Commonspace

Writing in Feyer, leaving unfinished wording as a provisional fragment to revisit later.

Commonspace Walkthrough

v0.1.19

A workspace for people and agents

Part play, part product; part research output, part tool I use to write and research; Commonspace is an AI-native workspace application designed and built for both humans and agents, acting as a hub where writing and research can be organized and pursued. Commonspace focuses on making it as easy as possible to create the data structures and tooling agents need to be effective, balancing their needs with the ways people explore, leave things unfinished, and develop ideas.

knowledge production and distribution

Your workspaceYour audiences

  • Your workspace
    • Document · What is a Workspace?
      • cites 43 Sources
        • Source · Distributed Cognition: Toward a New Foundation for Human-Computer Interaction Research Hollan et al., 2000. Highlights: 6. Proposes distributed cognition as the footing for studying people and computers together: thinking is spread across people, tools and time, so the thing to study and design for is the whole working environment.
          • Highlight · In interaction with objects in the world, history of use is sometimes available to us in ways that inform our interactions with them. For example, a well-worn section of a door handle suggests where to grasp it. A new paperback book opens to the place we last stopped reading. The most recently used pieces of paper occupy the tops of piles on our desk. The physics of the world is such that at times the histories of… Source: Hollan et al., 2000.
          • Highlight · A third tenet of the theory of distributed cognition is that the study of cognition is not separable from the study of culture, because agents live in complex cultural environments. This means, on the one hand, that culture emerges out of the activity of human agents in their historical contexts, as mental, material and social structures interact, and on the other hand, that culture in the form of a history of… Source: Hollan et al., 2000.
          • Highlight · For the design of work environments, this means that work materials are more than mere stimuli for a disembodied cognitive system. Work materials from time to time become elements of the cognitive system itself. Just as a blind person’s cane or a cell biologist’s microscope is a central part of the way they perceive the world, so well-designed work materials become integrated into the way people think, see, and… Source: Hollan et al., 2000.
          • Highlight · Minds are not passive representational engines, whose primary function is to create internal models of the external world. The relations between internal processes and external ones are far more complex, involving coordination at many different time scales between internal resources—memory, attention, executive function—and external resources—the objects, artifacts, and at-hand materials constantly surrounding us.… Source: Hollan et al., 2000.
          • Highlight · For example, an examination of memory processes in an airline cockpit shows that memory involves a rich interaction between internal processes, the manipulation of objects, and the traffic in representations among the pilots. A complete theory of individual memory by itself is insufficient to understand how this memory system works. Furthermore, the physical environment of thinking provides more than simply… Source: Hollan et al., 2000.
          • Highlight · The first of these principles concerns the boundaries of the unit of analysis for cognition. In every area of science, the choices made concerning the boundaries of the unit of analysis have important implications. In traditional views of cognition the boundaries are those of individuals. Sometimes the traditionally assumed boundaries are exactly right. For other phenomena, however, these boundaries either span too… Source: Hollan et al., 2000.
        • Source · The Intelligent Use of Space Kirsh, 1995. Highlights: 4. Classifies the ways people arrange the space around them to make work easier: layouts that simplify choices, layouts that simplify what must be perceived, and movements that save computing in the head.
          • Highlight · To see an action as available for choice is to notice that it is afforded by the current situation. An affordance of a situation, crudely stated, is a way that a situation lends itself to being used [15]. An empty container affords filling, an active television affords viewing, and a hammer affords striking. Because of the relational nature of affordances, we need to tie them to the action repertoire of an agent. A… Source: Kirsh, 1995.
          • Highlight · Let us say that an environment E, is well designed for a particular task T, and a particular agent A endowed with skill S, and rules R, if, at every choice point, the environment provides sufficient physical and informational resources to locally determine A’s choices. On the one hand, this will require E to be rich in task useful physical structures, for instance, E ought to have tools and surfaces which make it… Source: Kirsh, 1995.
          • Highlight · To understand why lining up the ingredients in well ordered neatly separated rows is clever, requires understanding a fact about human psychophysics: estimation of length is easier and more reliable than estimation of area or volume. By using length to encode number she created a cue or signal in the world which she could accurately track. Laying out slices in lines allows more precise judgement of the property… Source: Kirsh, 1995.
          • Highlight · The view I shall present is that whether we are aware of it or not, we are constantly organizing and re-organizing our workplace to enhance performance. Space is a resource that must be managed, much like time, memory, and energy. When we use space well we can often bring the time and memory demands of our tasks down to workable levels. We can increase the reliability of execution, and the number of jobs we can… Source: Kirsh, 1995.
