Switching Tracks

[frivoal]

We now have 3 tracks: the Rec Track, the Note Track, and the Registry Track. One thing we said we wanted to look into is the ability for a document to switch tracks.

We could disallow track-switching except through copy-and-paste and starting a new document with the same content.

However, I suspect that's probably too stringent, and I think there are valid cases for wanting to switch tracks.

[frivoal]

I'd suggest adding a paragraph or so saying that you can switch tracks, with a few restrictions. Maybe something like:

Given a [=Group decision=] to do so,
[=Working Groups=] can republish a [=technical report=]
on a different track than the one it is on,
under the following restrictions:

• A [=technical report=] that was previously on the [=Recommendation Track=]
  and that has already switched tracks
  cannot not switch back to the [=Recommendation Track=]
  (unless it was a [=Recommendation Track=] [=technical report=]
  discontinued into a [=Note=] under a pre-2021 version of the Process).
• A [=technical report=] that is or was
  a [=W3C Recommendation=], [=W3C Statement=], or [=Patent Review Draft=]
  cannot switch tracks.

[=Technical reports=] that switch tracks start
at the initial [=maturity level=] of the new track.

[dwsinger]

The last sentence basically says "slide down the snake to the start square again" which is exactly what you'd do in the absence of anything, i.e. copy-paste and start again, no?

[frivoal]

Almost. The only (intended) nuance is that you get to keep your identity: same url, same shortname, etc.

[css-meeting-bot]

The Revising W3C Process CG just discussed #509.

The full IRC log of that discussion

<plh> Topic: #509
<plh> https://github.com//issues/509
<plh> GitHub: https://github.com//issues/509
<plh> Florian: there is a proposal on the table for #509.
<plh> ... (summarizing the proposal)
<plh> ... doing rec-tracks, then note, then rec-track could introduce submarine patents
<plh> ... ... would be confusing to allow it
<plh> ... not to be encouraged
<plh> fantasai: if you switch back rather than restarting, you retain previous commitment. but this is confusing and people might not understand it. But better than restarting at FPWD
<dsinger> q?
<plh> ... we could allow for that and the patent implication are a bit complicated
<plh> david: this is a case that shouldn't arise
<fantasai> s/better/strictly better/
<wseltzer> q+ re Patents
<plh> Ted: having been in a # of groups. when a group can discontinued, it goes into Notes. The intent is that the Note can be picked up by a subsequent working group
<plh> Florian: you can move to discontinued Note (a parked state)
<plh> Ted: then you need to inform the groups about this
<dsinger> q?
<plh> ... we have groups in this case right now
<plh> Florian: it's not in the current process yet
<plh> s/discontinued Note/discontinued Draft/
<dsinger> ack ws
<Zakim> wseltzer, you wanted to discuss Patents
<plh> wseltzer: our goal is to understand the provenance of the contributions and make informed decisions. You don't need to prohibit
<plh> ... if the contributions are fine, why not?
<dsinger> q?
<plh> Florian: we can move to "should not"
<plh> ... I don't want to encourage folks
<plh> ... "should not", with someone to check what you're doing
<plh> dsinger: don't need to solve this today
<dsinger> q?

[TallTed]

Almost. The only (intended) nuance is that you get to keep your identity: same url, same shortname, etc.

In that case, I would change the last paragraph to say so --

[=Technical reports=] that switch tracks start at the
initial [=maturity level=] of the new track, while
retaining any established identity (url, shortname, etc.).

[frivoal]

@TallTed This sounds like a useful clarification. A bit of an annoyance is that url or shortname are implied by the process rather than actually clearly defined anywhere, but I think that as a parenthetical, it works out OK, as the main sentence is clear on its own.

Happy to include that if others are in support as well.

