Closed frivoal closed 6 months ago
dwsinger
commented
6 months ago I suspect that there are some wrinkles for depending on whether there were changes while the document was off the Rec. Track. One might argue that the second FPWD is actually a new document, starting its own progression, for example. There may also be wrinkles depending on whether it's the same WG with the same composition.
frivoal
commented
6 months ago I suspect that there are some wrinkles for depending on whether there were changes while the document was off the Rec. Track.
The specific downside of doing that, which is why it's a bad idea, is that any company who has left the working group while the document was off the rec track isn't made to commit to any material introduced into the Note, since it was not at that point a Working Draft, and therefore the provisions of the 3rd paragraph and bullet list in https://www.w3.org/Consortium/Patent-Policy-20200915/#sec-exclusion-resign don't kick in on that latest draft (they apply to the latest thing that was called a WD).
So this is indeed not an ideal thing to do, as iterations as a Note could leave a window for people to participate but not commit. So don't do that. Fortunately, most of the time when something is switched to a Note, that's not to iterate on it, but to park it, especially because discontinued draft used not to be a thing, and the only way we had to park REC track documents was to turn them into Notes.
One might argue that the second FPWD is actually a new document, starting its own progression, for example.
Well, the Process is explicit that identity is preserved, so that argument would not be according to process:
Technical reports that switch tracks […] [retain] any established identity (url, shortname, etc.).
The Patent policy handles the instantiation case just fine (because it just ignores the intermediate off-rec track publication). People who have left the group before the document was reinstated don't get committed to stuff that is added after they left, which is also what would happen if nothing out of ordinary had happened.
Breaking the continuity between the two parts would cause more problems, as it could be seen as extinguishing the patent commitments from earlier participants, without reinstituting them on the new draft if they have since left the working group, even though from a normative point of view, the spec still covers the exact same things.
So I think the Process is fine as it is from a normative point of view, and works well with the Patent Policy (because they were deliberately designed with that intent, and the effect is the same as following the provisions of https://www.w3.org/2023/Process-20231103/#abandon-draft). But an editorial clarification would be welcome.
caribouW3
commented
6 months ago "A technical report that is or was a W3C Recommendation, W3C Statement, or Patent Review Draft cannot switch tracks."
In my understanding, a FPWD is a patent review draft, therefore the new publication when returning on a REC track can't be a FPWD (moreover, a "Second First Public WD" sounds broken).
frivoal
commented
6 months ago In my understanding, a FPWD is a patent review draft
It's not. CRS are Patent Review drafts, as stated in https://www.w3.org/2023/Process-20231103/#candidate-recommendation-snapshot, and so is the combination of the existing Recommendation with the proposed amendments, as stated in https://www.w3.org/2023/Process-20231103/#change-review. But other stages are not. FPWD is relevant to the Patent Policy, but differently so, not as a PRD.
css-meeting-bot
commented
6 months ago The Revising W3C Process CG just discussed #877, and agreed to the following:
ACTION: fantasai to notify PSIG
What happens to a document which switches away from the REC track (e.g., to Note), and then back to the REC track?
Typically, nowadays, you're not really supposed to do that, and instead if you want to discontinue a REC track document, you should use https://www.w3.org/2023/Process-20231103/#abandon-draft and publish a Discontinued Draft without switching tracks. But that provision is new, so there are already a number of documents that have been parked as Notes, some of which could be switching back. There's now a should against switching away from the REC track that, but (1) it's only a should, and (2) this should is pretty new. So we still need to consider the case.
We have:
https://www.w3.org/2023/Process-20231103/#switching-tracks, which says:
and https://www.w3.org/2023/Process-20231103/#maturity-stages, which lists:
Interestingly, FPWD is not identified as a maturity stage, but is simply the name given to the first Working Draft of a technical report. This view is further strengthened by https://www.w3.org/2023/Process-20231103/#transition-reqs: advancing to a new maturity stage involves a Transition Request, and there's no Transition Request between FPWD and WD, so they are the same stage.
So, if a document switches to the REC track (for example, from Note):
I think that's what the Process means. And from the point of view of the Patent Policy, I think that's what makes the most sense: The provisions of the Patent Policy make no exception for a document being temporarily published off the REC track. The only things it concerns it self are: documents being published as FPWD, documents being published as Patent Review Drafts, Members joining and leaving the group, so it doesn't have an expectation that a document would be an FPWD more than once.
So, all in all, I think the Patent Policy and Process are coherent and that the way they are set up is good.
However, I think this interpretation is not entirely obvious when reading some sections out of context. Regrettably, it is easy, if you're only reading https://www.w3.org/2023/Process-20231103/#switching-tracks to assume that "FPWD" is the REC track's initial maturity stage, which could lead us to publish multiple "First" Public Working Drafts of a same technical report. This is undesirable, as that would trigger unnecessary Calls for Exclusion.
Therefore, I suggest adding a Note to https://www.w3.org/2023/Process-20231103/#switching-tracks, saying something like: