Hi all, I have just submitted a new individual Internet-Draft:
NAIM defines a canonical JSON representation (with a derived human-readable Markdown view) as a structured semantic intermediate layer between natural language descriptions and YANG data models (RFC 7950). The main motivation is to address the semantic gap in current AI-assisted YANG authoring workflows. Direct conversion from natural language to YANG often leads to ambiguity in critical modeling aspects such as: - configuration versus operational state distinction - list key identification - operational constraints and preconditions - visibility conditions (if-feature / when) - cross-node relationships NAIM makes these semantics explicit, machine-verifiable, and suitable for both automated processing and human review before deterministic YANG generation occurs. This draft specifies the canonical JSON structure, Markdown rendering rules, node template families, type representation, visibility/deviation declarations, and interoperability expectations. It does not prescribe specific AI models, prompt strategies, or internal tool workflows. I would greatly appreciate any feedback, comments, or suggestions from the working group. Thank you in advance for your time and valuable feedback. Best regards, Chong Feng Ruijie Networks Email: [email protected] <[email protected]> 于 2026年4月17日周五 下午10:52写道： > A new version of Internet-Draft draft-feng-netmod-naim-00.txt has been > successfully submitted by Chong Feng and posted to the > IETF repository. > > Name: draft-feng-netmod-naim > Revision: 00 > Title: NAIM: A Canonical Data Format for AI-Assisted YANG Modeling > Date: 2026-04-17 > Group: Individual Submission > Pages: 28 > URL: https://www.ietf.org/archive/id/draft-feng-netmod-naim-00.txt > Status: https://datatracker.ietf.org/doc/draft-feng-netmod-naim/ > HTML: https://www.ietf.org/archive/id/draft-feng-netmod-naim-00.html > HTMLized: https://datatracker.ietf.org/doc/html/draft-feng-netmod-naim > > > Abstract: > > This document defines the NAIM (Natural AI Interface Modeling) > Document format, a canonical JSON representation and a derived human- > readable Markdown view designed to serve as a structured semantic > intermediate representation between natural language descriptions and > YANG data models [RFC7950]. > > NAIM addresses a recognized gap in the YANG authoring workflow: > direct conversion from natural language to YANG is error-prone > because essential modeling semantics — including configuration versus > state distinction, list key identification, constraint expressions, > operational preconditions, and cross-module relationships — are > routinely absent or ambiguous in natural language input. NAIM makes > these semantics explicit, structurally consistent, and machine- > verifiable before YANG generation occurs. > > This document standardizes the canonical JSON structure, Markdown > rendering rules, node template families, type representation, > visibility and deviation declarations, and interoperability > expectations for tools that produce and consume NAIM artifacts. It > does not standardize AI model selection, prompt engineering > strategies, dialogue policies, implementation-internal workflow > logic, or runtime execution systems. > > > > The IETF Secretariat > > >
_______________________________________________ netmod mailing list -- [email protected] To unsubscribe send an email to [email protected]
