What will we be working on in the coming decade? This talk will consider a wide range of possibilities, ranging from predictable impacts of new technologies to difficult matters of perspective and policy that influence the development and deployment of technology, to quixotic ambitions. The opinions will be those of the presenter, if even he endorses them.

Block languages are extensively used to introduce programming to children. They replace the complex and error prone syntax of textual languages with simple shape cues that show how program elements can be combined. In their present form, blind learners cannot use them, because they rely on graphical presentation of code, and mouse interactions. We are working on a nonvisual blocks language called Pseudospatial Blocks (PB), that supports program creation using keyboard commands with synthetic speech output. It replaces visual shape cues for language syntax, the key feature of block languages, with filtering of program elements by syntactic category.

How can the broader field of computer science research be harnessed to address challenges and opportunities in accessibility? This poster summarizes the findings of a workshop, sponsored by the Computing Community Consortium (USA), that brought together computer scientists, representatives of disability advocacy organizations, people from industry, and government employees to develop an agenda for strategic research. Members of the ASSETS community may find the reports useful in organizing collaborative projects with others in the computer science community.

Interactive simulations are increasingly important in science education, yet most are inaccessible to blind learners. We demonstrate an accessible version of a simulation, Balloons and Static Electricity, that illustrates responses to key challenges in providing screen reader support: the need to describe unpredictable sequences of events, the manipulation of objects that act as both controls and displays, and the management of descriptions of changes in the state of the simulation as well as of the state of the interactive object, itself. Meeting these challenges requires extending current practices for verbal description of visual interactive content.

We describe progress in creating an inclusive version of an interactive simulation that presents the dynamic electric field produced by a moving charge, using an audio presentation. The results illustrate opportunities and challenges in making this simulation usable for blind users.

Interactive simulations are becoming increasingly important in education, but these tools are not accessible to many learners. Here, we share some goals to guide the development of accessible simulations, and summarize early findings in development efforts. We also highlight challenges and opportunities where the unique expertise of the ASSETS community is needed. We hope to interest readers in participating in the growing community designing, developing, and researching interactive simulations for learning.

Computational thinking aims to outline fundamental skills from computer science that everyone should learn. These skills include problem formulation, logically organizing data, automating solutions through algorithmic thinking, and representing data through abstraction. One aim of the NSF is to integrate these and other computational thinking concepts into the classroom. This paper introduces a tool called the Simulation Creation Toolkit wherein users apply high-level agent interactions called Computational Thinking Patterns (CTPs) to create simulations. Programming at the Computational Thinking Pattern level allows users to directly create agent interactions in a simulation by employing generic icons acting out a scientific phenomenon they are trying to represent. The Simulation Creation Toolkit aims to preserve the computational thinking benefits of simulation creation while enabling higher-level implementation of agent behaviors. Initial study data collected from sixth grade students with no prior programming experience indicates that students can work the mechanics necessary to create simulations in the classroom environment using the Simulation Creation Toolkit.

School districts across the country are embracing 21st century skills, and grappling with how to teach these to their K-12 students. At the same time, computer science educators are grappling with how to broaden participation. These two dilemmas are related, in that computer science can be used to teach many of the 21st century skills, and bringing computer science to all K-12 students can help broaden participation.

People with cognitive disabilities that affect their memory, attention, and comprehension can become overwhelmed when using technology---just as cognitively-demanding situations like driving or multitasking can hinder technology use for most people. However, appropriately-designed technology can assist in overcoming cognitive disabilities and cognitive limitations. This workshop seeks to bring together researchers and practitioners with design experience in the many areas of cognitive disability and cognitive limitation to exchange, evolve, and develop strategies for design. Workshop participants will present key lessons from their own experiences, and workshop activities will employ claims-based design strategies toward identifying, comparing, and mapping approaches for addressing cognitive disabilities and limitations.

People with cognitive disabilities need careful matching to find appropriate employment. However, it can be difficult for them to articulate their worries and concerns in a timely and useful manner. This work demonstrates how appropriately-designed technology can assist in accommodating cognitive disabilities, providing avenues to assess their concerns about work environments. A mobile application was developed with targeted repeated multimedia surveying to assess work concerns, with temporal- and geo-tagged answers stored for review by a personal assistant, employer, job coach, or person with a cognitive disability. An expert review provided feedback to ensure an appropriate application.

