Chartering has been perceived as a laborious process. Some IETF areas have created an area-wide working group with an open charter to handle one-off documents that don't have a clear home, in part because chartering a working group for these efforts seems too difficult and time consuming. However, such area-wide working groups can end up being a dumping ground for efforts that really require the work of a focused group of people. As an alternative to having the Applications Area Working Group (appsawg) handle some of this work, in 2012 the Applications ADs started working with the idea of quickly chartering small, tightly scoped, short-lived working groups. As of early 2014, we have done three working groups this way (imapmove, qresync, and jcardcal), and it has worked extremely well.
The imapmove working group was a good first try on this, as it started with a reasonable writeup of a problem that had been on the table in the IMAP community for quite some time. Similarly, the goal of qresync was to update and merge a couple of specifications for which it's been known for some time that updates were needed. The jcardcal working group actually had some significant development to do, but the group contained sufficient expertise and modeled their work on some well-understood prior technology.
Even though chairs were selected who could push the work forward, that did not require having only experienced folks in those slots. For qresync, a long-time participant who had never chaired was used, and in jcardcal, we paired a brand new participant with an experienced one.
The quick chartering process can be kept down to something on the order of four to five weeks: a week or two to build the charter and have brief discussion on it, a week of Internal Review by the IESG and IAB with initial approval on a telechat, a week of External Review by the community, and a week of IESG review of the final charter with final approval on the next telechat. The key is keeping that first step, the initial charter proposal, short — which is why this works with non-controversial topics for which a tightly scoped charter can be agreed upon quickly, with no rat holes.
After that, with a small number of clear deliverables, a group committed to working on it aggressively, and chairs who are prepared to push on it and not allow time to elapse unproductively, we can get work done in a few months, rather than a few years. For imapmove, this was the actual timeline:
That's about four months from the charter announcement to the IETF-wide last call of the document. After document approval, we waited until the RFC was published before closing the working group.
We encourage ADs for other areas to use this model, chartering short-lived, tightly scoped working groups, where appropriate, instead of longer-lived working groups that take longer to charter and years to complete their work.
The content of this page was last updated on 2014-04-22. It was migrated from the old Trac wiki on 2023-02-17.