w3c
Simplify the text on liaisons
E.g. the WHATWG MOU etc was more political handholding than actually necessary for any participation/work in practice.
This is a pretty disappointing assessment. W3C and WHATWG had struggled for years to find a collaboration model and the MoU has been highly successful in repairing a fractured community.
I think the question of whether past MoUs have accomplished a lot and the question of whether the text in the current process is useful are reasonably separate.
Regardless of whether you think Liaisons and MoUs are useful (I think they can be, but again I think that's besides the point), I believe:
- The current text on liaisons doesn't really mandate nor ban anything. It does say that Liaisons must be coordinated by the Team, but without an constraining definition of what a Liaison is, that is not terribly actionable. Therefore from a normative point of view, I think it is decently close to a no-op, and would like to remove it altogether. Jeff thinks there's some value in that statement, and would like to preserve that bit.
- The current text on MoUs requires that they go through AC ballots first, makes the result of an AC appeal binding, and bases this on a somewhat high level, but still decently clear definition of what MoUs are. This is may or may not be a good idea, but it is not a no-op. For now, I don't suggest changing it.
W3C does receive Liaison statements from other standards bodies from time to time, and there is a process for dealing with them (i.e., coordination by the Team, typically by notifying the relevant WGs and/or IGs). Should the Process document not cover this? If we're lacking a definition, I'm sure we can come up with one.
I'm less concerned about liaison documents than I am about preserving the ability to have liaison representatives attend meetings (each way).
@dwsinger That's an interesting point, but I am far from convinced that the current text clearly provides for that. How about adding an explicit statement to that effect in https://w3c.github.io/w3process/#group-participation instead?
@frivoal isn't this text deleted? it's the place where memberships and specifically mutual memberships are mentioned:
from very informal (e.g., an individual from another organization participates in a W3C Working Group, or just follows its work) to mutual membership, to even more formal agreements. Liaisons are not meant to substitute for W3C membership.
It's there, but only as an example of a very informal relation. If that just says that we may use invited experts (or even have other organizations become members and send reps) as the vehicle for liaisons, we'd continue to be able to do that without that sentence. Maybe it's trying to say that the team can invite liaison reps to wg meetings separately from them being WG participants under normal rules, but if that's the case, it's not doing it very well: it's only framed as an example, and says nothing about patent policy implications. And then we have "Liaisons are not meant to substitute for W3C membership", which further points to it being the former meaning.
So again, I don't think this text does anything. I support the existence of Liaisons, and if we want to formalize who can or must do what with regards to them, we should write that down, but that's separate from trimming no-op text.
No argument from me that the old text is confusing, with apparent expectations in examples that are not reflected in the normative text, and so on.
Is this a 'high nail' for this year?
Is this a 'high nail' for this year?
Well, there's a stated goal to reduce complexity and page count, to make the hole thing more digestible. This is an isolated chunk, so it doesn't do much for complexity, but if we do want to get the page count down, looking into removing sections that seem to do nothing seems worth a try.