Engaging Computer Science in Traditional Education (ECSITE, pronounced "excite") is a 5-year program that began in 2009 to bring computer science into traditional K-12 classrooms. Rather than seeking to draw students into computing courses, we bring computing into the courses that students are already taking. To date, these have included art, biology, health education, mathematics, and social studies courses as well as a Native American focus program. Middle school and high school students are introduced to computational thinking and computer science concepts including algorithms, graph theory, and simulations in interdisciplinary contexts, mirroring the ways in which computing technologies are utilized in research and industry. Teachers report that students increase their understanding and perception of computer science, and that participating K-12 teachers increase their knowledge about computing and will continue to include the computational curriculum after their involvement with ECSITE.

This paper explores the potential benefits of using mobile webpages to present simpler web content to people with cognitive disabilities. An empirical analysis revealed that the majority of popular mobile sites are smaller than their desktop equivalents with an average of half the viewable content, making them a viable method for simplifying web presentation.

Bangaten is a new version of Banga [2,3], a smart phone application that supports word finding practice, a form of therapy for people with aphasia. Early user experience shows that Bangaten offers useful cross-platform operation, on both Android and iPhone devices, including remote management of a client's device. Bangaten demonstrates the growing usefulness of emerging HTML5 technology for implementing assistive technology applications, while also illustrating some remaining limitations.

Bangagears is a new version of Banga, a smart phone application that supports word finding practice, a form of therapy for people with aphasia [1]. While Banga was implemented as a native application, a program specific to a particular kind of phone, Bangagears uses the emerging HTML5 technology to operate, in principle, on many different kinds of phones and other Web platforms, and to offer simpler development and deployment. Lessons from Bangagears will be useful to other developers of applications for people with disabilities

Banga is a software system that uses the Android open source software platform for mobile phones to support word finding practice, a form of therapy for people with aphasia. By connecting to a Web site, Banga provides for remote monitoring and management of the therapy, making it easier for patients to participate in treatment, and for clinicians to deliver it.

The ioRules system uses model-based techniques to solve two classes of software quality assurance problems: policy satisfaction, and regression. In a policy problem, the user wants to know whether a program being evaluated satisfies a policy that is critical in the application domain. In a regression problem, the user wants to know whether a modified version of a program implements the same functionality as some original baseline. The system also allows the end user to browse a model of their program, viewing its behavior flexibly from multiple perspectives.

People often use powerful tools to manage the documents they encounter, but very rarely to store the mental knowledge they glean from those documents. Popcorn is a personal knowledge base: an experimental interface and database designed to store and retrieve a user's accumulated personal knowledge. It aims to let the user represent information in a way that corresponds more naturally to their mental conceptions than simply text would, in part by making heavy use of transclusion: sharing items among multiple contexts. This paper describes the design rationale for the system, contrasting it with related efforts, and presents the results of deploying it to a group of volunteers who used it in real-world settings. The results, while revealing some limitations in the tool, and some challenges in coping with knowledge reorganization, suggest that the analysis underlying the design is useful, and that Popcorn is a powerful and effective tool for a variety of intellectual work.

No abstract available.

This paper proposes a transformation of the Cognitive Walkthrough (CW), a theory-based usability inspection method that has proven useful in designing applications that support use by exploration. The new Cognitive Walkthrough for the Web (CWW) is superior for evaluating how well websites support users' navigation and information search tasks. The CWW uses Latent Semantic Analysis to objectively estimate the degree of semantic similarity (information scent) between representative user goal statements (100-200 words) and heading/link texts on each web page. Using an actual website, the paper shows how the CWW identifies three types of problems in web page designs. Three experiments test CWW predictions of users' success rates in accomplishing goals, verifying the value of CWW for identifying these usability problems

(2000). Introduction to This Special Issue on New Agendas for Human-Computer Interaction. Human–Computer Interaction: Vol. 15, New Agendas for Human-Computer Interaction, pp. 69-74.

ABSTRACT considered simply, but a more complex evaluation of the We have adapted the programming walkthrough technique to help design computer-supported educational activities in elementary school science. We present examples from a case study which illustrate ways in which design of an educational activity is similar to and different from design of a user interface. We have found that the walkthrough approach is useful in this new setting, and that it sheds new light on the general task-centered orientation to design.

