Charter Review

Jeffrey: The only point I know of where we disagree is whether we ask the WG to discuss the distinction between JSON and JSON-LD, whether we ask they do not, or we do nothing, or we note that the TAG disagrees. How do people feel?

Martin: I raised this because it's mentioned in the charter. By doing so, they did themselves a disserrvice because the spec is good on this point. I'd suggest they duck the issue. The work they're doing has done a good job.

https://www.w3.org/TR/2020/REC-json-ld11-20200716/#interpreting-json-as-json-ld

Jeffrey: Missing something in spec that says "here is how you decide to use this instead of JSON".

Sarven: This is where I'm not sure that the charter needs to address this. I like the idea about guidance, if they want to incorporate that into the spec or other documentation, but from the point of whether they need to make the case for JSON-LD at this point. EIther that case was already made, starting from incubation, etc... Lots of stuff already uses it. Don't need to rejustify JSON-LD and its existence. Maybe that guidance can be highlighted in this next spec version, that's fine.

Jeffrey: Fracture was mentioned in the review request. VC split into two because of this JSON vs -LD "split". I will probably make a comment as an AC rep about this; the fact that it has split a working group makes it our concern. That's why it should be in the charter. Not just because it exists, but they should document what should cause someone to use JSON or -LD in their work.

Sarven: Does the content/paragraph capture the recommendation we are trying to make?

Jeffrey: Sarven is saying that they should clarify. I'm saying that it should be in scope. The question is what should the TAG recommend they do.

Sarven: Does this explain why JSON-LD exists?

Martin: Question is, given an application, how would you decide that JSON-LD is the right choice, as opopsed to raw JSON. What's the processing model? All those questions need to be more thoroughly addressed in an application spec. Section 6.1 discusses how to treat JSON as JSON-LD.

Jeffrey: that question is that one I'd suggest they try to answer in their spec. They say that if you have linked data here is how you serialize that in JSON. That's not the use case that JSON-LD claims. Maybe consider the new use case that has appeared.

Sarven: It would be a good service to the community to include that in the spec. THen other people can use that in their work. To eliminate or elevate disputes. I'm OK with that. It's clear enough that we can . WIthin what we were thinking. Guidance to that extent is fine. I was more worried about justification.

Sarven: Would like to draft that. Do we need approval of that sentence. Was planning to post already.

Jeffrey: Would like to double-check, but feel like just the two of us are the only ones who need to approve it.