AFRINIC-34 | Public Policy Meeting

1) Introduction

The AFRINIC-34 Public Policy Meeting (PPM) was held on 17 & 18 November 2021, 09h00-13h00 UTC in an online format. During these 2 days, the Policy Development Working Group(PDWG) discussed proposals pertaining to the management of Internet Number Resources and the Policy Development Process.

The minutes of the session have been documented in this document.

The video recording for the Day 1 PPMv(17 Nov 2021) is published at:

The video recording for the Day 2 PPMv(18 Nov 2021) is published at:

2) Welcome, Introduction & Agenda Overview

Delegates were welcomed to the meeting (by Vincent Ngundi, PDWG Chair). The agenda was presented and there were no modification requests.

The agenda is as follows:

9:10 - 9:15 -- Welcome, Introduction & Agenda Overview
9:15 - 9:35 -- The AFRINIC PDP & Building Consensus
9:35 - 9:50 -- Policy update from other RIRs
9:50 - 10:00 -- Policy Implementation Experience Report
10:00 - 10:10 -- Questions & Answers
10:10 - 10:20 -- TEA BREAK
10:20 -11:10 -- PDP Working Group (WG) Guidelines and Procedures Policy Proposal
11:10 - 12:00 -- IPv4 Inter-RIR Resource Transfers Policy Proposal Comprehensive Scope
12:00 -12:10 -- TEA BREAK
12:10 - 13:00 -- Public Information Policy Proposal
13:00 -- Closing Remarks for Day 1

3) The AFRINIC PDP & Building Consensus

Vincent Ngundi, PDWG Chair, shared the guidelines to participate in the AFRINIC PPM. He advised all participants to adhere to the AFRINIC code of Conduct by being respectful, professional at all times thus acting in the best interest of the community.

He also pointed out that harassment, intimidation, and offensive behaviours will not be tolerated. Vincent Ngundi then explained the AFRINIC Policy Development Process based on these key points:

Definition of the AFRINIC Internet Number Resource Policies
How the Policy Development Working Group (PDWG) is composed and coordinated
The roles of the PDP Co-Chairs
The Policy Development Process (PDP)
The Policy Development Process (PDP) Principles

Vincent Ngundi also went through the PDP simplified process. He also encouraged the community to refer to AFRINIC’s Consolidated Policy Manual (CPM) which is a central point of accessing all AFRINIC policies.

He added as well that the manual is updated once new policies are ratified and implemented.

Next, Vincent Ngundi presented on Building and determining Consensus which he mentioned is the key role of the Co-chairs of the PDWG.

Reference was made to AFRINIC’s Consolidated Policy Manual (CPM), specifically, Section 3.0 which highlights the objective of the PDP, and Section 3.4.2 of the CPM which stipulates that the Chairs determine whether consensus has been achieved during the Public Policy Meeting.

Vincent Ngundi elaborated on the three elements which help in building and determining consensus, namely:

RFC 7282 which provides a guide to building and determining consensus, developed for the IETF which is an environment similar to ours
Moderation of Policy Proposal Discussions - the role of Co-chairs are defined as:
1. to identity objections and contentious issues regarding the policy proposal
2. to track open issues that are yet to be addressed by the Author(s) and participants
Roles of the Co-chairs in Building Consensus
1. to direct the working group towards areas that are contentious
2. to encourage participants to focus and seek consensus on contentious areas
3. to ensure that concerns raised through the staff impact assessment are addressed

Also, Vincent Ngundi clarified that the objective of the Co-chairs is to always aim for rough consensus, if not consensus.

He also raised the following:

Rough consensus is not built/determined through a VOTING mechanism
Rather by ensuring that all objections/concerns are adequately addressed
Look/seek consensus throughout the process (for each contentious issue)
No VOTING mechanism was applied at any point in time (to avoid “vote stuffing”)
100 people for and 5 people against might not be “rough consensus”. If a minority of participants have a valid objection, that objection must be dealt with before rough consensus can be declared
5 people for and 100 people against might still be a rough consensus, as long as there are no valid objections that have not been addressed.

4) Policy Update from other RIRs

Madhvi Gokool from the AFRINIC Policy Liaison Team provided an update on policies from the other RIRs. She took us through the agenda which was as follows:

Proposals that have been discussed at other RIRs
RIR comparative policy matrix

Madhvi explained that the APNIC EC endorsed the four proposals that reached consensus during APNIC 52, via an electronic ballot and asked the APNIC Secretariat to implement the proposals.

At RIPE NCC, there are currently no active policy proposals. A revised version of the RIPE PDP is under discussion at the moment.

LACNIC has a total of 7 policy proposals under discussion. No consensus has been reached on these proposals during the LACNIC meeting held in October 2021.

ARIN had 6 policies under discussion in their public policy meeting in October. The below-listed 3 proposals currently have as status draft policy:

Clarifications to Sections 8.3, 8.4, and 8.5.6
Update ISP and End User References for 2022 Fee Schedule
Remove Circuit Requirements

The RIR Comparative Policy Overview provides a comparative overview of implemented policies across the Regional Internet Registry (RIR) system and is updated on a quarterly basis. It is not a policy statement by the RIRs but serves as a reference for the global Internet community.

5) Policy Implementation Experience Report

Presentation URL

Neelam Matoo Bhorut, AFRINIC Staff, presented the Policy Implementation Experience Report (PIER).

She mentioned that the purpose of the PIER is to provide feedback to members and the community regarding recently implemented policies and experiences faced by hostmasters while handling requests governed by currently implemented policies.

She stated that the Lame Delegations in AFRINIC Reverse DNS - AFPUB-2017-DNS-001-DRAFT-02- policy has been fully implemented this year.

The DNS lameness test runs on a daily basis. When an nserver record is found to be lame on consecutive days, the registered contacts receive 4 notifications. On day 11, a remark is added on the domain object; the nserver is still found to be lame and the lame nameservers are removed after 30 days.

She also presented the statistics about the number of lame delegations. A considerable drop has been observed.

The experiences and challenges faced by Hostmasters were as follows:

1. Section 6.8.2 from the CPM - IPv6 prefixes which are not announced after 12 months of issuance are considered non-compliant and should be reclaimed after 3 notifications. But she pointed out that this reclamation exercise has not yet started and there is an automated mechanism in the development phase.

2. Section 12 from the CPM - Anycast Assignments; The current CPM text is not consistent with the final version 4 of the policy that was ratified. It refers to version 2 of the proposal, thus missing references such as BCP126, BCP169, or their respective successors. The implementation was not aligned with the final interpretation by PDWG.

She as well pointed out the below actions to be taken by AFRINIC:

Update CPM with the appropriate text
Update list of pre-CPM policies to reflect version
Publish interpretation of Assigned Anycast policy on AFRINIC website
Implement a mechanism to evaluate multiple requests for anycast purposes to check against abuse

6) Questions and Answers

Highlights of the Q&A session are as follows:

JORDI PALET MARTINEZ from the IPv6 company had one point regarding section 6.8.2 from the CPM on the IPv6 assignment. For him, AFRINIC should reclaim IPv6 resources that have not been announced; as mentioned in the CPM and RSA. AFRINIC has provided help to make it possible so for him it's time to start reclaiming unannounced IPv6 blocks.
Regarding the Anycast assignments, JORDI PALET MARTINEZ suggested that staff explain much better, maybe on the mailing list? whether the text we have in the CPM is not consistent with the policy that was adopted or what?
For the first question related to IPv6, JAMES CHIRWA, Head of Member Service at AFRINIC, agreed that resources should be reclaimed and because there is a huge volume their plan was to have an automated process because simply the procedure is already given in the policy itself. They have expected it to go live with the update of MyAFRINIC version 2 which has slightly been delayed. He also said that the volume that they have could have added a lot of pressure and this is why they went for the automated approach.
On the Anycast part, James Chirwa from AFRINIC stated that the CPM text is not the text that was in version 4 of the policy that went through ratification. This is one thing that they have captured in due course and this is why they are bringing it forward to say that we must rectify the CPM to have the right text and also on the implementation because the text that was implemented was on the wrong version 2. JAMES added that there are some important references that are missing and that is what they are trying to clarify.
Vincent NgundI thanked James and Jordi, and asked whether Jordi’s concerns have been addressed or if they were willing to take them on the mailing list?
JORDI PALET MARTINEZ said that he will be happy to continue the discussion on the mailing list but wanted to add something else; In the case of IPv6, there may be a valid justification for not to announce it. This should be considered if that is the justification from the resource holder because the difference between IPv6 and IPv4 is that there are no private addresses. In some situations, they may be an allocation, difficult to believe, but at least an assignment that is made for an internal method. That is a perfectly valid reason
JAMES CHIRWA responded to Jordi by saying that the policy is clear on that, as a provision was provided for members to check whether the usage will be internal or external so we are aware of it is announced.
ELVIS wanted to ask a question to the co-chairs on their presentation about determining consensus. How do you basically determine what is a valid and invalid objection? Because some objections might seem valid to some people but to others it is invalid so how do you determine that in order to be able to come to a consensus?
Vincent Ngundi responded to ELVIS; the whole principle behind determining if something is objective or not is the technical backing that it has. Let me tell you an example of what is not a valid objection. You say you do not support this policy proposal and you leave it like that, you must explain why you do not support it. That is the best example I can give you. Unless you have a follow-up on that, we can give you an opportunity for a follow-up. The floor is yours. Go ahead, Elvis. ELVIS agreed that Vincent Ngundi had addressed its concerns about determining consensus
Prof Nii from Ghana.com- AFRINIC Member, was wondering if the working group wants to handle an overload of proposals. Looking closely, one author has four proposals. Maybe the working group or the co-chairs might advise the author to not flood us so down, so we can concentrate on solving really hard problems that are in the interest of AFRINIC
Responding to Prof Nii, Vincent Ngundi said that Prof Nii is facing some of the issues they face as co-chairs. As a follow-up to the last public policy meeting, co-chairs were asked to seek ways of engaging the PDWG. One of the reasons they held a webinar was because they realised that there might be too many proposals that may not necessarily be the case because they realised that some of the proposals were actually conflicting. Which might put us in a very difficult situation. The purpose of the webinar was to see whether authors can find it within themselves to come together and co-author proposals to avoid a situation where we have conflicting proposals.
Vincent Ngundi continued by saying that we need to think about how we go about developing policy proposals, and finding ways of ensuring that we avoid conflict as much as possible. Obviously, if we cannot agree then PDP can allow anybody to submit their one policy proposal. We have to ensure that we have the end in mind so that we don't find ourselves in such a predicament.
Prof Nii commented on Vincent Ngundi's response by saying that he can appreciate the first part, but it is again, part of his question was that he thinks authors should also understand the situation we are in, and not try to please us with volumes of proposals. Prof Nii suggested that what you need is a few good proposals and time to spend on those proposals.
Alain Aina from Digital Intelligence Services explained that this was an old problem we had with the PDP that we were trying to solve long ago by proposing what you call PDP bis. We were told not to fix what was not broken. The PDP was not broken, everything was fine, so maybe it is time for us to reconsider fixing the PDP so that we can address some of these issues. We don't have clear systems in place to evaluate and accept proposals. The PDP we have now is old and has not had an update since 2011 when the system was built.
Alain continued by saying that, Jordi said when he was responding to James that we do not have a private IP with IPv6. We do have a ULA, but I Understand the point he was making. Do not just recall the numbers, make sure that some of them will not be seen in the public routing table, but I do think we have ULA which is almost the same.
In response to Alain Aina, Vincent Ngundi said that what we need to do to avoid this scenario, as co-chairs we can prepare policy proposals, is for the PDWG to find a way of incorporating the issue of interest groups into the PDP so that we avoid such scenarios. We really need to find a way of institutionalising it. We need to make sure that we do not have conflicting and unnecessary policy proposals being presented.
In response to Alain’s comment, JORDI PALET MARTINEZ said, If we tell the people that ULA is the same as private addresses, we are doing it very wrong. ULAs are restricted to /48 and a network can be bigger than that. ULAs are meant for not doing the translation. There is not something like NAT in IPV6. There is network prefix translation, which is in experimental protocol, that cannot be used in production networks. I cannot agree with that, it is a technical question, but there is a lot of differences between private addresses in IPv4 and ULAs in IPv6. There is no valid justification in most of the cases, maybe 99.9% for having a public block for IP addresses inside of a network, and not announcing it. But in the case of IPv6, it is very clear.
PAUL HJUL from Crystal Web (Pty) Ltd wanted to comment on the assertion made a little bit earlier that maybe the PDP needs to be thrown out. He thinks the PDP at the moment might be in a little bit of difficulty and so on, but certainly, the policy manual and so on are in a much better state than many other things. He doesn't think we should adopt the attitude that the system is broken, let's throw everything out and allow people to come up with whatever they are wanting.
If anything, PAUL thinks we have a far stronger foundation for the consolidated policy manual and anything else that is in AFRINIC. He doesn't think we should let the PDP be thought that there is an excuse to rewrite policies upended and not from a community top-up approach.
Vincent Ngundi understanding of policy should be there to help the community and not the other way around. The whole idea is to try and improve it. Not that it's entirely broken but just to improve it based on the experience we have had especially in the past two years. That is where we are coming from. It is upon us to ensure that we develop policies that work for us as a community.
Darwin Da Costa agreed with Vincent Ngundi and added that we are improving day by day, month by month as we move on with constructive feedback from the community. We always take this into consideration together with colleagues. This is a process, mentioning it is broken is not the right word to use but we are making some positive steps and I would like to thank the comment from Professor Nii, whereby we need to focus clearly on proposals which will at the end solve problems within the African internet continent. That was something that I wanted to add to that.

