The W3C JSON-LD Community Group

Go Back


W3C Logo

JSON-LD CG

Minutes for 2024-02-07

Topic: Announcements and Introductions

Gregg Kellogg is scribing.

Topic: YAML-LD

Anatoly Scherbakov: --Shares screen--
Anatoly Scherbakov: I created issue json-ld/yaml-ld#131 for script tags in HTML.
https://github.com/json-ld/yaml-ld/issues/131 -> Issue 131 YAML under `<script>` HTML tag unsupported in the spec (by anatoly-scherbakov) [bug]
... I also created json-ld/yaml-ld#132, although PR Preview is having problems.
https://github.com/json-ld/yaml-ld/pull/132 -> Pull Request 132 (closes #131) Embed YAML-LD into HTML `<script>` tag (by anatoly-scherbakov) [blocked-by-recharter]
Anatoly Scherbakov: Changes involve spec changes and some tests.
... We need to un-indent the code block to process a YAML stream.
... I have a Python implementation, that also runs JSON-LD tests, as it's a superset of JSON.
... This PR can't be merged due to the status of work in the WG. There is a charter issue that needs to be resolved.

Subtopic: updated WG charter

Benjamin Young: We haven't done any work on a charter.
Gregg Kellogg: Current charter only allows work on notes, not rec-track.
Benjamin Young: Plan is a re-charter of https://www.w3.org/2023/01/json-ld-wg-charter.html
Benjamin Young: Needs to be voted on by membership.
... I think we can continue to work on the document, but there are different patent policy considerations
Ted Thibodeau Jr.: We can vote that the document is a final report (which we have done).
... Theoretically, it should stay static. When the WG picks it up, we should stop work.
Benjamin Young: I think the work fell through the cracks with the WG.
Benjamin Young: We'll do a WG meeting in four weeks on March 6th.
Benjamin Young: I was trying to find out where the charter lives.
Anatoly Scherbakov: What is the WG process if re-chartered? Does the WG copy a snapshot?
Gregg Kellogg: Typically it is just merged over.
Anatoly Scherbakov: So, we can make PRs and merge later.

Topic: CBOR-LD

David I. Lehn: I haven't done any work on the spec.
Gregg Kellogg: We probably need to publish a final report.
Niklas Lindström: The https://github.com/json-ld/charter repo is inactive, and this is the current charter (mentions YAML and CBOR)? https://www.w3.org/2023/01/json-ld-wg-charter.html
Gregg Kellogg: Maybe could be published as a draft report
Benjamin Young: How many links do we want to publish?
... Do we want to issue both as notes?
Gregg Kellogg: I'd say list both as draft reports.
Benjamin Young: Maybe should move PA's version to the json-ld org, so it's not confusing.
ACTION: pchampin to move JSON-LD in CBOR draft to json-ld org.
Sorry, I don't know what repository to use.
Benjamin Young: Repo to pchampin's spec https://github.com/w3c/json-ld-cbor/
Benjamin Young: Question is do they need to be finalized for the WG to take them up?
Benjamin Young: Implementation https://github.com/digitalbazaar/cborld
Gregg Kellogg: It would be good to have issues added to CBOR-LD to indicate needed changes to match implementations.
Benjamin Young: It takes some time to make those updates, and time is short. I referenced a link to the implementation.
Gregg Kellogg: I'd like to see something with a bit more of a summary for how implementation diverges from spec.
David I. Lehn: For the current stuff, before we tag it, I want to double-check to make sure URLs are up to date.
... Not sure if people are familiar with how the spec works.
Gregg Kellogg: Editorial changes, including editors notes, should be fine. normative changes may need more agreement.

Topic: JSON-LD Issue Discussion

David I. Lehn: Do we need a process for merging tests.
For example w3c/json-ld-api#591
https://github.com/w3c/json-ld-api/pull/591 -> Pull Request 591 Add test for `@context` redefinition. (by davidlehn) [test:missing-coverage]
David I. Lehn: Looked into this, and it looked like an error.
... It was a bug in JSON-LD where it was too generous.
... I think our implementation would have ultimately failed.
Niklas Lindström: https://w3c.github.io/json-ld-api/#algorithm-0 (step 4 and 5)
David I. Lehn: When we don't have tests it can be hard to tell.
Gregg Kellogg: I think its an issue that we don't specifically linked from normative statements to tests.
Niklas Lindström: +1 For test coverage for each normative statement
Niklas Lindström: And +1 for predictable naming of those ...
Gregg Kellogg: Not too difficult, but time consuming.
David I. Lehn: On the same topic, if all implementations did coverage reports, that would be useful.
Gregg Kellogg: I would say that test changes that have developer approval can be merged, as long as they don't require a normative change to the spec.
ACTION: dlehn to tag reviewers on tests to be considered for merging.
Sorry, I don't know what repository to use.
Gregg Kellogg: Also work should go forward on the charter where people can contribute.

Topic: Open Discussion

Gregg Kellogg: Meet again on February 21.