
In a sense, our text for the next few pages is that "the meek should inherit . . . "; or, at least, if they don't get the whole earth, they deserve a fair shake in computing. Hopefully, the title we've chosen is self-explanatory; but if there were to be a subtitle, perhaps we'd choose, "A Funny Thing Happened on the Way to Interactive Computing."
One can't deny the effectiveness of video arcade games in reachipg users! Just loop at the number of quarters pushed into the slots, the time spent by people of widely differing abilities, and the number of repeat encounters with the systems. At least part of the success is due to the ease of getting started (the first play of the game gets one comfortable with the procedures), the high degree of visualization of controls and results, and the responsiveness overall. Other factors will be taken up by the panelists.Review of the home computer market shows what can be accomplished by an easy-to-use accounting aid through advertising store demonstrations, and word of mouth. Visicalc has sold over a million dollars! Attendees will have an opportunity to try some of these impressive applications before and after the session.
It has been demonstrated that interactive natural language dialog is remarkably unruly, with many misspellings and grammatical errors. Although progress has been made in getting computers to process pristine English text, the day when computers will be able to process unlimited interactive natural language dialog is still very far off.The vast majority of the effort that has gone into designing interactive natural language systems has concentrated on the computer half of the human-computer dyad. Our approach concentrates on the human half. Specifically, the goal of our research is to define a human engineered subset of natural language that retains all of the user-oriented benefits of unrestricted natural language dialog, while greatly reducing the processing burden that true natural language interaction places on the computer. This paper is a preliminary examination of the possibility that these criteria may be satisfied by simply asking usres to be concise.
CONFER emerged from a concern with small group governance in both its communications and decision making dimensions. This context will be described as well as the principles operationalized in the CONFER system: individual equality, freedom, privacy and flexibility, and the facilitation of individual participation. CONFER is based on the proposition that effective communication is an active process for all concerned. This activity is strongly encouraged in CONFER through a number of mechanisms designed to facilitate interaction between the user and the CONFER system, as well as interaction among all the users of the system. As well, growth of the system over time is promoted by interaction of the system designer with the user community. The development of Merit's use of CONFER was aided greatly by our continuing close relationship with Dr. Parnes and by the participation of many members of the Merit staff in the CONFERence for CONFER development that he eventually began. As we identified new aspects of CONFER use, we usually also were able to think of changes to CONFER that would facilitate them. For example, as we began to realize how important was the general conversation among our users--a critical need for the isolated user in particular--we led the lobbying effort for changes to make the voting structure more interactive. In response, CONFER over the years has radically altered the default form of participant comments from its original formal and anonymous style to a conversational model. Similarly, as CAUCUS grew to enormous size we asked for (and got) aids in managing, and then in re-starting, large Conferences. In a remarkably short time, CAUCUS became a general meeting place for many who were not Merit users at all, simply because it offered a place for users to meet one another and share problems and solutions. Eventually CAUCUS gave birth to a general-purpose CONFERence for users of one of the host systems--an idea that has since been implemented elsewhere as well. The growing use of CAUCUS as a communications medium for Merit staff eventually resulted in the birth of a CONFERence for that specific purpose. Before the use of CONFER there had been relatively little communication between the staff members at the central site and those located at the host computing centers except for occasional face-to-face meetings that were largely reportorial in content. Once CAUCUS showed the staff that network issues could be discussed in real time and that input from the host staffs could have weight, the pattern of decision-making began to change. With the inception of the MNET:STAFF CONFERence, and the resulting privacy for internal discussions, the role of the staff members--both centrally-located and at the hosts--continues to grow as we expound, lobby, and negotiate on network issues while (or even before) they are happening. It is no exaggeration to say that we members of the Merit staff could not do the job we do without CONFER. Merit's primary goal is to facilitate the use of remote computer resources for a heterogeneous and dispersed community of users. Without CONFER we would be working harder to accomplish much less of this objective.
The presentation deals with two questions:1) What makes games so captivating?2) How can the same features (that make computer games captivating) be used to make other user interfaces more interesting and enjoyable to use?First, three empirical studies are described. These studies analyze which features of several computer games are most important in making the games enjoyable. Then a set of heuristics for incorporating these features in other user interfaces will be outlined. The heuristics are organized in three categories: challenge, fantasy and curiosity.