Vincent Ngundi:

Session closed. Break time.

After the Q&A sessions, PDWG Chair Vincent Ngundi presented the Draft Policy Proposals that will be discussed during the AFRINIC-34 PPM, namely:

Proposal #1: PDP Working Group (WG) Guidelines and Procedures Policy Proposal
Proposal #2: IPv4 Inter-RIR Resource Transfers Policy Proposal Comprehensive Scope
Proposal #3: Public Information Policy Proposal

He further gave an overview of how the discussion will be carried out for each proposal and encouraged the participants to adhere to the timeframe presented.

7) Proposal #1: PDP Working Group (WG) Guidelines and Procedures

ID: AFPUB-2020-GEN-002-DRAFT04
Proposal URL - https://afrinic.net/policy/proposals/2020-gen-002-d4
Presentation URL - https://afrinic.net/ast/pdf/policy/6-17Nov-AF34-PDP-WG-Guideline.pdf

7.1) PDWG Chair introduction of the DPP and Discussion Flow

Darwin Da Costa announced the proposal which will be discussed and gave an overview of the flow of discussion.

8 minutes by the authors
2 minutes by the policy liaison team
5 minutes for the contentious area
12 minutes for the q/a whereby the open mic will be contained as well
And last we will announce our final decision

Darwin Da Costa mentioned that this policy proposal was authored by Alain Aina and Maina Noah and asked them to present their policy.

7.2) Author’s presentation of DPP

Maina Noah, after introducing himself, passed the flour to the co-author Alain Aina to do the same. Maina Noah mentioned that the policy is on its version 4 and started to present their policy by highlighting how it improves the current PDP as shown:

The proposal addresses the problems by defining clear and explicit working group guidelines and procedures.
The proposal amends the policy development process section 3.3 of the AFRINIC CPM and 3.5

Noah also said that they are only addressing rules and guidelines which are attached to the PDP so their objective is not to change the PDP.

The summary of the proposal is to define the working group guidelines and procedures with these key points:

To serve as a guideline on how the PDWG shall operate.
Defines clear roles and responsibilities for the PDWG co-chairs.
Defines clear procedures for the working group administration.
Defines the appointment process of co-chairs
- Consensus-based appointment
- Secret Ballot (College)
- Interim Appointment
Recall and/or resignation of co-chairs
Individual Behaviour of members of the working group

The author presented some changes from version 3 to version 4 on the following :

We clarify paragraphs of section 3.3 and amended
The proposal obsolete section 3.5.3 of the CPM
The CEO (instead of board chair) shall lead WG consensus in absence of co-chair (sections 3.3.3, 3.3.4 and 3.3.5)
Direct that code of conduct appeals be handled by the CEO (instead of board chair) (section 3.3.10.1)
Update the conditions for Co-chair recall ( section 3.3.4)
Added provisions for the composition of the recall committee (section 3.3.4)
Clarified the basis for CEO decisions on appeals against WG posting privileges suspension (section 3.3.10.1)

The authors further explained on the Appointment of PDWG Co-chairs that the working group appoints co-chairs for a two-year term and makes decisions by consensus and shall also appoint co-chairs by consensus. The working group makes its decisions by consensus as defined in the PDP and finally, elections should be used as a last resort solution through ranked-choice (IRV) election.

The co-chair's appointment should be done as follow:

We, therefore, suggest the appointment of co-chair be done by;
- Consensus led by a co-chair (Preferred)
CEO leads WG absent co-chairs
Voting is done by the College of Past;
- Past Co-Chair (who completed a term without being recalled)
- Past Board Directors
- Past CEOs
Secret ballot by means of ranked voting

Also, conditions to recall a co-chair has been modified in their proposal as shown:

Instead of support from 5 PDWG members as is current, we will now require 10 PDWG members from different organizations who are active in the PDP by means of their participation on the RPD list
The Board shall form a recall committee comprised of:
- One CoE member
- One GC Member
- One Community Member

Before the author ends he mentioned that this proposal clarifies how to deal with Individuals Behaviors on these points:

In case individual behaviour is disruptive to working group engagement.
Co-chairs shall caution such behaviour openly once or twice.
Co-chairs caution shall openly expose what is being warned against and persistent behaviour shall invoke moderation/ suspension.
Co-chairs decisions on any moderation/suspension of posting privileges will be subject to appeal to the CEO

7.3) Staff Impact Assessment

Presentation URL - https://afrinic.net/policy/proposals/2020-gen-002-d4#impact

Brice ABBA from the AFRINIC secretariat presented the impact assessment of the proposal. He stated that the following points are derived from the staff’s interpretation and understanding of version 4 of the proposal.

This policy proposal brings in some changes to the functioning of the Policy Development Working Group (PDWG) as follows:-

PDWG Co-chairs appointment and responsibilities
Clarifies what needs to happen if one or both co-chairs are recalled
Clarifies how to proceed should a co-chair resign
Mentions that if the working group finds itself without a co-chair, AFRINIC CEO will lead the consensus process.
Describes the operations of the PDWG such as the moderation of the PDWG discussions and sessions, individual behaviours in public policy meetings
Appeals - Someone whose posting privileges have been suspended can appeal against these decisions
The Board appoints interim co-chairs within a prescribed timeline
Clarifies on those eligible to be in the voting register, should the last resort online secret ballot be used to select the PDWG Chair(s).
In the case of recall of PDWG Chair(s), the proposal mentions the recall needs to be supported by at least 10 other persons from 10 different organisations and that these persons must have been subscribed to the working group mailing list for at least one(01) year and attended at least one (1) AFRINIC Public Policy
Meeting during the last two (2) years, either in-person or remotely.
The proposal also mentions the criteria that the AFRINIC Board shall use to appoint the members of the Recall Committee, as well as guidelines for the consideration of a recall and a 4-week timeline for the recall committee to have done its work.
Active participation of candidates for the PDWG Chairs position will be determined by the PDWG when assessing the candidate.

Impact on the meeting Platforms

Restriction of participation to be also implemented via the AFRINIC meeting platform for online/hybrid events.

Online voting Platform

An online voting platform will be required to host the online secret ballot.

The voting register is defined in the proposal and consists of past PDP WG co-chairs, past board of directors chairs, and past CEOs who completed at least one term and have not been recalled. The list of names is available and accessible to the online voting platform will be subject to identity verification

Impact on mailing Lists

PDWG Chairs will be subscribed to members-discuss mailing lists as observers.

Financial Assessment

Elections or Voting Platform is outsourced and therefore the budget needs to be planned accordingly.

