This introduction discusses the background and motivations for this special issue on Personal Information Management (PIM). We summarize accepted papers and outline future issues for PIM research.

Abstract In a controlled lab experiment, we compared how in-person and robot-mediated communicative settings affected attitudes towards communicators and discourse phenomena related to conversational negotiation. We used a mock interview within-participants experiment design where each participant (mock interviewee) experienced both types of communication with the same experimenter (mock interviewer). Despite communicating with the same person, participants found the in-person interviewer to be more likable, more capable, more intelligent, more polite, more in control, and less awkward than the same person using a telepresence robot. Behaviorally, we did not detect differences in participants’ productions of discourse phenomena (likes, you knows, ums, uhs), laughter, or gaze. We also tested the role of communicative expectations on attitudes towards communications. We primed participants to expect that they would be talking to a person via telepresence, a “disabled” robot-person combination using telepresence, or a person in person (between-participants). We did not find differences arising from people’s expectations of the communication.

How do Millennial and Gen-Z young adults decide between competing social media when choosing where to post? Previous research argues that decisions can be modeled based on these users' Goals, and the Affordances, Features1, and Social Norms of those media. To evaluate this model, 19 participants were given different self-presentation scenarios and asked to choose between Twitter, Snapchat, Instagram, and TikTok. Participants also constructed a flowchart to represent their mental model of factors influencing their choices. Data suggest revisions to the model; rather than reflecting Affordances and Features, we found that Social Norms and Goals primarily drove participants' choices. Participants choose media based on a direct match between Goals and existing Social Norms. Only when they failed to detect such a match did participants consider Affordances and Features. We present a revised conceptual model based on these results and discuss social media design and theory implications.

Many recent breakthroughs in medical diagnostics and drug discovery arise from deploying machine learning algorithms to large-scale data sets. However, a significant obstacle to such approaches is that they depend on high-quality annotations generated by domain experts. This study develops and evaluates BioLumin, a novel immersive mixed reality environment that enables users to virtually shrink down to the microscopic level for navigation and annotation of 3D reconstructed images. We discuss how domain experts were consulted in the specification of a pipeline to enable automatic reconstruction of biological models for mixed reality environments, driving the design of a 3DUI system to explore whether such a system allows accurate annotation of complex medical data by non-experts. To examine the usability and feasibility of BioLumin, we evaluated our prototype through a multi-stage mixed-method approach. First, three domain experts offered expert reviews, and subsequently, nineteen non-expert users performed representative annotation tasks in a controlled setting. The results indicated that the mixed reality system was learnable and that non-experts could generate high-quality 3D annotations after a short training session. Lastly, we discuss design considerations for future tools like BioLumin in medical and more general scientific contexts.

The COVID pandemic and social distancing have induced radical disruptions to work and learning practices. Working from home (WFH) has reduced offline interactions while increasing digital communica...

As global rates of anxiety and depression increase, we observe millions of downloads of mobile apps addressing mental health that adopt 'freemium' charging models offering complex combinations of free and paid features. We explore potentially negative outcomes of deploying such freemium designs to vulnerable MH populations. We assess outcomes for 41 frequently downloaded mental health apps, by combining thematic analysis of 41K user reviews with audits to validate app descriptions and reviews. We propose a new analytic framework identifying three types of negative outcomes for freemium deployments: first vulnerable users currently experiencing crises may feel pressure to download unhelpful or costly apps. Second unintuitive descriptions of complex apps can lead to inappropriate treatments and expensive subscriptions. Third, limited duration offers may result in incomplete treatments or unexpected charges. We discuss ethical considerations and broader HCI consequences. We describe implications for mental health app design, including the need for greater transparency around free versus paid features, and propose new approaches to provide vulnerable users with validated information about these commonly downloaded apps.

Conversational agents (CAs) are increasingly ubiquitous and are now commonly used to access medical information. However, we lack systematic data about the quality of advice such agents provide. This paper evaluates CA advice for mental health (MH) questions, a pressing issue given that we are undergoing a mental health crisis. Building on prior work, we define a new method to systematically evaluate mental health responses from CAs. We develop multi-utterance conversational probes derived from two widely used mental health diagnostic surveys, the PHQ-9 (Depression) and the GAD-7 (Anxiety). We evaluate the responses of two text-based chatbots and four voice assistants to determine whether CAs provide relevant responses and treatments. Evaluations were conducted both by clinicians and immersively by trained raters, yielding consistent results across all raters. Although advice and recommendations were generally low quality, they were better for Crisis probes and for probes concerning symptoms of Anxiety rather than Depression. Responses were slightly improved for text versus speech-based agents, and when CAs had access to extended dialogue context. Design implications include suggestions for improved responses through clarification sub-dialogues. Responses may also be improved by the incorporation of empathy although this needs to be combined with effective treatments or advice.

Imbuing robots with personality has been shown to be an effective design approach in HRI, promoting user trust and acceptance. We explore personality design in a non-anthropomorphic voice-assisted home robot. Our design approach developed three distinct robot personas: Butler, Buddy, and Sidekick, intended to differ in proactivity and emotional impact. Persona differences were signaled to users by a combination of humanoid (speech, intonation), and indirect cues (colors and movement). We use Big Five personality theory to evaluate perceived differences between personas in an exploratory Wizard of Oz study. Participants were largely able to recognize underlying personality traits expressed through these cue combinations in ways that were consistent with our design goals. The proactive Buddy persona was judged as more Extravert than the more passive Sidekick persona, and the Butler persona was perceived as more Conscientious and less Neurotic than either Buddy or Butler personas. Users also had clear preferences between different personas; they wanted robots that mimicked but accentuated their own personality. Results suggest that future designs might exploit abstract cues to signal personality traits.

We use personality theory to compare self-presentation between multiple Instagram accounts, investigating authenticity and consistency. Many studies claim social media promote inauthentic self-presentation focused on socially desirable traits. At the same time, affordances suggest that self-presentation should be relatively consistent within one social medium. For 88 participants, we examine personality traits for 'real Instagram' ('Rinsta') versus 'fake Instagram' ('Finsta') accounts, comparing these with people's offline traits using mixed-methods. Counterintuitively, we find Finsta accounts often present socially undesirable traits. Furthermore, different accounts on the same social medium reveal quite different styles of self-presentation. Overall Finstas are more Extraverted, less Conscientious, and less Agreeable than Rinstas, although equally Neurotic as offline. Interviews indicate trait differences arise from differing audience perceptions. A large anonymous Rinsta audience promotes a carefully curated self. In contrast, a small but trusted Finsta audience can engender more authentic, but negative self-presentation. We discuss design and theory implications.

Abstract File sharing is an integral component of modern work. Files can be shared either using Group Information Management (GIM), where collaborators exploit a common repository (e.g., the cloud), or Personal Information Management (PIM), where files are sent via email attachments, and collaborators store files individually in personal collections. Given the recent prevalence of GIM, we compare the effects on retrieval for PIM versus GIM collections. We examine the effects of various theoretically motivated factors relating to collection size, properties of the target file, and user workload. In our study, 289 participants accessed 1,557 of their own shared files in a naturalistic setting. Results indicate that factors relating to collection size, file versions, and user workload negatively affect the retrieval of GIM more than PIM files, indicating that PIM is more scalable than GIM. Testing a very different population, we confirm previous findings that failure percentages of GIM are approximately double those of PIM. We discuss possible theoretical explanations, specifically how factors that hinder retrieval exacerbate the general GIM problem of retrieving files organized by other people. Overall, PIM's greater scalability has practical implications for fast‐growing organizations such as startups when choosing their sharing policies.

It is essential that users understand how algorithmic decisions are made, as we increasingly delegate important decisions to intelligent systems. Prior work has often taken a techno-centric approach, focusing on new computational techniques to support transparency. In contrast, this article employs empirical methods to better understand user reactions to transparent systems to motivate user-centric designs for transparent systems. We assess user reactions to transparency feedback in four studies of an emotional analytics system. In Study 1, users anticipated that a transparent system would perform better but unexpectedly retracted this evaluation after experience with the system. Study 2 offers an explanation for this paradox by showing that the benefits of transparency are context dependent. On the one hand, transparency can help users form a model of the underlying algorithm's operation. On the other hand, positive accuracy perceptions may be undermined when transparency reveals algorithmic errors. Study 3 explored real-time reactions to transparency. Results confirmed Study 2, in showing that users are both more likely to consult transparency information and to experience greater system insights when formulating a model of system operation. Study 4 used qualitative methods to explore real-time user reactions to motivate transparency design principles. Results again suggest that users may benefit from initially simplified feedback that hides potential system errors and assists users in building working heuristics about system operation. We use these findings to motivate new progressive disclosure principles for transparency in intelligent systems and discuss theoretical implications.

As we increasingly delegate important decisions to intelligent systems, it is essential that users understand how algorithmic decisions are made. Prior work has often taken a technocentric approach to transparency. In contrast, we explore empirical user-centric methods to better understand user reactions to transparent systems. We assess user reactions to transparency in two studies. In Study 1, users anticipated that a more transparent system would perform better, but retracted this evaluation after experience with the system. Qualitative data suggest this arose because transparency is distracting and undermines simple heuristics users form about system operation. Study 2 explored these effects in depth, suggesting that users may benefit from initially simplified feedback that hides potential system errors and assists users in building working heuristics about system operation. We use these findings to motivate new progressive disclosure principles for transparency in intelligent systems.

