The W3C JSON-LD Community Group

Go Back

W3C Logo

JSON-LD CG Telecon

Minutes for 2025-08-06

Present
Gregg Kellogg, Ted Thibodeau Jr., Piotr Sowinski, Niklas Lindström, Pierre-Antoine Champin, David I. Lehn
Chair(s)
Gregg Kellogg, Benjamin Young
Scribe(s)
Gregg Kellogg, Benjamin Young
Agenda
https://www.w3.org/events/meetings/04e937c0-e751-4c7e-9ba4-5782e0b3d272/20250730T120000/
Topics
  1. Announcements and Introductions
  2. Charter

Topic: Announcements and Introductions

Gregg Kellogg is scribing.
Piotr Sowinski: Hi, I'm Piotr Sowinski.
... I represent a startup working on streaming knowledge graphs.
... I'm a CG member at this time, and mostly observing.

Topic: Charter

https://github.com/json-ld/json-ld-wg-charter/pull/10 -> Pull Request 10 Rework as the Linked Data Formats WG (by BigBlueHat)
Benjamin Young: We've been spinning on the group name, and an action list for cleanup.
... Let's focus on the name, to start.
... Earlier, the opinion was "Linked Data Formats", but this was reconsidered on the last call.
... We wanted to be clear that this was about things more directly related to JSON-LD.
... Perhaps "JSON-LD Linked Formats WG".
Benjamin Young is scribing.
Gregg Kellogg: What threw people off was that there are other Linked Data Formats which do not use the JSON-centric processing model--such as Turtle
Niklas Lindström: +1 To gkellogg re "not all kinds of formats; more like-json"
Pierre-Antoine Champin: +1
... so presenting that JSON heritage seems important
David I. Lehn: Do we need to come up with a name and fill in a backronym?
Ted Thibodeau Jr.: `{JSON|CBOR|related} Linked Data Formats WG`
Niklas Lindström: I generally agree; I thought about talking about the "compact" feature, but haven't figured out how to frame that.
... We should focus on the "JSON-ness" of it.
... We should probably stay with JSON-LD.
Benjamin Young: That's the intention of the group.
Gregg Kellogg: It would not be helpful to take JSON-LD out of the name
... but maybe adding to it would help
Niklas Lindström: "JSON-like" ...?
... somehow opening it up to other formats
Pierre-Antoine Champin: I suggested "JSON-similar object notation linked data formats".
... If we find something less convoluted, it might be good.
Niklas Lindström: JSON = "JSON-subset of object notations" ;P
Ted Thibodeau Jr.: `Backronym WG`
Benjamin Young: I think that the CG is techincally the "JSON for Linking Data CG".
... We changed that for the WG.
Ted Thibodeau Jr.: We were also trying to get some distance from RDF.
Gregg Kellogg: Turtle is layered on top EBNF
Benjamin Young: The group could be anything as long as it includes JSON-LD.
... We need a way to distinguish our formats from traditional RDF formats.
... It's the "INFRA-based RDF formats".
Benjamin Young: Maybe: JSON-LD [and] Linked Data Formats Working Group
Benjamin Young: Thoughts on this.
Gregg Kellogg: I think the `and` helps open it up other formats
Niklas Lindström: I think that "and linked data formats" is problematic, because that might include Turtle.
Gregg Kellogg: How about JSON-LD Working Group?
Niklas Lindström: I like that
Gregg Kellogg: I think the work to be done is to cast JSON as not just a line format, but as a data model
... which I think is how people already think of it
... CBOR and YAML do provide additional capabilities--which I think we should be open to
... really, it's JavaScript, and the ability to represent these structures there
Niklas Lindström: "JSON[-serializable (in-memory?) structure]"
... obviously not just JavaScript
Gregg Kellogg: JSON-LD needs to be in there
Ted Thibodeau Jr.: `JSON-LD++ WG`
... whether it's helpful or not to add more words, I don't know
Ted Thibodeau Jr.: (But are those really escaped spaces or plus signs?)
Pierre-Antoine Champin: I agree with what Gregg said, the short name is establishged.
... I'd be fine with JSON-LD, and explain the related formats in the context.
... I'm worried that "and Linked Data Formats" might include Turtle, but the fact we're starting with JSON-LD does set a direction.
Niklas Lindström: Maybe "and related"... ? (ambiguous; indicates intent..)
Benjamin Young: So, we're "JSON-LD WG".
Ted Thibodeau Jr.: "And blessed"
Pierre-Antoine Champin: I'm a close -1 on keeping the same name. We want to signal that we're about other formats.
David I. Lehn: Since naming is hard, may be worth sitting on this for a bit longer if possible. i know it takes me days/weeks for good name options to pop in my head.
... A change of name might signal this to the rest of the world.
Benjamin Young: We need to justify why CBOR-LD and YAML-LD are there.
Gregg Kellogg: In 1.1 we did set the basis for related formats
... that's part of why we used the internal reperesentation
... and we call out that this is meant to extend to CBOR
... so this shouldn't be too surprising to those following along
... maybe we just need more research in IETF land
Ted Thibodeau Jr.: `JSON-LD cousins WG`
... we can't be the first ones to try and attempt this naming
Benjamin Young: Any other thoughts?
Benjamin Young: For the moment, I'll go back to "JSON-LD WG".
https://github.com/json-ld/json-ld-wg-charter/pull/10 -> Pull Request 10 Rework as the Linked Data Formats WG (by BigBlueHat)
Gregg Kellogg: We're adding me as co-chair
Ted Thibodeau Jr.: `5 Crazy LD Formats WG -- and you'll love #3!`
... we've had different opinions about how detailed the scope should be
... ivan first suggested we be very specific, but we apparently overdid it
... this charter will also sort of split the work of the group
... JSON-LD and other formats