Legal Assessment

The first paragraph of the proposed policy reads as follows – “The Policy Development working group (PDWG) provides an open public forum to discuss Internet number resources policies and related topics of interest to AFRINIC and the Internet community in the AFRINIC service region”.
At the outset, it should be recalled that the PDP is an emanation of section 11.3 of the AFRINIC bylaws which provides that:
“For the purpose of subsection 11.2 a Public Policy Meeting means a meeting open to the community wherein proposals for policies for a proper and responsible usage and Management of Internet number resources are discussed and agreed within the framework of the Policy Development Process (PDP) defined by the Regional Internet community and ratified by the Board.”
As such, the policy as styled is vague in as much as the mandate of the PDWG is limited to propositions and discussions of policies pertaining to IP number resources management. Any other discussion, albeit related to AFRINIC, is thus inadmissible.
Under paragraph 3.3.3 of the proposed policy (5th paragraph), reference is made to – “Any natural person residing in a country from the AFRINIC service region is allowed to volunteer”.
It is relevant to recall that section 11.3 of the bylaws provides as follows:
“For the purpose of subsection 11.2 a Public Policy Meeting means a meeting open to the community wherein proposals for policies for a proper and responsible usage and Management of Internet number resources are discussed and agreed within the framework of the Policy Development Process (PDP) defined by the Regional Internet community and ratified by the Board.”
However, it is also acknowledged that notwithstanding the provisions of section 11.3 of the bylaws, i.e. for AFRINIC to have a Regional Internet community, it is an acceptable practice at AFRINIC to allow persons not necessarily residing with the AFRINIC service region to subscribe and participate in its PDWG. Hence, refraining persons not residing in the AFRINIC service region from being appointed as PDWG’s co-chairs would be unfair in these circumstances.
If the intention of the authors is that the PDWG should be owned and controlled by persons residing within the AFRINIC service region, then the mischief should be addressed at the source, thus giving full effect to the provision of section 11.3 of the bylaws so that only person residing within the AFRINIC service region be entitled to subscribe and participate in the PDWG and all others may appear as observers only.
Under paragraph 3.3.3 of the proposed policy (10th paragraph), reference is made to “If no consensus can be reached and more than one candidate is being evaluated, then an online secret ballot to appoint the new co-chair will be held within two weeks after the PPM. The secret ballot shall be opened to past PDPWG co-chairs, past board of directors chairs, and past CEOs who completed at least one term and have not been recalled”
There is no legal rationale that for the purpose of finding consensus, it is the working group that decides but in case of an election, the decision-makers are persons excluding the same working group. There is no logic in having 2 separate pools of decision-makers on the same issue of selecting the PDWG’s co-chairs. The authors are recommended to review this aspect of their proposition.
Under 3.3.10 of the proposed policy, reference is made for an appeal to be heard by the Chief Executive Officer. It is added that section 3.5 of the CPM already provides for an appeal against the decision or action committed by the co-chairs. Hence, the current proposal has the effect of making the CEO sits as an alternate appeal venue to the existing Appeal Committee. It is simply duplicity to the existing structure. Besides, even if there was any legal soundness to that proposal (at least for the sake of argument) but as previously raised in the case of the board chair, it should not be forgotten that the CEO is a member of the board so that imposing any additional responsibility to that function can only be done in consultation with the board of directors.

The above are the fundamental issues that the authors and PDWG need to discuss and resolve.

7.4) PDWG Chairs Summary

The PDWG chairs noted the following concerns and their status (Addressed or still pending)

Pending concern. A clause in section 3.3.3 as stated below states that a candidate for the co-chair position must have attended At Least one meeting in person over three years to qualify for this position.
In the era of Covid, it is possible that in-person meetings might not come back or might take a few more years. If this policy gets ratified as it is. It disqualifies everybody from contesting for this position.
Pending concern. In section 3.3.3, there is a voting process that doesn’t involve the entire WG stipulated. I think it is awkward at least that a voting process is proposed but the entire community is not allowed to exercise their franchise as community members.
Pending concern. The proposition that the CEO serves as the arbiter of acceptable speech is untenable, moreover, the entire approach of the section is badly framed and crafted:
It speaks of an "appeal" to the CEO and then speaks of "complainers". This is badly ambiguous and unworkable. If a person is alleged to violate the code of conduct somebody complains (who logically can be referred to as the complainer, but should be called a complainant) to the co-chairs. The co-chairs then consider the complaint and make a decision (which decision is taken without affording audition to the person who against whom the complaint is made) which if adverse against a person results in posting rights suspension. That person (who could be called an appellant) can appeal to the CEO but is he now complaining (a complainer) about the co-chairs. What of the situation where the co-chairs decline to act? Do the original complainers have a right to go to the CEO? This problem is all the more severe if the CEO is engaged in discussions on the group and is the complainant alleging conduct violations
Pending concern. In 3.3.3, since the co-chair selection process is to take place in a meeting, In the event that one nomination is received, with the CEO leading the consensus process to determine the next co-chair and there is a lack of consensus, there is no stated guideline as to who chairs that current ongoing PPM.
Pending concern. In regard to section 3.3.8, Community members invest time and money to attend meetings virtually or physically. I do not think it is best to have a clause that can possibly adjourn a meeting that travels, permission from work, preparation for an event must have taken place by community members.

7.5) Open Microphone Discussion on the Proposal

Elvis from Nigeria said that this policy proposes a restriction on participation to AFRINIC members. He wants to point out that this policy is restrictive. what happens to the universality that is in section 11.3 of the bylaws?
Cheken Chetty from a university in South Africa. is strongly opposed to this policy as it gives more power to AFRINIC’s CEO. For Cheken the community should have more power rather than the executive.
Lamiaa C. from Morocco is opposed to this policy because it does not consider their feedback. it gives more authority to the CEO. It's a way to limit community participation to the mailing list.
Darwin Da Costa would like to understand the 'mailing list limitation' Lamiaa’s were expressing. She said she will develop in the chat.
Jordi from the IPv6 company listed the following points on the proposal just presented: (also in the mailing list https://lists.afrinic.net/pipermail/rpd/2021/014024.html )
1. 3.3 the PDP is not just "guidelines" it is strict rules, it can't be subjected to interpretations and then we come into trouble. CPM has rules.
2. 3.3.1 is very wrong to assign only one co-chair to each policy. Both of them shall be co-leading it, so in case one of them fail, the other one can follow without further delay 3.3.3 there is no determination of who can participate in consensus to elect chairs or vote if that's the case.
3. 3.3.6. rejecting inputs is discriminatory. If someone was too busy to participate in part of the discussion he/she has the right to come in at any time during the discussion. May have inputs that have not been considered before and in any case, he has the right to express an opinion. That section must be completely deleted.
4. 3.3.8. Is bad to have a mention to specific sections of the bylaws, they can change (they should, in fact, they must not encroach the community and PDP).
5. 3.3.9 Lack of consensus definition. Many folks don't read RFCs - there is no clear view of if the consensus is determined also in the mailing list discussion. Lack of last-call definition Lack of clear timings for all the processes.
Prof Nii from ghana.com question is about how we are reading the bylaws in bits and creating a situation where it appears we don't share the same vision of the PDP is being bombed. Section 17 of the same bylaws says living and operating in the African region that is how it defines community. If you are not living or operating in the region by this bylaw really you are not in focus. We should be honest. The openness we are discussing has more to do, not everybody can come and participate in the PDP which itself is defined in the bylaws. The bylaws should be superior to the PDP. I am wondering whether we are missing that?
Prof Nii continued by saying that his concern is that that kind of open community list email address is only meant to be used for consensus. We are supposed to work by consensus…
Paul form, crystal web, wanted to know whether the objections and communication that has been received on the mailing list in the last days are being viewed as objections as it currently stands and in light of the plethora of objections that have been shown in the session. For him, we are wasting our time with a policy that will not be able to find consensus and that this discussion is becoming a little bit futile.

The authors were given a chance to respond to the questions:

In response to Paul, Alain Aina said that we are not necessarily here to get consensus to approve this; this is a policy and this meeting is to discuss policy. This is not to just oppose but to also propose an amendment of how it is coming. We focus on: I support or I oppose. This community has to work together. If you are not used to this community you get scared when you read all the comments. It is not our role to respond to all, it is not authors versus the community.
The author stated that what Jordi said is not correct to only assign one co-chair. If you read the proposal, it says for each policy proposal one co-chair must be assigned as a primary contact. Both co-chairs, however, shall actively participate in consensus-building. This is an example it is just a matter of how we read things. We should read things positively and try to build and not try to kill or to destroy.
There's also another comment on adjourning the meeting. If you read the text again, we said that if we are in a situation where both co-chairs cannot attend, the working group shall appoint a temporary chair to lead. However, If the working group cannot appoint a temporary chair, the board of directors shall appoint a temporary chair. If the BoD fails, then the meeting has to be adjourned. You can't have PPM without co-chairs, this is how the process is. You need a co-chair to determine consensus
The author wanted to comment on the legal Impact, for him, some of them are personal opinions. The proposal states that to be co-chair you must reside in the region, this is the closest definition we have to an Internet community in Africa. Legal said it is unfair they haven't said you have violated a rule. Finally, this committee has the final decision and the legal team is trying to deny the working group to discuss some topics on this working group. We said this working group is an open forum to discuss Internet resource policy and related topics of interest. As a working group, sometimes we go the policy route but we are also allowed to discuss anything related to policy, the policy is anything related to IP management.
NOAH MAINA thanks Alain Aina for his comments to the questions asked and wanted to thank, most importantly, all of the participants for their countless comments and obviously these are issues that have been shared. Until today we had received only comment in terms of the present version, and I attempted to respond to them last night. But we have seen that there are so many others, which I believe some would be related to misinterpretation. I think what we can do as authors is to engage with the working group, to see how we can resolve some of them
The author said that they received comments from Laniaa, Paul, Hytham, Elvis and Jordi and that they have taken note of all of them. He will summarise them with his co-author and they will see how best they can improve. If there's anything to do with misinterpretation they will try their best to see if they can fix those.
For the co-author, we are not here to seek consensus but to see how we can forge forward and basically fix that part of the PDP process which we as authors believe there is some fixing that needs to be done. And that is the intention of this proposal, to be honest. Just to thank everybody for the comments, believe me, we will respond. Thank you so much.

7.6) PDWG Chairs decision

After deliberation, the PDWG Co-chairs made the following decision:

“Having considered the discussions in the RPD mailing list and the number of concerns, the authors have not addressed the concerns raised by the PDWG and the authors acknowledged that and were actually seeking feedback. The Co-chairs have determined that rough consensus has not been reached.
The draft policy proposal, therefore, goes back to the mailing list and authors are encouraged to address all the concerns”

8) Proposal #2: IPv4 Inter-RIR Resource Transfers Policy Proposal Comprehensive Scope

ID - AFPUB-2019-IPv4-002-DRAFT07 Proposal URL

Presentation URL:

8.1) PDWG Chair Introduction of the DPP and Discussion Flow

The PDWG Co-chair mentioned that this proposal was authored by Jordi Palet Martínez and invited the latter to present the policy.

8.2) Author’s presentation of DPP

Jordi Palet Martinez briefed the community on the problem statement on the following points:

AFRINIC is lagging behind in the IPv4 market and this is negative to the region which is leading the region to a situation of discrimination and scarcity of addresses, not only in the RIR itself but in the region's market.
New businesses can’t be established in the region, due to the lack of addresses.
Additionally, it is important to highlight that the deployment of IPv6, in some cases, may require small blocks of IPv4 addresses for transition mechanisms, or significantly increase the costs thereof, and many AFRINIC entities could, therefore, be at a serious disadvantage if they do not have access to a global market, as it is currently the case.
Legacy holders in the AFRINIC region can’t transfer their resources out of the region and remain dormant.
Under the table agreements for transfers lead to the loss of the history of the registration.
Even if a member qualifies for the transfer of its resources, if the transfer fails due to the recipient not meeting stipulated policy conditions, the member can be penalised for stockpiling and the unused resources will be subjected to return to or recovery by AFRINIC.