Computerized Conferencing Systems are a new communications medium which offer an opportunity for geographically dispersed groups of persons to communicate regularly and to interact and make decisions as a group. They also include an embedded language which makes it possible to program and conduct fully automated experiments on group communication processes. The system thereby becomes a "laboratory without walls" which can be transported to the subjects.This paper reviews the methodological aspects of the first of a series of controlled "laboratory" experiments on the effects of the medium on small group communication processes. It focusses on the potentials for use of such systems as a tool for conducting communication experiments. The findings of the experiment itself will be reported more fully elsewhere.
CONFER, first used at Wayne State University in 1979, has proven itself to be an extremely useful and adaptable communications medium. Some examples of present CONFER applicatons at WSU are: Computing Center Staff CONFERence Presently used to coordinate communication between the 180 staff members of the CSC. Communications between various departments isolated by distance and responsibility has been improved. Specialty CONFERences CONFERences exist on such varied topics as school transportation, text processing, nursing education, microcomputers and instructional technology. Project Management Used by the University's PLATO development staff for cross-campus management decision-making and communication. CONFER has reduced the need for staff meetings, has served as a "tracking device" for personnel appointments, and has kept a detailed log of project decisions. CONFER has also been used for CSC project development, notably for the design of an MTS Help Facility. Academic Communications CONFER has been used to facilitate communication between students in an undergraduate Computer Science course, for graduate students in Instructional Technology, and for individual student projects. This summer, a CONFERence will be implemented which will manage course communications for 300 students in a freshman Computer Science course.
A study is presented of a database system interface in which an entity (a concept) and the relationships in which it is involved are displayed to the user: the user is permitted to move about in the database by selecting entities related to the current one displayed. The database system is intended as a personalized database (PDB) for a scientist, student, manager, or anyone who has a need for a fast mechanism for storing and organizing a wide variety of information. The study is exploratory recording baseline times and types of behavior for a variety of personal information management tasks performed by one individual. Data entry, information retrieval, and browsing behavior are examined and contrasted to behavior with more conventional storage media.
Members of the Merit staff first met Robert Parnes in the fall of 1975 and began participating in his experimental CONFERence shortly thereafter. It soon became evident that CONFER could help us provide consultation to our users, who were distributed over a large part of southeastern Michigan, and in January of 1976 Merit started what we believe to be the first CONFERence open to the general public, MNET: CAUCUS.Five years later CAUCUS is still alive and well, and we still use it to provide help to a widely-dispersed user community--in fact, with the advent of Telenet service later that same year, and with Telenet's subsequent expansion of service to Canadian and overseas networks, our users are spread all over the world. But we have learned over the years that computer conferencing is good for much more than simply facilitating the user-consultant relationship. As we gained experience with CONFER we found that it gave us a solution to problems that were so basic we had simply taken them as part of the environment. CONFER also provided a medium for communication among consultants--the Merit staff--and among users.
A prerequisite for the design of better systems - in terms of human interface- is knowledge of its users, their problems and behavior. Within the context of a larger project comparing several large statistical program packages, attempts have been made to attack the problem of "knowing the user'. Among traditional methods like surveys, different ways of automatic data collection have been tried and their strengths and weaknesses can be discussed. A particularly powerful tool proved to be a logfile which is automatically updated each time certain software is used. It contains individual level data about size of job and data set, control cards and -statistical procedures used, types of errors and more. This gives valuable insights about:--the structure of the user community.--styles of package use,--weak points of packages.In my paper I will discuss some general problems of recording and analyzing user information, and will present data fre~ the logfile described. This should he considered as an example in the methodological discussion as well as a substantive contribution to the analysis of SPSS-use and -users.
