The W3C JSON-LD Community Group

Go Back

W3C Logo

JSON-LD WG

Minutes for 2024-10-16

Present
Benjamin Young, Ted Thibodeau Jr., Gregg Kellogg, Anatoly Scherbakov, Niklas Lindström, Pierre-Antoine Champin, Ivan Herman, David I. Lehn
Chair(s)
Benjamin Young
Scribe(s)
Gregg Kellogg, Benjamin Young
Agenda
https://www.w3.org/events/meetings/67486b62-8fda-478f-a379-9f8a297266fb/20241016T120000/
Topics
  1. Announcements and Introductions
  2. Issue Discussion

Topic: Announcements and Introductions

Gregg Kellogg is scribing.
Benjamin Young: "Will be held in November 2025 in Kobe Japan & online."
Benjamin Young: Next TPAC will be in Kobe Japan, in November 2025.

Topic: Issue Discussion

Benjamin Young: Primary objective is towards rechartering, so we need to label the issues accordingly.
... We may need new tags to surface issues related to the charter.
Pierre-Antoine Champin: I had an action to create lables in the repos for class 1-4 changes, which I just did.
Pierre-Antoine Champin: Instead of having an exhausive list of changes in the charter, we'd like to have a link to a stable list.
Benjamin Young is scribing.
Gregg Kellogg: Class 1 & 2 changes are considered errata
... and we have a tag for that already, so maybe we can avoid retagging
... Class 3 are slightly different
Pierre-Antoine Champin: Some Errata are non-editorial, so they are class 3 changes.
... For me, the errata tags are used for things that were raised as such according to the process.
... There is some redundancy.
Benjamin Young: This is a great one to look at when considering issues.
... There are cases where ErratumRaised is also used with spec:editorial.
... The "Future Work" section is something to discuss, particularly w3c/json-ld-api#604.
https://github.com/w3c/json-ld-api/issues/604 -> Issue 604 Using flatten() on HTML input having a single script element with an object with `@graph` results in named graph (by gkellogg) [spec:enhancement] [ErratumRaised]
... I'd suggest we go through the "Future Work" list and classify what we can.
... We may make a category for things we don't intend to address.
... The Errata column would update the errata document.
... We also have some open PRs. If merged, they would go into the next addition.
Gregg Kellogg: Let's talk tests for a minute
... if we're making changes to an edited spec and there are corresponding tests, then those will need to change together before the next release
Benjamin Young: Do we want to block PRs until there are corresponding tests?
Gregg Kellogg: There are cases where we have tests, but no spec text to go with
... dlehn ? did you do some of these?
David I. Lehn: There are some changes related to rdf:JSON.
... For example, w3c/json-ld-api#559.
https://github.com/w3c/json-ld-api/pull/559 -> Pull Request 559 Add JSON literal tests. (by davidlehn) [spec:substantive] [test:missing-coverage]
... I wanted feedback on the tests before getting into spec updates.
... There are a number of discussion points still open on that PR.
Benjamin Young: Do we have a way to say that it is still pending spec changes.
... This one references w3c/json-ld-api#560.
https://github.com/w3c/json-ld-api/issues/560 -> Issue 560 Various `@json` processing issues. (by davidlehn) [test:missing-coverage] [ErratumRaised]
Benjamin Young: I think we need a new issue to add corresponding spec changes.
Ivan Herman: I'm a bit worried that if we consider just merging things.
... I'm not sure what the plans are for what are class 3 changes. The naming is up to the process doc editors.
... In my mind, class 1-2 changes are the same; we can make the changes and re-publish.
... Class 3 is separate; do we plan to go through the changes in our existing charter, or should they be postponed to the new charter?
... If we want to do something in this charter we need to be careful.
Gregg Kellogg: I did look through the changes we've made so far, and they seem suitable, but we should likely have the group reconsider them in light of the charter
Ivan Herman: I have the pleasure of doing this classification with another group
Ivan Herman: The class 3 changes need ins/del; if we re-publish we need to publish them as candidate changes.