The author explained how his proposal addresses the problem statement on the following:

The proposal aims at maintaining the current intra-RIR transfers.
Enables bidirectional, compatible and reciprocal transfers with all the other RIRs.
Facilitate a dynamic in the market and by increasing the offer and making it transparent, reduce prices.
Enables both legacy holders and resource members in the AFRINIC region to transfer resources.
At the same time, ensures that resources being transferred from an AFRINIC resource member have been used according to the documented justified needs as per the RSA/CPM conditions, and not stock-piled. There is also a grace period for justification of the usage of the resources, in case of a failed transfer due to the recipient failure. Not including that will actually disallow transfers, as it is obvious that a member willing to transfer is no longer justifying the resources and will be subjected to returning them or a recovery

For the author, reciprocity is the key, otherwise, the proposal doesn’t make sense.

Jordi informed the community that he is aware of IP resources that were transferred outside the AFRINIC region, but these Proposals make inter RIR transfers formalised and known instead of under the table as it is right now

The author further read through the proposed texts; Due diligence text has been added after Staff Assessment was provided. This Policy does not cater for Mergers and Acquisitions since this is already covered by existing text, another policy need to be proposed to cover and cater for Mergers and Acquisitions

8.3) Staff Impact Assessment

https://afrinic.net/policy/proposals/2019-ipv4-002-d7#impact

Madhvi Gokool from the AFRINIC secretariat presented the impact assessment of the proposal.

8.3.1 AFRINIC Staff Comments on clarity of policy

This policy states that after a successful transfer pre-check AFRINIC can’t perform any reviews and audits on this member for a period of 12 months. The author is requested to clarify the reasoning behind 'validity of pre-checks is 12 months
In case of pre-check failure, can the author clarify if AFRINIC can take the resources back as the member is not compliant.

8.3.2 Impact on Systems

1. The transfer tool on MyAFRINIC and NMRP will require further adjustments to accommodate inter-RIR transfers
2. Introduce an automated tool to monitor the direction of the resources in order to easily manage 5.7.6
3. RPKI ROAs
4. Reverse DNS (majority /8s)
5. Internal Ticketing system
6. Change transfer business rules to add the possibility of inter-RIR transfers.
7. Add cross-RIR verifications.
8. Update the status of Legacy Resources after a transfer in WHOIS and MyAFRINIC.
9. Keep an audit trail including pre-check results.

8.3.3 Impact on Processes and procedure

Complete process and procedural reviews will be undertaken including the handling of inter-rir transfer of IPv4 resources

8.3.4 Impact on MS Operations

Resource Transfer evaluations are resource-intensive and registry functions will ensure to automate as much as possible, however, the evaluation task would still require additional staffing needs to facilitate speedy and timely delivery.

8.3.5 Contractual agreements

Revision of transfer Agreement & Registration Services Agreement may be required

8.3.6 Impact on Registry Functions

Implementation on the requested dashboard
1. Changes in preconditions and checks
2. New workflow for Inter RIR transfer
Coordination with other RIRs to enable incoming and outgoing transfers
1. How the request shall be received and treated.
2. How AFRINIC shall send the request to other RIRs for proper processing
Handling Legacy and resulting statuses

8.3.7 Impact on Finance

The financial impact is forecasted to be high and negative as:-

Resources from the AFRINIC Pool can be transferred in outgoing transfers to other RIRs. AFRINIC will lose members to other RIRs in outgoing transfers.
AFRINIC will gain revenue in incoming transfers from the other RIRs(including those with legacy status as the latter lose legacy status after being transferred)

8.3.8 Legal Assessment

Coming back to the present proposed policy, the author aims at establishing the mechanism to allow transfers of IPv4 resources to/from other regions and to align AFRINIC with a market that purportedly already exists and in which, according to the author, AFRINIC is lagging.
The decision of allowing, or not, an inter-RIR transfers of IPv4 resources from and to the AFRINIC region is not strictly a legal one. In fact, it is purely and simply a business decision to be taken judiciously and prudently both by the PDWG and the Board of Directors having regard to the directors’ duties provided in the Companies Act, i.e. to act in the best interests of the company. Acting in the best interests of the company in this context means considering the real financial impact of such policy for AFRINIC so that the sustainability and business continuity of AFRINIC, both as a company and RIR, is not compromised.
Further, it is observed that the scope of the proposed policy is not limited to non-legacy IPv4 resources, but also extends to legacy resources. Therefore, it is important to highlight that, as a matter of law, legacy resource holders existing within the AFRINIC’s service region are not contractually bound by AFRINIC’s adopted policies such that these policies have no direct effect on legacy resource holders, and it is up to those legacy-holders to adhere to AFRINIC’s policies. Thus, the author must bear in mind that obligations impacting on legacy resource holders may not necessarily achieve the intended results if the legacy resource holders refuse to opt for voluntary registration with AFRINIC.
The other question arising relates to outbound transfers of resources. It is understood that the intended transfers will be channelled through AFRINIC. Therefore, other than simply setting out the conditions for transfers, AFRINIC’s role in the whole process must also be adequately defined. In this respect, it is unclear as to whether AFRINIC’s role in the process would be limited to facilitating the administrative aspect of the intended transfers only with or without such legal responsibilities attached thereto, more so that AFRINIC will be relying on representation made to it when attending to similar requests. To address this issue, it is proposed that the burden of conducting such adequate due diligence be placed on both the source holder and the intended recipient, and that AFRINIC’s role should be limited to acting as a facilitator only without bearing any legal responsibility whatsoever in that process.
Moreover, while it is observed that legacy resources will lose their status upon being registered with AFRINIC (viz inbound transfers), it is not clear as to whether the receiving party will be required to sign an RSA with AFRINIC. Although one may presume that this is the intent of the author, yet it is imperative that same be clarified as well as whether AFRINIC will still be able to execute its RSA with the obvious risk of the concerned IP number resources being reclaimed by AFRINIC in case of a subsequent breach of the RSA, despite that the recipient organisation would have most probably paid good consideration (financial value) for such transfers.

8.4) Author response to Staff assessment

Basically, there are several points that I'm responding to the impact analysis because I believe that the staff is copied from the previous version so it is outdated. I believe I have resolved the issues. The first point is that the impact analysis says that:

This policy states that after a successful transfer pre-check AFRINIC cannot perform any reviews and audits during 12 months. My text is clear about that. In order to verify that the resources being transferred were allocated, assigned and used in accordance with the requirements in section 5.7.2.1, we need to define a time that ensures that if the transfer fails the resources can find another recipient.
In case of pre-check failure, can the author clarify if AFRINIC can take the resources? AFRINIC did not initiate a recovery, reclamation process within this period. Outside of that. You can and you must.
Legal assessment seems to be copied from the previous version so it is outdated.
The financial impact is forecasted to be high and negative. This is incorrect because as I just presented there is a positive impact and a negative possible impact. Real data from all of the registries shows that most of the transfers come from ARIN so the impact will be more positive than negative.
And then regarding the reciprocity, I explained before, I haven't changed anything. Since version 1. That can modify the reciprocity status.

The co-chair gave the chance to the AFRINIC secretariat to comment on the author's response to the staff Impact Assessment.

Madhvi Gokool from the AFRINIC secretariat explained that based on the feedback that they have received from all four RIRs, the policy is reciprocal. The impact assessment on the website will be updated shortly to include the final response that we have received.

8.5) PDWG Chairs Summary

The PDWG chairs noted the following concerns and their status (Addressed or still pending):

1. Pending concern. Suggest the period of time to be modified in sections 5.7.2.2 and 5.7.2.3 from 16 months to 18 or 24 months
I do not favour coming to a consensus on a policy with a silly clock with a plan to amend the clock later. I think we should come to a consensus on a complete policy including timing. I propose 24 months, but I would find 12 or 18 months acceptable. If the community can’t come to a consensus on the timing, then you don’t have a consensus on the policy since the timing is part of the policy proposal.

2. Pending concern. … I object to the policy so long as it contains the provision in section 5.7.5, unless it is modified as follow:

Section 5.7.5 to be:

AFRINIC may deny a registrant as the source of a transfer only if one or more of the following applies:

Clear evidence that the resources were fraudulently obtained.
Registrants are not current on their AFRINIC fees.
There is some question or dispute as to whether the registrant is the resource holder of record.
There is an unresolved third-party claim to the registration.

If one of the above does not apply, then AFRINIC must approve the registrant as a source of the transfer and proceed unless there is a problem with the recipient.

3. Pending concern. ...If the African market is already in shortage, then why is there still a free pool? If there is a shortage, then the bad policy is protecting the free pool from being issued to legitimate needs.

4. Pending concern. ... Honestly this seems like a lot of steps just to transfer resources from one RIR to another. Since the region currently does not have one, you'd think this would solve problems but instead, this allows for greater AFRINIC interference in a supposedly interference-free transfer. This is still not addressed. Additionally, the staff themselves has pointed out glaring vague and open-ended definitions that should also be addressed. I would definitely support a policy that has less "interference" with RIR than the current one.

8.6) Open Microphone Discussion on the Proposal

Alain Patrick Aina -
- If AFRINIC is not the owner, why are LIRs being called owners? Resources belong to everyone, not RIRs or LIRs
- On the financial statement, see the LACNIC and other region statistics
- Mentioned that the author had a safeguard in previous editions
- I will share more details on the mailing list of why he considers the policy Bogus
Arnaud A. A AMELINA
- Most questions were taken up by other intervenants
- We have a small portion of the limited resources compared to other regions
- We cannot accept such a policy since the resources are already limited
- I would Agree to bring more resources in the region via Legacy resources transfers into the region
- I reject such policies to make resources leave the region
Professor Nii Quaynor
- Has same concerns as Gregoire with the problem statement
- Are we not rushing instead of taking this in a responsible manner
- Serious concerns that the policy mentions that “AFRINIC is a facilitator only” We should not ponder on the rights of the organisation
- You cannot take AFRINIC away from the process
- AFRINIC needs to approve the transfers
Gregoire EHOUMI Community member
- Has issues with the problem statement such as the mention of IPv4 market in the region
- AFRINIC is not an IPv4 broker
- 5.7.2 there is no limit or requirement for source
- This is negative for AFRINIC and we will kill AFRINIC with these transfers
- Jordi said that transfers are from ARIN inbound this is a guess, we need supporting data and details
Cheken Chetty - University in South Africa
- AFRINIC is only the manager of the resources
- They should not interfere with Resource transfers
- The policy should not give more authority to AFRINIC since it is only mere managers of the resources
Oluwabunmi EGBEYEMI researcher in university in Nigeria
- AFRINIC should not have the authority to approve or reject transfers
- Transfers should happen without AFRINIC interference
- They are not owners of the IP resources but guardians only
- Opposes this policy
ELVIS IBEANUSI - Nigeria - University Student:
- I don’t think this policy will solve anything
- It will not change or solve anything
- AFRINIC does not have a need for authority with every transfer
- Opposes the policy
Mark Elkins
- Does not think we are rushing since this has been discussed since Pointe Noire
- Thinks we should be equal with the world to make space move around
- Resources stockpiled at AFRINIC should be made available to others
- Think AFRINIC resources should be made available worldwide to allow the Spacelink project for example to have resources from the region to allow connection to the remote location
James Chirwa - From AFRINIC Member Services
- Estimate that was run, as raised by Alain
- Our estimate was over the last 12 months Q3 2020 to 2021
- Q3 of 2021 shows different data that we considered for transfers within the region
- Legacy, no intra RIR has happened since the Intra RIR policy was noted
- All put in consideration, in the first two years we will have a slight decline in revenue
- In the third year is where we will have a noticeable impact between revenue and operational cost resulting in an increase in cost per member

