search

w3c / json-ld-bp

JSON-LD 1.1 Best Practices Note
https://w3c.github.io/json-ld-bp/
Other
21 stars 7 forks source link

Update spec status and/or publish? #43

Open gkellogg opened 2 years ago

gkellogg commented 2 years ago

These document was never published, but the specStatus indicates its a WG-NOTE. This causes issues when people look at the ED and see a reference to a non-existing note.

IMO, the group clearly intended to publish this as a NOTE, and it has been referenced. It also gets confused with the former CG note, which it obsoletes. IIRC, the WG is still capable of publishing this as a NOTE, and even with open issues, it should be done.

Noted in discussion on Framing: https://github.com/w3c/json-ld-framing/issues/133#issuecomment-1229728032 (w3c/json-ld-framing#133)..

In the interim, the specStatus should be DNOTE (Draft Group Note).

gkellogg commented 2 years ago

Note, I changed status to DNOTE, and even tried ED and unofficial (sorry, intended to do this in a PR); it still references the /TR version as last published. Not sure why it does that. @iherman?

iherman commented 2 years ago

To be honest, @gkellogg, I don't really know. What happens if you set the status to base and you do not set the short name?

gkellogg commented 2 years ago

ReSpec must be getting status from somewhere else. Commenting out the shortName doesn't change anything. Setting specStatus to "base" (locally) generates some errors:

W3C Working Group documents can't use "base" for the specStatus configuration option. How to fix: Pleas see specStatus for appropriate status for W3C Working Group documents. (plugin: "w3c/defaults") Documents with a status of "base" can't be published on the W3C's /TR/ (Technical Report) space. How to fix: Ask a W3C Team Member for a W3C URL where the report can be published and change latestVersion to something else. (plugin: "w3c/headers")

Note that respecConfig does not contain a latestVersion, either. Using "unofficial" just reports the issue about latestVersion. I believe that the correct status for this should be DNOTE, as it is, in fact, a Draft Group Note, but not sure what to do about the Latest published version, though.

Although I believe it's still find to use a shortName, it's pretty odd that removing it doesn't cause it to go away.

iherman commented 2 years ago

Although I believe it's still find to use a shortName, it's pretty odd that removing it doesn't cause it to go away.

I agree. I would go as far as saying that, for documents marked as "undefined" or "base", this is a respec bug.

(I did play around with my local copy and it did not make me any wiser. The only hack I could see is to run a post generation script that removes that header.)

I believe that the correct status for this should be DNOTE, as it is, in fact, a Draft Group Note, but not sure what to do about the Latest published version, though.

The only proper way ahead is for the WG to officially publish this document as a Draft Note. The WG can do this, and that would remove all ambiguities.

Note, b.t.w., that the charter of the WG runs out... tomorrow! Something should be done about it...

Cc @plehegar

gkellogg commented 2 years ago

Ouch! I thought the group was chartered to go quite a bit longer as a maintenance group. It will certainly have things to do to respond to RDF-star changes, and the YAML-LD activity (described in the charter) has really ramped up. Best if we can keep the group going then to establish a charter for a new group.

cc @BigBlueHat

BigBlueHat commented 1 year ago

The JSON-LD WG was rechartered on January 19th, 2023 until January 31, 2025. Consequently, we're in a good position to formalize the note as a publication.