large, unused basement space was chosen for a second lab because it could easily be converted without extensive remodeling. Wiring and walls were much cheaper when installed in undeveloped space than when remodeled in existing rooms. The savings went toward carpeting and comfortable wood furniture. My point is this: Choose space carefully because remodeling is very expensive. Rather than spending money moving walls, how much more satisfying it is to spend limited funds on computers and printers, or on design features and furnishings that make the lab an inviting environment.

Because we worked with other faculty in designing the labs at SIUC, we decided to have 32 micros to accommodate average class sizes in other departments. This many computers take a large space, especially when desktop workspace is figured in for large machines (e.g., IBM PCs). We would certainly have preferred a smaller, more intimate space, with 20 to 25 machines, but this was an issue of compromise in our marriage. It meant we found ourselves teaching in a very large space, which called for voice projection none of us were accustomed to and left us with a feeling of being far from our students.

When planning space and remodeling, it is important to remember that both machines and people generate heat, so independent cooling systems are important (and costly). Fred Kemp, now of Texas Tech University, notes the difficulty of cooling a basement space at The University of Texas at Austin that was not designed as a lab:

In order to keep the room cool, we ran the air conditioning all night on full blast. In the morning, the room was like an icebox, but by the early afternoon, it would begin to heat up badly. The monitors gave off the most heat, and we turned them off when not in use. We have used a window fan in the doorway to keep the room bearable, which it does, but the noise adds to the room's already considerable noise level. (F. Kemp, personal communication, Nov. 1988)

Once again, I would argue that an English department's best alliance is with other units on campus, especially with those experts who know from experience how to keep machines happy and how much it costs.

People who compose on computers spend a long time at it, and comfortable, well-designed furnishings are important. Desktops should be 26 to 27 inches high, like a typing table, not 29 to 30 inches, like most tables. Chairs should provide back support, as secretarial chairs do. If the chairs have coasters, all the better, because people are always moving around to work together. Extra chairs can allow students to work together at single stations.

Ambient light should allow reading of books or notes, but not reflect in the monitor screens. Windows need some kind of treatment to prevent glare, especially early and late in the day. Computer rooms are noisy places, not particularly conducive to the concentration that writing demands; sound-absorbing ceiling tile, carpet, and drapes can help. Coat racks and bookshelves can conserve precious workspace and keep bulky winter coats from preventing movement around the lab.

Everyone who has taught in a lab setting has strong preferences for room layout. What someone considers best depends to a large extent on beliefs about how composition is best taught, and these beliefs vary from teacher to teacher. Although I cannot recommend a single layout as best, I can review the various possibilities, note the advantages and disadvantages of each, and perhaps help readers consider the relevant issues and form their own ideas.

The Traditional Classroom Grid
The most obvious layout is a simple row-and-column grid based on the traditional arrangement of desks in schools. The space is a familiar one, a default setting for many people. It places the teacher at front and center, directs the students' attention toward either their screens or the front of the room, and places students next to each other for consultation. Teachers can move laterally across the rows to work with individual students. This traditional layout can use long tables (which are cheaper than individual stations), and the wiring can be run along the backs of the tables from station to station, nicely concealed within the vanity panels. Four to six stations can share the same printer through local cabling to a switch box.

The major drawback to the grid is that it places the computer between the teacher and the student. This creates a physical barrier, one that can be quite formidable with large computers with large CPU's, or monitors. Students must crane their necks to see the teacher, and the teacher ends up on tiptoes, trying vainly to maintain the eye contact that is so critical to holding attention.

Even more limiting is the psychological barrier--there is no underestimating the attention draw of the glowing screen. Talking heads in the front of the room are simply no competition. Fred Kemp describes this screen magnetism well:

The computer-based classroom is the hardest for traditional instructors to swallow, for it strikes hard at the comforting tyranny of the traditional classroom . Computers soak up interest; they are compelling devices. The usual pontificating instructor doesn't have a chance trying to orate against a room full of computers. The students are far too interested in what is happening right there in front of them to suspend their interest and stare at somebody talking in front of the classroom (F. Kemp, personal communication, Nov. 1988).

