W3C home > Mailing lists > Public > public-w3process@w3.org > November 2014

Re: Issue-129 Coordination Groups

From: Michael Champion (MS OPEN TECH) <Michael.Champion@microsoft.com>
Date: Tue, 25 Nov 2014 02:15:35 +0000
To: Stephen Zilles <szilles@adobe.com>, "public-w3process@w3.org" <public-w3process@w3.org>
CC: Judy Brewer <jbrewer@w3.org>, "Phil Archer (phila@w3.org)" <phila@w3.org>
Message-ID: <1416881734896.80879@microsoft.com>
I remain unconvinced that Coordination Groups need to be specified in the Process Document.  How the Domain Leads, team contacts, and chairs do their jobs and coordinate with each other is a job for W3C management to define, monitor, and adjust.  The other functions mentioned below could be handled by an Interest Group with the same scope as existing Coordination Groups.


> the bits uniquely relevant to Coordination Groups might boil down to just a few sentences,

> and provide the streamlining the Process TF seeks.


The goal is to simplify the actual structure and operation of W3C, not just streamline the document itself.


> the main differences are that CGs do not have specified Deliverables and do not have a fixed Duration
> (at which a renewal is required). There is also a difference in how these categories of Groups are
> terminated. ...
> Some other differences are that participation in a CG seems to be limited ...


A couple of questions:

- What do existing Coordination Groups do that could not be done by an Interest Group with the same scope? (I think the answer is in what it does NOT do, e.g. have specific deliverables, terminate, allow a broader set of participants)

- What harm would come to W3C if what is now a Coordination Group  became an IG with specific deliverables, had to be renewed periodically, and allowed broader participation by interested members?



________________________________
From: Stephen Zilles <szilles@adobe.com>
Sent: Monday, November 24, 2014 3:39 PM
To: public-w3process@w3.org
Cc: Judy Brewer; Phil Archer (phila@w3.org)
Subject: Issue-129 Coordination Groups

I have been carrying on an e-mail discussion with Judy Brewer on her concerns with removing Coordination Groups from the Process Document. This discussion raised some topics that that had (to the best of my knowledge) not been raised before on this list. I have, therefore, copied to this list the section of our interchange that covered the those topics.

Steve Z (SZ)
And implicitly, Judy Brewer (JB)

[JB] Thoughts going forward:

    - The Process Document section on Coordination Groups seems to have a number of redundancies with the description of WGs and IGs. Perhaps that entire section could be streamlined by modularizing the common items that are shared between all three flavors of groups. The bits uniquely relevant to Coordination Groups might boil down to just a few sentences, and provide the streamlining the Process TF seeks.
[SZ] This is a reasonable suggestion. I have looked at the sections on WGs and IGs versus the section on Coordination Groups. The removal of Activities and Good Standing eliminate some prior differences. All of these Groups require a Charter that is sent to the AC for Review to be created. In terms of Charter requirements, the main differences are that CGs do not have specified Deliverables and do not have a fixed Duration (at which a renewal is required). There is also a difference in how these categories of Groups are terminated. WGs and IGs cease when their duration expires or the Director closes them. CGs cease when, “there is no longer a perceived need for coordination.”
Some other differences are that participation in a CG seems to be limited to: the Chair, the chairs of the coordinated Groups, Team representatives (usually the Team Contact) and Invited Experts.
[JB] In our case, Invited Experts generally participate only if they have status as a Chair, or, by invitation, as a Facilitator of a Task Force; their IE status is immaterial. Similarly to how Phil handles his Coord Group though, we occasionally invite someone outside of these roles if they have relevant information for a particular discussion.

[SZ] Finally, the sentence, “Where a Coordination Group's scope covers two groups with unresolved disputes or tensions, it is the first locus of resolution of these disputes.” seems to give the only power of CGs beyond being a mechanism for coordination.
[JB] I think that's right. In practice though, when there has been an unresolved dispute between two WAI groups, the escalation path has been to me as domain lead, and I've mediated it outside of the Coordination Group meeting.

[Continuing]

[JB]   - With W3C's expanded scope of work, there seems to be a need for more rather than less coordination. This is both within given activities, to ensure that quality aspects of spec development including testing get carried out well, but also across different activities and domains including the newer verticals. Maybe what W3C needs at this time is a more fluid set of options for coordination -- one that allows stably chartered coordination bodies where effective, along with encouragement of more agile forms of coordination where there is the opportunity or need.
[SZ] I think there is a need for dialog among Groups with common interests. Such dialog may include coordination or even joint activities (such as the CSS/SVG Joint FX Task Force). But, I think that experience gained over the last 20 years has shown the CGs are only one mechanism by which this dialog is achieved and for many W3C participants, they are not the favored mechanism.
[JB] We use joint task forces extensively, but for a different purpose than WAI CG; generally for intensive joint development of requirements, use cases, techniques, specs.

[SZ] I would, however, support a paragraph in Section 6 that talks about Coordination as being important, gives suggestions on how that might be achieved, including such mechanisms as coordination groups (lowercase), and defines an escalation path when coordination fails. Today, the escalation path is all too often a Formal Objection. I think there are steps, such as mediation by a Team Contact or Domain Lead, that might solve some of these problems prior to a Formal Objection.
[JB] I'm curious what the incidence of formal objections is in W3C. In the accessibility area I'm aware of the PFWG FO to HTML WG after their decision on a PFWG spec; and of the recent Apple FO on longdesc, an HTML extension spec. I don't have a clear picture of the frequency of FO's across W3C.

My impression is that mediation by Domain Leads is more common, either by a single DL when both groups are in the same domain, or by two DLs when a dispute crosses two domains. In my experience, such mediation is relatively common, and nearly always successful.

I would also support your suggested paragraph on the importance of coordination with suggestions of multiple approaches for achieving that. Since neither Phil's nor my Coordination Group seems to be using Coordination Groups for resolving disputes, I see no need to maintain a provision for that.

I don't know what your timeline is for closing this (other than a presumed "as soon as possible.") In respect of the strong interest from some WAI chairs and team contacts for maintaining a chartered Coordination Group, and given that this change would affect eight WAI groups, I'd like to get their confirmation that the write-in-coordination option works for them. Our next meeting would be Wednesday Dec 3.
Received on Tuesday, 25 November 2014 02:16:10 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:35:13 UTC