Skip to main content
Knowledge mapAFS-UI

You are here. See how this question connects to other ideas.

Select a node to open its page · Expand to read within the map

Knowledge mapFollow a connection. Understand a question.
← Learning

LEARN / AFS + AFS-UI

From context to interface.

A task can appear as a card, a row or a tool’s target. Learn how people and agents can work with the same object as its presentation changes.

Start with AFS

Choose a starting point

  1. 01
    Start with AFS

    Understand what a task can name, access and operate.

    New to AFS · resources and addresses5 questions
  2. 02
    Read an interface from both sides

    Connect what a person sees with the structure an agent can use.

    After AFS · connect views and objects5 questions
  3. 03
    What should endure?

    Follow the architectural questions behind the keynote.

    Architecture · agreements that should endure6 questions
  4. 04
    From protocol to site

    Understand an interface tree, its protocol and how a website presents it.

    Developers · distinguish protocol and website implementation9 questions
  5. 05
    Understand neighboring approaches

    Put protocols, SDKs, hosts and design tools in their respective places.

    After AUP · compare responsibilities8 questions
  6. 06
    From finding a resource to operating it reliably

    Distinguish addresses, capabilities, updates and failures in a shared working environment.

    After AFS · resource behavior and reliability6 questions

A TALK THROUGH THE IDEAS

What endures when everything evolves?

From an X Window memory to shared context, addresses and displays—and research questions that can test those choices.

Explore the connections

Each explanation stands on its own. “Builds on” links show the background it uses.

01

A world you can address

Resources, paths and operations

  1. What does AFS organize?

    AFS gives a task a named, inspectable view of the resources it can work with.

  2. What does a path give an agent?

    A path identifies a resource within a namespace, so operations can refer to it explicitly.

  3. How does a service enter AFS?

    A provider implements resource operations; a mount places that provider in a namespace.

  4. Why do operation semantics matter?

    Naming a resource is useful only when consumers can understand what operations on it mean.

  5. Does shared context mean everyone sees everything?

    A Small World presents the resources relevant to one observer, rather than exposing the whole system.

  6. Where does a mount put a resource?

    An entry path and the resource’s original location are different things.

  7. Does an address tell you what you can do?

    Location, supported operations and caller authority are separate questions.

  8. How does another display learn about a change?

    Reading the same address and receiving changes are separate steps.

  9. What happens when two people edit at once?

    A shared address does not decide how stale edits are handled.

  10. Why distinguish search from query?

    Finding relevant content differs from selecting records by explicit conditions.

  11. Should a failed operation be retried or handled differently?

    Missing, unsupported, denied and conflicting operations need different responses.

02

An interface you can inspect

Views over operational context

  1. What does AFS-UI add?

    AFS-UI treats the interface as a projection over operational context that humans and agents can work with.

  2. What does the agent see behind a button?

    A visual control can correspond to named structure, state and actions, rather than only a region of pixels.

  3. Can the view change while the object stays the same?

    Separating context from projection lets multiple views refer to the same operational objects.

  4. How are a display and a session different?

    A display presents and accepts interaction; a session groups the operational state associated with an interaction context.

  5. What would looking behind the interface reveal?

    Inspection should expose the actual structure of the active view and its relationship to state.

03

Ideas that endure

Boundaries that survive change

  1. What did X Window teach us?

    Computation and the place where a person sees and controls it can be separated.

  2. What makes separate parts composable?

    Compatible resource addresses and operations let parts cooperate without sharing every implementation detail.

04

The interface protocol

Nodes, bindings, events and capabilities

  1. What is AUP?

    Agentic UI Protocol describes interface structure and its connections to state and actions.

  2. What is in an AUP tree?

    Read identities, types, properties and children through one task card.

    Builds on ↑What is AUP?
  3. How do src, bind and propBind differ?

    Reading resources, writing input and updating properties are distinct connections.

  4. What happens after a click?

    Trace an interface event to an explicit operation target.

  5. Can every device render the same interface?

    Capability declarations and degradation determine how a structure reaches its target.

    Builds on ↑What is AUP?
  6. Which primitives does AUP define?

    Eighteen core types provide an entry into the interface vocabulary.

    Builds on ↑What is AUP?
05

Building for the Web

Sites, themes and implementation boundaries

  1. What is Web Device?

    Organizing content and interfaces into a website is another layer of work.

    Builds on ↑What is AUP?
  2. How do AUP and Web Device divide responsibilities?

    One describes interface contracts; the other organizes a Web site implementation.

  3. What are Web Device components?

    Distinguish core primitives, Web-only types and theme components, then browse the default theme.

06

The primitive vocabulary

Eighteen types, one question at a time

  1. view: How do you group content?

    view organizes child nodes and expresses their arrangement.

  2. text: How does text become a node?

    text expresses words, heading levels and supported formats.

  3. media: How do you express images, audio and video?

    media expresses media kind, source and alternative information.

  4. input: How do you receive user input?

    input describes an input control; change represents value changes.

  5. action: How does a button express an action?

    action expresses a triggerable action or navigation entry.

  6. overlay: Where does an overlay keep its open state?

    overlay expresses dialogs and other overlaid interfaces, with explicit open or closed state.

  7. table: When should data be a table?

    table uses column definitions and row data for comparable records.

  8. time: How does time differ from a schedule?

    time expresses dates, clocks, countdowns and related time information.

  9. chart: How do chart semantics differ from a chart library?

    chart expresses data visualization; Web drawing has its own implementation.

  10. map: Is map an arbitrary spatial graph?

    map addresses geographic locations, map centers and markers.

  11. calendar: How do you express scheduled events?

    calendar organizes events into month, week, day or agenda views.

  12. chat: What does chat in the core vocabulary mean?

    chat is a core type, but needs a supporting session target.

  13. rtc: Does a reserved type mean a completed capability?

    rtc reserves a name for real-time communication; it is not yet a usable calling control.

  14. explorer: How do you browse AFS resources?

    explorer describes a resource-browser subsystem; the runtime must support its expansion and presentation.

  15. editor: Does an editor automatically save resources?

    editor supports code or text editing; saving depends on explicit connections.

  16. canvas: Are canvas strokes shared data?

    canvas provides freeform drawing; the application connects saving and sharing.

  17. surface: How do you embed a resource or interface?

    surface provides an embedding boundary for resources or remote AUP.

  18. afs-list: Can the same resources have different list views?

    afs-list presents AFS resource collections as lists, grids and other arrangements.

07

Neighboring approaches

Protocols, SDKs and design workflows

  1. How should generative UI approaches be compared?

    Compare responsibilities, addresses, state and interaction contracts.

  2. What does A2UI solve?

    Render agent-provided interface descriptions through client-supported components.

    Builds on ↑What is AUP?
  3. What are MCP Apps?

    Bring interactive HTML interfaces from MCP tools into supporting hosts.

  4. How do MCP-UI and MCP Apps relate?

    Separate the standard from implementation tooling and legacy compatibility.

    Builds on ↑What are MCP Apps?
  5. Are AG-UI and A2UI the same layer?

    Agent–application event exchange differs from interface description.

    Builds on ↑What is AUP?
  6. Is Flutter GenUI SDK another protocol?

    It provides implementation tools for generative interfaces in Flutter.

  7. Where does Google Stitch fit?

    Compare its design workflow separately from runtime protocols.

  8. Does every path address the same kind of thing?

    Compare scope, resolution and available operations.