[Vote] Should artifact type move out of annotations

[sudo-bmitch]

Voting seemed to work for picking a direction in this group, so here's a vote for #59. Should we move the annotation to be a first class type in artifact-spec and depend on the config media type in image-spec? Vote by giving a thumbs up to one of the two comments below. We'll leave this open for 1 week, and go with the community's decision on the 2022-07-19 meeting.

[sudo-bmitch]

Yes, get rid of the artifact type annotation and make it a dedicated field in artifact-spec, plus depend on the config media type in image-spec.

[sudo-bmitch]

No, use the annotation as described in proposal E today.

[jonjohnsonjr]

I'm uncomfortable voting on this without a definition of artifactType somewhere.

[sudo-bmitch]

I'm uncomfortable voting on this without a definition of artifactType somewhere.

It would follow the same IANA usage we have with the config media type.

https://github.com/opencontainers/artifacts/blob/main/artifact-authors.md#defining-a-unique-artifact-type

My plan was to formally define that in a proposed spec of this is the direction we pick. Does that cover what you're looking for?

[jdolitsky]

My understanding is that a new artifactType field should be expected to have same format/validation as mediaType

[SteveLasker]

EDIT:
I moved my comment here to the discussion on #59, leaving this for a pure vote, while #59 is the discussion.

[jdolitsky]

We've decided to go with the first option for now:

Yes, get rid of the artifact type annotation and make it a dedicated field in artifact-spec, plus depend on the config media type in image-spec.

[sudo-bmitch]

@silvin-lubecki @chris-crone A concern raced in the meeting is that #59 removes support for Docker Hub. Are there plans for Hub to support arbitrary config descriptor media types?

[sudo-bmitch]

Docker says they'll have support for OCI Artifacts in Hub soon, so I'm retracting my No vote.

[sudo-bmitch]

From today's meeting, we've moved forward with #59