For pt.I see ibid., vol.32, no.2, p.83-93 (1991). The authors explore the two-year Danish CIM/GEMS project in computer-integrated manufacturing (CIM) at the Technical University of Denmark to show how writing is an act of technology and knowledge representation, a vehicle of their transfer to a user community, and, if successful, an accommodation of technology to its users. They address creators and users of system documentation who need documentation for CIM implementation. The authors argue that documentation is often better for representing and explaining a CIM system than the actual system itself, and they recommend that documentation production be viewed not as a separate, end-of-project activity but as an integrated part of technical development. Planned and regular documentation production can in fact be a stimulus and aid to technical development, possibly even shortening the project life cycle.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

The authors describe and classify different hypertext documentation approaches by surveying a number of existing systems. They comprehensively list features of each system but do not numerically compare systems because of the subjective nature of evaluation. Instead, the data provided should be used as a basis for local comparisons by the reader. The classification breaks systems down according to publishing type, i.e. those designed for both publishing and accessing hypertext, either publishing or accessing documentation, and those that use general-purpose hypertext systems for document production and presentation. These categories are further subdivided according to functionality and specific system information.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

It is noted that computer users often do not take advantage of system shortcuts; instead they rely on familiar but less efficient methods for getting their work done. The authors examine a system that incorporates system-initiated online help to advise users of shortcuts for menu navigation. Subjects received either no advice messages or one of three advice message types, varying in timing and specificity, and subjects were assessed on their subsequent use of shortcuts. Results revealed that users who receive system-initiated advice take greater advantage of shortcuts and are able to complete their work more quickly than users who receive no system-initiated advice messages. Further, a majority of users reacted positively to the messages, finding them helpful in introducing shortcuts.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

A three-stage design analysis that can be added to a prewriting strategy to produce a more efficient and portable computer user manual is presented. The three stages are: preliminary analysis, defining who the audience is, what they presently know about the subject, what they need to know, and how they will react to the information presented; needs assessment, a systematic effort that gathers opinions and ideas from a variety of sources regarding performance problems or new systems and technologies; and design for learning, which moves from gathering information to analyzing audience needs, defining learning objectives and designing the manual itself. It is shown that the three-stage process justifies a writer's work by citing specific cost factors and casts the writer in the role of a communication analyst who is capable of finding motivational, environmental, and cost issues in the organization.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Electronic communication systems, such as a fiber-optic and telephone-delivered videotext service a videoconferencing capability, and a fax network, that have been used to improve the speed and quality of communication to 15000 employees dispersed throughout the state of Florida are described. The ways in which foreseeable changes in the state's demographics, the company's business environment, and emerging media technologies will interact to affect the evolution of the company's employee communication are described.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

To discover the similarities and differences between primary and secondary computer manuals, and to account for the popularity of the secondary texts, two best-selling books for word processing and spreadsheet programs are compared to documentation supplied by the manufacturer. A heuristic for analyzing software documentation based on cognitive and rhetorical principles is developed and applied to the corporate documentation for (WordPerfect 5.0) in contrast to Stewart's Using WordPerfect 5 from Que, and the corporate documentation from 'Lotus 1-2-3' in contrast to Gilbert and Williams's 'The ABC's of 1-2-3 from Sybex.' It is shown that the trade texts from Que and Sybex contain more conceptual background information than the corporate documentation and differ in their rhetorical stance: the writers provide a richer context by giving more examples for applying the software; the writers provide global and structural frameworks; the writers use persuasive marketing techniques to ease the reader's anxieties and remind them of the software's benefits; and the writers identify themselves.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

The documentation effort undertaken by the Danish CIM/GEMS (computer-integrated manufacturing: general methods for specific solutions) project to drive, direct, and support its CIM-technology development and transfer activities is described. The project recognized that in the phases of CIM development-from planning through maintenance-the primary deliverable is not the CIM system itself, but the documentation used to describe, specify, justify, and support its development and use. As a result, the project organized itself so that it would transfer to its industry participants the information needed to build CIM systems, and not CIM technology alone. Because this information best resides in well-written and complete documentation, the CIM/GEMS project adopted a management-by-document approach, treating documentation activities as part of system development and assigning professional personnel to the task. As an archive of CIM-system development, the documents captured the project's in-progress work and the knowledge of developers who performed that work. As a deliverable, the documents were the principal means for carrying CIM knowledge and experience off-site.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

