Last updated: June 17, 2004
The efforts to address the structural problems in the IETF have been organized in three phases:
The efforts to start working on the last phases have spawned multiple efforts, with individual lifetimes, control patterns and agendas. This overview tries to call out what the current efforts are, and who is in charge of them.
This section lists the current activities that have separate "identities". Some one-shot activities are mentioned only under the "problem list" that follows next.
The Problem Statement WG has produced two documents - draft-ietf-problem-issue-statement-05 and draft-ietf-problem-process-04. Both have been approved by the IESG for publication as Informational. -issue was approved on February 5, 2004, and published as RFC 3774 in May. The -process document was approved on April 15, 2004. When the RFCs are published, this WG will be closed; this phase of the work is finished. Thanks to all who participated!
draft-alvestrand-ietf-mission is a draft mission statement for the IETF. It is currently (June 2004) at version -02, and is being discussed on the email@example.com mailing list. It has been sent to IETF Last Call for BCP status. This activity should be finished before San Diego.
An Advisory Committee, with Leslie Daigle as chair, was formed in August 2003 to investigate the relationships between the IETF and other bodies ("organizational structure", as opposed to the "procedural structure" that deals with the way WGs, the IESG and the document development process works); part of that investigation has direct impact on this item. This group presented its findings at the Minneapolis IETF Plenary, and its report was published as RFC 3716. The committee is now finished.
The IETF Chair and the IAB Chair are working to propose an implementation of the recommendations from this report - Leslie Daigle is giving monthly reports on the effort on the firstname.lastname@example.org list. The following documents are available:
The then-current drafts of these documents was presented at the plenary in the Seoul IETF (March 2004), and received a generally favourable reception.
In Barcelona, mid-May 2004, the ISOC BoT approved a motion releasing funds for hiring a consultant to help with the process of planning and implementing the administrative restructuring.
The EDU team (website at http://edu.ietf.org/) has provided valuable materials and training for the last several IETFs. It has achieved the goals in the current charter, and is drawing up a revised charter. See the email@example.com list for the discussion, and to submit input!
This group is focused on early review (a good deal before AD review).
The WG chairs intend to go through the available documents with a view to getting experiments started - starting with trying to start an experiment based on SIRS.
Important part of the experiment design: Specify what information we expect to get from them!
We have an I-D surveying the various ideas and inputs to the WG.
So far, there is no resolution of any of the issues that have been raised on the list.
The PROTO team (http://mip4.org/proto/) has identified a number of "pilot projects" to deal with making it possible for WG chairs to deal with a number of issues that so far have depended totally on AD cycles to get done. We hope this will make processing faster.
A team for collecing and channeling community input on IT support functions for IETF work is in the process of being set up by Henrik Levkowetz. This will be announced soon.
The IESG has changed the way it processes RFC Editor documents - see draft-iesg-rfc-editor for details. This document is in Last Call (expires ....)
In June, the IESG is running an experiment with an altered review process for informational and experimental documents from WGs - details were sent to the ietf-announce list on June 7.
An individual proposal - draft-klensin-july14 - has received good feedback for proposing a system of "experimental procedures" that the IETF can implement. The Last Call is expired, and we are working to get the final approval in the IESG finished. It will be adopted as BCP.
A proposal for alternate consensus methods - draft-hardie-alt-consensus - is being put forward as an experiment, and has been last-called to be published under the "july14" rule. This might help in certain situations where a WG gets into a very wedged mode, and needs to make a decision in order to move forward.
draft-wasserman-rfc2418-ml-update has been sent to Last Call for BCP.
This section uses the "IETF Problem Statement" document as a structuring mechanism. Its intention is to show that there is activity covering all the identified parts of the problem area.
The logical response to this is to formulate the IETF's mission in terms that the community can agree with.
See "IETF Mission Statement" above for the status of this effort. While this is not the be-all and end-all of common understanding, it is a step forward.
At the Vienna IETF, the COACH BOF, led by John Loughney and Bernard Aboba, tried to address this issue, focusing on changes to procedures and the introduction of tools. It was not clear from the BOF how one should go forward from there.
There is also the SIRS experiment, run by Brian Carpenter and Dave Crocker, that attempts to get more review into the IETF process, that could be seen as coming under this topic. See ICAR above.
Among the various tools and procedures suggested and tried recently, issue tracking has come in fairly high on the list of things that seem to have some success. At the Minneapolis IETF, part of the "topical" WG Chairs training session was devoted ot issue trackers. Two reasonably large issue trackers based on different technologies - RT (rt.psg.com, contact Rob Austein) and Roundup (roundup.machshav.com, contact Steve Bellovin) have been set up, and are being used by some working groups.
