Internet-Draft M. Wasserman, Editor Document: draft-ietf-problem-process-01.txt Wind River Expires: December 2003 June 2003 IETF Problem Resolution Processes Status of this Memo This document is an Internet-Draft and is in full conformance with all provisions of Section 10 of RFC2026 [RFC2026]. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet- Drafts. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html. Abstract This document suggests processes to address the problems identified in the IETF Problem Statement. This document decomposes each of the problems described in the problem statement into a few areas for improvement, categorizes those areas into longer-term and near-term problems, and suggests processes to address each area. Copyright Notice Copyright (C) The Internet Society (2003). All Rights Reserved. Wasserman, Editor Expires December 2003 1 IETF Problem Resolution Process June 2003 Table of Contents Status of this Memo...............................................1 Abstract..........................................................1 Copyright Notice..................................................1 Table of Contents.................................................2 1 Introduction..............................................3 2 IETF Purpose and Core Values..............................3 2.1 Non-Core Values...........................................5 3 Building on our Success...................................6 4 Problem Decomposition.....................................7 4.1 Decomposition of Mission Problem..........................8 4.2 Decomposition of the Engineering Practices Problem........8 4.3 Decomposition of the Complex Problems Problem.............8 4.4 Decomposition Standards Hierarchy Problem.................9 4.5 Decomposition of the Engagement Problem...................9 4.6 Decomposition of the Management Scaling Problem..........10 4.7 Decomposition of the Working Group Practices Problem.....11 4.8 Decomposition of the Preparedness Problem................12 5 Process Recommendations..................................12 5.1 Near-Term Improvements...................................12 5.1.1 Suggestions to Improve WG Quality Processes..............14 5.1.2 Suggestions to Increase the Use of Tools.................14 5.1.3 Suggestions to Improve Training..........................14 5.1.4 Suggestions to Increase WG Chair Communication...........15 5.1.5 Suggestions to Improve Maintenance of Standards..........15 5.2 Longer-term Improvements.................................15 5.2.1 IETF Improvement Working Group...........................15 5.2.1.2 Working Group Oversight and Management...................16 6 Conclusion...............................................17 7 Security Considerations..................................17 8 Normative References.....................................18 9 Informative References...................................18 10 Acknowledgements.........................................18 11 Editor's Contact Information.............................19 12 Appendix A: Suggested Charter for the Improvement WG....19 Wasserman, Editor Expires December 2003 2 IETF Problem Resolution Process June 2003 1 Introduction This document suggests processes to address several problems facing the Internet Engineering Task Force (IETF) that have been described in the IETF Problem Statement [IETFPROB]. This document begins with a short discussion of the purpose and core values of the IETF, and it offers a reminder of the many good things about the IETF that we donĘt want to lose in the process of solving our problems. We then decompose each of the problems described in the problem statement into a few areas for improvement, and organize those areas for improvement into two categories: - Areas that can be addressed in the near-term, via discrete, minimally disruptive changes or improvements. - Areas that would require longer-term efforts to address, such as fundamental changes to our organizational structure or standards-track processes. It is suggested that the IETF work on these two classes of improvements in parallel, so that we can enjoy some near-term benefits while more fundamental, longer-term improvements are being carefully considered and executed. Concrete suggestions are included for how we can begin or continue work on near-term improvements. The document then offers recommendations for how to initiate, organize and manage our longer-term improvement effort. 2 IETF Purpose and Core Values As we consider how to address the problems with the IETF processes and organizational structure, it is important to keep in mind the things about the IETF that we don't want to change -- our sense of purpose, and the core values that give the IETF its unique identity. It is not the role of the Problem Statement Working Group (WG) to document IETF consensus regarding the purpose and core values of the IETF -- that role is recommended for a future WG. However, the words of our past and present leaders may help to set the right tone for this effort. The IETF has a rich history and tradition, full of memorable quotes that capture our spirit and values. Two of the most memorable are: "We reject kings, presidents and voting. We believe in rough consensus and running code." -- Dave Clark "Be conservative in what you send, liberal in what you accept." -- Jon Postel Wasserman, Editor Expires December 2003 3 IETF Problem Resolution Process June 2003 At two IESG plenary meetings in 2002, the chair of the IETF, Harald