The data collected in electronic databases by government agencies, universities, and private organizations could be of use to a public increasingly interested in participating in public policy formation. However, because most users lack skill in communicating with computers and because institutions resist the dissemination of databases, this information is not widely accessible to the public in forms it can use. The authors propose establishing a public access 'data hub' to process requests for database access, download a copy of the requested data from the original source database (databases would not reside at the hub), and display the data in the form requested. The 'bridge' between user and data would be an interface to process the request for data and then shape and interpret the data for the user. A single interface would allow users access to information that resides on many different computers without their having to learn to 'talk' to each different system. Helping to design and implement this interface would be the task of the professional communicator.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

It is pointed out that many organizations have decentralized their operations to respond more quickly to rapid changes in markets and technologies. Decentralization has legitimized crossfunctional communication and decision making at relatively low levels in these organizations as a means of improving the quantity and timeliness of information. However, decentralization can adversely affect an organization's communications unless a new communication system is designed as part of the altered communication process. It is suggested that working within a new approach to organizational change, called sociotechnical design, technical communicators can help define performance standards for new communication systems, including those integrated with computer-mediated information systems. The objectives of sociotechnical design are consistent with the language-action perspective of the modern office. That perspective asserts that language organizes work relationships and actions by means of conversations (or transactions) in which requests and promises are made among individuals and groups. These conversations are structured by procedures, policy guides, handbooks, training materials, and a host of other communications.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

The author describes the design and creation of a task-oriented user guide for students doing computer-assisted writing at San Diego State University. The project incorporated prototyping and task analysis techniques borrowed from systems analysis. A database provided a task dictionary which described 35 characteristics for each of the 66 tasks identified as necessary to perform simple word processing. The author discusses definition of user needs, selection of the documentation format, prototyping as a method for documentation development, putting the prototyping approach to work, devising a method for task analysis, and implications of using database techniques to task analysis.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

It is noted that existing editorial style sheets/guides do not address the unique needs encountered in developing computer-delivered information. The author presents guidelines that supplement the stylesheets/guides. These guidelines suggest ways to handle the organization and presentation of information, paragraphs, sentences, questions and responses, language, punctuation, and screen design.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Many well-written online documents fail because their visual design does not account for the many differences between the paper page and display screen, nor do they take advantage of the display power of the computer. The author provides guidelines for practitioners based on extensive research on computer display. It is concluded that the visual design of the online document should be crafted to match its use, ensuring displays that are scanned quickly and read reliably.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

The so-called computer explosion has had a number of undesirable effects on corporate communications. One of these is that well-established and integrated communication systems have been replaced by poorly designed and mismatched computer systems. This has caused a number of communication functions to be performed and administered by amateur communicators rather than by communication professionals. The effects on people and output have been drastic, even affecting management credibility. Professional communicators are exhorted to reestablish themselves as the experts on corporate communications so that the undesirable effects due to the computer explosion and the amateur communicators can be corrected. Computers can then assume their proper role in corporate communications.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

The authors introduce the concept of hypertext and focus on how hypertext can facilitate writing activities. A survey of the capabilities of existing hypertext systems and current research activities is included.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

The multidimensional nature of usability (including tradeoffs between dimensions) is described, stressing the linkage of the concept to the effective use of documentation by a particular group of readers in performing a certain set of software-related tasks under realistic environmental constraints. Major issues in the evaluation and design of usability are examined. It is argued that the sampling of realistic tasks and target readers is often very difficult, particularly in a laboratory setting, and that additional emphasis must be placed on using multiple observation periods, standardized measures and materials, and multimethod assessments of usability.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

A discussion of the real-world environment of usability testing in the computer industry leads into a discussion of how the real world constrains the usability process and how the usability specialist can cope with the constraints. The focus is on two major constraints that result from the need to get information to the interface designers early enough so that they can use it: the need to work with incomplete materials, and the short amount of time usually allotted for a usability test iteration. The design of a new documentation model at Microsoft is considered as an example.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