We investigate self-perception in social media through the lens of personality theory. Two mixed-methods studies involving 148 participants examine if people self-report different personality traits in social media compared with their offline traits. We first compare offline and Facebook traits, finding that on Facebook people are less Neurotic, Open and Agreeable. A second study compared offline, Facebook and Snapchat traits, replicating and extending our initial results. Again Facebook personality was less Neurotic and less Open than offline. In contrast, Snapchat personality was more Extravert than both offline and Facebook and more Open than Facebook. Interviews indicate how personality differences arise from social media affordances. Anxiety about audience judgments leads people to curate posts to appear less Neurotic on social media, but the transience of Snapchat promotes greater Extraversion than offline and Facebook. We discuss theory and design implications.

Upgrades are a common experience for people who use technology for any length of time. Upgrades may bring clear benefits, and some users look forward to new versions of software. However for others, upgrades may be a burden. Upgrades bring costs, imposing a need to refamiliarize ourselves with a new version after mastering the old one. It is important to understand such costs as failure to upgrade can make users vulnerable to security risks. We first review multiple studies showing users' dislike of upgrades resulting from changes introduced to the interface. Then we propose a framework to analyze the nature of upgrade costs, explaining their origins using well-validated cognitive psychology concepts including: cognitive automation, negative transfer, cognitive maps, procedural memory and locus of control. We use these concepts to explain the psychological costs of frequent interface changes while sketching some approaches to addressing them. Although new versions are necessary for progress, this paper provides software developers with a deeper understanding of the sometimes overlooked costs of frequent interface changes.

Successful online communities accumulate large amounts of long-term content. However there has been little quantitative, theoretically-motivated exploration of how communities organize such content, nor which community members take responsibility for active organization. We examine one aspect of long-term content organization through link behavior, also exploring role differences between enterprise community leaders and members in the context of life-cycle community models. We first classify how content is linked within posts, identifying usage patterns that organize information within and outside communities. We next present an exploratory quantitative analysis of 2,010 communities including 428,476 posts and 1,246,570 links. We show paradoxically that although mature communities accumulate substantial content, organizing that content using links decreases over time. Further analyses suggest that this arises from a recency bias, with communities being focused on current content. Our results also challenge descriptive lifecycle community models, which propose that regular community members adopt greater responsibility over time. We explore explanations for our findings and implications including new tools that encourage responsibility for active organization, as well as methods for members to revisit critical content.

Stephanie Lukin, Pranav Anand, Marilyn Walker, Steve Whittaker. Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers. 2017.

We explore the Examined Life, informing the design of reflective systems to promote emotional well-being, a critical health issue. People now have increasingly rich, digital records of highly personal data about what they said, did, and felt in the past. But social science research shows that people have difficulty in tracking and regulating their emotions. New reflective technologies that promote constructive analysis of rich personal data potentially offer transformative ways that individuals might better understand themselves and improve well-being. However, there are important system design challenges in supporting effective reflection about personal data. We explore fidelity in recording and representing past personal mood data, and forecasting future actions, feelings, and thoughts. Much prior personal informatics work has been dedicated to past-centric tools for recording and capture. In contrast, forecasting examines how we might use such past data to inform and motivate our future selves, providing recommendations about remedial actions to improve future well-being. Fidelity addresses both how and what reflective systems should show people about their pasts, in particular whether we should filter negative past experiences. To inform reflective system design, we examine forecasting and fidelity in controlled field trial interventions that explore two novel system designs for presenting and reflecting on mood data. We detail findings from 165 participants, 4,693 participant logfiles, 65 surveys, and 15 user interviews. Our novel forecasting system, EmotiCal, uses past mood data to model and visualize future user moods with the goal of encouraging participants to adopt remedial new behaviors to regulate negative moods before they occur. Such forecasting both improved mood and subsequent emotional self-awareness compared with controls who simply monitored their past. Consistent with system goals, interview responses also indicated that participants generated important insights into behaviors that affect their moods. Our second intervention examined filtering; it assessed the impact on well-being of recording and revisiting past experiences containing negative emotions. We compared participants who were encouraged to record and reflect on positive versus negative experiences. Long-term measures of happiness and ruminative behaviors improved by recording and reflecting on positive but not negative experiences, although this depended on the intensity of the negative experience. We discuss general design and theory implications for future systems that support monitoring, reflection, and forecasting to facilitate productive examination of our emotional lives.

Maintaining work focus when on a computer is a major challenge, and people often feel that they use their time ineffectively. To improve focus we designed meTime, a real-time awareness application that shows users how they allocate their time across applications. In two real-world deployments involving 118 participants, we examined whether greater awareness of time use improves focus. In our first deployment, we provided awareness information using meTime, to both office workers and students. Exposure to meTime reduced use of social media, email, browsing and total time online. However increased awareness didn't affect time spent in productivity applications. A second educational deployment largely replicated these results and showed that meTime also reduced users' perceptions of their ability to focus effectively. Changed perceptions were associated with higher class grades. We discuss practical and theoretical implications as well as design principles for use of time applications.

Research on small team collaboration repeatedly shows that 'distance matters'. More recent work has refined this concept of distance to develop geographic dispersion measures to explain the negative effects that team configuration has on productivity and interaction. Dispersion measures explain why teams distributed across multiple time zones, or across multiple sites, have more coordination difficulties than collocated teams with a single remote member. Although larger online communities are increasingly used in enterprises, few studies have examined the effects of dispersion on community behavior. We studied 1206 online enterprise communities (OECs), each using a set of collaborative tools. We present new data showing counter-intuitively that OEC dispersion does not affect content generation or contribution inequality, even when restricting community size to those that most resemble small teams (with 3-12 members). We found that previously documented negative effects of geographic dispersion seem to be reduced in enterprise communities regardless of size. Our results provide additional support to prior case studies suggesting that online communities can mitigate geographic dispersion by providing resources that support coordination and resource sharing.

Technology increasingly allows us to capture and revisit rich digital records of our lives, processes which we call Technology-Mediated Memory (TMM). We explore whether TMM alters unmediated remembering and also whether such changes affect psychological well-being. Human memory biases promote well-being by adaptively editing our memories, making them more positive. In contrast, TMM often provides rich records of what people actually did and felt, which could disrupt adaptive edits. To explore this, we developed a smartphone-based personal TMM application, Echo, that allows participants to record and later reflect on everyday events. In a month-long deployment, 64 users made over 3200 recordings and reflections. We found that although Echo TMM alters how we remember, these changes remain adaptive. Instead of compromising adaptive biases, Echo TMM helps well-being and benefits are sustained long-term. Logfile analysis shows that participants use Echo strategically to prospectively edit by initially reporting events positively to anticipate future viewing. Participants also distance themselves from past negative events by reflecting more positively than at recording. We discuss design and theoretical implications.

People increasingly live their lives online, accruing large collections of digital possessions, which symbolically represent important relationships, events, and activities. Most HCI research on bereavement focuses onretainingthese significant digital possessions to honor the departed. However, recent work suggests that significant digital possessions may complicate moving on; they function as both comforting and painful reminders but currently provide inflexible methods for disposal. A few works have investigated the disposal of digital objects as a means ofletting go. To better understand this we interviewed 10 psychotherapists who employ rituals of letting go to help patients overcome loss in situations such as a divorce, a breakup, or a stillbirth. Patients disposed of either natural artifacts or symbolic personal possessions through actions such as burning, burying, or placing in a body of water. Therapists noted that people increasingly have digital possessions, and that the act of deletion does not offer the same cathartic sense of release as disposal of material artifacts. Based on the analysis of this grief therapy, we propose a new conceptual framework for rituals of letting go that highlights temporality, visibility, and force. It provides a vocabulary to talk about disposal. We then offer design implications connecting the rituals of letting go to the disposal of digital things. Based on our interviews and analytic framework, we propose novel technologies that better connect the embodied nature of letting go rituals to the process of digital disposal.

Preventable behaviors contribute to many life threatening health problems. Behavior-change technologies have been deployed to modify these, but such systems typically draw on traditional behavioral theories that overlook affect. We examine the importance of emotion tracking for behavior change. First, we conducted interviews to explore how emotions influence unwanted behaviors. Next, we deployed a system intervention, in which 35 participants logged information for a self-selected, unwanted behavior (e.g., smoking or overeating) over 21 days. 16 participants engaged in standard behavior tracking using a Fact-Focused system to record objective information about goals. 19 participants used an Emotion-Focused system to record emotional consequences of behaviors. Emotion-Focused logging promoted more successful behavior change and analysis of logfiles revealed mechanisms for success: greater engagement of negative affect for unsuccessful days and increased insight were key to motivating change. We present design implications to improve behavior-change technologies with emotion tracking.

Automated coaching systems offer a convenient, cost-effective way to reduce stress, which can be a serious health issue. However, one concern with such systems is compliance; users fail to achieve daily stress reduction goals because goals are too easy or too difficult. To address this, we built DStress (Design for Stress), a theoretically grounded system that sets adaptive goals in three coaching dimensions: Exercise, Meditation and Accessibility. DStress modifies goal-difficulty based on the individual's immediately previous performance. In a 28-day deployment with 65 users, DStress reduced scores on one direct measure of stress almost in half, significantly more than two other non-adaptive coaching strategies. However, on a second direct stress measure, no improvement was found. There were also no improvements on other indirect stress measures. Analysis of 2842 user-generated reports suggests our findings were the result of DStress balancing compliance against the degree of challenge of the goals it would set.