Advances in both hardware and software continue to make it possible to design user oriented systems more easily. Because we have not had a language for describing the user orientation of computer systems, a variety of interpersonal metaphors have been used to aid in the comparative evaluations of systems. Recent cultural history has shaped the semantics of computer systems. Out of the turbulent, liberal strains of the 1960s emerged the movement to humanize computer systems. During the self-centered backlash of the 1970s the term friendly became a computer household word. During the 1980s we need to grow beyond a concern for friendliness alone and build systems that are considerate.Consideration supercedes friendliness in at least three major ways, First, it goes beyond satisfaction by focusing upon attempts to help and assist others. Secondly, it requires that a person take the role of another and take the other's needs into account. Thirdly, to be considerate is to be courteous and, most importantly, respectful. In these respects, the metaphor of the considerate system points to the essence of user orientation wltbout sacrificing other critical system features such as productivity. In fact, truly considerate systems will facilitate productivity because of improved communication clarity, greater tolerance for user errors and idiosyncrasies, and increased availability of options, i.e., user-directed socio-computer interaction.Designing and developing considerate systems is not easy and requires considerable time and effort. Representative users must he involved in the selection of system features and in the process (formative) evaluation as well as the outcome (summative) evaluation. Consequently, there is a very necessary and essential role for the social scientist in the development of present day socio-computer systems.
CONFER is used by a variety of organizations which have nodirect contact with any MTS site (the CONFER host computer system).In many cases, the main contact of the CONFER user is with thethird party vendor or consultant who is supporting the networkingeffort. Several problems arise when the third party relationship isnot broadened to include the institutional consulting anddocumentation efforts. Such problems include:1. Overload on thetime of the third-party consultant.2. Production of customdocumentation which parallels the institutional documentation.3.User perceptions that they are not "ready" for institutionaldocumentation.4. Development of abbreviated cognitive maps of thesystem and its user network.5. Users exchange superstitious viewsof how the system works.While it is not yet clear that theabsorption of third-party, mediated networks will solve suchproblems, it is an obvious first step to a solution.MTS Help FacilityA version of CONFER is tentatively planned for use as theprincipal Help Facility at WSU. A subset of CONFER capabilitieswill be employed to decrease demands on the system.Features of the Help Facility will include agenda stringsearches, execution of computer assisted instruction lessons fromwithin tbe CONFERence, a "suggestion box" facility, and usermodelling features intended to decrease user frustration andmaximize efficiency. Existing help files will be called from theHelp Facility to increase the redundancy of available help to theuser.The agenda used will be multi-dimensional, allowing the user tosearch for help within subject areas, type of information and skilllevel. For example, the user would be able to view all items whichserve as tutorials on statistical packages with the CONFER command:AGENDA TUTORIALS+STATISTICS. Similarly, the user will be able tosearch item descriptors or text for character strings, such asusing WHICH "SPSS" to locate all items dealing with SPSS.Requests for help will be sent directly to various CSCdepartments, and responded to either via CONFER or telephone. Usersuggestions and requests will be screened on a continuing basis byCSC staff, and changes to the information contained in each itemwill be made as a result of these suggestions. Materials developedby users will be solicited for inclusion in the Help Facility.
This session described how CONFER, a computer-based conferencing system at the University of Michigan, was developed with participation of users, and what impact the system has on communities of users. At this conference it may be especially interesting to discuss various ways in which help provided to new users has evolved. Some extrapolations may be made for other than electronic communication aids: orientation and training, on-line reference information, on-line consultation, etc.
Users of interactive systems typically must deal with numerous interactive interfaces, including especially the text editor and the system command interpreter. Unfortunately, the various interfaces too often have differing and even conflicting conventions. This paper suggests that an enhanced text editor can serve as the interactive interface for most purposes. For example, consider the file directory instead of choosing among half a dozen or more system commands to view and modify it, the user can edit an image that represents the directory. Deletion, renaming, and movement to another directory are easily accomplished with ordinary editor commands. Other system commands can he supplanted by a mechanism of "creation sequences" for files. Rather than execute the creation sequence, the user simply asks to view the file resulting from it.To facilitate this form of interaction, the text editor must include some novel features. It must permit structured files; where the structure can be a field structure within records or a hierarchical structure between records. A suitable editor is sketched.
