RFC #564: Do typed ports belong in the ADR-0047 package spec?

[znerol2]

Moved from #601 so the question can be discussed before any work is scheduled. The original issue is closed and links here.

---

The question

A connection between two media functions can only be checked if each function declares typed ports. Should those declarations be added to the ADR-0047 package spec as an extension, or defined somewhere else?

What the plan proposes

• Section 7 (step 3) puts typed ports in the design schema, with two-way validation. Typed ports are machine-readable compatibility properties such as essence, formats, resolutions, frame rates, PTP requirement and 2022-7 support. This would let a design surface reject impossible connections before deployment.
• Section 3.4 notes that a stub already exists. The one live catalog entry declares mxl_flows with produces and consumes, and no code reads it. The plan proposes growing that field rather than adding a parallel declaration.
• Section 3.4 warns about a vocabulary collision. ADR-0047 already uses "ports" to mean IPAM service ports. The plan proposes a distinct name, such as interfaces or media_ports, from the first commit.
• Section 4 states the binding rule: a package may declare anything, but the platform binds any declaration that has effects outside the function's own pod.
• Whether this step needs an amendment to ADR-0047 is part of the question.

What would change the answer

• A decision that ADR-0047 should stay limited to packaging and trust, with function semantics in a separate spec.
• Evidence that a typed-port schema cannot stay vendor-neutral enough for a package spec.
• A different name or shape for the existing mxl_flows field.

Links

• Discussion: https://github.com/orgs/dmfdeploy/discussions/564
• Plan: https://github.com/dmfdeploy/dmfdeploy/blob/main/docs/plans/DMF%20Media%20Workload%20Design%20Canvas%20Plan.md

This is a request for comment. Answering it does not start any work.