Alvestrand, presented his view of the purpose and core values of the IETF. These presentations may serve as a useful basis for focusing on our mission and core values. At the IESG plenary in London in July 2002, it was stated that the purpose of the IETF is to "produce high quality, relevant, and timely technical standards for the Internet". Our organizational structure and processes should be judged by how well they help us to achieve that mission. At the following IESG plenary in Atlanta, Georgia in November 2002, five core values were presented [COREVAL]: "Cares for the Internet" As its name implies, the Internet Engineering Task Force (IETF) focuses on Internet-related activities. We care about the Internet, and our standards work and operational activities are intended to improve the utility, scalability and availability of the Internet. The Internet isn't value-neutral, and neither is the IETF. We want the Internet to be useful for communities that share our commitment to openness and fairness. We embrace technical concepts such as decentralized control, edge-user empowerment and sharing of resources, because those concepts resonate with the core values of the IETF community. These concepts have little to do with the technology that's possible, and much to do with the technology that we choose to create. The IETF community also cares about making the Internet model a viable business proposition. People who choose to offer Internet products and services that fit with our core values should be able to do so with maximum benefit and minimum amount of fuss. The IETF community wants the Internet to succeed because we believe that the existence of the Internet, and its influence on economics, communication and education, will help us to build a better human society. "Technically Competent" We pride ourselves on our technical competence, and our processes are intended to ensure the high technical quality and utility of our standards and other documents. "Open Process" Openness is a core attribute of the IETF. Our standards and other documents are developed in an open process, which allows us to achieve wide input and review. Wasserman, Editor Expires December 2003 4 IETF Problem Resolution Process June 2003 Anyone can participate in defining Internet standards in the IETF. We do not require corporate membership. We make final decisions on mailing lists, not at face-to-face meetings, so anyone with Internet access can contribute. All IETF documents are freely available, whether they are active working documents or finished specifications. Individuals as well as working groups may submit Internet-Drafts for consideration as Internet standards. "Volunteer Core" With some honorable exceptions, the IETF community consists of people who are employed elsewhere, and much of our IETF work is directly related to the business of our employers. However, many of us take on additional roles in the IETF, beyond those directly sponsored by our employers. We participate in the IETF as individuals, because we want to work for the good of the Internet community and its inhabitants. The IETF community is committed to the continued success of the Internet, not to the continued success of the IETF itself. IETF is only worthwhile if it can effectively produce high quality, relevant standards that benefit the Internet. Openness and individual participation are both parts of an interlocking structure that is the strength of the IETF. The openness permits all segments of the Internet community to participate, without demanding that they meet any qualifying criteria, such as belonging to a member company. The individual participation allows us to focus on a wider set of "success criteria" than the health and well-being of our individual employers. Ultimately there is no conflict between the volunteer nature of the IETF and employer-sponsored participation, because we believe that the long-term survival and growth of the Internet benefits ourselves, our societies and our employers. "Rough Consensus and Running Code" It is an inherent part of the IETF culture that we base our decision making on rough consensus of the community, developed through open discussion. We also value running code as an indication of specification quality and completeness, and we require interoperable implementations for promotion in the standards process. 2.1 Non-Core Values Understanding our core values will also help us to understand the long-standing features of the IETF that we can change without compromising our values or sacrificing our unique identity. Wasserman, Editor Expires December 2003 5 IETF Problem Resolution Process June 2003 During the November 2002 IESG Plenary, Harald Alvestrand also presented the following "non-core values" [COREVAL]: - The division into WGs and Areas - The three-step standards process - The ASCII format for RFCs and I-Ds - The format of IETF meetings - The structure of WG mailing lists - The powers of the IESG and IAB These things were designed to help us achieve our goals in a way that is consistent with our core values. If they are no longer effective, we can and should change them. 3 Building on our Success While focusing on our operational problems, we shouldn't forget that the IETF is a very successful organization. We are responsible for some of the most widely used communications standards in the world, and we have contributed to the creation and growth of the Internet, one of the greatest technical and social achievements of our time. It is