        • Source · Dimensions of Integration in Embedded and Extended Cognitive Systems Heersmink, 2015. Highlights: 2. Sets out the dimensions along which a person and an artifact can be integrated, such as how reliably it is at hand, how far it is trusted and how transparent it is in use, so that integration becomes a matter of degree.
          • Highlight · Cognitive states and processes are thus conceptualized as hybrids or amalgamations of neurological, bodily, and environmental objects and processes (Wilson 2004; Rowlands 1999, 2010). On this view, some of the vehicles of cognitive processes are sometimes not located in the brain, but in the cultural and artifactual environment. It is thus a metaphysical claim about the location of human cognition, implying rather… Source: Heersmink, 2015.
          • Highlight · The goal of this paper is to develop some of the tools to conceptualize this complementary integration between agents and artifacts. It does so by proposing a multidimensional framework, including the dimensions of information flow, reliability, durability, trust, procedural transparency, informational transparency, individualization, and transformation. The proposed dimensions are all matters of degree and jointly… Source: Heersmink, 2015.
        • Source · Extending Minds with Generative AI Clark, 2025. Highlights: 2. Answers the fear that generative AI will make people think less by recalling that humans have always built thinking systems out of tools, and asks what it takes to do that well with AI.
          • Highlight · Recent neurocomputational work in the area known as ‘predictive processing’ (or ‘active inference’) has shown just how it is that brains like ours are poised to become woven into these larger tapestries of brain, body, and world. According to this emerging picture, we learn about our worlds by constantly trying to predict the sensory consequences of our own actions14,15. As we do so brains like ours become expert at… Source: Clark, 2025.
          • Highlight · Reaching just beyond the body, we exploit the rich world of familiar old-school static media. Even today, many of us will resort to pen and paper, scribbling madly as we try to think through a problem in math, life, or philosophy. More prosaically, we may sketch and re-sketch the design of a new kitchen on a paper napkin. These familiar loops through external media slowly become part and parcel of how many of us… Source: Clark, 2025.
        • Source · Weak Bundle, Strong Bundle: How AI Redraws Job Boundaries Garicano et al., 2026. Argues that AI's effect on an occupation depends on how costly it is to pull the tasks AI can do apart from the rest of the job, and not only on which tasks those are.
        • Source · A Rich Landscape of Affordances Rietveld & Kiverstein, 2014. Widens affordances from reaching and grasping to the whole range of skilled human activity: what an environment offers depends on the abilities available in a form of life.
        • Source · Endogenous Task Bundling, Skills and Automation Gans, 2026. Treats the grouping of tasks into jobs as something the labor market decides, and works out how technology shifts where one job ends and the next begins.
        • Source · The Extended Mind Clark & Chalmers, 1998. Argues that when something outside the head, such as a notebook, plays the part memory would play, it is part of the mind: the environment takes an active role in cognition.
        • Source · O-Ring Automation Gans & Goldfarb, 2026. Studies automation when the quality of every task multiplies into the final product, so handing some tasks to machines changes how workers spend their time on the rest.
        • Source · What Can I Do Here? A Theory of Affordances in Reinforcement Learning Khetarpal et al., 2020. Brings affordances into reinforcement learning: an agent that knows which actions are feasible where can plan and learn models faster than one that assumes every action is always available.
        • Source · The Ethnography of Infrastructure Star, 1999. A call to study the unexciting systems other work runs on. Infrastructure means different things to different groups, goes unnoticed while it works, and takes its own methods to observe.
        • Source · Explaining Multistability: Postphenomenology and Affordances of Technologies de Boer, 2023. Asks why a technology can be put to many uses but not to just any use, and answers with affordances: what a thing offers the people using it bounds the uses that settle.
        • Source · How Well Does Agent Development Reflect Real-World Work? Wang et al., 2026. Compares where agent benchmarks concentrate with how human work is spread across the labor market, to see how well agent development represents real jobs.
        • Source · ReAct: Synergizing Reasoning and Acting in Language Models Yao et al., 2023. Introduces ReAct, a way of prompting a language model to alternate between reasoning in text and taking actions, so each informs the other.
        • Source · Workflows and Automation Trammell, 2026. Suggests tasks are bundled into jobs because doing one makes a worker better at related ones, and shows that this can make automation pay off little at first and sharply once a whole cluster of tasks is covered.
        • Source · SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering Yang et al., 2024. Introduces SWE-agent and the agent-computer interface: commands and feedback designed for a language model, which improve how well it can resolve issues in real code repositories.