A number of research-based guidelines for incorporating instructional diagrams into documentation are discussed. A survey that was conducted to determine the extent to which existing computer-aided-design (CAD) documentation follows these guidelines is described. The results of the survey indicate that diagrams in CAD documentation do not, for the most part, follow the guidelines. Avenues of further research are suggested.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

A database management system for organizing the program of a conference is described. The system runs on an IBM-compatible personal computer with hard disk using DBASEIII-PLUS. The system is used for logging papers submitted for review, writing letters of acknowledgement, distributing and tracking papers in the review process, writing letters that notify authors of paper disposition, communicating with session chairmen, and automatically compiling and printing the resultant camera-ready program and author mailing labels. The system has been successfully used for the 1988 IEEE International Conference on Acoustics, Speech, and Signal Processing, a conference of 1500 submitted papers with a program of 750 papers presented in 56 sessions over four days. Use of the system saved six weeks and about $3000 compared to the time and costs of this conference in 1987.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Issues in desktop publishing are discussed as they relate to the use of computer databases. The design and development of software to automate the production of conference programs by linking databases to a desktop publishing system is described. The discussion covers: the redesign of the existing conference program; specifying styles in Ventura Publisher; setting up dBASE II databases; writing the conference program generator software; and using Ventura Publisher to produce camera-ready copy.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

The card-sort technique and cluster analysis were used for determining an effective organization for a help menu in Unix EMACS. Similarity data were gathered for with a card-sorting task using EMACS commands, and a hierarchical cluster analysis of the data was performed. The results indicate that differences among novices, intermediates, and experts appear with computer-based concepts such as windows and buffers, but that the sorts are more similar than they are different. It is argued that cluster analysis may aid designers in determining a functional organization, but that in the domain of this study, this organization may not help users bridge the mapping from real world tasks to computer tasks.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Methods of displaying computer-generated graphic images are examined. Several alternative technologies are evaluated. Principles of operation are given for each of these advanced technologies. These color display technologies include screen cameras, film recorders, LCD projection devices, and RGB projectors. A brief survey of commercially available devices is presented.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

It is argued that technical writing and computer programming are based on similar principles, perhaps because both are mechanisms for efficiently organizing and communicating complex information. Both technical writing and programming are multistep processes involving planning, drafting, and revising. Moreover, six key aspects of computer programming (modularity, modifiability, user interface, failsafe presentation, style, and debugging) have analogs among such technical writing principles and practices as report sectioning, audience analysis and adaptation, and editing. These similarities are discussed and a course that builds on these likenesses to teach technical writing to programming students is described.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

A student perspective is provided on the process of teaching technical writing as a process parallel to computer programming. The idea that each follows a recognizable series of discrete steps is presented, and the steps that parallel each other are contrasted.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Characteristics of user-developed software applications and their documentation requirements are reviewed, and the results of a study undertaken to identify factors affecting the level of documentation required by these applications are reported. The results provide a framework to guide computer-using management in formulating standards for a broad spectrum of such applications.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

The author provides guidance in responding to US federal government procurements that include quantitative and qualitative evaluation criteria. Systematic analysis of the relative importance of all evaluation criteria in a request for proposal can optimize page allocations and ensure balanced treatment of customer concerns, thereby achieving a maximum proposal score. A method is presented for determining relationships among evaluation criteria, and thus for making meaningful allocations of pages for proposal sections and subsections. The model lends itself to desktop computer implementation.

OVER the last few years, the technology of electronic technical publishing has become increasingly powerful and flexible; professional technical communicators, as well as other professionals in technical fields, now can expect as a matter of course to develop documents at computer terminals, incorporate tabular and graphic material, format them using a wide range of visual attributes, store them electronically, and produce them on sophisticated computer-supported printing equipment.

“Working alone on an Independent Learning Program can prove to be a lonely experience; it can also take a long time.” That's how Ron Blicq, the principal author of this learning guide, acknowledges what anyone who has ever attempted an independent learning course must surely think about the prospects for actually finishing. This time, however, it will be different; this learning guide was prepared with your loneliness in mind. Stopping just short of making a house call, black bags in hand, Blicq and his associates have gone to a great deal of trouble to give their students a mindset that will make them want to keep going, in spite of all the usual distractions and demands on their time. This is both a comprehensive course in professional communication skills and a carefully thought out independent learning guide.