A new generation of innovative, highly visual children's programming environments is under development.In this paper, we consider the instructional requirements for children learning to program in a visual environment.Based on our year-long experience using Apple Computer's Kid-Sim/Cocoa prototype [2] and the results of a year-end assessment, we conclude that the children failed to grasp many aspects of the program operation.The children readily mastered drawing and animating characters in imaginary worlds, but struggled to achieve more complex behaviors.Lack of explicit instruction on program functionality hindered these children in their attempts to create more sophisticated science programs.We explore the prospects for more effective instruction and suggest some guidelines for designing visual programming environments.

Users who have worked with just one or two pieces of application software on a computer system are often faced with the need to use a new piece of software on the same system.Consistency between program interfaces is intended to make the new program easier to learn in this situation, but how "consistency" should be defined is not always clear.We present a model of analogical reasoning that describes how users rely on interface consistency to induce correct actions in a new situation.Versions of the model are implemented in ACT-R and Soar.The model yields a clearer and more principled understanding of design guidelines that recommend interface consistency.

Abstract The programming walkthrough is a method for assessing how easy or hard it will be for users to write programs in a programming language. It is intended to enable language designers to identify problems early in design and to help them choose among alternative designs. We describe the method and present experience in applying it in four language design projects. Results indicate that the method is a useful supplement to existing design approaches.

As policymakers and technology planners respond to the growing activity in human—computer interaction, a broad perspective may be helpful. This article offers a top-down view of current activities and suggests opportunities and challenges for the continued growth of HCI. Partnerships among universities, corporations, government agencies, and professional societies are proposed. Infrastructure needs to support this new discipline are outlined.

Traditional programminglanguage design has focussed on efficiency and expressiveness, with minimal attention to the ease with which a programmer can translate task requirements into statements in the language, a characteristic we call "facility."The programming walkthrough is a method for assessing the facility of a language design before implementation.We describe the method and its predictions for a graphical programming language, ChemTrains.These predictions are contrasted with protocols of subjects attempting to write their first ChemTrains program.We conclude that the walkthrough is a valuable aid at the design stage, but it is not infallible.Our results also suggest that it may not be enough for programmers to know how to solve a problem; they must also understand why the solution will succeed.

No abstract available.

The value of theoretical analyses in user interface design has been hotly debated. All sides agree that it is difficult to apply current theoretical models within the constraints of real-world development projects. We attack this problem in the context of bringing the theoretical ideas within a model of exploratory learning [19] to bear on the evaluation of alternative interfaces for walk-up-and-use systems. We derived a “cognitive walkthrough” procedure for systematically evaluating features of an interface in the context of the theory. Four people independently applied this procedure to four alternative interfaces for which we have empirical usability data. Consideration of the walkthrough sheds light on the consistency with which such a procedure can be applied as well as the accuracy of the results.

The NoPumpG prototype [7,8] suggested that the spreadsheet model of computation could simplify the creation of some types of interactive graphical application when compared with other approaches. We report here experience in developing an enhanced follow-on system, NoPumpII, and describe three applications developed using it. We conclude that (1) the potential advantages of the spreadsheet model are realized in this application experience, (2) revisions to the prototype design have permitted an increase in the complexity and scale of applications, and (3) there remain limitations in the current design which, if redressed, would further enlarge the scope of application. More generally we conclude that alternative computational models are an important area of exploration for HCI research.

Progress in scientific visualization could be accelerated if workers could more readily find visualization techniques relevant to a given problem. The authors describe an approach to this problem, based on a classification of visualization techniques, that is independent of particular application domains. A user breaks up a problem into subproblems, describes these subproblems in terms of the objects to be represented and the operations to be supported by a representation, locates applicable visualization techniques in a catalog, and combines these representations into a composite representation for the original problem. The catalog and its underlying classification provide a way for workers in different application disciplines to share methods.>

Although the practical importance of user interface technology is now well established, the proper role of research in the development of the technology and the kind of research that is appropriate remain in question. This article takes stock of some of the competing positions and proposes an agenda, identifying areas of work that might command some consensus despite the widely varying viewpoints represented in the research community. The major initiatives proposed are understanding goals and preferences, broadening applied cognitive theory, supporting innovation, and credit assignment.