Response time is one of the key components of the human interface in an interactive computer system. This study evaluated two different response times and their impact on learning effectiveness. Using a counterbalanced experimental design (2**2 combinations of 2 response times), this study measured completion times, lesson mastery, error rates, and attitude. Data were obtained from student questionnaires.The Control Data PLATO Computer-based Education system provided the environment for the study. The system was connected to two networks with different response time characteristics. The means of the two response times tested were .25 sec (response time A) and 1.3 seconds (response time B). The covariate analysis of variance and chi square tests were used to show the significant difference between the two response times (p < .05), giving the following results:1. The subjects using the shorter response time finished the lessons significantly faster than the subjects using the longer response time.2. The number of subjects that mastered the lessons was significantly higher for the subjects using the shorter response time.3. The performance of subjects using the shorter response time for time dependent tasks was significantly better than the subjects using the longer response time. However, for time independent tasks, the subjects using the longer response time performed significantly better.4. The subjects using the faster response time showed significantly more favorable attitudes toward the response time experienced than the subjects using the slower response time.In conclusion, the shorter response time (A) was more efficient for learning and was more favored by students.
Direct manipulation is a style of interaction which has been used by implementers of widely varying systems. Direct manipulation permits novice users access to powerful facilities without the burden of learning to use a complex syntax and lengthy list of commands. Display editors use direct manipulation more than line editors. Form-fill-in is more direct than tag fields and delimiters. Spatial data management is more direct than query-by-example, which is more direct than SEQUEL. Computer arcade games and Visicalc are further examples.Direct manipulation involves three interrelated techniques:1. Provide a physically direct way of moving a cursor or manipulating the objects of interest.2. Present a concrete visual representation of the objects of interest and immediately change the view to reflect operations.3. Avoid using a command language and depend? on operations applied to the cognitive model which is shown on the display.
Designing and providing subject access to works of art has traditionally been very subjective. This is due to the fact that there are no standardized word lists which can be expected to meet the scope of all art collections. Whereas tailoring subject terms to the scope of a given collection is the most practical approach for the curator in charge of the collection, the researcher, who may not be an art specialist, often is frustrated when the listing of subject terms does not include the terms relevant for his/her purposes. This presentation will explore how the development of a thesaurus resolves the conflict of subject vocabulary. Specific examples will be drawn from four computer projects at the National Museum of American Art. Each of the projects varies in scope, yet a single subject classification guide has been developed for purposes of providing subject access to the contents of each project.Because a separate subject word list was not originally designed for each project, a thesaurus is now being developed which will allow for a listing of terms not used for indexing but which are relevant to both the scope of each project and anticipated researcher needs.From a practical viewpoint, the presentation will demonstrate how the computer can be used to generate terminology to be included in the thesaurus.
This project management system is designed to provide research administrators and department executives the capability to handle the financial accounting needs for 30 or more separately budgeted projects/departments on a computer with 64K of RAM and two floppy disks. The system generates a wide variety of flexible reports including: a Cost Accounting Summary showing previous period expenses, current period expenses, total expenses, budgeted amount, encumbrances, and remaining balance by line item; a Budgetary Summary which shows for the current period and year-to-date the actual expenses, the budgeted amount, the variance between actual and budget, and the percent of budget by line item expended; an Income Statement showing revenue and expenses; an Expense Report; and a Transaction Register; and more.
Results are reported showing the changing pattern of command use by introductory business data processing students. Using the ability of the University of Calgary's Honeywell Multics Operating System to tailor a command and response environment, a subset of commands and responses (called GENIE) was set up in a user-friendly environment to facilitate novice students programming at CRT terminals. Frequency and time of usage of all commands was metered and changing patterns of usage emerged as the semester progressed. For example, "help" usage -- which was originally quite extensive and broad -- limited itself over time to questions only about specific topics. Reluctance to use an "audit" facility to capture an interactive session disappeared as the commands for such usage were likened to a movie camera taking pictures over a student's shoulder. It is further noted that specific emphasis was placed on simplifying commands and reducing options.The whole idea of a restricted command environment is compared to the "abstract machine" concept of Hopper, Kugler, and Unger who are building a universal command and response language (NICOLA, a NIce Standard COmmand LAnguage). GENIE is seen as an example of what such an abstract machine could be if the Multics operating system were viewed as a basic or "parent" abstract machine. Interactive environments such as Multics provides are viewed as essential to providing a satisfactory timesharing system for the various, but frequently intermittent uses, in the social sciences.
