W3C home > Mailing lists > Public > public-w3process@w3.org > March 2015

Re: When to close a Working Group?

From: Wayne Carr <wayne.carr@linux.intel.com>
Date: Wed, 11 Mar 2015 12:59:14 -0700
Message-ID: <55009E92.7040109@linux.intel.com>
To: "Michael Champion (MS OPEN TECH)" <Michael.Champion@microsoft.com>, Jeff Jaffe <jeff@w3.org>, Harry Halpin <hhalpin@w3.org>, Revising W3C Process Community Group <public-w3process@w3.org>
This is already in the Process 6.2.8 
https://dvcs.w3.org/hg/AB/raw-file/356968630e27/cover.html#GeneralTermination 
<http://www.w3.org/2014/Process-20140801/#GeneralTermination>

Current: "A Working Group or Interest Group charter specifies a duration 
for the group."  It then goes on to say: "The Director, subject to 
appeal <http://www.w3.org/2014/Process-20140801/#ACAppeal> by Advisory 
Committee representatives, MAY close a group prior to the date specified 
in the charter in any of the following circumstances:"  and then it 
lists reasons like "There are insufficient resources to produce 
chartered deliverables or to maintain the group, according to priorities 
established within W3C."

If it turned out a WG was chartered to do something completely useless 
or harmful or if the WG has no chance of success, the above is 
sufficient to close it if the Director and AC agree (AC because they can 
appeal and override).

That seems clear that the Working Group closes when the Charter expires 
(the Director can no longer close it after that for the reasons it can 
usually be closed early), but some WGs don't currently act like that is 
true - and that isn't good.  We should change the language to explicitly 
say the WG closes when the charter expires.

Suggested Change: "A Working Group or Interest Group charter specifies a 
duration for the group.  The group closes on that date if the charter is 
not extended or the Working Group is not re-chartered."  (with links for 
extended and re-chartered)

Related issue:  As people noted below, the AC review can decide not to 
renew charters.  That is a reason not to allow long Charter extensions, 
rather than re-chartering where an AC Review happens. Extensions should 
only happen if the WG is so close to completion that review by the AC 
isn't useful or if there was an unexpected delay in rechartering.  
Extension shouldn't be allowed to exceed a brief period like 3 months 
beyond the AC approved duration.





On 2015-03-10 22:08, Michael Champion (MS OPEN TECH) wrote:
>> I'm not sure what the problem is, and I'm not sure it is a team problem.
> I should have phrased what Jeff is reacting to as "to the extent there is a problem here, the team is already empowered to address it."
>
> I agree with Jeff's analysis of the 3 scenarios below.  I interpreted Harry's message as describing scenario #2, and in that case the logical thing is to let the WG continue through their chartered period, then the team can decide whether or not to invest in it going forward.  Or the AC ballot may reveal that there is not enough member support for the WG to continue it.
>
> Scenario #3 is more of  a potential problem, but I agree it's hypothetical unless there are examples that don't come to mind.
>
> So, it would be good to hear of cases where the team thinks they need help from the process to do the right thing and stop wasting resources and reputation on WGs that are unlikely to produce a useful standard.  If laying out the specifics is too delicate a conversation to have in a public forum, the AB might be a better venue than this CG, but hopefully we can publicly discuss facts in a tactful manner.
>
> ________________________________________
> From: Jeff Jaffe <jeff@w3.org>
> Sent: Tuesday, March 10, 2015 5:55 PM
> To: Michael Champion (MS OPEN TECH); Harry Halpin; Revising W3C Process Community Group
> Subject: Re: When to close a Working Group?
>
> On 3/10/2015 7:01 PM, Michael Champion (MS OPEN TECH) wrote:
>>> How do we close a chartered Working Group when its clearly gone off the rails and won't produce a useful standard?
>> I'm sure it's a sensitive topic, but can you give some examples?
>>
>> The obvious way to close a chartered WG is to wait for its charter to expire and then not renew it.  I don't recall that happening very often.  The Web Services Architecture WG was one "success" story of the team and chairs deciding to pull the plug on an effort that wasn't on the road to success, but that was 10 years ago.  Also note that there is a current AC ballot open in which  someone :-)  formally objected to rechartering the WG until it shows it can reach the milestones identified in its existing charter.
>>
>> So, my immediate reaction is that this is a team problem: if a WG doesn't show progress, the team shouldn't devote resources to getting its charter extended or renewed. And arguably it's not fair to close a WG until it has had the time period identified in its charter to reach its objectives, so it's not clear that there needs to be a mechanism to shut down WGs before their charters run out.
> I'm not sure what the problem is, and I'm not sure it is a team problem.
>
> There are several cases:
>
> 1. The WG believes it has nothing productive to do.  I think that if the
> WG has a consensus to close we would just close it.  I don't think we
> need new process for this.
>
> 2. The WG believes that it is doing something useful but others
> disagree.  Here I kinda agree with Mike - given that the WG signed up to
> work on the Charter - it is probably more effort than it is worth to try
> to close it early.  Give the WG a chance to achieve chartered
> deliverables and don't renew it if there is no progress. This case does
> not need any new process.
>
> 3. The WG is doing something so worthless and is so egregiously
> consuming W3C resources that it must be stopped right now!  I'm doubtful
> that this comes up often enough to make it worthwhile to create a
> process for this.
>
>> Another way to look at this: Let's think about how to not charter WGs until there is a spec with a community behind it plus some implementation plans by browsers, frameworks, or tools... and likely IP holders ready to make RF patent commitments. One way to do this would be to more strongly encourage potential WGs to incubate their spec in a Community Group to get at least an initial draft of a spec written down and socialized among implementers and users, and do a Final Specification Agreement to identify those who will make patent commitments.  In short, if we are more selective about chartering only WGs that already have draft specs and  viable communities behind them, we don't have to worry as much about pulling the plug on those that don't get real support.
>>
>> I haven't thought about this much, but perhaps there should also be a WG->CG transition mechanism so WGs that don't get members or implementers can easily reboot as CGs to give them a chance to refocus and demonstrate they're ready to get back on the Recommendation track.
>>
>> -----Original Message-----
>> From: Harry Halpin [mailto:hhalpin@w3.org]
>> Sent: Tuesday, March 10, 2015 3:08 PM
>> To: Revising W3C Process Community Group
>> Subject: When to close a Working Group?
>>
>> I'd like to bring up what I think is an open issue: How do we close a chartered Working Group when its clearly gone off the rails and won't produce a useful standard?
>>
>> For example, a Working Group could be chartered with lots of member support, but later the members withdraw or don't implement. This sort of Working Group seems to be a waste of W3C member resources. However, it seems unclear how to actually end a Working Group.
>>
>> My preference would be that there could be an internal review by W3C if there is a motion by members, staff, or chairs to close the WG, and if the W3C review shows that the worries are justified even if the WG is chartered, then a vote goes to the AC to close the WG. All existing rec-track docs then transform to WG Notes before closure.
>>
>>      cheers,
>>         harry
>>
>
Received on Wednesday, 11 March 2015 19:59:46 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 11 March 2015 19:59:47 UTC