Teachers can cope in this setting--by speaking in loud, demanding tones or insisting in the most teacherly fashion on attention, please. Some assertive teachers will insist on eye contact or on students turning off the monitors during discussion. But these are essentially coping strategies dictated by a classroom design that places an attention draw between students and teachers.

A second drawback to the grid is its inefficient use of open space. The rows must be widely spaced if the teacher or students are to move around freely without encroaching on each other's space. The grid distributes all open space equally, which is fine for individual work, but not good for collaborative activity. The space in even a large room is quickly consumed because so much of it is taken up by dead space at the rear of the tables and between machines.

Facing Rows
Arranging the computers in facing rows, perpendicular to the front of the room, makes a somewhat more efficient use of space than traditional rows and columns. Long tables can accommodate machines facing in from both sides. Aisle space is conserved, and the wiring can be dropped from the ceiling or run behind the rows. While some students may work facing the walls, others will see faces when they look up. The arrangement closes the distance between writers and encourages interchange. Students can turn to look sideways toward the front station, and the teacher's sightlines down the facing rows are better than in the traditional arrangement. It would seem an arrangement better suited to the teacher's gaining and holding attention, though Jay Funston of James Madison University finds it difficult to get student attention and believes facing students head-on, in the traditional grid, is preferable. (J. Funston, personal communication, Oct. 1988) Perhaps neither arrangement particularly facilitates full group interaction.

Clusters
Clusters of three or four computers arranged in the available space (as islands) make a lot of sense. The arrangement itself suggests that writing is collaborative. Students look at other students, not the back of someone's head. And because open space is usefully conserved, movement is unrestricted. While it is easy to get next to other writers, a sense of privacy is preserved.

The hardware in the middle of the group does obstruct sight lines and isolate group members. Tommy Barker, of Texas Tech University, partially solved this problem by building a lab with IBM System 25s, a machine similar in size to the Macintosh. Small machines allow people to see each other and the teacher. Barker likes the cluster arrangement because he can move about in the middle of the classroom, and because students can look toward him from various angles when he needs to address them. He also believes that the clusters allow students to choose a space comfortable to them, out in the public middle or in a private side or corner. (T. Barker, personal communication, Nov. 1988)

Full group instruction is definitely not facilitated by this arrangement, but that may not be all bad. The arrangement puts the emphasis on workshopping. Clusters also work very well as drop-in labs. They are visually interesting and provide for something all designers of good public space know: that people like to look at other people, especially from a position of privacy. The intervening islands of equipment allow for this. Imagine, if you would, why restaurants, with their clusters of diners, are pleasant places, feeling both private and public.

Around the Perimeter
A fourth arrangement, one that I've seen work very well, places the computers around the perimeter of the room, so that students work facing the walls. The advantages are several. Best of all, the barrier between teacher and students is gone--if the teacher needs to talk to everybody, the students can simply turn their seats around and face center. Computers have a strong draw, but turning students 180 degrees is an effective counter. A circle is the ideal discussion seating, one that democratizes discussion by giving everyone equal status. It is a group-forming structure, because students can speak directly to other students, all of whom have faces, and see the effects of their comments. Donna Singleton, who has taught in four different layouts at Southern Illinois University at Edwardsville, strongly prefers their U-shaped classroom, with computers on three walls and work tables along the fourth wall of windows. What is important here is the fit between the layout she prefers and her own sure sense of what should happen in classrooms:

I am very much in favor of student-centered classrooms, a workshop environment, classes devoted to composing, and group activities at the computer. Some of what makes my classes "work" has to do with a group sense, an inexplicable "gelling" that comes from talking and sharing informally. I have taught in a crowded, cluttered basement lab and a long, narrow room with the machines on the outside walls. But the classes in the U-shaped room always seemed to "gel" better and quicker, and the students seemed to have more fun and get to know one another better. I looked forward to those classes. In the other rooms, there is no place for a short person like me to stand or sit and carry on a discussion. In the U-shaped lab, the students can turn away from the machines and it's like talking in a lounge or something, very informal and relaxed. (D. Singleton, personal communication, Oct. 1988)

I have seen this peripheral arrangement work equally well, whether in a second-grade class or in a college class.