Recent research has developed analytics that threaten online self-presentation and privacy by automatically generating profiles of individuals' most personal traits-their personality, values, motivations, and so on. But we know little about people's reactions to personal traits profiles of themselves, or what influences their decisions to share such profiles. We present an early qualitative study of people's reactions to a working hyper-personal analytics system. The system lets them see their personality and values profile derived from their own social media text. Our results reveal a paradox. Participants found their personal traits profiles creepily accurate and did not like sharing them in many situations. However, they felt pressured by the social risks of not sharing and showed signs of learned helplessness, leading them to share despite their misgivings. Further, they felt unqualified to significantly modify their profile contents due to a surprising trust in the "expert" algorithm. We explore design implications for hyper-personal analytics systems that consider the needs and preferences of the people being profiled, suggesting ways to enhance the control they feel and the benefits they reap.

Many teams are now adopting shared repositories for their work. Such adoption is paradoxical, however, as past research has repeatedly shown major co-organizational barriers; teams cannot agree a common organizational scheme, making it difficult to retrieve information organized by others. Another barrier is email competition; email provides a reliable alternative for distributing files that are then personally organized. To address this paradox, we explored how 27 participants actively using shared repositories overcome these barriers in a qualitative study. We found teams addressed co-organization using 4 strategies. First they create ContentMaps that provide explicit structure to organize shared information. Participants also co-organize using implicit strategies based on task structure, expertise, and tool affordances. Greater shared repository use also leads to a changed role for email. Versioning problems mean email is not used for distributing attachments, instead for task management. We present technical implications suggesting how new tools might be better integrated with email facilitating these continued email uses.

Individual differences are prevalent in personal information management (PIM). There is large variation between individuals in how they structure and retrieve information from personal archives. These differences make it hard to develop general PIM tools. However we know little about the origins of these differences. We present two studies evaluating whether differences arise from personality traits, by exploring whether different personalities structure personal archives differently. The first exploratory study asks participants to identify PIM cues that signal personality traits. While the aim was to identify cues, these cues also proved surprisingly accurate indicators of personality. In a second study, to evaluate these cues, we directly measure relations between structure and traits. We demonstrate that Conscientiousness predicts file organization, particularly PC users' desktops. Neurotic people may also keep more desktop files. One implication is that systems might be customized for different personalities. We also advance personality theory, showing that personal digital artifacts signal personality.

Online communities are successful only if they achieve their goals, but there has been little direct study of goals. We analyze novel data characterizing the goals of enterprise online communities, assessing the importance of goals for leaders, how goals influence member perceptions of community value, and how goals relate to success measures proposed in the literature. We find that most communities have multiple goals and common goals are learning, reuse of resources, collaboration, networking, influencing change, and innovation. Leaders and members agree that all of these goals are important, but their perceptions of success on goals do not align with each other, or with commonly used behavioral success measures. We conclude that simple behavioral measures and leader perceptions are not good success metrics, and propose alternatives based on specific goals members and leaders judge most important.

A central tenet of HCI is that technology should be user-centric, with designs being based around social science findings about users. Nevertheless a repeated but critical challenge in design is translating empirical findings into actionable ideas that inform design, or generating implications for design. Despite various design methods aiming to bridge this gap, knowledge informing design is still seen as problematic. However there has been little empirical exploration into what design researchers understand by such design knowledge, the functions and principles behind their creation. We report on interviews with twelve expert HCI design researchers probing the roles and types of design implications, and the process of generating and evaluating them. We synthesize different types of design implications into a framework to guide their generation. Our findings identify a broader range than previously described, additional sources and heuristics supporting their development as well some important evaluation criteria. We discuss the value of these findings for interaction design research.

Most studies of social tools examine usage of each tool in isolation. Instead, we explore how online communities (a) combine multiple social tools, and (b) use social tools together with external tools. Based on interviews with community leaders and quantitative analysis of 128 online community spaces, we explored the combined use of six social software tools--wikis, blogs, forums, social bookmarks, social file repositories, and task-management tools. We contribute a detailed characterization of how enterprise online communities combine multiple social tools, adding to our understanding of community behaviors: Communities combine social tools to curate and organize complex information spaces. When combined, each tool is used for limited "core" functions; thus 'social' features are not always leveraged for every tool. Leaders and members divide labor by tool boundaries. Our results suggest that an important overlooked aspect of social media concerns how different tools can be effectively combined. While most prior work on communities emphasizes end user content, we identify additional important design activities where community participants curate and organize pre-existing content from multiple tools to serve their community needs.

People who are collaborating can share files in two main ways: performing Group Information Management (GIM) using a common repository or performing Personal Information Management (PIM) by distributing files as e‐mail attachments and storing them in personal repositories. There is a trend toward using common repositories with many organizations encouraging workers to use GIM to avoid duplication of files and management. So far, PIM and GIM have been studied by different research communities, so their effectiveness for file retrieval has not yet been systematically compared. We compared PIM and GIM in a large‐scale elicited personal information retrieval study. We asked 275 users to retrieve 860 of their own shared files, testing the effect of sharing method on success and efficiency of retrieval. Participants preferred PIM over GIM. More important, PIM retrieval was more successful: Participants using GIM failed to find 22% of their files compared with 13% failures using PIM. This may be because active organization aids retrieval: When using personally created folders, the failure percentage was 65% lower than when using default folders (e.g., My Documents), and more than 5 times lower than when using folders created by others for GIM. Theoretical reasons for this are discussed.

As people document more of their lives online, some recent systems are encouraging people to later revisit those recordings, a practice we're calling technology-mediated reflection (TMR). Since we know that unmediated reflection benefits psychological well-being, we explored whether and how TMR affects well-being. We built Echo, a smartphone application for recording everyday experiences and reflecting on them later. We conducted three system deployments with 44 users who generated over 12,000 recordings and reflections. We found that TMR improves well-being as assessed by four psychological metrics. By analyzing the content of these entries we discovered two mechanisms that explain this improvement. We also report benefits of very long-term TMR.

Online communities are increasingly being deployed in enterprises to increase productivity and share expertise. Community leaders are critical for fostering successful communities, but existing technologies rarely support leaders directly, both because of a lack of clear data about leader needs, and because existing tools are member- rather than leader-centric. We present the evidence-based design and evaluation of a novel tool for community leaders, Community Insights (CI). CI provides actionable analytics that help community leaders foster healthy communities, providing value to both members and the organization. We describe empirical and system contributions derived from a long-term deployment of CI to leaders of 470 communities over 10 months. Empirical contributions include new data showing: (a) which metrics are most useful for leaders to assess community health, (b) the need for and how to design actionable metrics, (c) the need for and how to design contextualized analytics to support sensemaking about community data. These findings motivate a novel community system that provides leaders with useful, actionable and contextualized analytics.

People are increasingly acquiring huge collections of digital possessions. Despite some pleas for 'forgetting', most theorists argue for retaining all these possessions to enhance 'total recall' of our everyday lives. However, there has been little exploration of the negative role of digital possessions when people want to forget aspects of their lives. We report on interviews with 24 people about their possessions after a romantic breakup. We found that digital possessions were often evocative and upsetting in this context, leading to distinct disposal strategies with different outcomes. We advance theory by finding strong evidence for the value of intentional forgetting and provide new data about complex practices associated with the disposal of digital possessions. Our findings led to a number of design implications to help people better manage this process, including automatic harvesting of digital possessions, tools for self-control, artifact crafting as sense-making, and digital spaces for shared possessions.

Folder navigation is the main way that computer users retrieve their personal files. However we know surprisingly little about navigation, particularly about how it is affected by the operating system used, the interface presentation and the folder structure. To investigate this, we asked 289 participants to retrieve 1,109 of their own active files. We analyzed the 4,948 resulting retrieval steps, i.e. moves through the hierarchical folder tree. Results show: (a) significant differences in overall retrieval time between PC and Mac that arise from different organizational strategies rather than interface design; (b) the default Windows presentation is suboptimal - if changed, retrieval time could be reduced substantially and (c) contrary to our expectations, folder depth did not affect step duration. We discuss possible reasons for these results and suggest directions for future research.

Collaboration personas are a tool that can be used to design for groups. Prior work posits that collaboration personas can improve tool adoption by helping designers create collaboration tools that are better targeted to the goals, needs, and interactions between members of collaborative groups. We present a comparative study of design and user experience practitioners who used both collaboration personas and individual personas. Participants conducted a cognitive walkthrough and provided redesign suggestions for a collaboration tool. Our results show that the focus of the cognitive walkthrough and redesign task differed, with collaboration personas showing more group focus. Collaboration personas led to a more complete discussion, as indicated by a greater amount of time spent on the task compared to individual personas. Despite prior experience and training with individual personas, collaboration personas were preferred and better supported the task, since they focused on groups of people and their interactions.