Author responses:

Commented on the staff response to the impact on financials
Mentioned that a big chunk was received by LACNIC from ARIN which happened as a special unique event and will never happen again. This is not possible in the AFRINIC service region since there is no such big owner of resources in the AFRINIC service region
The policy takes around 1 year to implement, the pool is draining and new entrants should benefit from more resources
Due diligence is not only the responsibility of AFRINIC only but all RIR involved
We cannot delete bureaucracy as checking and validation is required
Resources transfers over the long term, the other RIRs are getting more resources since ARIN is a major Holder of Legacy resources
Implementation takes one year, we are not rushing
Yes, AFRINIC is a facilitator but we need to use the policies and proper market rules
If the policy is a danger for AFRINIC, Board can take the power of bylaw and endorse or not the policy
Again reiterated that there is a shortage on the AFRICAN market
Concerning objections from Owen, Using timeline from existing CPM to ensure that there is consistency in the periods mentioned in CPM
Understands that Owen is defending his customer, the Author is working in the interest of the community and believes that the policy is protective to AFRINIC and its members

Co-chairs advised that the online Q&A’s subjects were addressed as part of the intervenants on the queue.

Madhvi AFRINIC Staff - Liaison team - provided Clarification

As mentioned in the staff assessment, AFRINIC will act as RIR and assess all requests as per policies
We want to clarify that the source and recipient would be interacting outside the purview of AFRINIC, we will not be a party to
Whatever agreements or promises made between the parties
This is why the source and recipient need to do their due diligence
Jordi summarised the clarifications as AFRINIC is not taking part in the transaction but will check that the transaction can be done following the rules
Madhvi agreed to the summary

8.7) PDWG Chairs decision

The decision of PDWG Co-chairs is No consensus due to a number of valid concerns not addressed yet during the online session and from the mailing list. The proposal will go back to the list for community inputs to further refinement and clarifications.

PDWG chairs will share more details for the decision on the meeting minutes

9) Proposal #3:Publication of Information Policy Proposal

ID- AFPUB-2021-GEN-001-DRAFT02
* Proposal URL
* Presentation URL

9.1) PDWG Chair Introduction of the DPP and Discussion Flow

The PDWG Co-chair mentioned that this proposal was authored by Jordi Palet Martínez and invited the latter to present the policy.

9.2) Author’s presentation of DPP

Jordi Palet briefed the community about the problem his proposal addresses:

Recent discussions in the RPD and unprecedented situations in AFRINIC demonstrated that some of the information provided for the justification of the requests for resources could be used maliciously and not discovered at a first sight when those resources are allocated/assigned.
This is especially harmful to the community and the AFRINIC region in general when resources are being exhausted as is the case now for IPv4 and must be avoided.
At the same time, it is understandable that the justification of a request for resources can be confidential as it can provide insights into a business plan to competitors. However, it is also obvious that in the Internet business, time passes very quickly and the need for confidentiality vanishes very quickly. It is difficult to believe, for example, that you request resources for the next 5 years and only in year 5 your business will be public.
In summary, the problem is that long-term confidentiality of a resource request justification is against the principles of openness, transparency and fairness.
The author explained how his proposal addresses the problem mentioned above:
This simple proposal looks for the publication of information about resource request justifications after a community agreed period of time unless there is a valid demonstration of the need to extend the confidentiality period.
Note that the information to be published is not the already existing resource registration in whois as per the actual CPM. This proposal looks for publication of the overall resource request justification (justification of the need), which allows understanding of the justified need is sustained at the publication time or not.

9.3) Staff Impact Assessment

Brice ABBA from the AFRINIC secretariat presented the Staff Impact analysis starting the below:

The Staff Impact Analysis is work done by AFRINIC team members, done to make the work more readable.

It is an assessment that contains both legal & financial findings.

AFRINIC operates by giving resources to our members, based on the justification given to us.

The interpretation of the staff is this policy also stated the “how” of the task to be done.

9.3.1 AFRINIC Staff Comments on clarity of policy

AFRINIC has issued ~6500 ASNs, IPv4 and IPv6 prefixes to its members since 2005
Around 5000 of the above have reached their 2-year anniversary to date.
The needed justifications are provided to AFRINIC through the different platforms made available to gather such information and are therefore kept in tickets that are logged as a result of a resource request.
The information submitted by AFRINIC Resource Members is bound to be confidential and used only for the evaluation of the resource requests in compliance with the resource policies in the policy manual.
AFRINIC neither keeps a summary of these justifications nor has sought the agreement of the AFRINIC Resource Member so that the same can be published.
The AFRINIC Resource Members to whom resources have already been issued have not provided any agreement/consent that AFRINIC can publish a summary of their justification.
The author has clarified that staff is being mandated by the proposal to summarise the needs and share the latter with the Resource Members in the warning issued 3 months before the 2-year anniversary date of a particular resource. For ~5000 resources that have reached their 2-year anniversary date(issuance date is up to Nov 2019), staff will have to examine ~5000 records and make a summary
Staff will therefore do more work than "process the possible responses of “non-disclosure” without requiring extra human resources" & additional human resources will be required, even if a phased approach is being suggested by the author.
The proposal as styled lacks an element in regard to AFRINIC explicitly seeking the consent of the Resource Members to publicly publish its justification summary and the consequences of a member refusing to comply.
The attention of the author is drawn to the statement "allowing the staff to process the possible responses of “non-disclosure” without requiring extra human resources" and that staff will need to process ~5000 records pertaining to the ~5000 resources that have reached their 2-year anniversary. To date, AFRINIC has also issued ~6500 resources to its Resource Members.
AFRINIC takes note of the author's comments that automation may not be possible and further mentions that the justified needs do not exist on AFRINIC systems at the moment in a single standardised way.
We would like to mention that AFRINIC expects resulting questions, clarifications, and disagreements in respect to the text to be published or consent.
On the other hand, in the case of new member requests, the organisations may be requested to furnish an additional summary of their needs justification, in addition to the detailed needs, and provide their consent that the said summary will be publicly published 2 years from the date they were allocated/assigned their IPv4 and/or IPv6, ASN resources.

9.3.2 Impact on the registry function

The attention of the PDWG and Author is drawn to Section 4(d) of the Registration Services Agreement "AFRINIC will comply with all applicable data protection and privacy laws of the Republic of Mauritius in its handling of data and information submitted to it by the Applicant in furtherance of an application for services and use thereof." The proposal as written exposes AFRINIC to a potential breach of section 4d of the Registration Services Agreement, should it be driven by the proposal to publish a public summary in the case where no feedback or consent is received from the member within 30 days.
The proposal will introduce an additional workload for the registry operations team to draw summaries for ~5000 support requests that have already reached the 2 years threshold. In addition, human resources will be required to handle the workload as well as manage the Resource members' subsequent requests.

9.3.3 Legal Impact

The present proposal aims at providing for the disclosure/publication of information pertaining to a Resource Member’s needs, i.e. information which would have been shared with AFRINIC during the pre-contractual stage.
It is apposite to state that all information exchanged between AFRINIC and an applicant (resource member) either prior or during the tenure of the RSA fall under the regime of confidentiality at common law such that AFRINIC cannot, without the express consent of the concerned resource member or pursuant to a Judge's Order, disclose this information to third parties. In fact, this is not something that can unilaterally be imposed on individual resource members by AFRINIC through the latter’s PDP framework. Accordingly, there is no legal soundness to this proposal.

9.3.4 Financial Assessment

This section will be updated once clarifications have been received and confidentiality concerns have been addressed.

9.4) Author response to Staff assessment

Jordi defending his proposed policy with regards to two (2) points in the Impact Analysis:

Point 1:Confidentiality

AFRINIC will not be sharing confidential details about resources members. The policy is proposing to make a summary without mentioning the confidential details of the member.

Point 2: Workload

The policy proposes to phase the implementation using the resources available at AFRINIC. If it means that only 10 summaries are published per month….and within 5 years the exercise is completed, the author said he is fine with it.

The author assumes that if this policy is adopted, AFRINIC will start with members having big blocks of IP addresses, shoes that might have consensus or artificial justifications.

9.5) PDWG Chairs Summary

The PDWG chairs noted the following concerns and their status (Addressed or still pending)

Pending concern. Forcing members to come forward with their business when AFRINIC is just there to manage resources and not act with authority is not promoting openness and transparency within the Community.
Addressed concerns: Why is there a need to publish this justification?
I believe only resources in dispute should be published, and this must only be invoked by a court order. Risks of implementing this include copycat justification and witch-hunting from the competition.
Addressed concern: Does giving 1 month to provide alternative text give room to provide a different justification, or to revalidate existing justification? How does the additional one year change the privacy status of the business blueprint?

9.6) Open Microphone Discussion on the Proposal

Darwin Da Costa explained the house rules for the Q&A section and gave the mic to the attendees.

Before proceeding to the Q&A the author wanted to clarify that AFRINIC has the information regarding the business making the request for resources.

The degree of level of information that the proposal is asked to be shared is just a summary.