The office of the future is defined. It is a work stationdominated by a micro computer which is in communication with morepowerful computers, large disks, printers, and other equipmentwhich can be shared. The large computers, disks, printers, and allthe rest already exist. Micro Computers or desk top computersalready exist. There are only two steps left in realizing theoffice of the future. One step involves electronics; establishinghigh speed communication between the desk top computer and all ofthe other equipment. The second involves programming; defining anddeveloping coherent software systems. This paper is about the wayin which these two problems were handled by the department ofpolitical science at the university of Iowa. Weeg Computing Center provides academic computing services forthe University of Iowa. They do this with a collection ofcomputers; there is an I.B.M. 370, 5 Prime 750s, 4 Hewlett Packard2000s, and a Vax. There are also disks, plotters, printers, and thewhole panoply of equipment one expects in a university computingenvironment. In addition, a Gandalf communication system makes itpossible to access any one of these computers at speeds up to 2400bits per second. These facilities are normally accessed with theaid of one of the approximately 750 terminals on campus. With thesupport of Weeg computing Center the department of politicalscience convinced the university administration that it would beworthwhile experimenting with desk top computers for an entiredepartment. Money was made available, and 37 computers werepurchased. Which brought us to the two problems: high speed (inthis case 2400 bits per second) communication; and a coherentsystem of programs. Commodore 8032s were purchased, and this meant that we needed aninterface which would translate into the serial signal the Gandalfsystem used. Several interfaces were available commercially, but wecould not make any of them work at 2400 bits per second. Nor couldwe get the terminal emulator programs available commercially toprint the screen at 2400 bits per second. The dominant speed inmicro to larger computer communication is 300 b.p.s., fast means1200 bits per second. An interface board was designed by theelectronics shop at the University for us, and a machine languageprogram was written which would handle sending and receiving at therequisite speed. In principle this seemed a straightforward task.In practice there was a stream of nitty gritty problems that seemedas though it would go on forever. But it was done. Three design considerations dominated the creation of thesoftware system. One, it had to be as simple to use as possible.Half of the potential users had never used a computer before. Thatimplied that the system must be as automatic as possible. Two, ithad to, initially, provide a terminal to each user plus handle wordprocessing and electronic mail for the department. Three, it had toallocate tasks between the desk top computers and the Weeg suppliedcomputers in a way that, within reasonable bounds, maximized thework done on the desk top computers and minimized use of centralresources. Two features of the system illustrate our attempt to make it aseasy to use as possible. First, signing on to the central computeris almost automatic. Each person's copy of the machine languageprogram contains their four codes required to sign on to thesystem. The individual can send these codes by pressing the controlkey and the up arrow or programs developed for this system willautomatically send these codes. Signing off is equally automatic.When the Prime on the other end has finished sending or receiving aprogram or data it automatically signs off. Second, all choices aremade from two lists. One list is on the Prime, and the other listis embedded in the software used on the CBM 8032. Physicalcontiguity of choices reduces the amount of recall necessary to usethe system. A good screen based text editing program enhances one's abilityto put words together. It does this by making it easy to write andrevise. However, there are many tasks, in addition to writing andrevising, that one needs to be able to carry out with the documentbeing written. You must be able to save and recall the text. It isvery nice to be able to append a second file to the one beingwritten. It is very nice to be able to save a segment of the filebeing written. And the text must be sent to be printed. All ofthese operations can be performed with the aid of the list ofcommands in the text editing program. One presses 's' and gives afile name; the programs do the rest. In addition, one would like tobe able to look at a list of files that have been created andsaved, delete files that are no longer needed, transfer a file tothe disk storage area of a collaborator for his or her review, senda file (program or program control information) to the batchprocessing system, and send a file directly from the disk to beprinted. All of these tasks can be performed using the list ofcommands on the Prime. Most of the operations one can imaginewanting to perform on a text file can be accomplished with one ofthe two lists. That makes the system easy to learn; it providescoherence. In this system the Prime serves as the storage device and themedium of communication. Programs and text files are stored on thePrime disk; there are no disks for the desk top computers. Inaddition, sending a file to be printed or sending a file to acollaborator plus electronic mail are handled by the Prime. Thedesk top computers handle the writing and revising which is donewith a program written in BASIC to run on these computers. Sincewriting is an exceedingly important part of the job of academicsthe computers are on all the time. But most of the time they arenot on to the central computing resources; they are working asstand alone computers. When you walk up and down the hall in the department ofpolitical science you see lots of computers going. Writing hasbecome much easier. And getting a final document produced has neverbeen as easy or as fast. The mini-micro connection works.