important to consider that: - Good times hide problems; - Bad times hide successes. In good times, it is easy to succeed despite operational inefficiencies, so organizations tend to ignore operational problems and focus on their success. In bad times, organizations can become overly critical of their own structure and processes, blaming the organization for problems that are actually caused by outside forces. We are currently suffering difficult times in the IETF and throughout the communications industry. The IETF should be careful not to unjustly blame our own organizational structure or processes for the effects of industry-wide changes such as: - Economic issues in the global communications industry, which are causing increased scrutiny regarding expenses and return-on-investment. These same factors are causing job changes and uncertainty for many IETF participants. - The commercialization of the Internet, which has drastically increased the financial impacts of standardization. - The convergence of the datacom and telecom sectors of the communications industry, which has led to an influx of experienced people into the IETF with a different culture and industry perspective. Although it is important to recognize and correct the serious organizational problems currently facing the IETF, many of these problems have existed for years, and the IETF has been successful Wasserman, Editor Expires December 2003 6 IETF Problem Resolution Process June 2003 in spite of these issues. We should not overreact to these issues with sweeping revolutionary changes to the IETF structure and processes. Instead, we should focus on developing a culture of continuous operational improvement through which we can evolve our organizational structure and processes to make them more scalable and effective. We should take this opportunity to develop the mechanisms and processes that we can use to continually monitor and improve our organizational effectiveness, both in good times and bad times. The IETF currently has a large amount of valuable work underway, and care should be taken not to disrupt or delay that work while we address our organizational problems. The IETF is also fortunate to have a large number of extremely talented and dedicated individuals that serve in formal and informal leadership roles throughout the organization. We should be careful not to alienate or disenfranchise our leaders and key contributors while making organizational or process changes. 4 Problem Decomposition The problem statement document lists seven root cause problems currently facing the IETF: - Participants in the IETF do not share a common understanding of its mission; - The IETF does not consistently use effective engineering practices; - The IETF has difficulty handling large and/or complex problems; - Three stage standards hierarchy not properly utilized; - The IETF's workload exceeds the number of fully engaged participants; - The IETF management structure is not matched to the current size and complexity of the IETF; - Working group practices can make issue closure difficult; and - IETF participants and leaders are inadequately prepared for their roles. Each of these problems can be decomposed into several areas for improvement, some of which can be addressed in the near-term while others require longer-term consideration. It is also important to note that the problem statement lists problems that have been reported by some members of the IETF. Although all of these problems are believed to exist, not all of these problems are present in all parts of the IETF, and some of these problems may in fact be symptoms of other problems. Wasserman, Editor Expires December 2003 7 IETF Problem Resolution Process June 2003 4.1 Decomposition of Mission Problem In order to determine the best organization and processes for the IETF to fulfill its mission and achieve its goals, we need to reach a common understanding of the mission and goals of the IETF. Although it should be possible to understand the mission and goals of the IETF with no disruption to our current processes, it would be most valuable as part of a longer-term effort to align the organization and priorities of the IETF with its mission. As part of understanding our mission, the IETF will need to identify our stakeholders and understand how we serve them. We will need to define the scope of the IETF, so that it is possible to determine what is in-scope and out-of-scope for the organization. We will also need to define our goals and priorities, and learn how to recognize and measure our own progress and success. Once we reach consensus on the mission and goals of the IETF, we should develop educational programs or resources to expose new participants to these principles, so that they can be quickly acclimated to the IETF culture. 4.2 Decomposition of the Engineering Practices Problem The IETF lacks effective engineering practices in four major areas: 1. Failure to clearly define the scope of the work, engineering trade-offs and acceptance criteria for each project. 2. Lack of effective mechanisms for issue tracking and/or document change control. 3. Lack of effective processes to ensure quality throughout the development of IETF work items, such as intermediate acceptance criteria or formal review processes. 