[frivoal]

Based on comments above and discussions during the call, here's updated proposed wording:

Given a [=Group decision=] to do so,
[=Working Groups=] can republish a [=technical report=]
on a different track than the one it is on,
under the following restrictions:

• A [=technical report=] that is or was
  a [=W3C Recommendation=], [=W3C Statement=], or [=Patent Review Draft=]
  cannot switch tracks.
• A [=technical report=] should not switch away from the [=Recommendation Track=]
  without due consideration of the Patent Policy implications
  and approval of the W3C’s legal counsel
  if the Working Group envisions a likelihood of returning to it later.

[=Technical reports=] that switch tracks start at
their new track’s initial [=maturity level=],
while retaining any established identity (url, shortname, etc.).

[wseltzer]

s/different than/different from/

[css-meeting-bot]

The Revising W3C Process CG just discussed Switching Tracks, and agreed to the following:

• RESOLVED: Merge PR for #509

The full IRC log of that discussion

<fantasai> Topic: Switching Tracks
<dsinger> Switching Tracks #509 <https://github.com//issues/509>
<fantasai> github: https://github.com//issues/509
<fantasai> florian: I'm happy with the text in the PR now
<fantasai> florian: Anyone else with comments?
<fantasai> dsinger: basically if you switch tracks, you start at the beginning
<dsinger> q?
<fantasai> florian: Yes, the only nuance is that there are Patent implementations for switching to/from REC
<fantasai> florian: Better than copy-paste, because maintain some commitments from before
<fantasai> florian: but it's tricky, so we want people to pay attention to that when switching to/from REC
<jeff> q+
<dsinger> q?
<fantasai> florian: "A [=technical report=] should not switch away from the [=Recommendation Track=]
<fantasai> without due consideration of the Patent Policy implications
<fantasai> and approval of the W3C’s legal counsel
<fantasai> if the Working Group envisions a likelihood of returning to it later."
<wseltzer> q+
<fantasai> florian: You can do it, but we want to avoid doing it accidentally
<dsinger> q?
<dsinger> ack jeff
<fantasai> florian: The check is on leaving the REC track, not returning, because if you need to return to REC track the best thing to do is return to it. But in that case ideally shouldn't have switched away from REC in the first place.
<fantasai> jeff: Curious if we've shared this one with PSIG and asked for any input?
<fantasai> florian: I don't believe we have. Also note this doesn't introduce a new possibility. You can currently go from REC to NOTE and back today.
<fantasai> florian: It's the separation of tracks that we created earlier that makes this switching tracks section necessary
<dsinger> q?
<dsinger> ack ws
<fantasai> dsinger: We currently mandate abandoned WDs to switch to NOTE
<fantasai> wseltzer: Am I correct that the reason we prohibit REC and PRD switching to NOTE is to avoid confusion about status under shortname?
<fantasai> florian: There's a designated status for abandoned RECs/PRDs (not a NOTE)
<fantasai> dsinger: You've completed the track, so
<fantasai> wseltzer: Why do we have that restriction?
<fantasai> florian: The most important is PRD
<fantasai> florian: Seems weird to switch from "we made a spec with patent commitments, but decided it shouldn't be" seems weird
<fantasai> wseltzer: I'd rather have us make it clear to people what the recommended things to do are, rather than prohobit things
<fantasai> wseltzer: If ancient piece of tech in this PRD, want to use it as exemplary text for something else
<fantasai> florian: This statement doesn't prohibit copy-paste. You can take the content of your PRD and make it something esle.
<fantasai> florian: but if you have a PRD and want to rescind it, can do that.
<fantasai> florian: If you have a PRD, and ppl have implemented it, and then you decide to make it a Note, that's ...
<fantasai> florian: It's weird, nobody wants it, don't do it.
<dsinger> q?
<fantasai> florian: and it's easy to relax restrictions later if we feel like it, but it's hard to undo the problem
<fantasai> florian: If you change PRD to NOTE, material in PRD is still under patent protection, but that would be very non-obvious from the most recently published NOTE
<dsinger> q?
<fantasai> wseltzer: Seems fine. Grammar suggestion.
<fantasai> dsinger: Consensus to include?
<fantasai> +1
<dsinger> q?
<fantasai> RESOLVED: Merge PR for #509