The arrangement consolidates open space, rather than dispersing it, so there is actually room to move and work in the middle of the room. The arrangement leaves room for small tables in the center, or work tables along one wall (and if they are on a window wall, all the better to encourage both vision and reflection). Writing workshops need two sorts of space, as Mary Peterson points out: "I have found that in teaching writing on computers one needs private space and group space, and it's nice to be able to move easily between the two." (M. Peterson, personal communication, Oct. 1988)

A major advantage of the peripheral arrangement is that the teacher can lead a group in coordinated activities. The leader's voice is projected from the center of the room, equidistant from each station, so there is no need to shout. The leader can immediately see when someone gets off track and help that person recover with a keystroke or two. A leader who can't see the screens is at a disadvantage, able to give directions but not able to confirm that people are on task.

One limitation to the arrangement is that individual writers face the wall when they look up during composing, and the walls are often very close. Does this close visual horizon limit the creative, idea-generating heuristic known to us all--staring into space? A second limitation is that the arrangement doesn't naturally structure students into groups; it strings them out and gives each person a neighbor or two. But the flexible, open space of the rest of the room can counter that difficulty.

Computers
Choosing hardware configurations is admittedly difficult. The choices are always constrained by budget and often shaped by individual familiarity with and preference for one system or another. There's no way to say, "Here is the best machine." The questions center on your purposes for using computers. What do you want to do? What are your program goals? Are you satisfied with simple word processing, or do you want graphics, color output, or powerful applications, such as hypertext? I will offer a few comments on the leading contenders--Apple Macintoshes and IBM PCs, with clones included.

When given a choice, students choose to use Macintoshes over DOS machines. Macintosh software is easy to learn because of the consistent interface of pull-down menus and mouse-executed commands. The graphics integration and font control allow students control of page layout and visual aspects of text, increasingly important in all kinds of writing, but especially in technical and business writing. With the exception of the powerful Macintosh IIs, the machines have a small footprint--they take up comparatively little space--and they are quiet, too.

Interesting courseware is beginning to show up from Apple's higher education grant programs, with various writing, instructional, and course management software now being marketed through Kinko's Copies. Apple's HYPERCARD is attracting a lot of attention in higher education; it is a promising environment for courseware development. (IBM's version of HYPERCARD, LINKWAY, will probably have been released by the time this article is published.) HYPERCARD is difficult to describe. It is software that is bundled with all new Macintoshes and intended to be a general desktop environment, combining such functions as calendars, address books, telephone dialers, and a host of other applications. It attempts to provide a freely structured, integrated text and graphics database that allows free and fast movement around and within large groups of information. It offers ways for users to organize a wide variety of information in nonlinear, freely associated stacks, using self-designed links that allow movement from one piece of information to another. It also includes drawing and word-processing tools and a scripting language for designing specialized applications. However, HYPERCARD certainly ups the ante on machine requirements, because the only reasonable way to run it is on a hard drive.