Prof Nii Quaynor has 2 comments
- Problem statement: I have a feeling this policy is hurried, as opposed to a policy that is “urgent”.
  - The problem statement states “the justification information could be used maliciously”
  - Prof Nii asked where the information was used maliciously? Or did the author mean “Could have been used” . The problem statement has to be very clear. The problem statement was not clear enough.
- Discussion on RPD: We have to know what should be presented as a policy.
  - Sometimes policies are seen as micromanaging the organisation.
  - And telling the organisation what to do about its confidential data and govern the organisation and the business.
Cheken Chetty | University from South Africa stated disagrees with this policy
- It violates and acts as surveillance on resource members by having confidential data monitored.
- The RSA already gives AFRINIC authority to validate information regarding resource members.
- And it also gives AFRINIC more authority to monitor their members shortly, which should not be the case, because AFRINIC only manages the AFRICAN IP resources and are not the owner of it.
Oluwabumi | Researcher, from Nigeria, stated disagrees with the proposal:
- The whole proposal is confusing, as well as the goal of the proposal. What are the goals this proposal wants to achieve?
- Why is there a need to publish all this information?
- What will the community benefit from it?
- If AFRINIC wants to promote openness, there are multiple ways. and safeguards must be ensured without compromising the data security of its members
Anthony, speaking from his own capacity
- The policy is something that he doesn’t understand, and one of the comments was not addressed.
- When a resource member has an agreement this agreement is bi-lateral,
- And this policy goes against the laws of Mauritius.
- “Why is Jordi writing this proposal?”
- “What is the ultimate goal of Jordi?”
Alain Aina | Internet Rechercher commented on the policy as shown:
- Let us not go and propose a policy at the very first round.
- It’s better to discuss first then we see if we propose a policy,
- If we agree that there is room for policy then we can work around that.
- Maybe it’s not a policy but rather a recommendation to staff or the board.
- If we try to sort everything by using the policy channel it is not correct, we lose time.
Questions from the Chat: Paul
It isn't clear to me whether this policy has retrospective application, will the disclosures be only in respect of the application for resources after the introduction or will it be a case for all members already done and are affected?
I believe a retrospective will raise serious questions? … and valid objections but forward operations.
Questions from the chat: Mafana Researcher from University South Africa.
I disagree, this proposal could be damaging to the business because of the publishing of information. It is counterproductive, regarding openness and transparency.
What if a member submits a draft full of loopholes?
Comment from Paul
While Paul has concerns about the legality of implementing the policy immediately. AFRINIC staff should start preparing these summaries. Even if the policy cannot be made mandatory, the exercise of preparing the summaries. I Implore the AFRINIC staff to start implementing this policy.

Author responses:

It would be good to prepare the summary even if we don't have a policy that is a perfectly valid internal process for the staff.
We want to make sure that the justification is not artificial. The only way to look at that is to have a summary.
According to Prof Nii comments, I am referring to what happened up to now and what can happen in the future. It may have already happened and we know that somebody has cheated AFRINIC.
I'm not a fan of micromanagement but when things are not clear the community can step in and decide how to do things.
To Alain, comments; We don't waste time having policy proposals. Policy Proposal is the way we have and the only way we have to get real input from the community.
To Mafana in the chat room, Yes, you can cheat the staff when you request the resource and try to cheat the staff when making the summary, I think it will be too obvious. With that, I think I am done. Thank you.

9.7 PDWG Chairs decision

Having considered the discussions on the rpd. mailing lists, the co-chairs have determined that the policy has not reached consensus, because of a number of valid concerns not unaddressed. The policy goes back to the list for further discussions.

The Co-chairs thanked everyone for their participation and ended the sessions.

2) Day-2 18 Nov 2021

The agenda for the second day of the Public Policy Meeting was as follows:-

09:00-09:50 Update of PDP - Policy Proposal
09:50-10:40 Policy Compliance Dashboard - Policy Proposal
10:40-10:50 TEA BREAK
10:50-11:40 AFRINIC Number Resources Transfer Policy - Policy Proposal
11:40-12:10 Open microphone
12:10-12:20 TEA BREAK
12:20-12:55 Open microphone
12:55-13h00 Closing Remarks for Day 2

10) Proposal #4: Update of PDP - Policy Proposal

ID: AFPUB-2021-GEN-002-DRAFT02
Proposal URL
Presentation URL

10.1) PDWG Chair introduction of the DPP and Discussion Flow

The PDWG Co-chair mentioned that this proposal was authored by Jordi Palet Martínez and invited the latter to present the policy.

10.2) Author’s presentation of DPP

The author started by mentioning that this proposal was originally sent in 2018 and due to expiration etc, it now has a different ID. He then highlighted the problems this policy intends to solve and they are as follows:-

The consensus definition is not clear.
Who constitutes the PDWG? Should participation be restricted to identified people? The PDP needs to work explicitly and without doubts of compliance in situations such as the Covid “new normal”.
It is the discussion for the consensus (and possible appeals) only the one in the meeting, or it should explicitly be inclusive (allowing people that can’t be in meetings) stating that the mailings list is part of it.
Shall the impact analysis be mandatory and what is the timing for that?
The Last-Call is not well defined.
Should the proposals expire if not updated and how often?
Is the actual PDP timing good enough or should be improved?
Can other online sessions be organized in addition to the formal Public Policy Meetings?
Shall the functions of the board on the PDP matters be better described?
Should the moderation of the discussions process be explicit?

The proposal covers most of the above based on personal experience in the RIRs and AFRINIC meetings.

The proposal is not trying to solve the following topics:-

PDWG Chairs Eligibility, conflict resolution and Acceptable Use Policy(AUP). These can be addressed in independent proposals and there was not sufficient time in this meeting to discuss all these.

He also mentioned that he will not be reading all the slides and that he is expecting that everyone has read the proposal and presented the following changes:-

Definition of rough consensus
Updates to Section 3.3 of the CPM and mentioned wording change derived from the staff impact assessment.
The following graphics show the change in the timing of the proposals. A minimum of 8 weeks for discussion of a proposal and allowing 2 weeks after the PPM for the Chairs to determine consensus.

10.3) Staff Impact Assessment

Presentation URL - http://afrinic.net/ast/pdf/policy/2-AF34-Impact-Assessment-18Nov.pdf

Madhvi Gokool from the AFRINIC secretariat presented the impact assessment of the proposal. Staff interpretation is documented on the website.

The impact is as follows:-

Secretariat duties
1. There is no limit to the number of proposals that can be put on an AFRINIC PPM agenda. Impact Assessments are comprehensively prepared and require the contributions of many internal stakeholders. The timing of these assessments to be prepared and published 1 week of the PPM needs to take into consideration the number of proposals on the agenda and the fact that updated versions of the proposal are also submitted by authors closer to the PPM
Legal
1. In reference to the policy proposal’s 2 week period to determine rough consensus, section 11.3 of the by-laws state that proposals for policies are discussed and agreed upon during the Public Policy Meeting. Therefore, a declaration of rough consensus (or not) must be made during the PPM itself, else it will not be consistent with Section 11.3 of the Bylaws.
2. The policy proposal’s last call amendments are inconsistent with section 11.3 of the by-laws. Last-Call must be restricted to changes to the policy that are purely editorial and non-substantial in nature.
3. Paragraph 3.4.5 as styled is inconsistent with sections 11.3, 11.4, and 11.5 of the by-laws. The question of the board of directors introducing its own adopted policy to the community by way of a DPP is questionable. In practice, and if section 11.4 is eventually triggered, the Chairperson of the board of directors would be required to submit a brief to the community at the next PPM substantiating the actions taken by the board of directors so that the said action may be endorsed by the community.
4. If any issues of the proposal are clarified, the timeline of implementation can be within 6 months of the Last Call as prescribed by the CPM.

10.4) Author response to Staff Assessment

He thanked the staff for working much better on the staff impact assessments and mentioned that the assessments could be provided this week.
The author did not agree with the legal inputs since he deemed them to be encroaching the community powers
11.2 and 11.3 don’t limit the time nor the format of the PPM and therefore the meeting can be kept open for as many weeks as needed.
Defining the last call can’t be discriminatory as participants of the PPM aren’t restricted to continue participating in the RPD. New objections that have not been raised can be done and can turn down consensus.
The by-laws don’t define what endorsement means and it doesn’t exist in the PDP. The only way to endorse something that the PDP has is through consensus.
By-laws are inconsistent with the intended purpose of the PDP and must be amended.

10.5) PDWG Chairs Summary

The PDWG chairs noted the following concerns and their status (Addressed or still pending):-

Addressed concern. “First, in general, the newly proposed version I feel is a longer version of the current version. There is no major modification made here. They are both the same thing, just one is more written out making the other look like a short summary but they both have the same meaning. Similarly, what's the essence of the additional one week at the end of the last call, in retrospect it means that the length of the last call is three weeks instead of two weeks in which every of the pure editorial comments and other modifications are made. There is no need for an extra one week"
Pending concerns relating to section 3.4.2 - “The reduction of the announcement of the meeting agenda on the RPD list from two weeks to one week makes no sense because it does not give the RPD enough time for discussion and less time to get a better grasp of the meeting's agenda. In the current version, it states that 'no change can be made to the draft policy within one week of the meeting' but good enough, there are two weeks. The proposed version estate the same thing but it only gives a one-week provision”
Pending concern to section 3.4.5. ”There is no necessity for the additional functions of the board of directors. Because it would be additional work to both the community and the board plus giving the board the opportunity to make temporary policy changes which would last until the next PPM might create an issue from it was created to the next PPM"
There is also a legal assessment in section 3.4.5 as well.

10.6) Open Microphone Discussion on the Proposal

Open microphone discussion on the proposal was open for 20 minutes.

The author stated that he already addressed the concerns raised. This was done in his presentation responding to the staff impact analysis.

Questions/comments from the Q&A were as follows:-

Oluwabunmi EGBEYEMI, a researcher from a University in Nigeria mentioned her disagreement with the shortening of the expiration of the PPD and questioned the purpose?
She is of the opinion that the PPV should be consolidated within one year, so it is approved to better serve the community.
Paul Hjul from Crystal Web had a follow-up question in regard to the legal assessment. He wanted to know whether the position being presented is that this proposal conflict with the bylaws, or whether the objection that is coming in the impact assessment is based on the idea that it does not align with the bylaws. He thinks that Jordi is 100% correct that the bylaws require that it is in the meeting process that the determination is made. So, the idea of adjourning the meeting for two weeks to afford the co-chairs an opportunity to apply their minds to something just makes perfect sense.
Elvis Ibeanusi mentioned that he is not in support of this policy. He thinks it gives the CEO some kind of…….[dropped off].
Alain Aina from Digital Intelligence Services had some points to raise:-
- He said that as per the author, putting some rules for the moderation of the working group can lead to discrimination. We know that every working group follows some rules to make progress.
- There is no discrimination because we have said that everything we're doing here is publicly archived. Anyone who joins the discussion can read the archives. There is no need for people to raise things we have discussed unless there is some new information or new data.
- He also noted something about the document saying that the chair should determine consensus. This is the wrong way of saying things. We never give the power to determine consensus, it is not what it is, the chair has the mandate to make sure that this working group has reached consensus. Again, we should make sure that when we write this, we follow the right procedure or it could be problematic.
- He commented on the topic of impact analysis. While he likes the idea that was said about fixing the PDP, this way of imposing staff analysis any time we change anything in the proposal is useless. Staff are being turned into one of the major players in the policy discussions.
- He suggested that now that we have extended the time for the policy to be discussed for eight weeks instead of four, why can't we say that we have four weeks to have a discussion, no need for staff assessment, and then after eight weeks we decided that the policy has matured, we go for another four weeks of review, and then after this period we request staff analysis. I think in this community we have lawyers, legal, we should not be going to staff all the time. We should not be asking for staff analysis at any time.
- The document must reach a certain level before we request staff analysis, we are the working group.
Professor Nii Quaynor from Ghana Dot Com, a member stated that:
- He thinks that the author may have failed to define the problem statement. The problem statement has more than eight or nine questions, that’s for the research study. If you want to solve a problem, you have to state that you have this problem, this evidence, and this is how to solve the problem. The author can be helped to better write problem statements.
- Section 3.3 begins to talk of identifying people on the PDP and even if the author is thinking of solving the problem, another problem may be created as the PDP should be open. It may affect the commitment to ICP-2.