Personas are a critical method for orienting design and development teams to user experience. Prior work has noted challenges in justifying them to developers. In contrast, it has been assumed that designers and user experience professionals - whose goal is to focus designs on targeted users - will readily exploit personas. This paper examines that assumption. We present the first study of how experienced user-centered design (UCD) practitioners with prior experience deploying personas, use and perceive personas in industrial software design. We identify limits to the persona approach in the context studied. Practitioners used personas almost exclusively for communication, but not for design. Participants identified four problems with personas, finding them abstract, impersonal, misleading and distracting. Our findings argue for a new approach to persona deployment and construction. Personas cannot replace immersion in actual user data. And rather than focusing on creating engaging personas, it is critical to avoid persona attributes that mislead or distract.

Meeting participants can experience cognitive overload when they need both to verbally contribute to ongoing discussion while simultaneously creating notes to promote later recall of decisions made during the meeting. We designed two novel cuing tools to reduce the cognitive load associated with note-taking, thus improving verbal contributions in meetings. The tools combine real-time automatic speech recognition (ASR) with lightweight annotation to transform note-taking into a low overhead markup process. To create lightweight notes, users do not generate the notes' content themselves. Instead they simply highlight important phrases in a real-time ASR transcript (Highlighter tool), or press a button to indicate when they heard something important (Hotspots tool). We evaluated these markup tools against a traditional pen-and-paper baseline with 26 users. Hotspots was highly successful: compared with handwritten notes, it increased participants' conversational contributions and reduced their perception of overload in the meeting, while improving recall of the meeting two months later. Highlighter also improved recall without compromising conversational contributions, although users found it more demanding.

This special issue focuses on new uses of digital media to help people remember in everyday situations. We begin this introduction by describing the field’s origins (personal memories past), using this to contextualise the papers presented here (personal memories present). We conclude by identifying a number of important research challenges that we feelmust be addressed by future work in this area (personal memories future).

Lifelogging is a technically inspired approach that attempts to address the problem of human forgetting by developing systems that ‘record everything’. Uptake of lifelogging systems has generally been disappointing, however. One reason for this lack of uptake is the absence of design principles for developing digital systems to support memory. Synthesising multiple studies, we identify and evaluate 4 new empirically motivated design principles for lifelogging: Selectivity, Embodiment, Synergy and Reminiscence. We first summarise 4 empirical studies that motivate the principles, then describe the evaluation of 4 novel systems built to embody these principles. The design principles were generative, leading to the development of new classes of lifelogging system, as well as providing strategic guidance about how those systems should be built. Evaluations suggest support for Selection and Embodiment principles, but more conceptual and technical work is needed to refine the Synergy and Reminiscence principles.

New enterprise tools (wikis, team spaces, social tags) offer potential benefits for enterprise collaboration, providing shared resources to organize work. However, a vast amount of collaboration still takes place by email. But email is problematic for collaboration because information may be distributed across multiple messages in an overloaded inbox. Email also increases workload as each individual has to manage their own versions of collaborative materials. We present a novel system, Topika that integrates email with collaboration tools. It allows users to continue to use email while also enjoying the benefits of these dedicated tools. When a user composes an email Topika analyzes the message and suggests relevant shared spaces (e.g., wiki pages) within the user's collaboration tools. This allows her to post the email to those spaces. An evaluation of Topika's suggestion algorithm shows that it performs well at accurately suggesting shared spaces.

The success of social computing has generated a host of workplace collaboration tools. However, adoption of these tools by entire groups is a major problem. One reason for the adoption problem is a lack of methods for considering collaborative groups in technology design. Even when designing collaboration tools, designers often employ methods that focus on individuals. This leads to tools that are not well targeted at the groups who will use them. To solve this problem, we propose the notion of collaboration personas, which are empirically derived descriptions of hypothetical groups, including details that inform the design of collaboration tools. Collaboration personas differ from individual personas in having (1) multiple, inter-related individuals playing specific roles; (2) a focus on collective goals and elaboration of individual goals that affect the collective goal; and (3) new attributes that characterize collaborative aspects of the group's work. We contrast collaboration personas with other design approaches and provide examples of how they can be used to design new collaborative tools that better meet the needs of typical groups.

We all spend time every day looking for information in our email, yet we know little about this refinding process. Some users expend considerable preparatory effort creating complex folder structures to promote effective refinding. However modern email clients provide alternative opportunistic methods for access, such as search and threading, that promise to reduce the need to manually prepare. To compare these different refinding strategies, we instrumented a modern email client that supports search, folders, tagging and threading. We carried out a field study of 345 long-term users who conducted over 85,000 refinding actions. Our data support opportunistic access. People who create complex folders indeed rely on these for retrieval, but these preparatory behaviors are inefficient and do not improve retrieval success. In contrast, both search and threading promote more effective finding. We present design implications: current search-based clients ignore scrolling, the most prevalent refinding behavior, and threading approaches need to be extended.

The rapid increase in smart phone capabilities has introduced new opportunities for mobile information access and computing. However, smart phone use may still be constrained by both device affordances and work environments. To understand how current business users employ smart phones and to identify opportunities for improving business smart phone use, we conducted two studies of actual and perceived performance of standard work tasks. Our studies involved 243 smart phone users from a large corporation. We intentionally chose users who primarily work with desktops and laptops, as these "non-mobile" users represent the largest population of business users. Our results go beyond the general intuition that smart phones are better for consuming than producing information: we provide concrete measurements that show how fast reading is on phones and how much slower and more effortful text entry is on phones than on computers. We also demonstrate that security mechanisms are a significant barrier to wider business smart phone use. We offer design suggestions to overcome these barriers.

Lifelogging technologies can capture both mundane and important experiences in our daily lives, resulting in a rich record of the places we visit and the things we see. This study moves beyond technology demonstrations, in aiming to better understand how and why different types of Lifelogs aid memory. Previous work has demonstrated that Lifelogs can aid recall, but that they do many other things too. They can help us look back at the past in new ways, or to reconstruct what we did in our lives, even if we don't recall exact details. Here we extend the notion of Lifelogging to include locational information. We augment streams of Lifelog images with geographic data to examine how different types of data (visual or locational) might affect memory. Our results show that visual cues promote detailed memories (akin to recollection). In contrast locational information supports inferential processes -- allowing participants to reconstruct habits in their behaviour.

Digital mementos are increasingly problematic, as people acquire large amounts of digital belongings that are hard to access and often forgotten. Based on fieldwork with 10 families, we designed a new type of embodied digital memento, the FM Radio. It allows families to access and play sonic mementos of their previous holidays. We describe our underlying design motivation where recordings are presented as a series of channels on an old fashioned radio. User feedback suggests that the device met our design goals: being playful and intriguing, easy to use and social. It facilitated family interaction, and allowed ready access to mementos, thus sharing many of the properties of physical mementos that we intended to trigger.

Many studies have explored social processes and technologies associated with sharing photos. In contrast, we explore the role of sound as a medium for social reminiscing. We involved 10 families in recording 'sonic souvenirs' of their holidays. They shared and discussed their collections on their return. We compared these sounds with their photo taking activities and reminiscences. Both sounds and pictures triggered active collaborative reminiscing, and attempts to capture iconic representations of events. However sounds differed from photos in that they were more varied, familial and creative. Further, they often expressed the negative or mundane in order to be 'true to life', and were harder to interpret than photos. Finally we saw little use of pure explanatory narrative. We reflect on the relations between sound and family memory and propose new designs on the basis of our findings, to better support the sharing and manipulation of social sounds.

People are often required to catch up on information they have missed in meetings, because of lateness or scheduling conflicts. Catching up is a complex cognitive process where people try to understand the current conversation without access to prior discussion. We develop and evaluate a novel Catchup audio player that allows time-travel. It automatically identifies the gist of what was missed, allowing people to join the meeting late and still participate effectively. In a lab study, we evaluated people's understanding of meetings they had partly missed, by asking questions about meeting content. We tested whether providing Catchup gist overcomes the potential disadvantage that people must join even later - because catching up takes time. Catchup users understood meetings 70% better than controls who simply joined late. They were more confident of their understanding and indicated a positive attitude towards the tool. We are currently exploring more general applications of the time-travel approach.

Abstract Folder navigation is the main way that personal computer users retrieve their own files. People dedicate considerable time to creating systematic structures to facilitate such retrieval. Despite the prevalence of both manual organization and navigation, there is very little systematic data about how people actually carry out navigation, or about the relation between organization structure and retrieval parameters. The aims of our research were therefore to study users' folder structure, personal file navigation, and the relations between them. We asked 296 participants to retrieve 1,131 of their active files and analyzed each of the 5,035 navigation steps in these retrievals. Folder structures were found to be shallow (files were retrieved from mean depth of 2.86 folders), with small folders (a mean of 11.82 files per folder) containing many subfolders (M=10.64). Navigation was largely successful and efficient with participants successfully accessing 94% of their files and taking 14.76 seconds to do this on average. Retrieval time and success depended on folder size and depth. We therefore found the users' decision to avoid both deep structure and large folders to be adaptive. Finally, we used a predictive model to formulate the effect of folder depth and folder size on retrieval time, and suggested an optimization point in this trade‐off.