4. Sufficient focus on milestones, and recognition or rewards for individuals or groups that achieve timely, high quality execution. Some of these areas (issue tracking and revision control) would require that tools are made available to WG chairs and editors, and that IETF participants (at various levels) are educated in how to use them. The other areas concern the formation and process management of IETF WGs, and would require documentation and adoption of effective engineering processes within IETF WGs. 4.3 Decomposition of the Complex Problems Problem The IETF has effective mechanisms for dealing with well-defined problems of limited scope. These problems are well handled in IETF WGs, where experts in a given technology can convene and solve the problems specific to one technology area. However, we are much Wasserman, Editor Expires December 2003 8 IETF Problem Resolution Process June 2003 less effective at resolving complex problems that affect more than one IETF WG or area. Today most communication between WG chairs, especially across area boundaries, goes through the IESG. Some inter-WG or inter-area communication problems could be alleviated by greater communication and coordination directly between the chairs of related WGs. There are some near-term efforts underway that are intended to increase communication between WG chairs. Other complex problems involve higher-level issues, such as unified architecture or highly-coordinated multi-area efforts. As part of any IETF reorganization, we should consider management structures that will allow us to achieve a better focus on architectural and cross-area issues. 4.4 Decomposition Standards Hierarchy Problem There are several problems with the IETF's three-step standards process. These problems can be grouped as follows: - The three standards-track steps are not used effectively within the IETF. - The IETF standards-track is not well understood by the users of IETF standards. - The current standards process does not make it easy for users to locate a set of related documents, such as an architectural framework and associated protocols. - The IETF does not have an effective way to maintain IETF standards. Major changes to the standards-track should only be considered as part of a longer-term process that includes an understanding of our mission and goals. However, there may be near-term changes that we could make to better maintain current IETF standards, or to make them more accessible to users. [OPEN ISSUE: Is this the correct balance between near-term and long-term changes to the standards-track? Or could we consider changes to the three-step process (i.e. to make it a two-step process) in the near-term?] 4.5 Decomposition of the Engagement Problem The engagement problem can be decomposed into three primary issues: - Some WGs do not have sufficient participation, and WG documents are often produced by very small groups of people, perhaps with limited expertise in some relevant areas. - WG documents are not adequately reviewed by people outside of the originating WG. Wasserman, Editor Expires December 2003 9 IETF Problem Resolution Process June 2003 - People lose interest in longer-lived WGs, especially when protocols take a very long time to develop. When too few people, or people representing too few areas of expertise, review WG documents this can result in poor quality output. We need to find ways to increase the effectiveness of document review at all levels. Quality processes based entirely on a gatekeeper at the end, whether that gatekeeper is the IESG or a WG review board, tend to result in a lower focus on quality by other participants. So, it is likely that instituting better quality processes throughout document development, including acceptance criteria and review at several stages, would increase the focus of WG participants on document quality. When the interest of document editors or key contributors starts to flag, this can cause serious problems for a WG. This most often happens when WGs are floundering, or when charters are so loose that WGs lose focus. It also happens when WG documents get delayed in AD review and/or IESG review for long periods with little feedback, or when the WG lacks consensus to progress its documents. Improvements to our processes for chartering, tracking or managing WGs could help to alleviate many of these problems. Training for WG chairs in how to manage difficult situations and disruptive contributors might help, as well. We also need to better understand what motivates people to become deeply engaged in the IETF and to remain engaged. It is possible that expanding the number of formal leadership positions and/or coming up with more effective ways to acknowledge our top technical contributors could encourage more people to become, and remain, deeply engaged in IETF 4.6 Decomposition of the Management Scaling Problem There are several issues grouped into the concept that the management structure of the IETF is not well matched to the size and complexity of the organization. One or two of these problems might be addressed by near-term solutions, but resolving the primary problem will require some type of IETF reorganization. There are four major areas for improvement that are grouped under this problem: - The current organization of the IETF does not scale. IESG members are running too many WGs, reviewing too many documents, etc. Most IESG members have dozens of direct reports (WG chairs, directorate members, etc.). In its current form, there are very few people who could do a good job as an IESG member, and the huge time commitment and responsibilities of this role make it very difficult to find qualified people who are willing to serve on the