Discovery of Agents With Names
charter-ietf-dawn-00-03
Block
Yes
No Objection
Andy Newton
Gunter Van de Velde
Jim Guichard
No Record
Christopher Inacio
Deb Cooley
Gorry Fairhurst
Ketan Talaulikar
Mahesh Jethanandani
Mike Bishop
Tommy Jensen
Summary: Has a BLOCK. Has enough positions to pass once BLOCK positions are resolved.
Ballot question: "Is this charter ready for external review?"
Roman Danyliw
Block
Block
(2026-08-31)
Sent
** Scope Discovery in the DAWN context is limited to a client requesting, "Find me an agent resource to interact with," within a local network, within an organization, or between organizations. [Roman] What is the scoping being provided by the text: “within a local network, within an organization, or between organizations”? -- Since “local network”, “organization” and “between organizations” are increasing larger scopes, is the text suggesting that different solutions are needed? -- what is the difference in scope between “the internet” and “between organizations”? -- If the solution is unified, isn’t this scope “the Internet”? ** Scope The WG work items will consider having confidentiality and integrity of all the information exchanged between the DAWN parties. [Roman] Since the WG only has to “consider” some security properties, does that mean they can be ignored? What scoping does this optional guidance provide. ** Out of Scope • Exchange of capability information beyond the minimum discovery information. [Roman] What information constitutes “minimum discovery information”? I’ll note the possible information listed in the Scope section was framed as “This *may* include communication about the following *example* attributes”, emphasis mine ** Out of Scope • Interoperable information schemas beyond discovering a transfer protocol. [Roman] Doesn’t the Scope section suggest that more information is needed than just identifying a “transfer protocol”. For example, it mentions capturing a “resource type”, “classification”, and “minimal requirements”? All of those seem like information elements beyond a transfer protocol.
Comment
(2026-08-31)
Sent
** Scope What is the agent resource's type and classification [Roman] What does “resource type” and “classification” mean?
Charles Eckel
Yes
Comment
(2026-08-28)
Not sent
This is a much improved version of the charter. It's helpful that "stable identifiers of AI agents, tools and skills" and "trust management... beyond secure content exchange" are clearly stated as out of scope — that keeps DAWN focused on discovery and leaves identity and authorization to other efforts. FYI for the group: WIMSE is currently mid-adoption of draft-klrc-aiagent-auth, which covers agent identity and authorization and itself points to OAuth for delegation mechanics.
Éric Vyncke
Yes
Comment
(2026-08-24 for -00-00)
Sent
The BoF chairs have worked on the charter with the dawn@ietf.org mailing list participants, and the dawn@ietf.org has converged to this current charter proposal.
Mohamed Boucadair
(was Block)
Yes
Comment
(2026-08-28)
Sent
Hi all, Thank you for the effort put into preparing this charter and the changes made so far [1]. I’m very supportive of this work. There is one pending point from my previous ballot [1]: # It is not clear what is the trigger of the discovery in DAWN and what is the underlying operational model. Specifically: ## Do we assume that the discovery is always triggered by a “name” that is known somehow to the client? Cheers, Med [1] https://author-tools.ietf.org/iddiff?url1=https%3A%2F%2Fdatatracker.ietf.org%2Fdoc%2Fcharter-ietf-dawn%2Fwithmilestones-00-00.txt&url2=https%3A%2F%2Fdatatracker.ietf.org%2Fdoc%2Fcharter-ietf-dawn%2Fwithmilestones-00-03.txt&difftype=--html [2] https://mailarchive.ietf.org/arch/msg/dawn/sufBO-6nP6G6VEtydtJbmfYrcts/
Andy Newton
No Objection
Gunter Van de Velde
No Objection
Jim Guichard
No Objection
Christopher Inacio
No Record
Deb Cooley
No Record
Gorry Fairhurst
No Record
Ketan Talaulikar
No Record
Mahesh Jethanandani
No Record
Mike Bishop
No Record
Tommy Jensen
No Record