You are here. See how this question connects to other ideas.
Select a node to open its page · Expand to read within the map
One question
What does a path give an agent?
A path identifies a resource within a namespace, so operations can refer to it explicitly.
- Namespacegives scope to→
- /work/reportresolves to→
- Resource
Consider the illustrative address /work/report. A reader and a tool can refer to that object without describing where it appears on a screen. A provider resolves the address to the underlying resource.
The path is not the content itself. Nor is it a semantic classification: putting something under /research does not explain its meaning to a model.
An address must be interpreted in its namespace and lifetime. The same string in two projected worlds can resolve differently. A stable contract is an explicit design choice, not a property granted by slash characters.
Read further
A namespace is the set of scope rules used to interpret an address. Think of a resource directory with a defined entry point: /work/report in project A need not identify the report found at the same string in project B.
A small example: a card displays “Meeting report,” while a tool reads the illustrative address /work/report. Moving the card into a right-hand column need not change the tool’s target. Moving or replacing the report itself is different: the application must preserve the old address, expose a new one, or explain that the reference has expired. The agreement concerns the object reference, not its screen coordinates.
Check your understanding
Why might moving a card between columns require no change to a tool?
The tool can address the object independently of its visual position, provided its resource address and operation contract remain valid.
Continue along a learning path
- Start with AFSStep 2 of 5
- What should endure?Step 2 of 6