In our Publishing Design Studio, we have chosen Mac IIs as our main machines because we will be doing work that demands powerful, well-integrated text and graphics: computer documentation; hypertext; graphic design; and computer-aided set, lighting, and costume design. Similarly, Gail Hawisher notes that the English Department at Illinois State University has chosen Macintoshes for its advanced writing classes. (They use IBM-compatible Zeniths for introductory composition.) Networked Macintoshes intended uses--technical and business writing, graphics, HYPERCARD applications, electronic mail--make them a good choice. (G. Hawisher, personal communication, Dec. 1988)

IBM PCs and their clones continue to present the conservative choice, with their wide market share in business and their attractive prices, especially with the steep educational discounts offered by Zenith and other companies. The IBM DOS--Disk Operating System--is the closest thing to a standard in the industry, so if students become used to working with these machines, they might have some advantage in the business world. DOS is the housekeeping program: It determines what you see when you start up a computer and how you communicate with the machine, its disk drives, and its memory. In other words, DOS is a command-driven interface: It presents the user with a simple prompt (A>), and the user enters commands such as DIR to see a directory of files or REN to rename a file. Such an interface is not particularly friendly to a new user because the screen presents only the prompt, requiring that the user know commands to work the machine. It is, though, a simple and fast way for an advanced user to work a computer.

In contrast to DOS is the Macintosh, or Windows, interface from MicroSoft. This interface presents pull-down menus and icons to the user. A mouse, a hand-size control separate from the keyboard, is used to select options from menus lined up across the top of the screen. Icons representing disks, programs, drawing tools, or whatever can be manipulated either with a click of the mouse or with keystroke combinations. Because the Windows interface exploits visual cues and because the various Macintosh programs use consistent menus, icons, and selection procedures, learning one application can carry over to learning another.

The Windows interface has proven attractive enough that IBM adopted it for its new line of computers, the PS/2 series. The low end of the new IBM line, the System 25, is attractive in some ways because it is cheap, compact, quiet, and it gives off little heat. The machine does not represent the new IBM architecture--it is essentially an IBM PC in a new package, but it does mimic some of the desirable features of the Macintosh interface. The dominance of DOS does not appear threatened, yet, by the promise of the new operating system and interface. Several years down the road, I assume we will be better able to evaluate the usefulness of IBM's new operating system, but most of the current advantages are potential, unrealized in software. Everything points to a convergence in interface design; the current ease-of-learning-and-use that pleases Macintosh users is increasingly part of IBM's systems.

Choices have to be made with respect to disk drives and memory. In terms of disk drives, 3 1/2" formats are becoming increasingly dominant because they store more data in a better designed package. During the transition, it makes sense to have at least a few machines equipped with both 3 1/2" and 5 1/4" drives. No matter what the choice, sufficient memory is important, which right now means 64OK RAM. The scramble among software developers to provide the most powerful features means that word processors now routinely require somewhere between 320K and 540K to run their spell checkers, produce fonts, and design pages. The evolution of word-processing software toward desktop-publishing software suggests that memory demands will only increase, making high-density disks and hard drives standard equipment.

Printers
Computers in labs can share dot matrix printers, using inexpensive switchboxes to cable several PCs to a single printer, preferably fewer than four PCs. However, the arrangement usually leads to bottlenecks in instruction, especially at the end of the hour. Laser printers are an attractive choice: Their output looks so good that it invariably increases the pride students take in their work and the energy they are willing to invest in page design. A single laser printer can quietly serve a large lab. However, a second laser printer can alleviate bottlenecks as well as offer a backup--printers do break down. Dot matrix printers can usefully produce drafts, with laser output saved for final products. No matter what the choice, keeping printers running, inked, and filled with paper is expensive. Computers increase the use of paper considerably, especially among novice writers who need to see frequent drafts on paper and who have a hard time getting the printers to do what they want them to do.

Large Screen Displays
Technological advancements are providing better products for large screen display than have previously been available. Earlier projection techniques--such as Sony's three-color projection on large, curved screens--simply did not provide the necessary resolution for displayed text, were not bright enough to be seen, and required straight-on viewing. Kodak and Sharp both have introduced reasonably priced ($1,000 $1,500) LED display plates that connect to a graphics monitor and, when placed on an overhead projector, project what appears on the monitor to the overhead screen. These classroom displays produce a decent image and can be very useful for demonstration learning. Before sending in a check for one of these devices, have the vendor demonstrate it in your lab, under your light conditions, from the vantage points of your seated students.

