You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
First, thank you for the excellent and detailed work on the A2A protocol specification.
As I'm planning some implementation, I have a question regarding the design of the tasks/get method. I noticed that it retrieves a task using only the taskId as a unique identifier.
From a server-side perspective, I'm envisioning a hierarchical data model where tasks are naturally scoped within a session or conversation, which corresponds to the contextId. The logical structure would be something like:
Session (identified by contextId) -> contains multiple -> Tasks (each with a taskId)
In this kind of architecture, it would be more direct and potentially more performant to locate a specific task by providing both the contextId and the taskId.
The current design, which relies on a globally unique taskId, seems to imply that the server must maintain a separate, global index or Task Manager to map a taskId back to its parent context. This adds a bit of implementation complexity and a potential extra lookup step.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
-
Hello A2A Team,
First, thank you for the excellent and detailed work on the A2A protocol specification.
As I'm planning some implementation, I have a question regarding the design of the
tasks/get
method. I noticed that it retrieves a task using only thetaskId
as a unique identifier.From a server-side perspective, I'm envisioning a hierarchical data model where tasks are naturally scoped within a session or conversation, which corresponds to the contextId. The logical structure would be something like:
Session (identified by contextId) -> contains multiple -> Tasks (each with a taskId)
In this kind of architecture, it would be more direct and potentially more performant to locate a specific task by providing both the
contextId
and thetaskId
.The current design, which relies on a globally unique taskId, seems to imply that the server must maintain a separate, global index or
Task Manager
to map a taskId back to its parent context. This adds a bit of implementation complexity and a potential extra lookup step.Please correct me if my thinking is wrong.
Beta Was this translation helpful? Give feedback.
All reactions