... We need a horizontal review for those changes, leading to an AC vote.
... Class 1 & 2 changes can be made and re-published without horizontal review, as they're considered minor editorial changes.
Benjamin Young: Do we know of any class 3 changes?
Gregg Kellogg: We could look at it now, but I looked at closed PRs after we reached recommendation
... and we probably need to do that again to make sure we didn't put in any class 3 changes
... and if we did, then we'd need to mark them per the process
Benjamin Young: It seems we need to sort out that before making further spec changes.
Ivan Herman: Test issues are not interesting for this purpose.
Gregg Kellogg: We noticed during YAML-LD development that `extractAllScripts()` had a different default
... https://github.com/w3c/json-ld-api/pull/606
https://github.com/w3c/json-ld-api/pull/606 -> MERGED Pull Request 606 The default for extractAllScripts in toRdf() should be true. (by gkellogg) [spec:enhancement]
... probably that's a class 3 change.
Ivan Herman: So, w3c/json-ld-api#606 is a class three change.
... EPub has gone through this effort.
Benjamin Young: Is this one a class-3 change? (probably)
Pierre-Antoine Champin: More generally, we're discussing how to make class-3 changes to 1.1. Would it be simpler to just put these into 1.2?
Ivan Herman: It's hard to judge, but don't forget that if we go through the traditional route, it will take a long time before it becomes a rec.
... If there's some urgency, we should consider doing now.
... It's a judgement call for each.
Benjamin Young: We also have to consider that we might not get a re-charter. In any case, we're looking at multiple years. We need to take them case-by-case.
... Is there more to do?
Gregg Kellogg: I think we need issues to remediate any already merged class-3 changes.
Benjamin Young: So, testing PRs don't necessarily need new spec work.
Benjamin Young: Look at w3c/json-ld-api#578.
https://github.com/w3c/json-ld-api/pull/578 -> MERGED Pull Request 578 remove surious paragraph in the description of frameExpansion (by pchampin) [spec:editorial]
... That is class-2.
Benjamin Young: W3c/json-ld-api#574 is also class-2.
https://github.com/w3c/json-ld-api/pull/574 -> MERGED Pull Request 574 Clarify that lexicographical order uses code point ordering. (by gkellogg) [spec:editorial]
Benjamin Young: W3c/json-ld-api#575 is also class-2.
https://github.com/w3c/json-ld-api/pull/575 -> MERGED Pull Request 575 Add"lint-ignore" class to some unused definitions to silence warnings. (by gkellogg) [spec:editorial] [class-1]
Benjamin Young: W3c/json-ld-api#566 seems more consequential. This would be a class-3 change.
https://github.com/w3c/json-ld-api/pull/566 -> MERGED Pull Request 566 Fix context processing for reverse terms (by niklasl)
Benjamin Young: W3c/json-ld-api#569 is marked as editorial. So, class-2.
https://github.com/w3c/json-ld-api/pull/569 -> MERGED Pull Request 569 Minor editorial rephrasing to clarify (by TallTed) [spec:editorial] [class-2]
Benjamin Young: W3c/json-ld-api#528 is substantive, so class-3.
https://github.com/w3c/json-ld-api/pull/528 -> MERGED Pull Request 528 Fix compacting property-based indexes. (by gkellogg)
Benjamin Young: W3c/json-ld-api#520 is class-2.
https://github.com/w3c/json-ld-api/pull/520 -> MERGED Pull Request 520 address issue #519 (by pchampin)
Benjamin Young: W3c/json-ld-api#388 is class-3.
https://github.com/w3c/json-ld-api/pull/388 -> MERGED Pull Request 388 Add tests and spec text for scoped-contexts on `@nest` alias. (by gkellogg) [defer-future-version]
Ivan Herman: The Errata label should be removed once the related change has been published.
Benjamin Young: Looks like those are all the API changes.
Ivan Herman: Meaning that no API changes have been made since 2021?
Benjamin Young: Yes.
Ted Thibodeau Jr.: It may be worth checking the commit history for direct merges.
Next, aggendum
Next, agenda item
Benjamin Young: Don't have time for further discussion. Continue with other docs next time.
Benjamin Young: Next several meetings will follow this format, until we've curated.
... Then we can start clarifying pending issues.