... and things we might do post-recommendation
... YAML-LD doesn't represent much work, but is there as a thing the group will need to focus on
... I'd suggest we just point to the project page from the charter
... even though it's dynamic
... and then write the rest of the charter in a vague enough way to focus on RDF1.2 alignment
... while still exploring these other formats
Pierre-Antoine Champin: I wanted to bring up two potential issues. Regarding Gregg's appointment as co-chair, this should be resolved in the next couple of days.
... When i asked for the announcement, it came up that we usually avoid having a chair be an editor.
... It's not forbidden be process, and we can argue that this is the situation we have, but maybe we want to find a third co-chair.
... Another thing regarding a living spec is that there was an email from Phillipe that some groups have experimented with it and have moved back to the old way.
... There is a way that uses the old way with out the "new features allowed" feature, as the tooling isn't there.
... Maybe we shouldn't commit to that.
Benjamin Young: Related, technically we're no longer a WG, as the group charter has expired.
... Question is how quickly we want to move along to rechartering.
... TPAC looms in November, so we have around 3 months.
... I'd like to come back to timing.
... There's keen interest in getting the charter going so that we could do something reasonable at TPAC.
... Without some action, there can't be a JSON-LD WG call at TPAC without a charter.
Pierre-Antoine Champin: I asked for a charter extension. It's not yet reflected in the published charter, but it will be reflected in a couple of days.
... Extending the charter doesn't need member approval, it's up to the team.
... That said, we announced to the members that we were working on a new charter.
... There is some urgency here; we should aim for TPAC to have the WG approved, or in the process.
... Before we send the charter to the AC we need horizontal review.
... We should hurry.
Gregg Kellogg: There is certainly tension between editor and chair roles
... so I would fully support stepping down as a chair if someone else will step up
... living specs do get in the way of errata handling
... we don't want needless forks, etc.
... I'd prefer to abandon all that and move to a JSON-LD 1.2 spec
... I don't think it taking longer to get a spec out is an issue
... if we do really need updates to JSON-LD to enable YAML-LD and CBOR-LD, then maybe there's a 1.2 release that's small
... small enough to just do those things
... and then move on to a 1.3 release
... as long as the charter doesn't restrict that
... I think we could get out of CR for the 3 documents in months given the work already done
... and then work on the bigger things like RDF 1.2
... other than it not being at CR yet...
... the work we need to do is already in hand
... I would like to see more JSON-LD-star work
... but the timing with the RDF group does make that problematic
Pierre-Antoine Champin: We can add new deliverables that include a new version of JSON-LD as long as it is in scope.
... The burden imposed by the process is that every version is a new deliverable and needs a new call for patent exclusion.
... The process described by Phillipe is about being able to work on multiple version.
Benjamin Young: We discussed going to a weekly cadence through August to try to get this moved along more quickly.
Pierre-Antoine Champin: +1 To a weekly cadence
... Any objections to going to weekly meetings until the end of August?
... By the 27th we should be celebrating.
... niklasl and TallTed are willing to serve as co-chairs. I'm also fine if gkellogg is also a chair.
Benjamin Young: Perhaps TallTed, niklasl, pchampin and I should have a call to go through the responsibilities.
Ted Thibodeau Jr.: I haven't chaired in a bit, I did for SHACL 1.0. I didn't get any training, just the documents on W3C, and relied on my co-chair to do much of the work.
Benjamin Young: By and large, the group is small enough to have a tight issue list.
Benjamin Young: It's going to come down to who joins the group and how contentious it might be.
Gregg Kellogg: Chairs, while opinionated, should ack more as a "Switzerland" when issues come up
... I don't think we can necessarily count on our group staying this size
... we may have other folks come back in
... it's always helpful to be slightly less personally invested in the spec text
... which is why also being an editor is problematic
... there's also external communication
... formal feedback, process to push forward, etc.
... editors don't normally contribute to
Benjamin Young: I'd like some redundancy, but as we re-charter, if it gets contentious, or things need to be shepherded along, it becomes a project management problem.
Gregg Kellogg: Niklasl has done quite a lot of contributing to the specs over the years
... that's not out of bounds, but it is something to be considered
... both niklasl and TallTed have been involved since the beginning
... I'd be happy to be on any calls if you want to discuss it
Ted Thibodeau Jr.: I'm less technically savy on JSON-LD than others, and will find important issues, but I'm not "fluent" in JSON-LD.
Benjamin Young: You need to know enough to keep the conversation moving.
... I'll schedule a weekly call through August, and get through Gregg's list and put the group name back to what it was.
... In a week, I hope we can look at a more complete charter.
... I think we'll look at a JSON-LD 1.2 that looks at CBOR and YAML needs and 1.3 for further changes.