        • Source · An Activity Theory Approach to Affordance Bærentsen & Trettvik, 2002. Re-reads Gibson's affordances through activity theory, arguing that HCI's troubles with the concept come from leaving out the activity an affordance belongs to.
        • Source · Organizational Learning and Communities-of-Practice: Toward a Unified View of Working, Learning, and Innovation Brown & Duguid, 1991. Finds that work as it is done differs from how manuals and job descriptions present it, and that the informal communities where people work practice out together are where an organization learns and innovates.
        • Source · The Theory of Affordances Gibson, 1979. The original statement of affordances: what an environment offers an animal, for good or ill, measured against that animal's size and capacities.
        • Source · Affordances Enable Partial World Modeling with LLMs Khetarpal et al., 2026. Uses affordances to have a language model build partial models of the world, covering only what matters for the task at hand instead of searching over a full one.
        • Source · Technology Affordances Gaver, 1991. Applies affordances to interface design by separating what an artifact allows from the information that reveals it, which yields perceptible, hidden and false affordances.
        • Source · Reflexion: Language Agents with Verbal Reinforcement Learning Shinn et al., 2023. Introduces Reflexion, in which an agent writes a short reflection on a failed attempt and keeps it in memory, so it improves over trials without any change to the model's weights.
        • Source · Do As I Can, Not As I Say: Grounding Language in Robotic Affordances Ichter et al., 2023. Grounds a language model's suggestions in what a robot can do: learned value functions score how likely each skill is to succeed from the current state, and the plan follows both.
        • Source · Affordance, Conventions, and Design Norman, 1999. Norman clarifies his own term: on a screen, designers mostly control perceived affordances and conventions, and showing that an action is possible differs from making it possible.
        • Source · Affordance as General Value Function: A Computational Model Graves et al., 2020. Casts affordances as predictions an agent can learn: general value functions that estimate where following a behavior from the current state would lead.
        • Source · Artifacts as Memory Beyond the Agent Boundary Martin et al., 2026. Environmental artifacts can hold memory for reinforcement-learning agents.
        • Source · Computer Environments Elicit General Agentic Intelligence in LLMs Cheng et al., 2026. A model's environment contributes to its abilities as an agent.
        • Source · Nature and Technology in History Schatzki, 2003. Widens history beyond human action: nature and technology are part of the arrangements social life consists of, so they belong in any account of how it changes.
        • Source · StagedWorkspace: A Versioned Workspace for Knowledge-Work Agents Hua et al., 2026. A versioned workspace tracks an agent's searches, edits, reviews, and submissions.
        • Source · Organizational and Economic Obstacles to Automation: A Cautionary Tale from AT&T in the Twentieth Century Feigenbaum & Gross, 2024. Follows AT&T's decades-long move from telephone operators to mechanical switching, to show how organizational and economic frictions slowed the adoption of a technology that already existed.
        • Source · From Language Models to Agentic AI: A Survey of Autonomous, Action-Enabled, and Collaborative LLM Agents Bajoria et al., 2026. This survey covers autonomous, tool-using, and collaborative language-model agents.
        • Source · Designing for instrument-mediated activity Béguin & Rabardel, 2000. Treats a tool as an instrument only once a user has built ways of using it, and asks what designing around that activity, and not around the object alone, would involve.
        • Source · AXIS: Efficient Human-Agent-Computer Interaction with API-First LLM-Based Agents Lu et al., 2025. Agents use application APIs first, turning to the interface when needed.
        • Source · Cognitive Architectures for Language Agents Sumers et al., 2024. CoALA describes language agents through memory, actions, and decision procedures.
        • Source · From language to action: a review of large language models as autonomous agents and tool users Chowa et al., 2026. This review covers language models that follow instructions, carry out tasks, and use tools.
        • Source · A New Cognitive Ethnography Hutchins, 2025. A framework for cognitive ethnography applies predictive processing to everyday activity.
        • Source · Redefining Affordance via Computational Rationality Liao & Holz, 2025. People infer what they can do by weighing likely outcomes within their cognitive limits.
        • Source · Agentic Large Language Models, a Survey Plaat et al., 2025. Surveys agentic language models under three headings: models that reason, models that act through tools and robots, and models that interact with other agents.
        • Source · On structural change: practice organizations and institutional logics Schatzki, 2024. Reads the institutional logics approach alongside practice theory to explain how the organization of practices changes.