IESG. - Current IESG members and other IETF leaders are overloaded. Wasserman, Editor Expires December 2003 10 IETF Problem Resolution Process June 2003 - The IETF selection processes have tended to select leaders (IESG, IAB and WG chairs) from the same small pool of people. The IETF needs to identify and develop additional leadership, and to delegate real authority and influence to a larger group. - The IETF is not effective at identifying and developing new leaders, and we lack sufficient recognition for the contributions of IETF participants. - One or two IESG members can block WG documents indefinitely (in AD review or IESG review). Some level of IETF reorganization is needed to improve in the first two areas. This should be undertaken as part of the longer-term improvement effort. In parallel with a longer-term IETF reorganization, however, some relief could be achieved by modifying IESG internal processes to remove the potential for one or two IESG members to indefinitely delay a WG document, either on purpose or due to work overload. The I-D tracker has already resulted in some improvement in this area, as it has created visibility regarding how and why a document is being delayed, but it may not have resolved all of the issues in this area. The IESG may also be able to take near-term steps, with community visibility and agreement, to transfer or delegate more work to WG chairs, to directorates, to the IAB, or to other to people in formal or informal leadership positions. If additional leadership positions are needed for this purpose, the IESG should consider creating them. The IESG could also help to expand the leadership pool of the IETF by actively seeking interested and qualified people for leadership positions, and by using more open processes for the selection of WG chairs and other influential positions. 4.7 Decomposition of the Working Group Practices Problem Although "rough consensus" is considered a core value of the IETF, consensus-based decision making works best in smaller groups with a common viewpoint and common goals. Somehow we need to resolve the apparent conflict between our core values regarding rough consensus, and our desire to be an effective organization with several thousand participants. Although consensus-based decision making has some inherent issues, there are some problems in the IETF that exacerbate these issues: - WG chairs may lack the skills and training to deal with common behavior problems that undermine or prevent consensus. - IETF participants are often unaware of how the IETF decision-making processes are intended to work. Wasserman, Editor Expires December 2003 11 IETF Problem Resolution Process June 2003 - WG chairs and participants often lack good conflict resolution skills. Each of these issues could be addressed through training or other educational resources. 4.8 Decomposition of the Preparedness Problem The IETF could benefit from training and educational resources that increase the preparedness of IETF participants and leaders at all levels. The IETF currently has formal training programs for new attendees and for new working group chairs. However, our current training programs could use some improvement. There are also several other groups who could benefit from training or other forms of development (web tutorials, on-line resources, references, mentoring, etc.), including continuing attendees, experienced WG chairs, document editors and IESG members. There is an effort underway to improve the IETF's internal education programs, and we recommend that it be continued. 5 Process Recommendations It is the overall recommendation of this document that we pursue near-term improvements to resolve IETF problems in parallel with a longer-term effort to reorganize the IETF and improve our standards processes. None of the efforts suggested in this document should be blocked pending the completion and publication of this document. Ongoing efforts should continue, and new efforts should start as soon as there is IETF consensus that they are worthwhile. In our improvement processes, we should attempt to focus our near- term improvements on areas that are less likely to be substantially modified by our longer-term efforts, thus minimizing the likelihood of making our own efforts obsolete. 5.1 Near-Term Improvements Many of the problems currently facing the IETF can be resolved, or mitigated, through near-term improvements to our current IETF organization and processes. Many of these near-term improvements are completely separable, and there is no reason to aggregate these efforts into a single IETF WG. It is also unnecessary that all of these changes be directed by the (already overworked) IESG. However, in order to prevent the chaos and confusion that could be caused by trying to change everything at once, it is recommended that we choose a few high priority areas for improvement and focus on making improvements in those areas. In choosing which areas to pursue first, we should consider the following criteria: Wasserman, Editor Expires December 2003 12 IETF Problem Resolution Process June 2003 - We should address our most urgent, important problems. - The areas chosen should be cleanly separable, to allow multiple improvements to be carried out in parallel with minimal interference. - We should maximize the benefit vs. the cost of making the improvements (i.e. look for low hanging fruit). - As much as possible, we should focus on improvements that are less likely to be completely invalidated by a longer- term reorganization of our management structure. This might be accomplished by focusing on improvements at the WG and participant levels, rather than at the IESG/IAB level. In the sections above, we have identified several areas that could benefit from near-term improvements, including: 1. Improve WG quality processes and the effectiveness of document reviews at all levels. 