[Agentproto] Re: Gunter Van de Velde's No Objection on charter-ietf-agentproto-00-01: (with COMMENT)

[Behcet Sarikaya <[e-mail]>]

Hi Charles,


On Thu, Oct 1, 2026 at 6:28 PM Charles Eckel (eckelcu) <eckelcu=
[e-mail]> wrote:

> Hi Gunter,
>
> Thanks for your feedback regarding what "protocol" means in the context of
> the agentproto charter (tracked in Issue #111).
>
> In reviewing our work-in-progress charter text (in GitHub but not yet
> uploaded to the datatracker), we believe this is answered by the following:
>
>

> "The agentic dialog management protocol neither replaces application-layer
> agent communication protocols nor defines a new transport protocol.
>

You are saying that it is a session layer
protocol.

Why not spell it out?

Behcet

> It defines dialog identifiers, lifecycle semantics, and the propagation of
> dialog context, together with bindings specifying how that context is
> carried over existing IETF protocols. A dialog context is defined
> independently of any particular application: a participant that understands
> the dialog context can correlate and maintain a dialog, and the protocol
> does not constrain how the application is structured or what content it
> exchanges."
>
> We also explored further clarifications in PR #118, but weren't able to
> reach consensus on specific text changes beyond what is already captured.
>
> Please let me know if you are ok with this, or if you feel an additional
> change is still needed.
>
> Thanks again,
> Charles
>
> On Sep 16, 2026, at 9:29 AM, Charles Eckel (eckelcu) <[e-mail]>
> wrote:
>
> Hi Gunter,
>
> Thank you for your support for this work.
>
> I created the following issue to track your question regarding the use of
> the term “protocol”.
> https://github.com/ietf-artarea/charters/issues/111
>
> Cheers,
> Charles
>
> On Sep 16, 2026, at 2:35 PM, Gunter Van de Velde via Datatracker <
> [e-mail]> wrote:
>
> ----------------------------------------------------------------------
> COMMENT:
> ----------------------------------------------------------------------
>
> I am a confident supported for this work in AI space.
>
> This proposal sits primarily in the application space, outside Routing, so
> my
> expertise in this area is limited.
>
> I found the use of the term “protocol” somewhat ambiguous. In a routing
> context, I would normally understand a protocol to mean a wire protocol
> with
> its own message framing, common header, and mandatory or optional fields or
> TLVs. I suspect that is not what is intended here.
>
> Could the charter clarify what “protocol” means in this context? In
> particular,
> is the WG expected to define a new application protocol or wire encoding
> carried over HTTP or QUIC, or a data model and serialization—for example,
> using
> XML or JSON—for exchanging information over existing protocols? Clarifying
> this
> would make the intended scope easier to understand.
>
>
>
> _______________________________________________
> Agentproto mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]
>