        • Source · Organizational Routines as a Source of Continuous Change Feldman, 2000. Shows that organizational routines, usually defined by their stability, change continually as the people performing them respond to how the last round went.
        • Source · Signifiers for Conveying and Exploiting Affordances: From Human-Computer Interaction to Multi-Agent Systems Lemée et al., 2024. Environmental cues help artificial agents discover and use available actions.
        • Source · What Work Does Generative AI Do? Bick et al., 2026. Uses the Real-Time Population Survey to measure how many U.S. workers use generative AI on the job, how that has grown since 2024, and how use differs by occupation.
        • Source · A Survey of Large Language Models Zhao et al., 2026. A broad survey of large language models: how they are pre-trained, adapted, used and evaluated.
      • experiments 2 Programs
      • links to 3 Documents
        • Document · Future of Workspaces The Future of Workspaces is a labor economics and policy focused research direction part of the larger exploratory Workspace Research Agenda, sitting alongside the design-focused Agent-Computer-Human-Experience (ACHE) and the philosophically-focused Agent Phenomenology research…
        • Document · ACHE Agent-Computer-Human Experience. Agent-Computer-Human Experience (ACHE) is a product design and software engineering focused research direction within the larger Workspace Research Agenda, exploring how we might best build interfaces and software that can dually support humans and Large Language Model (LLM)…
        • Document · Agent Phenomenology Agent Phenomenology explores Large Language Models (LLMs), LLM agents, and our human use of them, through a framework that draws on Heideggerian phenomenology, post-phenomenology, distributed cognition, extended mind theory, and philosophical accounts of delegation,…
      • uses 2 Datasets
        • Dataset · O*NET Occupational Information Network. The U.S. Department of Labor's occupation database covers tasks, skills, knowledge, and work context.
        • Dataset · RPS Real-Time Population Survey. A recurring survey tracks work and generative AI use among U.S. working-age adults.
  • Your audiences
    • Document · What is a Workspace?
      • queries 43 Sources
        • Source · Distributed Cognition: Toward a New Foundation for Human-Computer Interaction Research Hollan et al., 2000. Highlights: 6. Proposes distributed cognition as the footing for studying people and computers together: thinking is spread across people, tools and time, so the thing to study and design for is the whole working environment.
          • Highlight · In interaction with objects in the world, history of use is sometimes available to us in ways that inform our interactions with them. For example, a well-worn section of a door handle suggests where to grasp it. A new paperback book opens to the place we last stopped reading. The most recently used pieces of paper occupy the tops of piles on our desk. The physics of the world is such that at times the histories of… Source: Hollan et al., 2000.
          • Highlight · A third tenet of the theory of distributed cognition is that the study of cognition is not separable from the study of culture, because agents live in complex cultural environments. This means, on the one hand, that culture emerges out of the activity of human agents in their historical contexts, as mental, material and social structures interact, and on the other hand, that culture in the form of a history of… Source: Hollan et al., 2000.
          • Highlight · For the design of work environments, this means that work materials are more than mere stimuli for a disembodied cognitive system. Work materials from time to time become elements of the cognitive system itself. Just as a blind person’s cane or a cell biologist’s microscope is a central part of the way they perceive the world, so well-designed work materials become integrated into the way people think, see, and… Source: Hollan et al., 2000.
          • Highlight · Minds are not passive representational engines, whose primary function is to create internal models of the external world. The relations between internal processes and external ones are far more complex, involving coordination at many different time scales between internal resources—memory, attention, executive function—and external resources—the objects, artifacts, and at-hand materials constantly surrounding us.… Source: Hollan et al., 2000.
          • Highlight · For example, an examination of memory processes in an airline cockpit shows that memory involves a rich interaction between internal processes, the manipulation of objects, and the traffic in representations among the pilots. A complete theory of individual memory by itself is insufficient to understand how this memory system works. Furthermore, the physical environment of thinking provides more than simply… Source: Hollan et al., 2000.
          • Highlight · The first of these principles concerns the boundaries of the unit of analysis for cognition. In every area of science, the choices made concerning the boundaries of the unit of analysis have important implications. In traditional views of cognition the boundaries are those of individuals. Sometimes the traditionally assumed boundaries are exactly right. For other phenomena, however, these boundaries either span too… Source: Hollan et al., 2000.
        • Source · Dimensions of Integration in Embedded and Extended Cognitive Systems Heersmink, 2015. Highlights: 2. Sets out the dimensions along which a person and an artifact can be integrated, such as how reliably it is at hand, how far it is trusted and how transparent it is in use, so that integration becomes a matter of degree.