Networks
In a lab of standard microcomputers configured without hard drives, a network demands a powerful central machine, a network server. In this central machine, power is important--at least a 20M hard disk and 2-4 megabytes of RAM. Our server in the IBM lab uses 60 megabytes to coordinate software and printing for about 60 machines. Legal use of software requires that programs without site licenses be used on the server in multiple copies, one copy for each person who will be simultaneously using the software; if a class of 24 students all use the software at the same time, 24 copies should be on the machine. This software storage eats memory fast, especially for programs like VENTURA PUBLISHER.

The server can direct file traffic in the lab, send programs and data files to and from individual stations, and control the flow of files to the laser printer. It can hold course assignments, syllabi, notes to the class, and various records. It can provide mailboxes so students can submit files for review and pick them up with comments from either the teacher or other students. And it can keep track of lab use, recording check-in and check-outs.

Networks are not cheap. They require the powerful network server (a computer dedicated solely to the network), cards and cables, software to run the network, and file-sharing software to run whatever instructional applications are specific to writing classes. IBM PCs require an installed card ($300-$400 per PC). Although networks are cheaper and easier to install on Macintoshes (an AppleTalk connector costs about $40), all networks are complicated and call for advanced expertise. But they open wonderful possibilities for electronic organization of classrooms.

One alternative to a network deserves mention. Simple, cheap video switching systems are available that let instructors show files on selected monitors. These are not networks, because there is no real file sharing. Rather, a screen image--the teacher's or one of the student's--can be shown on all or selected screens. Thus, a student's text can be viewed by the whole class or by a small group, who can make suggestions orally for revision and watch as the writer makes changes. The viewers can only see the text. They cannot make changes because all they see is an image of the writer's screen--the actual file does not exist in their RAM. But the system can work well for setting up several small groups simultaneously or for conducting full-class peer editing. Such systems demand only a simple switch box and video cable and are easy to operate.

The centerpiece for writing classes is certainly word-processing software. Two programs clearly dominate the market. WORDPERFECT and WORD (in that order). Both are very powerful; each tends to leapfrog the other in each new version. But many other programs serve perfectly well, depending on what teachers and students need to do. More powerful programs will allow editing in columns, drawing lines and boxes, using different fonts, outlining, importing graphics, inserting hidden text that does not print but can be read for editorial purposes, or inserting revision marks that don't change the original text but allow the writer to accept or reject editorial suggestions. The site licenses available for most university campuses can be a real relief from having to worry about illegal copying or having to require that every student buy an expensive program. Shareware is another option. Such programs as PC-WRITE still represent good value, though what the author does not make on the program is made up in the price increases on the manual.

It is important to use full-fledged programs. Some publishers offer low-priced educational versions of popular programs. These educational versions have often been crippled to restrict file length, or print obnoxious little messages on each page, or offer only a subset of the full program's options. Student writers may begin with low end demands, but they deserve programs that grow with them as they discover increasingly sophisticated uses of word processing.

Beyond word processing, writing classrooms can benefit most from standard applications software. Most word processors now come with integrated spell checkers and thesauri, both of which have good uses in writing classes. Many have outliners; a few even work well enough to be worth learning. Students should also be encouraged to use drawing and charting programs to incorporate visuals in their texts. The so-called grammar checkers continue to get mixed reviews, especially for grammatically naive students. It is possible to structure good lessons on style around text analyzers that offer counts of passives, "to be" verbs, nominalizations, and prepositional phrases. But simply turning students loose or requiring them to use such programs as GRAMMATIK is probably ill-advised. Generating text-analysis data and stock advice is easy; interpreting and applying the information is tricky.