2. Increase the availability and use of issue tracking and document sharing/revision control software in the IETF. 3. Improve training and resources for IETF leaders and participants at all levels. 4. Improved communication between WG chairs to identify and resolve inter-WG and inter-area problems. 5. Consider IETF processes or structures to better maintain IETF standards. 6. Modify IESG-internal processes to make it impossible for one or two IESG members to indefinitely delay a document. 7. Modify IESG processes to delegate more responsibility to WG chairs, to directorates, to the IAB or to people in other formal or informal leadership positions. 8. Modify the WG chair selection processes to widen the group of people considered, and consider ways to develop more leaders for the IETF. 9. Initiate regular AD review of WG milestones and progress. Applying the criteria outlined above, it would make the most sense to address areas 1, 2, 3, 4 and 5 through immediate short-term efforts. These are high-priority issues, they are sufficiently separable to be pursued in parallel, they place minimal additional burden on the IESG, and they are the least likely to be affected by an IESG/IAB-level reorganization of the IETF, or by subsequent changes to the standards-track document process. Specific recommendations for how to proceed in each of these areas are made in the following sections. Wasserman, Editor Expires December 2003 13 IETF Problem Resolution Process June 2003 The IESG should consider internal changes to address areas 6, 7 and 8. Area 9 would require a substantial time commitment from IESG members, so it is not suggested that near-term improvements be pursued in this area, unless the IESG believes that the near-term benefits would justify the effort. 5.1.1 Suggestions to Improve WG Quality Processes A working group should be formed in the General Area of the IETF to make improvements to the WG quality processes, including: The WG (re-)chartering process, the quality processes used by IETF WGs, and the effectiveness of IETF reviews at all levels. It should be the goal of this WG to improve the quality and timeliness of WG work output. This WG would be chartered to resolve the non-tools- related portions of the Engineering Practices problem (section 4.2) the WG-related portions of the Engagement Problem (section 4.5), and the non-training-related portions of the WG Practices problem (section 4.7). A great deal of efficiency and synergy can be achieved by adopting common processes throughout an organization. However, it is a strength of the IETF that WG chairs are given a great deal of latitude to choose their own processes and tools, based on the size and nature of their WGs. So, in general, processes and tools should be made available to WGs and WG chairs, not forced upon them. 5.1.2 Suggestions to Increase the Use of Tools Ideally, the proliferation of tools within the IETF would be accomplished via grass-roots efforts, organized by participants within the IETF. One example of this type of effort is the recent adoption of Jabber for use during IETF meetings. However, it is also possible that the IESG could designate functional leaders for specific tools-related efforts and support those leaders in organizing those efforts. It also might be helpful for the IETF to set-aside some technical and systems resources, to make useful tools available to WGs and participants throughout the IETF. These efforts should resolve the tools-related portions of the Engineering Practices problem (section 4.2). 5.1.3 Suggestions to Improve Training The current WG chairs and newcomer's training efforts should be continued and expanded as possible to cover training for other groups. This effort is expected to address the Preparedness problem (section 4.8), and the training-related portions of the Mission Problem (section 4.1) and the WG Practices problem (section 4.7). Wasserman, Editor Expires December 2003 14 IETF Problem Resolution Process June 2003 5.1.4 Suggestions to Increase WG Chair Communication Some efforts are already underway to allow WG chairs to meet each other, and to given them opportunities to establish communication channels. These efforts include WG chair socials and training sessions for experienced WG chairs. These efforts should be continued. The IESG could help to promote chair-to-chair communication by encouraging direct communication between WG chairs when multi-WG issues arise. However, most of the responsibility for establishing effective chair-to-chair communications channels lies with the individual WG chairs. We should stop relying on the IESG to resolve inter-WG issues, and start communicating with each other directly regarding inter-WG issues. These efforts may help to alleviate the Complex Problems problem (section 4.3), although a comprehensive solution to that problem would probably require some changes to the IETF management structures. 5.1.5 Suggestions to Improve Maintenance of Standards The IETF should consider proposals to improve the way that IETF standards are maintained. It might be possible for the IESG to document and implement a mechanism to maintain IETF standards without the need for a WG to enact this change. This effort should address the maintenance-related portions of the Standards Hierarchy problem (section 4.4). 