Users find it hard to delete unimportant personal information which often results in cluttered workspaces. We present a full design cycle for GrayArea, a novel interface that allows users to demote unimportant files by dragging them to a gray area at the bottom of their file folders. Demotion is an intermediate option between keeping and deleting. It combines the advantages of deletion (unimportant files don't compete for attention) and keeping (files are retrieved in their folder context). We developed the GrayArea working prototype using thorough iterative design. We evaluated it by asking 96 participants to 'clean' two folders with, and without, GrayArea. Using GrayArea reduced folder clutter by 13%. Further, 81% of participants found it easier to demote than delete files, and most indicated they would use GrayArea if provided in their operating systems. The results provide strong evidence for the demotion principle suggested by the user-subjective approach.

Lifelogging' technology makes it possible to amass digital data about every aspect of our everyday lives. Instead of focusing on such technical possibilities, here we investigate the way people compose long-term mnemonic representations of their lives. We asked 10 families to create a time capsule, a collection of objects used to trigger remembering in the distant future. Our results show that contrary to the lifelogging view, people are less interested in exhaustively digitally recording their past than in reconstructing it from carefully selected cues that are often physical objects. Time capsules were highly expressive and personal, many objects were made explicitly for inclusion, however with little object annotation. We use these findings to propose principles for designing technology that supports the active reconstruction of our future past.

Multitasking during teleconferences is becoming increasingly common: participants continue their work whilst monitoring the audio for topics of interest. Our previous work has estab-lished the benefit of spatialised audio presentation on improv-ing multitasking performance. In this study, we investigate the different spatialisation strategies employed by subjects in order to aid their multitasking performance and improve their user experience. Subjects were given the freedom to place each par-ticipant at a different location in the acoustic space both in terms of azimuth and distance. Their strategies were based upon cues regarding keywords and which participant will utter them. Our findings suggest that subjects employ consistent strategies with regard to the location of target and distracter talkers. Further-more, manipulation of the acoustic space plays an important role in multitasking performance and the user experience. Index Terms: multitasking, spatialisation, teleconference 1.

Identifying highlights in multimedia content such as video and audio is currently a very difficult technical problem. We present and evaluate a novel algorithm that identifies highlights by combining content analysis with Web 2.0 data mining techniques. We exploit the fact that popular content tends to be redundantly uploaded onto community sharing sites. Our "social summarization" technique first identifies overlaps in uploaded scenes and then uses the upload frequency of each video scene to compute that scene's importance in the complete video. Our user evaluation shows the reliability of the technique: scenes automatically selected by our method are agreed by experts to be the most relevant.

We all encounter many documents on a daily basis that we do not have time to process in their entirety. Nevertheless, we lack good tools to rapidly skim and identify key information from within such documents. This paper develops and evaluates Interactive Compression (IC) techniques that allow users to dynamically configure the amount of information they view in a document, e.g. by automatically removing unimportant information from view (Excision) or by making important information more salient (Highlighting). We explore IC techniques in the context of meeting transcripts that are typically unstructured - making it difficult to isolate relevant regions and extract key information. We demonstrate the superiority of IC compared with an unmodified text control. In contrast to traditional summaries, our results show extensive use of interactive, as opposed to fixed compression level, summarization. They also show the value of word- as opposed to utterance-based compression. There are also trade-offs between different IC designs. Excision allows users to scan documents faster than Highlighting but at the expense of overlooking relevant sections of the document.

Current technology makes it possible to capture huge amounts of information related to everyday experiences. Despite this, we know little about the processes by which people identify and manage mementos - objects which are directly meaningful to their memories. Among the millions of objects people encounter in a lifetime, few become such reminders of people, places or events. We report fieldwork where participants gave us a tour of their homes describing how and why particular objects become mementos. Our findings extend the existing digital memory literature; first our participants didn't view their activities as experiential 'capture', nor were mementos limited to pictorial representations of people and events; instead they included everyday objects. Furthermore, mementos were not only displayed and shared, but also integrated into everyday activities. Finally there were complex relations between house location and memento type. We discuss the theoretical and technical implications of our work.

We extend the notion of social tagging to construct social summaries of complex multimedia materials. Our system allows students to apply time-indexed multimedia tags such as handwritten annotations or photos to different parts of lecture recordings. These tags can be used to straightforwardly access different parts of the lecture. The social component of the interface presents information about which tags are most frequently accessed by others: allowing students to infer those parts of the lecture of most interest to others. We demonstrate the utility of the approach in a 6 week fieldwork study. Social summaries are used much more than corresponding systems that do not provide social information. In addition, social tool use was correlated with high course marks.

Teleconference participants often multitask: they work on a text-based ‘foreground’ task whilst listening in the ‘background’ for an item of interest to appear. Audio material should therefore be presented in a manner that has the smallest possible impact on the foreground task without affecting topic detection. Here, we ask whether dichotic or spatialised audio presentation of a meeting is less disruptive than the single-channel mixture of talkers normally used in teleconference audio. A number of talker location configurations are used, and we examine how these impact upon a text-based foreground task: finding all letter ‘e’ occurrences in a block of text. Additionally, we examine the effect of cueing the listener to direction or gender and record listener preferences for audio presentation style. Our results suggest that spatialised audio disrupts the foreground task less than single-channel audio when direction or gender is cued.

Efficient browsing of speech recordings is problematic. The linear nature of speech, coupled with the lack of abstraction that the medium affords, means that listeners have to listen to long segments of a recording to locate points of interest. We explore temporal compression algorithms that attempt to reduce the amount of time users require to listen to speech recordings, while retaining the important content. This paper implements two main approaches to temporal compression: artificial speech rate alteration (speed-up) and unimportant segment removal (excision). We evaluate the effectiveness of these approaches by having listeners rate comprehension and listening effort for different types of temporal compression. For different compression levels, we compare performance of various implementations of speed-up and excision as well as techniques based on semantic features and acoustic features. Our results indicate that listeners prefer low compression levels, excision over speed-up, and algorithms based on semantic rather than acoustic features. Finally, listeners were negative about hybrid algorithms that used speed-up to indicate missing regions in an excised recording.

Traditionally users access their personal files mainly by using folder navigation. We evaluate whether recent improvements in desktop search have changed this fundamental aspect of Personal Information Management (PIM). We tested this in two studies using the same questionnaire: (a) The Windows Studya longitudinal comparison ofGoogle DesktopandWindows XP Search Companion, and (b) The Mac Studya large scale comparison of MacSpotlightandSherlock. There were few effects for improved search. First, regardless of search engine, there was a strong navigation preference: on average, users estimated that they used navigation for 56-68% of file retrieval events but searched for only 4-15% of events. Second, the effect of improving the quality of the search engine on search usage was limited and inconsistent. Third, search was used mainly as a last resort when users could not remember file location. Finally, there was no evidence that using improved desktop search engines leads people to change their filing habits to become less reliant on hierarchical file organization. We conclude by offering theoretical explanations for navigation preference, relating to differences between PIM and Internet retrieval, and suggest alternative design directions for PIM systems.

Our lives are full of memorable and important moments, as well as important items of information. The last few years have seen the proliferation of digital devices intended to support prosthetic memory (PM), to help users recall experiences, conversations and retrieve personal information. We nevertheless have little systematic understanding of when and why people might use such devices, in preference to their own organic memory (OM). Although OM is fallible, it may be more efficient than accessing information from a complex PM device. We report a controlled lab study which investigates when and why people use PM and OM. We found that PM use depended on users' evaluation of the quality of their OM, as well as PM device properties. In particular, we found that users trade-off Accuracy and Efficiency, preferring rapid access to potentially inaccurate information over laborious access to accurate information. We discuss the implications of these results for future PM design and theory. Rather than replacing OM, future PM designs need to focus on allowing OM and PM to work in synergy.

We present an empirical study of teams that revealed the amount of extraneous individual work needed to enable collaboration: finding references to other people, finding files to attach to email, managing incoming email attachments, managing the variety of files used in shared activities, and tracking what work is owed to others. Much of this work involves finding recently accessed objects that are needed again in the user's current task focus. These observations led to the design of Recent Shortcuts, a tool to help support coordination by making recently used objects easily accessible. Recent Shortcuts enables quick access to people (including groups of people), received attachments, files, and file folders that the user interacted with recently for re-use in the user's current context. Recent Shortcuts makes it easy to use these objects across applications with no additional user input and minimal changes to the user's applications or work practice. Early user experiences with a working prototype led to an extension that integrates recently accessed objects across multiple devices.

Although speech is a potentially rich information source, a major barrier to exploiting speech archives is the lack of useful tools for efficiently accessing lengthy speech recordings. This paper develops and evaluates techniques for temporal compression - reducing the time people take to listen to a recording while still extracting critical information. We first describe an exploratory study that identifies novel excision techniques that remove unimportant words or utterances from the recording. We then develop a new method for evaluating how well temporal compression supports users in forming a general understanding of a recording. Applying this method, we demonstrate that excision techniques are generally more effective than standard compression techniques that simply speed up the entire recording.

We present a novel technique for time-compressing speech, semantic compression, which uses an ASR transcript to determine which elements of the speech are presented. We carried out an exploratory user study comparing semantic compression to other novel types of time-compression techniques. We found that users feel they have a greater understanding of recordings compressed using semantic techniques than those compressed using acoustic-based techniques. An approach of using high playback speeds to indicate the location of 'insignificant' speech is not favoured by listeners, who prefer to have such segments removed from recordings.