There are multiple other things that can be done, including improvements to the I-D tracker and other tools for handling information that the IETF handles centrally; the Secretariat and the IETF leadership are working on these. A team has been set up (the PROTO team) to look at the tools issues related to workflow, starting with the particular issue of improving the workflow for documents passed between the IESG and the WGs - this was announced on Jan 15 to the ietf-announce list.
As the IETF mission document says, complexity is a fact of life. We are creating more interlinked protocols, and the protocols we create and manage have far greater impact on the real world than previously. The result is that our problems grow more complex; this is something we cannot change, but that we need to learn to address.
Addressing this seems to require more thinking across areas and in architecture. The IAB has attempted to bring effective thinking to bear on some problems, resulting in documents like the OPES considerations and meetings like the IAB open meeting on addresses and their usage in San Francisco; the SIRS experiment is also intended to bring more cross-area review.
Addressing the issue of review directly, the IESG initiated an activity in Minneapolis under the name of "ICAR" - "Improved Cross Area Review", with the mailing list "firstname.lastname@example.org". This became a working group in February - see above.
The NEWTRK BOF in Minneapolis, chaired by Scott Bradner, tried to address the issue of standards hierarchy update. It recommended that a working group be formed, and the chair solicited volunteers for a design team to write a proposed charter. For details, see the meeting minutes. This became a WG in March - see above.
The list for the effort is email@example.com .
Suggestions to increase motivation for participants have been floated. One
can also imagine reducing the workload, but since many participants' engagement
is linked to specific work items, reducing scope reduces participants too; not
clear that this will converge.
If we can change some of the other problems (such as effective management, more effective review, or addressing large and complex problems effectively), this will have impact on this item too.
No current activity.
In Minneapolis November 2003, the IESG presented a proposal to address the managements structure mismatch by drawing on the resources of the working group chairs - a layer of the IETF management structure that is already present, and consists of more than 200 people. This led to extensive debate, and to the formation of the MPOWR mailing list (firstname.lastname@example.org) - more details on this below under 2.7.
An Advisory Committee, with Leslie Daigle as chair, was formed in August 2003 to investigate the relationships between the IETF and other bodies ("organizational structure", as opposed to the "procedural structure" that deals with the way WGs, the IESG and the document development process works); part of that investigation has direct impact on this item. This group presented its findings at the Minneapolis IETF Plenary, and its report was published as (currently) draft-iab-advcomm-01.txt. This will be published as an RFC; the committee is now finished.
The IETF Chair and the IAB Chair are working to propose an implementation of the recommendations from this report - Leslie Daigle is giving monthly reports on the effort on the email@example.com list, and a draft will be published before the Korea I-D deadline.
The EDU team, formed in the spring of 2003 under the leadership of Margaret Wasserman, has tried to bring more resources to bear on WG Chairs training, and also to add training for document editors. The team, which is being managed in the General area, has a charter approved by the IESG, and is running its website at edu.ietf.org.
Note that the EDU Team has to focus on training people to work within the current structure; if the IETF structure changes, the EDU team will follow up with changes to its curriculum, but it is not expected to initiate change.
The idea that WG chairs should have both the power and the responsibility to make sure their WGs conduct orderly debates and come up with technically sound solutions is fairly well embedded in the structure, but is not explicit in RFC 2418. A proposed set of revisions to RFC 2418 (draft-wasserman-rfc2418-update) turned out to be more controversial than the IESG initially expected, and has uncovered a clear lack of IETF consensus on what the present role of WG chairs is. Debate continues within the scope of the MPOWR mailing list (abbreviation for Management Positions - Oversight, Work and Results), including debate over practices to deal with the "disruptive participant" problem - a perennial staple of IETF discussions. This may turn into a working group; at least a BOF in Korea is expected.
The EDU Team has been charged with continuing and improving training and
educational resources for participants and leaders.
Currently, the EDU Team efforts are focused on training and education for participants, WG chairs and document editors.
The following is a brief history of what's happened since October 2003, and an extremely sketchy version of expected timelines for the various change processes.
Note: There are interactions between the various components. There have been arguments both in favour of doing "one thing at a time" and stick to procedures we know are problematic until other changes are implemented, and in favour of doing "everything at once", attempting to get to a stable end-state as soon as possible, limiting the time of disruption that's inevitable. At the Minneapolis IETF, the people who spoke at the plenary mike seemed to be strongly biased towards "going slow" - not attempting to do everything at once.
The implementation plan will no doubt be modified according to both perspectives, as the details grow clearer.
At the moment the discussion on these matters is carried out in multiple places.
All lists are monitored by some subset of the IESG members. Mail may also be sent to the IESG directly.
Other mailing lists that are relevant to specific problems:
All these lists obey the "-request" convention; send a message to <listname>-request@<domain> with the word "subscribe" in the body, and the right thing should happen.