As computers move onto the desks of all kinds of workers, and into all aspects of the work we do, online documentation becomes increasingly important. Documentation for a computer system or program contains the information that a user needs to know in order to operate the system or program. Online documentation refers to the effort to deliver some or all of that documentation using the same computer system that the user is learning about. This paper surveys many of the issues involved in designing online documentation: choosing a goal for online documentation, understanding the constraints on a design, and understanding the implications of a design for writers and users, mostly from the perspective of a designer or project planner concerned with issues in both software development and document development.

The UK Data Protection Act 1984 covers the collection, holding, manipulation and distribution of data concerning living individuals by computerized systems from within the United Kingdom. Individuals who visit the UK fall within the bounds of the Act as soon as they arrive. Although they may be exempt from the Act in relation to the data that they personally control within the UK, they could immediately be liable in connection with data held overseas-perhaps by their employer, client, or any other entity for whom they could be construed to be an agent. The author covers the basic elements of the Act: the definition of data, personal data, data subject, data user, computer bureau, processing, preparing the text of documents, transfer of data, and the territorial scope of the Act. He covers who might have to register and why, how to register, and other implications of the Act.

The author explores why computer users often do not use the HELP function offered by many programs. Three explanations are frequently offered for not using HELP: Asking for help: may break the user's flow of thought; disorients the user as far as navigation from screen to screen is concerned: and often does not provide the answer the user needs. Experience suggests that people have a low tolerance for unhelpful help, and that their initial expectations are quickly revised downward. What people think they want, what people expect, and implications of why users want context-dependent help are discussed. A conversational model for help is suggested. The application of a social, conversational model to human-computer interaction accounts for several phenomena that have been observed: people react badly to poor response time, people also react badly to very quick response time, and people react emotionally to programs.

Not too long ago, publication tasks such as layout, font selection, merging of text and graphics, and copyfitting were addressed only by communication professionals. Now, programs like Pagemaker by Aldus and Ventura Publisher by Xerox make these issues just another personal computer application, like spreadsheets, calendars, and word processing. The impact of desktop publishing (DTP) is already being hotly debated, and we open this new IEEE Transactions section on communication technology with an article that has significant implications for this debate.

When flutist James Galway was named to the principal's hair in the Berlin Philharmonic, it is said, his father wrote him from Ireland to the effect that this was all very nice, but when was he going to get a real job? When are you going to get a real job, my communicator friend? And you there, you the engineer, aren't you just playing games at the computer and calling it report-writing?

In an interview with Dr. J. Peter Kincaid, the value of readability formulas and computer editing systems to the engineer who writes on the job is explored. Dr. Kincaid developed the Kincaid Readability Formula, the standard used in judging the reading levels of Department of Defense manuals. The presentation is in a question and answer format.

The editor has invited me to serve as associate editor of the Transactions for papers on communication issues in software engineering. This will include papers on the design of user manuals, the development of online help and tutorials, and the documentation of software development. I am delighted to accept the position, because I see these issues as central to the continued advancement of computer applications.

There is a move in technical writing today toward a personalized, `user-friendly' writing style which is strikingly evident in many computer textbooks and instructional manuals. The author presents the results of a study to determine which style of writing people prefer in a computer tutorial, given the choice between formal, moderately friendly, and extremely friendly texts, and whether their level of computer expertise affects their selection.

A discussion is presented of test conducted at the IBM Human Factors Laboratory in Kingston, New York, to compare the productivity of people using computer-manual, task oriented vs. product-oriented information. The failure rate was 310% higher with the product-oriented information. The conservatively calculated overall productivity gain attained by task-oriented information was 41%.

Collaboration between the programming and documentation departments may be the key to writing good user manuals. Although time constraints and the computer culture stand in the way of collaboration, writers and programmers can overcome these problems with respect, good humor, and careful thinking. The authors describe an informal but successful system developed over the last three years at a software company.

Computer adventure games and operating systems have something in common: each presents to its players or users alien worlds that they can explore. The major difference between games and operating systems is in the way they communicate with people; operating systems tend to communicate poorly, while adventure games communicate so well that players need minimal introductory or reference documentation. What is it about adventure games that fosters ease of use? How might those techniques be applied to the design of operating system interfaces? The author offers answers for both questions.