5.2 Longer-term Improvements There are two major areas where we should consider longer-term efforts to improve the IETF: - Organizational structure - IETF standards-track process These two areas cannot be completely decoupled, as the roles and responsibilities of the IETF leadership are largely defined in terms of the standards process, and vice versa. Also, the standards-track process and the roles of IETF leadership are both largely defined within the same documents (RFC 2026 and RFC 2418). Therefore, a new organizational structure and any required changes to the standards-track process should be determined and enacted by a single WG, called the IETF Improvement WG (improve). The WG is encouraged to work on these issues in parallel, where possible. 5.2.1 IETF Improvement Working Group Wasserman, Editor Expires December 2003 15 IETF Problem Resolution Process June 2003 An IETF Improvement WG should be formed to resolve the Mission Problem (section 4.1), the Complex Problems problem (section 4.3), the Standards Hierarchy problem (section 4.4), the Management Scaling problem (section 4.6), and the longer-term portions of the Engagement Problem (section 4.5) This group should be empowered to make changes to the Internet Standards Process [RFC2026], The IETF WG Guidelines [RFC2418], the Nominations Committee process [RFC2727] and the charters of the IESG and IAB [IESG-CH, RFC2850], as needed to correct the problems with our current organization and standards track processes. However, the group should not make any changes in these areas until the goals and potential impacts of those changes have been carefully analyzed. 5.2.1.1 Working Group Charter and Deliverables The IETF Improvement WG will focus on three areas: - Reaching community consensus on the mission, core values, scope and goals of the IETF. - Improving the scalability and effectiveness of the IETF's management structure. - Improving the utility of the IETF's standards track document processes. This WG will follow a two-phase process. In the first phase, the WG will undertake an effort to reach community consensus regarding the mission, core values, scope and goals of the IETF. During this phase, we will also learn how to recognize and measure the success of the IETF, and we will generate performance baselines that can be used to evaluate later changes. In the second phase, the WG will determine what changes, if any, are required to the organizational structure and standards-track processes of the IETF to enable us to efficiently and effectively achieve our mission and meet our goals, while remaining consistent with our core values. Phase Two tasks will not be started until the deliverables for Phase One have been completed by the WG and submitted for publication. [OPEN ISSUE: Do we really want the Phase Two tasks to wait until after Phase One is completed? Or can work on these phases proceed in parallel?] A proposed charter for the IETF Improvement WG can be found in Appendix A. 5.2.1.2 Working Group Oversight and Management The IETF Improvement WG should be managed by the WG chair(s), using standard IETF practices and procedures, as defined in RFCs 2026 and 2418 [RFC2026, RFC2418]. In keeping with current IETF processes, Wasserman, Editor Expires December 2003 16 IETF Problem Resolution Process June 2003 the chair(s) of this WG will be chosen and managed by the General Area Director (AD). It is recommended that the General AD employ a selection process that allows for community visibility and input, similar to the process that was used to choose the chairs of the Problem Statement WG. To ensure that there is community consensus regarding the charter of this WG, the charter for the IETF Improvement WG will be developed within the Problem Statement WG and included in the final version of this document. We recommend that the IETF WG and document processes defined in RFCs 2418 and 2026 [RFC2418, RFC2026] be used for the oversight of the IETF Improvement WG. In particular: - The WG should be formed in the General Area of the IETF, with the General AD serving as the "responsible AD". - The documents produced by this WG should be submitted to the IESG for approval and publication, according to the usual IETF processes. - Any process documents that are produced by this group and approved by the IESG will be reviewed by the ISOC Board of Trustees, using the processes established for previous IETF process updates. - If necessary, any appeals based on the processes or output of this WG would be handled according to the appeals procedures defined in RFCs 2418 and 2026. 6 Conclusion The IETF has problems, and we need to work to solve those problems, both via focused near-term improvements and via a longer-term effort to build an IETF organizational structure and processes that can better handle our current size and complexity. However, the IETF is also an effective organization with a long tradition of excellence, and core values that we donĘt want to compromise in the course of improving our organization and processes. So, any major changes undertaken in the IETF should include an articulation of the IETF's mission and our core values, so that we can ensure that we build an organization that can carry out our mission in harmony with our core values. Working together, we can resolve the problems currently facing the IETF and make the IETF an even more effective, successful and fun place to work. 7 Security Considerations This document contains suggestions for processes that the IETF could use to resolve process-related and organizational problems with the IETF. Although the structure and quality of the IETF's Wasserman, Editor Expires December 2003 17 IETF Problem Resolution Process June 2003 processes may have an affect on the quality of the IETF's security- related work, there are no specific security-related issues raised in this document. 