Hi Brian,
Thank you very much for the helpful comments.
> I think it would be better to do it in JSON, however, or CDDL if your target
> protocol is GRASP
The structure definitions in this document are intended to provide a
machine-readable description of Service Intent with explicit typing for each
field (e.g., uint, tstr, bool). The current notation follows a CDDL-like style
to constrain the data format and avoid ambiguity. The origin of intent is
outside the scope of this draft. A Service Intent object may originate from
various sources, such as a local service orchestrator, an application-specific
ASA, or other management logic within a node. If the working group considers
that Service Intent should be distributed and synchronized between nodes, the
draft could further define a GRASP Objective carrying the intent object.
Otherwise, if the primary use case is interaction between components within a
node, a JSON-style representation could also be considered.
> I appreciate the idea of "Intent Scope" but I think it introduces hard
> problems - how do domains aquire a name, how is the border identified, can
> domains overlap, and how do hosts know reliably which domains they are in
In the current version of the document, the scope definition assumes the
existence of an Autonomic Domain but does not attempt to define how domains
obtain human-readable names or how domain boundaries are identified. Within the
ANIMA architecture, domain identification mechanisms are already defined in
related specifications. For example, RFC 8995 (BRSKI) introduces the concept of
a domainID that provides a stable and globally unique identifier for an
autonomic domain.
Therefore, the intent scope in this document aligns with the existing notion of
an Autonomic Domain defined by the ANIMA architecture, while the mechanisms for
domain identification and bootstrapping are handled by protocols such as BRSKI
and ACP. We will consider adding clarifying text referencing these mechanisms
in the next revision.
Best regards,
Longwei Zhu
From: Brian E Carpenter
Date: 2026-03-07 07:14
To: Anima WG; draft-zhu-anima-service-intent
Subject: Re: I-D Action: draft-zhu-anima-service-intent-00.txt
Hi,
I think it's a good idea to work out how "intent" can be objectified. I think
it would be better to do it in JSON, however, or CDDL if your target protocol
is GRASP.
I appreciate the idea of "Intent Scope" but I think it introduces hard problems
- how do domains aquire a name, how is the border identified, can domains
overlap, and how do hosts know reliably which domains they are in? (These
questions are not answered by RFC 8799, but it may explain why the problem is
hard.)
Regards/Ngā mihi
Brian
On 03-Mar-26 01:15, [email protected] wrote:
> Internet-Draft draft-zhu-anima-service-intent-00.txt is now available.
>
> Title: Definition of Service Intent in Autonomic Networks
> Authors: Longwei Zhu
> Bizhu Wang
> Sheng Jiang
> Name: draft-zhu-anima-service-intent-00.txt
> Pages: 10
> Dates: 2026-03-02
>
> Abstract:
>
> While ANIMA Intent enables goal-oriented control within an Autonomic
> Domain, emerging services (e.g., AI inference) require a common,
> interoperable representation for expressing service-level objectives
> and constraints that span network, compute, and storage resources,
> rather than connection-centric descriptions. This document defines
> Service Intent for Autonomic Networks by specifying a structured
> semantic model and a concise format with identification, scope,
> versioning, and lifecycle semantics.
>
> The IETF datatracker status page for this Internet-Draft is:
> https://datatracker.ietf.org/doc/draft-zhu-anima-service-intent/
>
> There is also an HTMLized version available at:
> https://datatracker.ietf.org/doc/html/draft-zhu-anima-service-intent-00
>
> Internet-Drafts are also available by rsync at:
> rsync.ietf.org::internet-drafts
>
>
> _______________________________________________
> I-D-Announce mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
_______________________________________________
Anima mailing list -- [email protected]
To unsubscribe send an email to [email protected]