Interruptions are a central aspect of working life. The prevalence of remote co-workers and the use of mobile technology mean that interruptions are more prevalent, and workers have to learn to manage availability. To understand general issues in availability management, we carried out a naturalistic study of how interruptions are handled in face-to-face situations. We found that availability management requires negotiation, that it is also highly dependent on awareness about the availability of others, and that it demands cognitive effort to shift attention to the interruption. On the basis of these observations, we developed a technology, named. The Negotiator, that embodies three main design requirements: (a) support for negotiation, (b) contextual information about when a recipient is available for a call, (c) lightweightness to reduce attention overhead. We carried out an experimental study of interruption management using this technology. The interface satisfied the original design requirements, that is, people, were able to use it effectively to negotiate times to talk, while successfully carrying out an intellectually demanding activity. Contrary to our expectations, however, people preferred to take responsibility for returning calls rather than delegating them, and they preferred to schedule calls as soon as possible rather than deferring them. We suggest that there are social reasons why people do this. They feel a social obligation to return calls as soon as possible so as not to inconvenience others and also to be responsible for making these calls themselves. They also take calls sooner to avoid having to remember future conversational commitments. We discuss the theoretical and technical implications of these findings.

Editing speech data is currently time-consuming and error-prone. Speech editors rely on acoustic waveform representations, which force users to repeatedly sample the underlying speech to identify words and phrases to edit. Instead we developed a semantic editor that reduces the need for extensive sampling by providing access to meaning. The editor shows a time-aligned errorful transcript produced by applying automatic speech recognition (ASR) to the original speech. Users visually scan the words in the transcript to identify important phrases. They then edit the transcript directly using standard word processing 'cut and paste' operations, which extract the corresponding time-aligned speech. ASR errors mean that users must supplement what they read in the transcript by accessing the original speech. Even when there are transcript errors, however, the semantic representation still provides users with enough information to target what they edit and play, reducing the need for extensive sampling. A laboratory evaluation showed that semantic editing is more efficient than acoustic editing even when ASR is highly inaccurate.

We present a conceptual framework for location-aware community systems and results from two studies of how socially-defined places influence people's information sharing and communication needs.

Modern work is a highly social process, offering many cues for people to organize communication and access information. Shared physical workplaces provide natural support for tasks such as (a) social reminding about communication commitments and keeping track of collaborators and friends, and (b) social data mining of local expertise for advice and information. However, many people now collaborate remotely using tools such as email and voicemail. Our field studies show that these tools do not provide the social cues needed for group work processes. In part, this is because the tools are organized around messages, rather than people. In response to this problem, we created ContactMap, a system that makes people the primary unit of interaction. ContactMap provides a structured social desktop representation of users' important contacts that directly supports social reminding and social data mining. We conducted an empirical evaluation of ContactMap, comparing it with traditional email systems, on tasks suggested by our fieldwork. Users performed better with ContactMap and preferred ContactMap for the majority of these tasks. We discuss future enhancements of our system and the implications of these results for future communication interfaces and for theories of mediated communication.

Spoken dialogue systems have a strong requirement to produce concise and informative utterances. While interacting over a phone, users must both understand the system’s utterances, and remember important facts that the system is providing. Thus most dialogue systems implement some combination of different techniques for (1) option selection: pruning the set of options; (2) information selection: selecting a subset of information to present about each option; (3) aggregation: combining multiple items of information succinctly. We first describe how user models based on multi-attribute decision theory support domain-independent algorithms for both option selection and information selection. We then describe experiments to determine an optimal level of conciseness in information selection, i.e. how much information to include for an option. Our results show that (a) users are highly oriented to utterance conciseness; (b) the information selection algorithm is highly consistent with user’s judgments of conciseness; and (c) the appropriate level of conciseness is both user and dialogue strategy dependent. 1.

This commentary reviews the existing research literature concerning support for talking about objects in mediated communication, drawing three conclusions: (a) speech alone is often sufficient for effective conversations; (b) visual information about work objects is generally more valuable than visual information about work participants; and (c) disjoint visual perspectives can undermine communication processes. I then comment on the four articles in the light of these observations, arguing that they broadly support these observations. I discuss the paradoxical failure of current technologies to support talk about objects, arguing that these need to be better integrated with existing communication applications. I conclude by outlining a research agenda for supporting talk about things, identifying outstanding theoretical, empirical, and design issues.

Mobile interfaces need to allow the user and system to adapt their choice of communication modes according to user preferences, the task at hand, and the physical and social environment. We describe a multimodal application architecture which combines finite-state multimodal language processing, a speech-act based multimodal dialogue manager, dynamic multimodal output generation, and user-tailored text planning to enable rapid prototyping of multimodal interfaces with flexible input and adaptive output. Our testbed application MATCH (Multimodal Access To City Help) provides a mobile multimodal speech-pen interface to restaurant and sub-way information for New York City.

Increasing amounts of public, corporate, and private speech data are now available on-line. These are limited in their usefulness, however, by the lack of tools to permit their browsing and search. The goal of our research is to provide tools to overcome the inherent difficulties of speech access, by supporting visual scanning, search, and information extraction. We describe a novel principle for the design of UIs to speech data: What You See Is Almost What You Hear (WYSIAWYH). In WYSIAWYH, automatic speech recognition (ASR) generates a transcript of the speech data. The transcript is then used as a visual analogue to that underlying data. A graphical user interface allows users to visually scan, read, annotate and search these transcripts. Users can also use the transcript to access and play specific regions of the underlying message. We first summarize previous studies of voicemail usage that motivated the WYSIAWYH principle, and describe a voicemail UI, SCANMail, that embodies WYSIAWYH. We report on a laboratory experiment and a two-month field trial evaluation. SCANMail outperformed a state of the art voicemail system on core voicemail tasks. This was attributable to SCANMail's support for visual scanning, search and information extraction. While the ASR transcripts contain errors, they nevertheless improve the efficiency of voicemail processing. Transcripts either provide enough information for users to extract key points or to navigate to important regions of the underlying speech, which they can then play directly

Current perceptions of Instant Messaging (IM) use are based primarily on self-report studies. We logged thousands of (mostly) workplace IM conversations and evaluated their conversational characteristics and functions. Contrary to prior research, we found that the primary use of workplace IM was for complex work discussions. Only 28% of conversations were simple, single-purpose interactions and only 31% were about scheduling or coordination. Moreover, people rarely switched from IM to another medium when the conversation got complex. We found evidence of two distinct styles of use. Heavy IM users and frequent IM partners mainly used it to work together: to discuss a broad range of topics via many fast-paced interactions per day, each with many short turns and much threading and multitasking. Light users and infrequent pairs mainly used IM to coordinate: for scheduling, via fewer conversations per day that were shorter, slower-paced with less threading and multitasking.

Much of our daily communication activity involves managing interpersonal communications and relationships. Despite its importance, this activity of contact management is poorly understood. We report on field and lab studies that begin to illuminate it.A field study of business professionals confirmed the importance of contact management and revealed a major difficulty: selecting important contacts from the large set of people with whom one communicates. These interviews also showed that communication history is a key resource for this task. Informants identified several history-based criteria that they considered useful.We conducted a lab study to test how well these criteria predict contact importance. Subjects identified important contacts from their email archives. We then analyzed their email to extract features for all contacts. Reciprocity, recency and longevity of email interaction proved to be strong predictors of contact importance. The experiment also identified another contact management problem: removing 'stale' contacts from long term archives. We discuss the design and theoretical implications of these results.

Recent work on evaluation of spoken dialogue systems suggests that the information presentation phase of complex dialogues is often the primary contributor to dialogue duration. Therefore, better algorithms are needed for the presentation of complex information in speech. This paper evaluates the effect of a user model on generation for three dialogue strategies: SUMMARY, COMPARE and RECOMMEND. We present results showing that (a) both COMPARE and RECOMMEND strategies are effective; and (b) the user model is useful.

Increasing amounts of public, corporate, and private audio data are available for use, but limited in usefulness by the lack of tools to permit their browsing and search. In this paper, we describe SCANMail, a system that employs automatic speech recognition, information retrieval, information extraction, and human computer interaction technology to permit users to browse and search their voicemail messages by content through a graphical user interface interface. The SCANMail client also provides note-taking capabilities as well as browsing and querying features. A CallerId server also proposes caller names from existing caller acoustic models and is trained from user feedback. An Email server sends the original message plus its transcription to a mailing address specified in the user's profile. 1.

This paper describes results of an experiment with 9 different DARPA Communicator Systems who participated in the June 2000 data collection. All systems supported travel planning and utilized some form of mixed-initiative interaction. However they varied in several critical dimensions: (1) They targeted different back-end databases for travel information; (2) The used different modules for ASR,NLU,TTS and dialog management. We describe the experimental design, the approach to data collection, the metrics collected, and results comparing the systems. 1.