8 Normative References [IETFPROB] E. Davies (ed.), "IETF Problem Statement", draft-ietf-problem- issue-statement-02.txt, June 2003 [RFC2026] S. Bradner, "The Internet Standards Process -- Revision 3", RFC 2026, BCP9, October 1996 [RFC2418] S. Bradner, "IETF Working Group Guidelines and Procedures", RFC 2418, BCP 25, September 1998 9 Informative References [RFC2727] J. Galvin, "IAB and IESG Selection, Confirmation, and Recall Process: Operation of the Nominating and Recall Committees", RFC 2727, BCP 10, February 2000 [IESG-CH] H. Alvestrand, "An IESG charter", draft-iesg-charter-03.txt, April 2003 [RFC2850] B. Carpenter, Ed. "Charter of the Internet Architecture Board (IAB)", RFC 2850, BCP 39, May 2000 [COREVAL] http://www.ietf.org/proceedings/02nov/slides/plenary-2/sld4.htm [DOCTRN] http://www.ietf.org/proceedings/01dec/slides/plenary-3/index.html 10 Acknowledgements The contents of this document were greatly influenced by members of the Problem Statement WG editorial team: Avri Doria, Dave Crocker, Elwyn Davies, Jeanette Hofmann, Melinda Shore, Rob Austein and Spencer Dawkins. In addition to the editorial team, the following people have provided useful feedback on earlier versions of this document: Harald Alvestrand, Randy Bush, Brian Carpenter, Leslie Daigle, James Kempf, John Klensin, John Loughney, Keith Moore. The initial text for the core values section is largely based on presentations and e-mail messages authored by Harald Alvestrand. Wasserman, Editor Expires December 2003 18 IETF Problem Resolution Process June 2003 "Good times hide problems; Bad times hide successes" is taken from a presentation by Tom St. Dennis, the former President and CEO of Wind River. 11 Editor's Contact Information Comments or questions regarding this document should be sent to: Margaret Wasserman Wind River 10 Tara Blvd., Suite 330 Phone: (603) 897-2067 Nashua, NH 03062 USA Email: mrw@windriver.com 12 Appendix A: Suggested Charter for the Improvement WG IETF Improvement Working Group (improve) Chair(s): TBD, as described above. Area Director(s): TBD, as described above. Mailing List: TBD DESCRIPTION: The IETF Improvement WG is chartered to make improvements to the management structure and processes of the IETF to address the fundamental organizational and process problems described in the IETF Problem Statement (RFC XXXX), according to the process described in the IETF Problem Resolution Process (RFC XXXX). The IETF Improvement WG will focus on two areas: - Improving the scalability and effectiveness of the IETF's organizational structure. - Improving the timeliness and utility of the IETF's standards track document processes. This WG will follow a two-phase process. Phase two tasks will not be started until the deliverables for Phase One have been completed by the WG and submitted for publication. Phase One: Understanding our Core Values and Mission In this phase, the WG will articulate and document the core values, mission, scope and goals of the IETF. We will also learn how to recognize and measure the success of the IETF, and generate performance baselines that can be used to assess the success of later changes. The deliverables for Phase One include: Wasserman, Editor Expires December 2003 19 IETF Problem Resolution Process June 2003 - A document describing the core values of the IETF that should not be compromised as a result of any reorganization or process changes. - A document describing the mission, scope and goals of the IETF. - A document describing how the IETF can recognize and measure our own success. - A set of performance baselines that characterize the recent performance of the IETF in achieving our mission and meeting and our goals. The output of Phase One will provide a framework that can be used to determine what improvements, if any, we should make to our organizational structure and standards track processes. It will also give us a means to understand and measure the success of any improvements that we do choose to implement. Phase Two: Organizing to Achieve our Mission and Goals without Compromising Our Core Values In this phase, the WG will document whatever improvements are needed to the IETF organization and processes to allow us to effectively achieve our mission and goals without compromising our core values. In this phase, the WG will: - Determine how the WG will identify, plan and execute any necessary improvements; - Solicit possible improvements from the community, and scour the problem-statement and solutions mailing lists for proposed improvements. - Evaluate, scope and prioritize a set of improvements designed to increase the effectiveness of the IETF's organizational structure and standards track processes; - Implement the improvements (most likely by publishing BCP RFCs); and - After a suitable time, reapply the metrics developed in Phase One to determine if the improvements have been successful. Goals and Milestones: TBD Wasserman, Editor Expires December 2003 20