          • Highlight · Cognitive states and processes are thus conceptualized as hybrids or amalgamations of neurological, bodily, and environmental objects and processes (Wilson 2004; Rowlands 1999, 2010). On this view, some of the vehicles of cognitive processes are sometimes not located in the brain, but in the cultural and artifactual environment. It is thus a metaphysical claim about the location of human cognition, implying rather… Source: Heersmink, 2015.
          • Highlight · The goal of this paper is to develop some of the tools to conceptualize this complementary integration between agents and artifacts. It does so by proposing a multidimensional framework, including the dimensions of information flow, reliability, durability, trust, procedural transparency, informational transparency, individualization, and transformation. The proposed dimensions are all matters of degree and jointly… Source: Heersmink, 2015.
        • Source · Extending Minds with Generative AI Clark, 2025. Highlights: 2. Answers the fear that generative AI will make people think less by recalling that humans have always built thinking systems out of tools, and asks what it takes to do that well with AI.
          • Highlight · Recent neurocomputational work in the area known as ‘predictive processing’ (or ‘active inference’) has shown just how it is that brains like ours are poised to become woven into these larger tapestries of brain, body, and world. According to this emerging picture, we learn about our worlds by constantly trying to predict the sensory consequences of our own actions14,15. As we do so brains like ours become expert at… Source: Clark, 2025.
          • Highlight · Reaching just beyond the body, we exploit the rich world of familiar old-school static media. Even today, many of us will resort to pen and paper, scribbling madly as we try to think through a problem in math, life, or philosophy. More prosaically, we may sketch and re-sketch the design of a new kitchen on a paper napkin. These familiar loops through external media slowly become part and parcel of how many of us… Source: Clark, 2025.
        • Source · Weak Bundle, Strong Bundle: How AI Redraws Job Boundaries Garicano et al., 2026. Argues that AI's effect on an occupation depends on how costly it is to pull the tasks AI can do apart from the rest of the job, and not only on which tasks those are.
        • Source · A Rich Landscape of Affordances Rietveld & Kiverstein, 2014. Widens affordances from reaching and grasping to the whole range of skilled human activity: what an environment offers depends on the abilities available in a form of life.
        • Source · Endogenous Task Bundling, Skills and Automation Gans, 2026. Treats the grouping of tasks into jobs as something the labor market decides, and works out how technology shifts where one job ends and the next begins.
        • Source · The Extended Mind Clark & Chalmers, 1998. Argues that when something outside the head, such as a notebook, plays the part memory would play, it is part of the mind: the environment takes an active role in cognition.
        • Source · O-Ring Automation Gans & Goldfarb, 2026. Studies automation when the quality of every task multiplies into the final product, so handing some tasks to machines changes how workers spend their time on the rest.
        • Source · What Can I Do Here? A Theory of Affordances in Reinforcement Learning Khetarpal et al., 2020. Brings affordances into reinforcement learning: an agent that knows which actions are feasible where can plan and learn models faster than one that assumes every action is always available.
        • Source · The Ethnography of Infrastructure Star, 1999. A call to study the unexciting systems other work runs on. Infrastructure means different things to different groups, goes unnoticed while it works, and takes its own methods to observe.
        • Source · Explaining Multistability: Postphenomenology and Affordances of Technologies de Boer, 2023. Asks why a technology can be put to many uses but not to just any use, and answers with affordances: what a thing offers the people using it bounds the uses that settle.
        • Source · How Well Does Agent Development Reflect Real-World Work? Wang et al., 2026. Compares where agent benchmarks concentrate with how human work is spread across the labor market, to see how well agent development represents real jobs.
        • Source · ReAct: Synergizing Reasoning and Acting in Language Models Yao et al., 2023. Introduces ReAct, a way of prompting a language model to alternate between reasoning in text and taking actions, so each informs the other.
        • Source · Workflows and Automation Trammell, 2026. Suggests tasks are bundled into jobs because doing one makes a worker better at related ones, and shows that this can make automation pay off little at first and sharply once a whole cluster of tasks is covered.
        • Source · SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering Yang et al., 2024. Introduces SWE-agent and the agent-computer interface: commands and feedback designed for a language model, which improve how well it can resolve issues in real code repositories.
        • Source · An Activity Theory Approach to Affordance Bærentsen & Trettvik, 2002. Re-reads Gibson's affordances through activity theory, arguing that HCI's troubles with the concept come from leaving out the activity an affordance belongs to.