* He said that the author described the process but missed out on a very important part which is proof of merit. If the person has a certain track record of having worked on PDP for a long time, it is different from someone who has just shown up with no understanding and telling us that we do not have any locus standi. We have to be careful how we describe these things.

The author was given a chance to respond to the questions:

The first objection has been already addressed.
The second one, you can send a preliminary agenda whenever you want to the list, and the chairs did that. He clarified that the final agenda should be fixed one week in advance. This is not creating the problem, because you can, and in fact, the chairs did this time, they said a preliminary agenda a lot of time ahead of the meeting. They started discussing with the authors, and then they sent the final agenda. What I'm saying is the same, but changing the timing so we are more efficient
The comment from Oluwabunmi the expiration time is for each version. He thinks she misunderstood that and explained that if the authors are not being proactive with the inputs from the list, then that version of the policy will expire. This forces the authors to be proactive.
He thinks Paul confirmed that he was right and does not see his comment as an objection. If you read all of the analysis impact from all of the policy proposals this time, there is a clear conclusion. Bylaws are wrong and are being misused as a throwing weapon against the community.
Regarding Alain’s comment, rules to make progress don't work. He asked Alain, which registry or which IETF working group is using them? None of those. Looking at the archives is not the point, the point is that if someone is not able to participate in the discussion-based week, and has an objection to something that has been addressed this week, he still has the right to raise the objection next week. The working group or the chairs will consider it or not, but the person has the right to say so. This is one way to prove that rules don’t work, and they are discriminatory.
He did not understand the point from Alain about consensus so he could not respond to that.
Regarding the impact analysis, he said that the facts prove that he(author) is right. If an author addresses the input from the community and impacts analysis, the next impact analysis will be simple because staff will just be deleting points from the impact analysis. This week, between this week and last week, the staff addressed the updated version of all of the impact analysis. So it works. It is a fact. That’s it.
Finally, in response to the comment from Prof. Nii, the author stated that the problem statement has not been defined by chairs in the list. That problem does not exist in my proposal. Regarding the identification of people, he stated that he was not saying to identify everybody participating in the working group, what I'm saying is identifying in case there is a suspicion that someone is using several emails. He thinks that is an obvious way for the staff to ensure that there is not any abuse. He thinks that this has been discussed in the list, and everybody must agree that there is a way to avoid that.
The way that part of the policy text is written, it is not assuming that it will always be possible to be done and that today, there are sufficient ways to check that.
Then, he did not understand the point from Prof Nii about the IETF. He said that his experience and the number of documents he has submitted to the five registries(more than 100) and the IETF(more than 250 documents) proves that he is talking from real experience.

Darwin Da Costa raised a question from the Q&A from Elvis Ibeanusi who was not able to speak up maybe because of connectivity issues. He said he strongly opposes the policy because of the provision that the CEO can step in and the PDWG chair is unable to do so. What if the CEO can split up the PDP in order to implement policies that will work in his favour? As we know, a policy must be discussed and known to the community. The CEO decides to speed up the process due to some unforeseen circumstances, wouldn't it be seen as an abuse of authority from the CEO? This policy will just give more authority to the CEO even in the PDP because he has the chance to step in the process itself.

Jordi Palet Martinez responded, saying that he is not sure if Elvis was reading his proposal or something else? He made no mention of the CEO in his proposal. He believes that Elvis is reading yesterday's one or maybe instead of CEO he is referring to the board. If he is referring to the board, this is something that already is in the bylaws. He(author) is trying to ensure is that because the bylaws talk about an endorsement, and we don't have such wording in our text, in our PDP, I am assuming – I'm not assuming it is the only way we have in the PDP to reach consensus if the board need to take an emergency policy is to reach consensus on that. What he is saying is, the bylaws say that he agrees with that, but we need to better define how we do that. That is it.

Alain Aina from Digital Intelligence Services clarified his point on the consensus. He said that the co-chairs are not given the right to co-chairs to determine consensus. Rather, the co-chairs are given the mandate to determine whether the working group reached a consensus. So when we are writing the text we should make sure we stay in that spirit otherwise it looks like the co-chair has a big power to determine consensus. The co-chair has the mandate to tell us if this working group reached a consensus on that. That is very important.

Alain Aina also raised another point - the idea of identifying people and asked why do we want to identify people here? People are only identified by the work and what they express. If it has a certain value for us we take; otherwise we go and so why are we trying to say we should identify people? Why should we say so? And then the problem statement issue. What the co-chair did was to ask questions. To go from questions to problem statements is wrong.

JORDI PALET MARTINEZ responded and said that he agreed with Alain on the consensus as he now understands what he meant. This proposal is not changing that. The co-chairs are determining if the working group reached consensus and are not determining consensus by themselves. Regarding identifying people this is just in case of abuse, that is it.

Prof Nii Quaynor from Ghana Dot Com also said that the document he has read had 9 questions in there and that the problem statement should not contain any questions therein. It should contain a clear statement that this is not good, evidence why this is not good and the approach that will be used. We are supposed to work by consensus. If they impede us in the way of people being able to freely contribute such as you are suggesting by identification, he thinks it is counter-productive. We ought to move away from that.

Paul Hjul mentioned that he has expressed these concerns in his email about the policies being plastering over deeper issues. He is not seeing anything coming forward as a valid objection to this policy being adopted and is trying to address the bylaw issues. And more importantly the interpretation of bylaw issues in a separate process. He can't understand the animosity that is being shown towards this proposal or its author that is coming up. The concerns about the CEO or the board misusing powers is quite separate from this proposal and what it seeks to achieve

Ashok Radhakissoon from AFRINIC responded to a statement made by Jordi which he considered very high-handed, especially regarding the staff giving the assessment, legal or otherwise. The role of the staff whenever it is examining a text is to highlight things that it considers, rightly or wrongly, needing to be flagged. When it is done, it is simply for that purpose. The purpose is not to restrict discussion within a working group. Our role is limited to our own interpretation of the bylaw, rightly or wrongly, these are community-approved documents and we think it is our duty to flag it whenever we feel the need for it. It is not to restrict discussion or whatever goes on in the working group.

Darwin Da Costa mentioned that the focus is to be kept on discussing problem statements on policy proposals and if the proposal is not good enough, what to do to make it better.

JORDI PALET MARTINEZ, responding quickly to Ashok said that I understand what he is saying. and congratulates the staff for the good job done in the list and in the meeting. The problem is that if you read some of the impact analysis, actually all of them, it is clearly going beyond suggestions from the legal part of the staff or from what the text of the authors is trying to do. In some cases, there is even a weapon like saying if the community has said that, it will not be ratified. That is in the case that we have already been in that situation, right? I don't think the way it is being worded is good enough. I am talking only about the legal part of the impact assessment. The rest is wonderful, really.

Responding to Paul, this is related to what Ashok said, it is not only against me. It is not only about this author, it is happening in all of the policy proposals. Even the policy proposals from Noah, Alain, have the same problem, they have weapons from the legal part of the impact analysis. Responding to Prof Nii, Jordi mentioned that he doesn't agree. If he reads the complete text of the proposal the problem statement is clear. The identification is in case of suspicious people that are using multiple emails on multiple personalities to participate and to hide or let's say put the balance in an objection more weight than if it is a single person.

Before going offline for deliberations, Darwin Da Costa encourage the PDWG to discuss the merits of any policy proposal by using the following guidelines:-

Do you agree with the problem statement and proposal as written?
Have you encountered an issue similar to the problem statement in the proposal?
Do you have any objections to the proposal as written. If yes, please state the sections and document your objections.
Are there any areas in the proposal that are ambiguous? If yes, please provide the changes you would like to propose.

10.7) PDWG Chairs decision

After deliberation, the PDWG Co-chairs made the following decision:

Having considered the discussions on the rpd and in the current public policy meeting. The co-chairs have determined that the policy has not reached consensus, because a number of valid concerns and objections remain unaddressed The policy goes back to the list for further discussions and community inputs, and refinement.

The Co-chairs thanked the authors & community for their engagement.

11) Proposal #5: Policy Compliance Dashboard - Policy Proposal

ID- AFPUB-2021-GEN-003-DRAFT02
Proposal URL
Presentation URL

11.1) PDWG Chair introduction of the DPP and Discussion Flow

Darwin Da Costa handed the microphone to the author to present on the policy. Jordi Palet Martinez was the presenter.

11.2) Author’s presentation of DPP

Jordi Palet Martinez from The IPv6 Company mentioned that:-

The registration SA does not set exact timings about whether staff should proceed in case of non-compliance.
When the AFRINIC community do policy proposals and they get consensus and then they are ratified, not all of the members are following all of those changes
Proposing to have an as much as possible automated way for updating members in their personal myAFRINIC page(not public to others) so that they know if they are still complying when a new policy is implemented
If there is a change applicable to all the members, those not following the policy discussions will not know and they will be lacking compliance
The proposed solution makes sure that they get warnings before actions are taken
Parts of compliance are severe, such as submission of falsified documents, if the staff or Board take a decision to immediately suspend the member, the author agrees with that decision.
Before taking action, members need to be informed that they are doing things wrong and AFRINIC should provide assistance if required.
The dashboard is consistent with the work being done on the new version MyAFRINIC and suggestions are being done.
Timing - AFRINIC cannot drop resources from one day to another, and it should ensure that the member has sufficient opportunities to correct the problem.
There is also a section for exceptions. For example, due to war or pandemic, members cannot fulfil their obligations, the proposal states that the Board may extend the revocation period.

