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

Re: process.next wish list

From: Tim Berners-Lee <timbl@w3.org>
Date: Wed, 15 Apr 2015 10:03:47 +0100
Cc: Wayne Carr <wayne.carr@linux.intel.com>, W3C Process Community Group <public-w3process@w3.org>
Message-Id: <2107D98F-E5AA-47D0-910A-B16A31E3835E@w3.org>
To: David Singer <singer@apple.com>

On 2015-04 -15, at 09:31, David Singer <singer@apple.com> wrote:

> Hi Wayne
> I donít disagree with any of this.  Just one point, though:
>> On Apr 15, 2015, at 0:02 , Wayne Carr <wayne.carr@linux.intel.com> wrote:
>> 2. Change the Process so a Working Group or Interest Group closes when the Charter expires.
>> A Working Group charter can be extended by an email from the Director extending it (subject to AC Appeal to override the extension).  A Working Group charter can be renewed by an AC Review and Director approval of a revised Charter.  If neither happen by the date the Charter for a WG expires, then the WG should close on the date of Charter expiration.   The current practice is that Working Groups with expired charters continue to work as if they have a charter.  That, in effect, removes the AC's ability to monitor and approve the continuation of Working Groups.  The AC can impact extensions and renewals, but can do nothing if WGs are allowed to continue operating without an expired Charter.  One negative consequence of this is that the policy for moving abandoned work to Community Groups requires that work be stopped on a specification.  In an expired group, what has happened is the group doesn't function, but it also doesn't close and release the specs.
> I agree itís a bad situation that we have groups operating on expired charters.

There is an underlying set of issues here. Simplistic version:

1. Groups work faster with a short charter.    If you don't have an aggressive charter which outlines the fastest the work cold possibly be done, people won't put the energy into it to make progress, and people will even be disinclined to join at all a multi-year effort.

2. Groups always take longer than expected to finish.  Well not always but very very often. The same people who would not have wanted to work on a long timescale later find lots of real good reasons why they want to investigate things more deeply, or liaise with other people, or elaborate simpler parts into more component parts, or get distracted etc etc.

3. Groups should now work without a charter agreed with the AC.

An possible solution to this may be to draw up an initial charter with milestones according to an aggressive schedule, but add on an equal amount of time for unspecified contingency.  The group reports to the staff and the AC how much of the contingency has been used up, and why.  But the extension of the charter is automatic. When they are near the end of they have to go to get a mandate to work for more time, with  an explanation and a complete new plan of the same form with contingency.

> But I think that we need to get much better at our practices before we make this rule change. Itís deplorable, but at the moment groups Ďrun off the end of the pierí, let their charter expire, and only then start worrying about renewing it. If we were to introduce this rule today, when charter expiry is all too common, chaos and panic would ensue. I donít want to be in the position of the forced closure of a WG that we all agree should have been re-chartered and still operating. That would be Ö embarrassing.

Maybe where a group is "just" late, and has not for example actually taken on a wider scope of work, then the rechartering should be quick and easy.

> If we get to the point where charter renewal is normally dealt with in the two quarters *preceding* expiry, then I agree, a WG that didnít follow that Ďnormal practiceí, or failed to get a charter agreed in that period, should indeed close.
> David Singer
> Manager, Software Standards, Apple Inc.

Received on Wednesday, 15 April 2015 09:03:59 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:51:28 UTC