        • Source · Organizational Learning and Communities-of-Practice: Toward a Unified View of Working, Learning, and Innovation Brown & Duguid, 1991. Finds that work as it is done differs from how manuals and job descriptions present it, and that the informal communities where people work practice out together are where an organization learns and innovates.
        • Source · The Theory of Affordances Gibson, 1979. The original statement of affordances: what an environment offers an animal, for good or ill, measured against that animal's size and capacities.
        • Source · Affordances Enable Partial World Modeling with LLMs Khetarpal et al., 2026. Uses affordances to have a language model build partial models of the world, covering only what matters for the task at hand instead of searching over a full one.
        • Source · Technology Affordances Gaver, 1991. Applies affordances to interface design by separating what an artifact allows from the information that reveals it, which yields perceptible, hidden and false affordances.
        • Source · Reflexion: Language Agents with Verbal Reinforcement Learning Shinn et al., 2023. Introduces Reflexion, in which an agent writes a short reflection on a failed attempt and keeps it in memory, so it improves over trials without any change to the model's weights.
        • Source · Do As I Can, Not As I Say: Grounding Language in Robotic Affordances Ichter et al., 2023. Grounds a language model's suggestions in what a robot can do: learned value functions score how likely each skill is to succeed from the current state, and the plan follows both.
        • Source · Affordance, Conventions, and Design Norman, 1999. Norman clarifies his own term: on a screen, designers mostly control perceived affordances and conventions, and showing that an action is possible differs from making it possible.
        • Source · Affordance as General Value Function: A Computational Model Graves et al., 2020. Casts affordances as predictions an agent can learn: general value functions that estimate where following a behavior from the current state would lead.
        • Source · Nature and Technology in History Schatzki, 2003. Widens history beyond human action: nature and technology are part of the arrangements social life consists of, so they belong in any account of how it changes.
        • Source · Designing for instrument-mediated activity Béguin & Rabardel, 2000. Treats a tool as an instrument only once a user has built ways of using it, and asks what designing around that activity, and not around the object alone, would involve.
        • Source · Agentic Large Language Models, a Survey Plaat et al., 2025. Surveys agentic language models under three headings: models that reason, models that act through tools and robots, and models that interact with other agents.
        • Source · On structural change: practice organizations and institutional logics Schatzki, 2024. Reads the institutional logics approach alongside practice theory to explain how the organization of practices changes.
        • Source · Organizational Routines as a Source of Continuous Change Feldman, 2000. Shows that organizational routines, usually defined by their stability, change continually as the people performing them respond to how the last round went.
        • Source · What Work Does Generative AI Do? Bick et al., 2026. Uses the Real-Time Population Survey to measure how many U.S. workers use generative AI on the job, how that has grown since 2024, and how use differs by occupation.
        • Source · A Survey of Large Language Models Zhao et al., 2026. A broad survey of large language models: how they are pre-trained, adapted, used and evaluated.
      • reproduces 2 Programs
      • reads 3 Documents
        • Document · Future of Workspaces The Future of Workspaces is a labor economics and policy focused research direction part of the larger exploratory Workspace Research Agenda, sitting alongside the design-focused Agent-Computer-Human-Experience (ACHE) and the philosophically-focused Agent Phenomenology research…
        • Document · ACHE Agent-Computer-Human Experience. Agent-Computer-Human Experience (ACHE) is a product design and software engineering focused research direction within the larger Workspace Research Agenda, exploring how we might best build interfaces and software that can dually support humans and Large Language Model (LLM)…
        • Document · Agent Phenomenology Agent Phenomenology explores Large Language Models (LLMs), LLM agents, and our human use of them, through a framework that draws on Heideggerian phenomenology, post-phenomenology, distributed cognition, extended mind theory, and philosophical accounts of delegation,…
      • explores 2 Datasets
        • Dataset · O*NET Occupational Information Network. The U.S. Department of Labor's occupation database covers tasks, skills, knowledge, and work context.
        • Dataset · RPS Real-Time Population Survey. A recurring survey tracks work and generative AI use among U.S. working-age adults.
Shared content

Research work packages diverse and extensive information, including source citations, datasets, related work, experiments, and the feedback of readers across revisions. While I am currently investigating a more ecological account of AI and LLM agents in the exploratory Workspace Research Agenda and its directions Future of Workspaces (AI exposure), Agent-Computer-Human Experience (design), and Agent Phenomenology (philosophy), this work also informs Commonspace, a research workspace hub aimed at making it easy to collect, relate, produce, and structure research both for working with agents as well as with others through agents.