Many important computer applications require that users be able to use them effectively with little or no formal training. Current examples include bank teller machines and airport information kiosks. Today successful systems of this kind can only be developed by iteration using costly empirical testing. This article aims to provide a theoretical foundation for the design of such systems, a model of learning by exploration called CE+. The theory incorporates assumptions from (a) the GOMS model and cognitive complexity theory (CCT) on the representation of procedural knowledge as productions, (b) the EXPL model on learning from examples, and (c) research on problem-solving processes for simple puzzlelike problems. Design guidelines for systems that can be learned by exploration, "design for successful guessing," are derived from the theory. These principles are compared with those developed by Norman (1988).

Easy learning of a user interface depends in part on users being able to generalize successfully about it. Philosophical doctrine, and some recent work in human-computer interaction, argues that causal analysis of interactions can support generalization. But neither the philosophical literature nor the HCI literature provides a rigorous theory of causal analysis adequate for problems in human-computer interaction. We propose such a rigorous theory here, and show how it accounts for two robust generalizations, using certain general assumptions. We then present evidence that these assumptions are accepted by people. Finally we compare this theory with other treatments of consistency.

Learning to control a computer system from limited experience with it seems to require constructing a mental model adequate to indicate the causal connections between user actions, system responses, and user goals. While many kinds of knowledge could be used in building such a model, a small number of simple, low-level heuristics is adequate to interpret some common computer interaction patterns. Designing interactions so that they fall within the scope of these heuristics may lead to easier mastery by learners.

Nine participants used a full screen computer text-editor (XEDIT) with an IBM 3277 terminal to edit marked-up documents at each of three cursor speeds (3.3, 4.7, and 11.0 cm/sec.). Results show that 9% of editing time was spent controlling and moving the cursor, regardless of cursor speed. The variations in cursor speed studied did not seem to act as a pacing device for the entire editing task.

Conventional spreadsheet programs offer a very convenient user interface for many quantitative tasks, but they are restricted to handling precisely-specified quantities and calculations. ASP is a generalized spreadsheet that extends the basic spreadsheet paradigm to encompass quantities which are not known exactly, and functions which are not known well enough to permit calculation. ASP works by propagating assertions about quantities and functions through the network of relationships that the spreadsheet defines.

Studies of people learning to use contemporary word-processing equipment suggest that effective learning is often "active," proceeding by self-initiated problem solving. The instructional manuals that accompany current word-processing systems often penalize and impede active learning. A set of instructional materials was constructed for a commercial word processor, specifically designed to support and encourage an active learning orientation. These "guided exploration (GE) materials are modular, task oriented, procedurally incomplete, and address error recognition and recovery. Learners using the GE materials spent substantially less time yet still performed better on a transfer of learning posttest than learners using commercially developed self-study materials. Qualitative analysis of aspects of the learning protocols of participants suggested that active learning mechanisms may underlie this advantage.

Nine participants used a full-screen computer text editor (XEDIT) with an IBM 3277 terminal to edit marked-up documents at each of three cursor speeds (3.3, 4.7, and 11.0 cm/s). These speeds occur when a user continuously holds down an arrow key to move the cursor more than one character position (i.e., in repeat or typamatic mode). Results show that cursor speed did not seem to act as a pacing device for the entire editing task. Since cursor speed is a form of system response, this finding is in contrast with the generally found positive relation between system-response time and user-response time. Participants preferred the Fast cursor speed, however. Overall, more than one-third of all keystrokes were used to move the cursor. We estimate that 9-14 percent of editing time was spent controlling and moving the cursor, regardless of cursor speed.

Computer text editors are powerful, but complex, tools.Particularly in the early stages of learning, the complexity of these tools can cause serious problems for users who are not experienced with computers.The problems of new users were studied by asking the users to think out loud while learning to use word-processing systems.In this paper several of the most typical and debilitating problems these users had understanding and following directions in using training manuals, as well as problems understanding and using interface functions to accomplish word processing tasks, are taxonomized and analyzed.Approaches for improving design features of the interface functions and the training methods used for learning are discussed.

There is growing interest in cognitive science in the the mental processes that underly learning and using computer systems (e. g., Bott {1}; Mayer, {2}; Card, Moran & Newell {3}). In this paper we report generalizations about the problems people who are not experienced with computers have learning to use a text-processing system. We are especially interested in unaided self-instruction, because of the practical interest in reducing the role of experienced personel in the training process. We analyze these difficulties in terms of the interaction between the cognitive characteristics of the learner, and the design of self-instruction, and the interface. Finally, we are also interested in implications of these problems for designing better training methods and computer interfaces that are easier to learn.