Given the complexity and functionality of todays software, task-specific, system-suggested help could be beneficial for users. Although system-suggested help assists users in completing their tasks quickly, user response to unsolicited advice from their applications has been lukewarm. One such problem is lack of knowledge of system-suggested help about the users expertise with the task they are currently doing. This thesis examines the possibility of improving system-suggested help by adding knowledge about user expertise into the help system and eventually designing an expertise-sensitive help system. An expertise-sensitive help system would detect user expertise dynamically and regularly so that systems could recommend help overtly to novices, subtly to average and poor users, and not at all to experts.<p> This thesis makes several advances in this area through a series of four experiments. In the first experiment, we show that users respond differently to help interruptions depending on their expertise with a task. Having established that user response to helpful interruptions varies with expertise level, in the second experiment we create a four-level classifier of task expertise with an accuracy of 90%. To present helpful interruptions differently to novice, poor, and average users, we need to design three interrupting notifications that vary in their attentional draw. In experiment three, we investigate a number of options and choose three icons. Finally, in experiment four, we integrate the expertise model and three interrupting notifications into an expertise-sensitive system-suggested help program, and investigate the user response. Together, these four experiments show that users value helpful interruptions when their expertise with a task is low, and that an expertise-sensitive help system that presents helpful interruptions with attentional draw that matches user expertise is effective and valuable.

Given the complexity and functionality of todays software, task-specific, system-suggested help could be beneficial for users. Although system-suggested help assists users in completing their tasks quickly, user response to unsolicited advice from their applications has been lukewarm. One such problem is lack of knowledge of system-suggested help about the users expertise with the task they are currently doing. This thesis examines the possibility of improving system-suggested help by adding knowledge about user expertise into the help system and eventually designing an expertise-sensitive help system. An expertise-sensitive help system would detect user expertise dynamically and regularly so that systems could recommend help overtly to novices, subtly to average and poor users, and not at all to experts.<p> This thesis makes several advances in this area through a series of four experiments. In the first experiment, we show that users respond differently to help interruptions depending on their expertise with a task. Having established that user response to helpful interruptions varies with expertise level, in the second experiment we create a four-level classifier of task expertise with an accuracy of 90%. To present helpful interruptions differently to novice, poor, and average users, we need to design three interrupting notifications that vary in their attentional draw. In experiment three, we investigate a number of options and choose three icons. Finally, in experiment four, we integrate the expertise model and three interrupting notifications into an expertise-sensitive system-suggested help program, and investigate the user response. Together, these four experiments show that users value helpful interruptions when their expertise with a task is low, and that an expertise-sensitive help system that presents helpful interruptions with attentional draw that matches user expertise is effective and valuable.

One of the key features of an online community network is that there is no central management authority; the community members themselves manage it. At the same time, for any application to be complete it must have a useful help system. So, for a community network to be completely run by the members, the task of creating and manipulating help documents must also be handled by the members/end-users. Previous studies about community networks show that extensive volunteer effort is one of the basic characteristics of a community network. Therefore a study about end-user authoring is possible in a community network. Minimalism is an instruction design method that helps users to learn about the system by performing real tasks. This study aimed at analyzing the possibilities of guiding the end-users to create a better minimalist help document than a more traditional and comprehensive one. The users' performance and preferences were used to compare the two approaches. The study also focused on users' preference to using minimalist help documents versus traditional help documents. The results indicated that it is possible to guide the end-users to create minimalist help documents. However, no significant results were found to conclude that the end-user authored minimalist help document would be better than an end-user authored traditional help document. The results also indicated that, although significant results were not found, the users seem to prefer a more traditional help, than a minimalist help, for a community network. The implications of the study and recommendations for future work are presented. / Master of Science

[pt] Este trabalho é baseado na teoria de Engenharia Semiótica para a qual a interface de uma aplicação é um ato de metacomunicação, uma mensagem do designer para o usuário, representando a visão do designer sobre o artefato por ele construído. Como o designer não pode estar presente fisicamente na interface, ele é representado por seu preposto. Este é dotado de uma capacidade comunicativa que lhe permite fazer um discurso completo e exclusivamente referente à conclusão final do designer sobre as necessidades, preferências, capacidades e oportunidades que ele entende que o usuário tem. Sendo impossível prever todas as interpretações que cada usuário pode dar para a aplicação, o designer precisa, através de seu preposto, explicar tudo o que fez (e por que motivo), e dar chance ao usuário de esclarecer suas dúvidas. A melhor alternativa de metacomunicação é sem dúvida o sistema de ajuda. Neste sentido, o trabalho aqui apresentado estende o modelo teórico atual da Engenharia Semiótica a fim de explicitar a presença do sistema de ajuda e o papel comunicativo do mesmo, objetivando fornecer aos designers de software ferramentas epistêmicas que possam apoiá-los na construção dos sistemas de ajuda de suas aplicações. Estas ferramentas permitem aos designers explorar o poder comunicativo da ajuda, auxiliando-os a refletir sobre os materiais disponíveis para sua elaboração (modelos de design, lógica de design, entre outros) e, a partir desta reflexão, auxiliando-os tanto na construção do discurso embutido na ajuda quanto na elaboração das possíveis formas de expressão do usuário frente a este discurso. / [en] This work is based on the Semiotic Engineering theory of Human-Computer Interaction. This theory views the application interface as a meta-communication act, a message from designer to user, representing the designers view of the artifact he developed. Since the designer cannot be present in the interface, he is represented by his deputy, who is endowed with a communicative capacity that allows it to carry out a discourse that is complete and exclusively regarding the designers final conclusions about the needs, preferences, capacities and opportunities that he understands the user has. Being impossible to predict all the interpretations that each user can generate for the application, the designer needs, through his deputy, to explain everything that he did (and why he did it), and to give the users a chance to clarify their doubts. The best alternative for metacommunication is, certainly, the help system. In this way, the work herein proposed extends the current theoretical model of Semiotic Engineering, making explicit the presence of the help system and its communicative role. The purpose of this extension is to provide software designers with epistemic tools to support them when constructing their application s help systems. These tools allow designers to explore help systems communicative power, supporting them in the reflection about the materials available for its elaboration (design models, design rationale, among others), thus supporting them both in the construction of the discourse embedded in the help system, and in the elaboration of the possible ways the user will be able to express himself within this discourse.