Good use continues to be made of such writing process software as Wresch's WRITER'S HELPER: STAGE II from Conduit. The program offers intelligent support for writing and, quite importantly, does not attempt to replace the word processor, but passes files to and from the chosen word processor at key points in the process. In a networked environment, one software option specifically designed for writing classes is the Daedalus Instructional System, developed at The University of Texas at Austin. [1] Another writer's environment, the Electronic Notebook, has recently been developed at the University of Southern California under Project Jefferson and is available through Kinko's. This HYPERCARD application includes facilities for class management, informational databases, notetaking, and writing. [2]

Most people I know who build good writing microlabs are hustlers. They look everywhere for money, ceaselessly hitting up vendors, local businesses, and employers of the university's graduates. They work on-campus through computer committees and make sure they know what's going on in computing affairs so they can have some say. They write often fruitless grant applications, whether for lots of equipment or a piece or two, casually tied to some research project. They squeeze the English department budget, badger the department chair and college dean, and talk whenever they can with university vice presidents or presidents, all of whom have at least some funds, even in this perennial night of state underfunding of higher education.

Sometimes, the lab builders connect with federal funding, often through entitlement programs aimed at enhancing minority success in higher education. Federal (especially National Science Foundation and Department of Education) and state grant programs for retraining of public school teachers can bring in equipment money. Sometimes special state funds become available for new or expanded university programs. The existence of the funds is often known only to university higher-ups, making it imperative that the lab builder stay in touch with the administrators and enlist their support.

Illinois State University (ISU), the only large university to fully computerize all its introductory composition classes, used start-up money from the computer center budget and computer-use-fees to build ten computer classrooms. A $30 course fee from each composition student (approximately 5,000 students per year) provides a sizable yearly budget. (C. Harris, personal communication, Oct. 1988) What's more, most students continue to buy computer-user cards in subsequent semesters because they get hooked on the machines in their composition courses. The recent decision at ISU is simply to have all students pay the fee each semester and not tie it to any particular courses.

One of the more unusual funding sources came our way at Southern Illinois University at Carbondale when money that had been embezzled by a former computing affairs director was ordered repaid by the courts. The new director wanted to spend the money in a highly visible way--to return it to the students-- and our microlabs were born.

Designing and building microcomputer classrooms is a lot of work. It takes at least one highly dedicated writing teacher, who sees the potential for improved student writing and improved classroom instruction. A team approach is better. The frustrations are many, and the traditional reward structures of the university (read "promotion, tenure, and merit pay") offer too little incentive to sustain an individual. The rewards come from seeing results as fruitful instruction takes place in new settings. But it takes a long time to get there, countless hours of planning, hustling, and problem solving. A team is much better equipped than an individual to work through the difficulties inherent in the process. The incentive to work toward such instructional environments comes from a sense of the computer's role in all our futures. These machines are and will continue to be the medium of communication for our students. The computer offers important new ways to work with students on their writing, and exploring new ways of teaching can keep us professionally alive. This sense of being explorers and innovators sustains those of us dedicated to realizing the potential of the microcomputer in the writing class.

Stephen A. Bernhardt is an Associate Professor of English at New Mexico State University in Las Cruces, New Mexico.

1. For information, write to Fred Kemp, Dept. of English, Texas Tech University, Lubbock, TX 79409-4530.
   

2. For a description, see Academic Computing, September 1988.
   

Barker, T. (1986). Guidelines for opening a multipurpose microlab in an English department. Collegiate Microcomputer, 4, 115-122.

McKenzie, L. (1988, October). The union of a writing center with a computer center: What to put in the marriage contract. Paper presented at the meeting of the Rocky Mountain Writing Centers Association Conference, Las Cruces, NM.

Schwartz, H. (1987). Planning and running a computer lab for writing: A survival manual. ADE Bulletin, 86, 43-47.

Selfe, C. (1987). Creating a computer-supported writing lab: Sharing stories and creating vision. Computers and Composition, 4(2), 44-65.