Increasingly, documentation for computer systems and products addresses diverse audiences, ranging from professional computer engineers to novice readers who have never used a computer. This diversity presents a challenge to the computer documentation professional: how to address the needs of both novice and sophisticated users? Effective user documentation must be procedure-oriented: how to combine this goal with the fact that different user audiences have different goals and face different problems? A discussion is presented of the challenge of audience diversity in computer documentation. Using case studies (i.e. an accounting product, a database management system, and a decision support system), it reviews and recommends different ways to meet the needs of various audiences.

Producing computer documentation today involves more than just writing. Today's computer documentation specialists must draw on a wide range of multidisciplinary skills to produce an effective documentation package. Beyond knowledge of the underlying computer technology and the necessary writing skills, they must also draw upon pertinent concepts from marketing, user psychology, graphic design, screen design, usability testing, and the product development cycle.

Computer user documentation tends to be written by computer people, those familiar with or involved with the computer's development, who organize manuals in terms of software structure. The author details problems caused by this approach, and ways in which they can be resolved.

User-testing of computer documentation is beginning to move into the laboratory. A properly conducted user test offers software producer much more information that cannot be obtained in any other manner. The author discusses the advantages and disadvantages of the previous ways of user testing such as testing in the marketplace, magazine reviews, and prerelease on-site testing (beta tests). In addition, a detailed case study of a laboratory-based usability test is presented.

Despite the acknowledged importance of reader feedback to computer documentation, few relevant, standardized, empirically validated measures for such feedback have been developed. Research involving hundreds of experienced computer users from more than 30 organizations indicates that there are at least three principal dimensions which readers use when evaluating documentation used in the performance of job-related software tasks: task relevancy, systematic arrangement, and credibility. A nine-scale measure of these dimensions is proposed for standardized use.

A telephone survey is a potential effective way to learn the preferences of technical documentation users. The sample must be carefully selected, however, to ensure that it is representative of the customers who will be using a given category of products and manuals. Surveys of two different populations of computer manual users have been shown to yield conflicting results. In one such survey, 84 individuals were interviewed that were drawn at random from the customer list of a manufacturer of computer peripheral controllers. The sample was 96.4% male and primarily included programmers (33.3%), technicians (32.1%), and engineers (22.6%). The author discusses the responses obtained from this sample.

The author discusses the results of a mail survey of 42 software-manual writers, editors, and their managers residing in the Minneapolis/St. Paul area. The survey analyzed the concept of task orientation in computer manuals. Analysis of the survey led to the following conclusions: software-manual writers, editors, and their managers overwhelmingly think task orientation is the best way to organize manuals. The type of manual you write (system or application) or the size of computer you write for does not influence the effectiveness of task orientation. When a writer organizes a manual by tasks, he or she most often uses an organizational strategy, where each major tasks is described in a chapter. Finally, respondents described problems encountered when writing task-oriented manuals, and techniques used to write effective task-oriented manuals.

The author argues that online information must focus on traditional reader expectations in the creation of layout features for specific tasks. Typical screen limitations must also be considered in terms of the limits they impose on typical user interactions and navigation techniques. To address both these issues, screen geometries must be applied to functional designs. In addition, information developers need to consider the educational value in the computer's novel capabilities, such as sound, motion, and time.

An analysis is presented of the assumptions and research methods behind computer manuals for programmers and system administrators. The author builds an objective critique of current manuals' disjointed and legalistic approach. He also relates documentation issues to the structure of computer systems and to methods of development. Although analytical and theoretical, this critique offers concrete examples of better documentation and ends with a list of suggestions for writers.

An introduction to text processing is intended for computer science students with a background in computer terminology and programming. The author discusses various aspects of the way a computer handles a particular document or text and how computer scientists can make the computer more efficient and user-friendly. The design of computer software is discussed from the standpoint of the programmer and of the user in both natural and programming languages.

Writing better computer user documentation teaches good documentation practices by means of a process model that reflects current industry procedures. The book is more philosophical than its highly prescriptive and pedantic predecessors, and its emphasis is on writing for the computer industry, not on writing in general with computers thrown in as an afterthought. Its major audience is data processing professionals, most specifically practicing technical writers, information planners, and writing managers, although the book might well find its way into college classrooms in some of the burgeoning masters' programs in technical writing.