We explored general issues concerning personal information management by investigating the characteristics of office workers' paper-based information, in an industrial research environment. we examined the reasons people collect paper, types of data they collect, problems encountered in handling paper, and strategies used for processing it. We tested three specific hypotheses in the course of an office move. The greater availability of public digital data along with changes in people's jobs or interests should lead to wholescale discarding of paper data, while preparing for the move. Instead we found workers kept large, highly valued papar archives. We also expected that the major part of people's personal archives would be unique documents. However, only 49% of people's archives were unique documents, the remainder being copies of publicly available data and unread information, and we explore reasons for this. We examined the effects of paper-processing strategies on archive structure. We discovered different paper-processing strategies ( filing and piling )that were relatively independent of job type. We predicated that filers' attempted to evaluate and catergorize incoming documents would produce smaller archives that were accessed frequently. Contrary to our predictions, filers amassed more information, and accessed it less frequently than pilers. We argue that filers may engage in premature filing : to clear their workspace, they archives information that later turns out to be of low value. Given the effort involved in organzing data, they are also loath to discard filed information, even when its value is uncertain. We discuss the implications of this research for digital personal information management.

Voicemail is a pervasive, but under-researched tool for workplace communication. Despite potential advantages of voicemail over email, current phone-based voicemail UIs are highly problematic for users. We present a novel, Web-based, voicemail interface, Jotmail. The design was based on data from several studies of voicemail tasks and user strategies. The GUI has two main elements: (a) personal annotations that serve as a visual analogue to underlying speech; (b) automatically derived message header information. We evaluated Jotmail in an 8-week field trial, where people used it as their only means for accessing voicemail. Jotmail was successful in supporting most key voicemail tasks, although users' electronic annotation and archiving behaviors were different from our initial predictions. Our results argue for the utility of a combination of annotation based indexing and automatically derived information, as a general technique for accessing speech archives.

We discuss findings from an ethnographic study of instant messaging (IM) in the workplace and its implications for media theory. We describe how instant messaging supports a variety of informal communication tasks. We document the affordances of IM that support flexible, expressive communication. We describe some unexpected uses of IM that highlight aspects of communication which are not part of current media theorizing. They pertain to communicative processes people use to connect with each other and to manage communication, rather than to information exchange. We call these processes "outeraction". We discuss how outeractional aspects of communication affect media choice and patterns of media use.

Certain generation applications may profit from the use of stochastic methods. In developing stochastic methods, it is crucial to be able to quickly assess the relative merits of different approaches or models. In this paper, we present several types of intrinsic (system internal) metrics which we have used for baseline quantitative assessment. This quantitative assessment should then be augmented to a fuller evaluation that examines qualitative aspects. To this end, we describe an experiment that tests correlation between the quantitative metrics and human qualitative judgment. The experiment confirms that intrinsic metrics cannot replace human evaluation, but some correlate significantly with human judgments of quality and understandability and can be used for evaluation during development.

The ability to automatically scan voicemail messages for content and caller identity cues would be a useful service. This paper describes a system which automatically files voicemail messages into caller folders using text independent speaker recognition techniques. Callers are represented by Gaussian mixture models (GMM’s). The speech for an incoming message is processed and scored against caller models created for a subscriber. A message whose matching score exceeds a threshold is filed in the matching caller folder; otherwise it is tagged as “unknown”. The subscriber has the ability to listen to an “unknown” message and file it in the proper folder, if it exists, or create a new folder, if it does not. Such subscriber labelled messages are used to train and adapt caller models. The system has been evaluated on a database of voicemail messages collected at AT&T Labs. A set of 20 callers from this database is designated as “ingroup”. Each of these callers has recorded at least 20 messages totalling 10 or more minutes in duration. A distinct set of 220 messages, each from a dierent caller, are designated as “outgroup”. Representative performance figures with threshold parameters set to ensure that outgroup acceptance is low compared with ingroup rejection are the following. The average ingroup message rejection rate is 11.0% and the average ingroup message confusion rate (matching the wrong caller) is 1.0%, while the average outgroup message accept rate is 2.7%.

We examine how differences in the accuracy of Automatic Speech Recognition transcripts affect users' ability to use these in tasks requiring the retrieval of speech "documents". We compare performance measures, processing strategies, and preference data for subjects using transcripts and speech data to perform a series of relevance judgment and summary tasks on transcripts with different levels of accuracy. Results show effects for transcript quality on solution accuracy, time to solution, amount of speech played for the task, likelihood of subjects abandoning use of a transcript, and subject perceptions of task difficulty, transcript utility, readability, and comprehensibility. 1. INTRODUCTION Research on Automatic Speech Recognition (ASR) generally assumes perfect transcription accuracy to be its holy grail. The more accurately a system can transcribe an utterance, the better that system performs (modulo time factors) Recently this assumption is being challenged, however. Different...

We identify a problem with the process of research in the human-computer interaction (HCI) community-an overemphasis on "radical invention" at the price of achieving a common research focus. Without such a focus, it is difficult to build on previous work, to compare different interaction techniques objectively, and to make progress in developing theory. These problems at the research level have implications for practice, too; as researchers we often are unable to give principled design advice to builders of new systems. We propose that the HCI community try to achieve a common focus around the notion of reference tasks. We offer arguments for the advantages of this approach as well as consider potential difficulties. We explain how reference tasks have been highly effective in focusing research into information retrieval and speech recognition. We discuss what factors have to be considered in selecting HCI reference tasks and present an example reference task (for searching speech archives). This example illustrates the nature of reference tasks and points to the issues and problems involved in constructing and using them. We conclude with recommendations about what steps need to be taken to execute the reference task research agenda. This involves recommendations about both the technical research that needs to be done and changes in the way that the HCI research community operates. The technical research involves identification of important user tasks by systematic requirements gathering, definition and operationalization of reference tasks and evaluation metrics, and execution of task-based evaluation, along with judicious use of field trials. Perhaps more important, we have also suggested changes in community practice that HCI must adopt to make the reference tasks idea work. We must create forums for discussion of common tasks and methods by which people can compare systems and techniques. Only by doing this can the notion of reference tasks be integrated into the process of research and development, enabling the field to achieve the focus it desperately needs.

Previous examinations of search in textual archives have assumed that users first retrieve a ranked set of documents relevant to their query, and then visually scan through these documents, to identify the information they seek.While document scanning is possible in text, it is much more laborious in speech archives, due to the inherently serial nature of speech.Yet, in developing tools for speech access, little attention has so far been paid to users' problems in scanning and extracting information from within "speech documents".We demonstrate the extent of these problems in two user studies.We show that users experience severe problems with local navigation in extracting relevant information from within "speech documents".Based on these results, we propose a new user interface (UI) design paradigm: What You See Is (Almost) What You Hear, (WYSIAWYH) -a multimodal method for accessing speech archives.This paradigm presents a visual analogue to the underlying speech, enabling visual scanning for effective local navigation.We empirically evaluate a UI based on this paradigm.We compare our WYSIAWYH UI with a visual "tape recorder", in relevance ranking, fact-finding, and summarization tasks involving broadcast news data.Our findings indicate that an interface supporting local navigation multimodally helps relevance ranking and fact-finding, but not summarization.We analyze the reasons for system success and identify outstanding research issues in UI design for speech archives.

Usenet may be regarded as the worl$s largest conversatioti appficatioq with over 17,000 newsgronps and 3 fion users.Despite its ubiquity and poptiari~, however, we know Etie about the natore of the interactions it suppo~.~s empirical paper investigates mass interaction in UseneL We anrdyse over 2.15 Mon messages from 659,450 posters, co~ected from 500 newsgoups over 6 months.We fit characterise mass interactio~presenting basic data about demographics, conversational strategies and interactivity-Using predictions tiom the common ground [3] model of interactio~we next conduct causal modetig to determine relations bemeen demographics, conversational strategim and interactiv~.We find evidence for moderate conversatiomd threading, but hge participation ine@ties in Usene4 with a d minority of participants posting a large proportion of messages.Contrary to the common gromd model and 'Etiquette" guidefies [8,10] we ho find that "cross-posting" to extemd newsgroups is hig~y fiequenk Our predictions about the eEects of demographics on conversatioti strategy were largely cotie~but we found disconfirming evidence about the relations befi'een conversational strategy and interactivity.Contrary to our expectations, boti cross-posting and short messages promote interactivity.We conclude that in order to e~lain mass interaction, the common ground model must be mowed to ticorporate notions of weak ties [5,q and communication overload [11,18].Ke~ords hlass interactio~Usene< conversatio~newsgroups, common groun~moderatio~FAQS, netiquette, empirid. lNTRODUC~ONUsenet maybe regarded as the worl&s largest and fmtest growing conversationrd appficatiom k 1988 there were fewer than 500 news~oups.bent estimatesvary, but at the time of our dab co~ection in Dec. 1996, there were Ptission 10de distal or hard copies ofall or pti ofthis ~vorkfor personal or classroom use is granted without f= protided that copies Zrenotmade or dis~%uted for profit or commercial advantage and that copies bear this notice and the fill citation on the fimt paga To copy othm~ise to republish, to post on sen'em or to redistribute to Iis& Tquires prior sptific permission and~ora fee CSCW 98 Seatie Wastinson USA ---_ -.

Despite the recent growth and potential utility of speech archives, we currently lack tools for effective archival access. Previous research on search of textual archives has assumed that the system goal should be to retrieve sets of relevant documents, leaving users to visually scan through those documents to identify relevant information. However, in previous work we show that in accessing real speech archives, it is insufficient to only retrieve "document" sets [9,10]. Users experience huge problems of local navigation in attempting to extract relevant information from within speech "documents". These studies also show that users address these problems by taking handwritten notes. These notes detail both the content of the speech and serve as indices to help access relevant regions of the archive. From these studies we derive a new principle for the design of speech access systems: What You See Is (Almost) What You Hear. We present a new user interface to a broadcast news archive, d...

