Order of precedence of Membership Agreement, Process document and normative references

[plehegar]

Clause 16 of the Membership Agreement defines the entire agreement to include the W3C Process. This would seem to include the documents referenced from the Process document. It is possible that the authors of these documents do not consider the impact their work will have on the entire Membership Agreement. Conflicts might exist between the Membership Agreement, Process document and referenced documents. The Process should define how such conflicts are resolved. This could be as simple as stating "in the event of a conflict between the Membership Agreement and the Process the Membership Agreement prevails. In the event of a conflict between the Process and referenced documents the Process prevails". Ideally such order of precedence definition should appear in the Membership Agreement but I understand there is not an appetite to alter that document at this time.

[nigelmegitt]

The way you've described this, @plehegar , the Membership Agreement includes the Process, so it's impossible for the Process to conflict with the Membership Agreement. I think you're wanting to express some precedence of a referencing document over the document(s) that it references, but if that's the case, shouldn't it be defined within the referencing document itself?

E.g. perhaps a lawyer might draft something for the Membership Agreement along the lines of:

Except where it is in conflict with this Agreement, the W3C Process is considered part of this Agreement.

Not sure that's an issue for the Process though?

[dwsinger]

(This issue was raised as part of a Formal Objection to Process 2021, but we're putting it into a issue here to track it and collect comments.)

[dwsinger]

We think this is a hypothetical question: IF there is conflict in the set of documents including the membership agreement and the documents it references, or that they reference, WHAT is the precedence order? i.e. how are those conflicts resolved? Now, we hope we never write such conflicts, and we're not aware of any today (and please, if we do have some, it would be good to surface them immediately). But such precedence statements are not uncommon. It's on PSIG's agenda; comments might be useful there, or perhaps direct to your PSIG rep or PSIG as a group.

[plehegar]

See also #572

[wseltzer]

PSIG recommends: "These policies are written with the intent that there be no conflict, but in the event that one exists, the Member Agreement takes precedence over this Process, which takes precedence over any other normative references. In matters dealing with patents, the Patent Policy takes precedence." minutes.

Added in https://github.com/w3c/w3process/pull/599

[nigelmegitt]

@wseltzer I find that wording begs a further question: over what does the Patent Policy take precedence? It could be read as either taking precedence over the Process or everything including the Member Agreement.

[dwsinger]

@wseltzer I find that wording begs a further question: over what does the Patent Policy take precedence? It could be read as either taking precedence over the Process or everything including the Member Agreement.

yes, it takes precedence over this process. and we probably need to say that all (member agreement, patent policy, process) each take precedence over the documents that each references.

[fantasai]

Message pinging PSIG again, 22 February 2023

[css-meeting-bot]

The Revising W3C Process CG just discussed Other Open Issues.

The full IRC log of that discussion

<fantasai> Topic: Other Open Issues
<fantasai> github: https://github.com/w3c/w3process/issues/574
<fantasai> florian: we have a question about precedence among Member Agreement vs Process vs Patent Policy
<fantasai> ... still no response from PSIG
<fantasai> ... so at this point we defer
<fantasai> ... it's a problem about existing text, not what we just introduced, so I'm comfortable deferring
<florian> q?
<plh> ack fan
<fantasai> [clarified that this goes into the DoC as Deferred due to lack of PSIG response for multiple months]

[plehegar]

[[ RESOLUTION: Given what is known at this time, PSIG recommends that the Process Document not define an order of precedence, but rather address each conflict as it is found on a case-by-case-basis. ]] https://www.w3.org/2023/09/11-psig-minutes.html

[plehegar]

[[ florian: We should talk about the precedence issue that we were blocked on PSIG for.

elika: PSIG concluded that they recommend not defining the precedence.

florian: I suspect we should follow their advice and close these things as no change.

plh: I didn't mean to make decisions today, so let's take a proposal that we close in two weeks.

florian: I will send this to the AB.

ACTION: florian: Forward PSIG conclusion to AB ]] https://www.w3.org/2023/09/27-w3process-minutes.html#t04

[rigow]

Please note that there is still an open action item within PSIG. This action item concerns the verification of the resolution with outside counsel. This is reflected in the "Given what is known". I would like to close this issue only once outside counsel has responded.

[frivoal]

ACTION: florian: Forward PSIG conclusion to AB

Done on September 28

No push back.

[plehegar]

Close with no action per 2023-11-08 meeting

[plehegar]

(feel free to re-open if there is new information)

[plehegar]

The Revising W3C Process CG just discussed #574.

The full IRC log of that discussion

<plh> subtopic: #574
<joshco> Github: https://github.com/w3c/w3process/issues/574
<joshco> plh: where are we? are we ready for a pull request?
<plh> Florian: PSIG and AB are fine with not defining order of precedence
<plh> fantasai: let's close the issue with noaction and re-open if we get new information
<joshco> fantasai: suggest closing the issue and see what comes
<plh> Resolved: close issue #574 and related pending PRs with no change
<florian> action florian to review PR 572 to see if there is any editorial bit to salvage before closing

[jwrosewell]

Please can you post the content of the following link?

https://lists.w3.org/Archives/Group/ab/2023JulSep/0173.html

[plehegar]

@jwrosewell , that link contains a question for the AB, asking their opinion of the PSIG decision. That link does not contain a response from the AB. Which information are you looking to make public exactly? It is the PSIG decision, the response from the AB, or simply the fact that the AB was asked ?

[fantasai]

More precisely, it asked if anyone in the AB had concerns with the PSIG resolution (which was quoted verbatim); and no one replied to say that they do.

It also contained links to this issue, the various pull requests, and the questionnaire results containing the text of your Process 2021 FO, so that anyone wanting to dig into the question would have all of the relevant info at hand.