Communication theories and technology have tended to focus on extended, formal meetings and have neglected a prevalent and vital form of workplace communication—namely, lightweight communication. Unlike formal, extended meetings, lightweight interaction is brief, informal, unplanned, and intermittent. We analyze naturalistic data from a study of work-place communication and derive five design criteria for lightweight interaction systems. These criteria require that systems for lightweight interaction support conversational tracking, rapid connection , the ability to leave a message, context management, and shared real-time objects . Using these criteria, we evaluate existing interpersonal communications technologies. We then describe an implementation of a system (TeleNotes) that is designed to support lightweight interaction by meeting these criteria. The interface metaphor allows communications to be based around desktop objects, resembling “sticky notes.” These objects are also organized into “desktop piles” to support conversational threads and provide mechanisms for initiating real-time audio, video, and application sharing. We conducted informal user testing of several system prototypes. Based on our findings, outstanding issues concerning theory and systems design for communication systems are outlined—in particular, with regard to the issue of managing conversations over time.

Article Free Access Share on Email overload: exploring personal information management of email Authors: Steve Whittaker Lotus Development Corp. Lotus Development Corp.View Profile , Candace Sidner Lotus Development Corp. Lotus Development Corp.View Profile Authors Info & Claims CHI '96: Proceedings of the SIGCHI Conference on Human Factors in Computing SystemsApril 1996 Pages 276–283https://doi.org/10.1145/238386.238530Published:13 April 1996Publication History 498citation11,623DownloadsMetricsTotal Citations498Total Downloads11,623Last 12 Months957Last 6 weeks131 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteView all FormatsPDF

This empirical study examines factors influencing the success of a commercial groupware system in creating group archives and supporting asynchronous communication.The study investigates the use of Lotus Notesm in a workplace setting.We interviewed 21 Notes users and identified three factors that they thought contributed to the successful use of Notes databases for archiving and communication.We then tested the effect of these factors on 15,571 documents in 20 different databases.Contrary to our users' beliefs, we found the presence of an active database moderator actually inhibited discussions, and reduced browsing.Further paradoxical results were that conversations and the creation of group archives were more successful in databases with large numbers of diverse participants.Conversations and archiving were less successful in smaller, more homogeneous, project teams.Database size was also important: a large database containing huge amounts of information was more likely to be used for further conversations and archiving, than a small one.This result again ran counter to users' beliefs that small databases are superior.We discuss possible reasons for these findings in terms of critical mass and media competition, and conclude with implications for design.

This 1.5 day workshop will bring together designers and researchers working on on-line communities, to discuss: (a) Existing understanding of real-world cummunities; (b) Experiences with the behaviour, implementation and design of on-line communities; (c) Lessons from traditional CSCW systems. We will develop design goals for such communities and identify oustanding research issues.

Article Back to the future: pen and paper technology supports complex group coordination Share on Authors: Steve Whittaker Lotus Development Corporation, One Rogers St, Cambridge MA Lotus Development Corporation, One Rogers St, Cambridge MAView Profile , Heinrich Schwarz Science, Technology, Society Program, Massachusetts Institute of Technology, Cambridge, MA Science, Technology, Society Program, Massachusetts Institute of Technology, Cambridge, MAView Profile Authors Info & Claims CHI '95: Proceedings of the SIGCHI Conference on Human Factors in Computing SystemsMay 1995 Pages 495–502https://doi.org/10.1145/223904.223971Online:01 May 1995Publication History 17citation560DownloadsMetricsTotal Citations17Total Downloads560Last 12 Months14Last 6 weeks0 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteGet Access

Article Free Access Share on Informal workplace communication: what is it like and how might we support it? Authors: Steve Whittaker Lotus Development Corporation, 1 Rogers St., Cambridge, MA and Office Appliances Dept., Hewlett-Packard Laboratories, Filton Road, Bristol, BS12 6QZ, UK Lotus Development Corporation, 1 Rogers St., Cambridge, MA and Office Appliances Dept., Hewlett-Packard Laboratories, Filton Road, Bristol, BS12 6QZ, UKView Profile , David Frohlich Office Appliances Dept., Hewlett-Packard Laboratories, Filton Road, Bristol, BS12 6QZ, UK Office Appliances Dept., Hewlett-Packard Laboratories, Filton Road, Bristol, BS12 6QZ, UKView Profile , Owen Daly-Jones Office Appliances Dept., Hewlett-Packard Laboratories, Filton Road, Bristol, BS12 6QZ, UK and Dept. of Psychology, U. of York, York Y01 5DD, UK Office Appliances Dept., Hewlett-Packard Laboratories, Filton Road, Bristol, BS12 6QZ, UK and Dept. of Psychology, U. of York, York Y01 5DD, UKView Profile Authors Info & Claims CHI '94: Proceedings of the SIGCHI Conference on Human Factors in Computing SystemsApril 1994 Pages 131–137https://doi.org/10.1145/191666.191726Published:24 April 1994Publication History 152citation9,908DownloadsMetricsTotal Citations152Total Downloads9,908Last 12 Months1,587Last 6 weeks338 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteeReaderPDF

Article Free Access Share on FILOCHAT: handwritten notes provide access to recorded conversations Authors: Steve Whittaker Lotus Development Corporation, 1 Rogers St., Cambridge, MA and Office Appliances Dept., Hewlett-Packard Laboratories, Filton Road, Bristol, BS12 6QZ, UK Lotus Development Corporation, 1 Rogers St., Cambridge, MA and Office Appliances Dept., Hewlett-Packard Laboratories, Filton Road, Bristol, BS12 6QZ, UKView Profile , Patrick Hyland Office Appliances Dept., Hewlett-Packard Laboratories, Filton Road, Bristol, BS12 6QZ, UK Office Appliances Dept., Hewlett-Packard Laboratories, Filton Road, Bristol, BS12 6QZ, UKView Profile , Myrtle Wiley Office Appliances Dept., Hewlett-Packard Laboratories, Filton Road, Bristol, BS12 6QZ, UK Office Appliances Dept., Hewlett-Packard Laboratories, Filton Road, Bristol, BS12 6QZ, UKView Profile Authors Info & Claims CHI '94: Proceedings of the SIGCHI Conference on Human Factors in Computing SystemsApril 1994 Pages 271–277https://doi.org/10.1145/191666.191763Published:24 April 1994Publication History 77citation1,466DownloadsMetricsTotal Citations77Total Downloads1,466Last 12 Months208Last 6 weeks26 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteeReaderPDF

No abstract available.

Recent trends toward telecommuting, mobile work, and wider distribution of the work force, combined with reduced technology costs, have made video communications more attractive as a means of supporting informal remote interaction. In the past, however, video communications have never gained widespread acceptance. Here we identify possible reasons for this by examining how the spoken characteristics of video-mediated communication differ from face-to-face interaction, for a series of real meetings. We evaluate two wide-area systems. One uses readily available Integrated Services Digital Network (ISDN) lines but suffers the limitations of transmission lags, a half-duplex line, and poor quality video. The other uses optical transmission and video-switching technology with negligible delays, full duplex audio, and broadcast quality video. To analyze the effects of video systems on conversation, we begin with a series of conversational characteristics that have been shown to be important in face-to-face interaction. We identify properties of the communication channel in face-to-face interaction that are necessary to support these characteristics, namely, that it has low transmission lags, it is two way, and it uses multiple modalities. We compare these channel properties with those of the two video-conferencing systems and predict how their different channel properties will affect spoken conversation. As expected, when compared with face-to-face interaction, communication using the ISDN system was found to have longer conversational turns; fewer interruptions, overlaps, and backchannels; and increased formality when switching speakers. Communication over the system with broadcast quality audio and video was more similar to face-to-face meetings, although it did not replicate face-to-face interaction. Contrary to our expectations, formal techniques were still used to achieve speaker switching. We suggest that these may be necessary because of the absence of certain speaker-switching cues. The results imply that the advent of high-speed multimedia networking will improve but not remove all the problems of video conferencing as an interpersonal communications tool, and we describe possible solutions to the outstanding problems.

No abstract available.

Conversation between two people is usually of MIXED-INITIATIVE, with CONTROL over the conversation being transferred from one person to another. We apply a set of rules for the transfer of control to 4 sets of dialogues consisting of a total of 1862 turns. The application of the control rules lets us derive domain-independent discourse structures. The derived structures indicate that initiative plays a role in the structuring of discourse. In order to explore the relationship of control and initiative to discourse processes like centering, we analyze the distribution of four different classes of anaphora for two data sets. This distribution indicates that some control segments are hierarchically related to others. The analysis suggests that discourse participants often mutually agree to a change of topic. We also compared initiative in Task Oriented and Advice Giving dialogues and found that both allocation of control and the manner in which control is transferred is radically different for the two dialogue types. These differences can be explained in terms of collaborative planning principles.

We conducted an empirical analysis into the relation between control and discourse structure. We applied control criteria to four dialogues and identified 3 levels of discourse structure. We investigated the mechanism for changing control between these structures and found that utterance type and not cue words predicted shifts of control. Participants used certain types of signals when discourse goals were proceeding successfully but resorted to interruptions when they were not.