11.3) Staff Impact Assessment

Policy Impact URL
Presentation URL

Policy Liaison Brice Abba mentioned that the impact on the registry function is as follows:-

New subprocesses are to be developed for the overall non-compliance workflow
The dashboard on the portal needs to be configured to send notifications as well as a periodical reminder of non-compliance to members
The portal needs to send notifications of persistent non-compliance after three months
Resource members follow-up in regard to persistent non-compliance are also required
It will also evolve as new policies are incremented
This proposal will enforce the contract and service agreement
RSA already empowers AFRINIC to initiate such reviews
The RSA also provides for the withholding and or the revocation of resources in cases where the concerned resource number fails to cooperate with AFRINIC

The legal assessment of the proposal was as follows:-

while the intention of the authors is evident, yet this proposal, as styled, has the effect of:
1. Encroaching on the internal management and operations of AFRINIC insofar as contract management of the Registration Service Agreement (RSA) is concerned.
2. The fact that AFRINIC will only be able to execute the provisions of the RSA upon “3 confirmed violations'' during a 12 months’ time frame makes it impractical and unrealistic inasmuch as breach(es) committed by resource members will differ from organisation to organisation. It is difficult to envisage how this proposal is workable in practice.
Besides, the RSA already empowers AFRINIC to initiate such reviews or investigations whenever it has good cause to do so. The RSA also provides for the withholding (suspension) and/or revocation of resources in cases where the concerned resource member fails to cooperate with AFRINIC. Therefore, the proposal that AFRINIC shall only trigger the termination process in cases where there is a persistent non-compliance on the part of the resource member is misconceived for the reason stated above.
Should the aforesaid policy proposal, as styled, reaches consensus, it is highly likely that the board of directors acting through its management will face difficulty to give full effect to the existing terms of the RSA since AFRINIC will be debarred from taking any actions under the RSA unless and until at least 3 violations on the part of the resource member have been recorded.
Further, by the time that a resource member ends up being persistently non-compliant, it is highly probable that the RSA which has an initial tenure of one calendar year would have automatically been renewed for another calendar year such that the identified breaches would then become ‘caduc’. Consequently, any identified breach(es) of the RSA on the part of the resource member would be deemed to have been regularised by the mere fact that the RSA has been renewed.

The policy can be implemented by Q3-2022

11.4) Author response to Staff Assessment

1. The PDP is able to manage the internal procedures of AFRINIC, same as in other registries
2. A similar proposal has been implemented in LACNIC but they don't have all of the legal assessments against the proposal
3. He agrees that micromanagement is not always good and the community has the right to draw the limit
4. The proposal is not amending the RSA but setting some details that are not in the RSA
5. The committee has the right to decide if a single breach, which can be resolved when notified, can be considered an RSA breach.
6. Certain cases of document fraud should not be allowed at any point as a mistake.
7. The proposal is not restricting AFRINIC to start an investigation at any time.
8. He is of the opinion that d) of legal assessment is incorrect because the timing provided is not conflicting with the RSA nor fixing the timing based on the RSA renewal time.

11.5) PDWG Chairs Summary

According to the PDWG Chairs, the contentious areas were as follows:-

The need for the implementation of a dashboard to monitor resource members seems to be intrusive in the privacy of resource members
There are more ways to promote transparency and openness within the community without compromising any sensitive information
Proposed a slight alteration to article 5. Contractual non-compliance such as unauthorized transfers, lack of payment or document fraud needs a fair hearing and considers all relevant circumstances in good faith.
Please also just confirm: My understanding is that this is based on data that AFRINIC already has, and no additional data will be required by AFRINIC from members.

The above have been addressed by the authors.

11.6) Open Microphone Discussion on the Proposal

1. Oluwabunmi EGBEYEMI - Researcher at a University in Nigeria mentioned that:-
* She does not agree with this proposal
* What are the safeguards that ensure there will be no data breach on the part of the members?
* The RSA exist for a reason and it should be able to provide a way for AFRINIC and the resource members to safely remind the members of the ideals and compliances

2. Alain Aina - Digital Intelligence Services mentioned that
* He supports the intention of this policy
* He asked Jordi, if he thinks that some of this dashboard data should be made public to the community?

3. Prof Nii Quaynor - Member - Ghana Dot Com mentioned that :-
* He likes compliance and does not see any surveillance issue
* His only challenge is that some things begin to cross the border - If staff is doing it, they have opportunity to engage and resolve. Why do we go to PDP to solve something that staff has already done? As I said previously there must be limits to what PDP can do? PDP is defined in the bylaws. He doesn’t want it to be superseding what is defined in RSA. Breaches are defined in RSA .He is concerned about scope creep of the PDP & that everything staff is doing is being micromanaged.
* What the policy is fixing is making sure that we have the opportunity to correct them

4. The author Jordi had the following responses:-
* Data is already available & sometimes is public(routing)
* Community can help that member
* Proposal asks to show each member its level of compliance - private.
* Lack of compliance will be made public (recovery of resources will have to be published)
* Staff doing the work their own away. Community is drawing the line that simple mistakes cannot be sanctioned.
* Automation is not perfect, hence automation as much as possible.
A further round of inputs was allowed by the PDWG chairs :-
5. Alain Aina asked if part of the dashboard should be published. He also clarified that we should clearly state the problem we are trying to solve when staff are doing the work.

6. Jordi as author responded that there is no need to publish. He also stated that :-
i) The proposal does not give more power to staff
ii) We are making sure how RSA is interpreted & applied in case of lack of compliance, is according to community decision & not staff’s decision which community cannot change.

7. Prof Nii Quaynor appealed to Jordi to come as peer to AfriniC & said that just as staff can make mistakes, so can the community. Community bottom-up process can also be hijacked. Encroaching on what staff do is to be avoided.
8. Mark Elkins intervened & mentioned that :-
i) he likes how the idea of this policy is there to assist members.
ii) Participants who are opposing are not members. I know this is an open community, but they have no skin in this game. I am very curious to understand what their motivation is

9. Sylvain ABOKA BAYA – cmNOG mentioned that the dashboard is a good thing, but he is not sure we can agree to push it from the policy. He does not support the proposal. He asked if there are any real issues with staff implementing such a platform.

10. Timothy AKINFENWA from a University in Nigeria mentioned that :-
* He does not have any problem with having a dashboard
* However, he would rather have something that would allow staff to be able to innovate and also put in their own idea based on what is already available in the RSA
* the major problem is what happens if a member fails to comply with policies

11. Paul HJUL from Crystal Web mentioned that :-
* He picked up security related issues in the Q&A. Nothing has come forward from that assessment that says this cannot be implemented in a way that preserves data privacy
* But we should not underestimate securities importance and privacy importance when it comes to implementing the policy
* He is a supporter of this policy with the amendment he had proposed.

Jordi Palet Martinez – The IPv6 Company
* The way the policy is designed already allows the staff to innovate
* Failure to comply is also described in the policy
* The PDP is the only way we have to take decisions and fix issues that are not clear on other documents
In response to Alain Aina, he said that:-
* The information is not public, it is only visible to its member
* Part of data will be made public because when AFRINIC recovers resources, it has to publish it. Everyone should know it has been recovered

11. Comments from chat; the statement of Keessun Fokeerah – AFRINIC were read by the PDWG Chair
* The reason for this policy is to empower members to be policy compliant and avoid surprises when members are non-compliant
* It will be implemented on myAFRINICv2
* Each member has access to their own data
* Phased implementation will be used to cover all provisions over time

11.7) PDWG Chairs decision

The policy has reached a consensus. Having considered discussions on the RPD mailing list and the current PPM. The authors have addressed the concerns raised by PDWG, the co-chairs have determined consensus has been reached. The policy proposal, therefore, moves to Last Call and the community can further comment in that period.

12) Proposal #6: AFRINIC Number Resources Transfer Policy - Policy Proposal

ID: AFPUB-2020-GEN-006-DRAFT02

Proposal URL
Presentation URL

12.1) PDWG Chair introduction of the DPP and Discussion Flow

Vincent Ngundi, PDWG Chair, announced the proposal and mentioned that the proposal was authored by Gregoire EHoumi, Noah Maina and Alain Aina.
He then gave an overview of the flow of discussion before giving the floor to Gregoire Ehoumi.

12.2) Author’s presentation of DPP

Gregoire Ehoumi mentioned that:-

1. IPv4 pool is expecting to run out soon .. and we need IPv4 to support their IPv6 deployments
2. The authors think there is a need to allow for unused IPv4 from other regions to move into the AFRINIC region without necessarily depleting the AFRINIC amount of IPv4 addresses. The reason being AFRINIC and AFRINIC service region needs more resources.
3. Current intra-RIR also allows all types of IPv4 addresses to be transferred including IPv4 from special blocks
4. Current intra-RIR does not allow ASN transfers
5. AFRINIC has 7.23 /8s with a low ratio of IPv4 addresses per inter-user.
6. The policy proposal defines a set of rules which he presented briefly.
7. The authors added some definitions & markings such as:

AFRINIC pool - marked as Regional
Special-purpose pool marked as Reserved
Legacy marked as Legacy
Others marked as Global

8. Presented an Overview of the rules as follows:-

9. Incoming transfers will bring more resources to the AFRINIC service region.
Note: “Reserved” is not part of the flow
They set several conditions (he did not cover them during the presentation) and some provisions

10. Regarding the compatibility/reciprocity: Last year, in regard to previous versions
* RIPE and LACNIC said proposal OK
* APNIC and ARIN have some reserves

11. Alain Aina mentioned that the authors are trying to improve the intra RIR transfer & respond to the need to bring in some numbers in the continent, while protecting the tiny pool of resources.We see there are current intra transfers in the region. This is a proposal which is solving a problem without breaking nothing

12.3) Staff Impact Assessment

Presentation URL

Madhvi Gokool from the AFRINIC secretariat presented the impact assessment of the proposal, mentioning that the interpretation and understanding have been published on the website.
1. AFRINIC had submitted some recommendations for editorial edits and definitions :
* the term "others" appears to be vague . we request the authors to have it clarified
* The term ‘any’ seems to be vague in section 3.3.5
* the type of transfer is ambiguous in section 3.4
* absence of 'hold down' time may lead to abuse of registry
* Under the conditions of the source & Section 3 , we suggested a rephrasing.

2. There are different areas of impact as this is an inter-RIR transfer
3. MyAFRINIC v2 --> resource tagging especially . This is being implemented on myafrinicv2 and specifications need to be revised in case the proposal reaches consensus.
4. Member Serice operations : impact on procedures and processes as there will be ASN transfers over and above IPv4 transfers
5. Human Resources : need for staffing in MS department
6. revision of transfer agreement and RSA may be required if this proposal reach consensus
7. MS tools : major rewrite of automated transfer tools to accommodate ASN transfers and integrate with other RIRs systems
8. Legal assessment is as follows :-
1. The decision of allowing, or not, inter-RIR transfers of IPv4 resources from and to the AFRINIC region is purely and simply a business decision to be taken judiciously and prudently both by the PDWG and the Board of Directors
2. One clarification regarding legacy resource holders: legacy resource holders existing within the AFRINIC’s service region are not contractually bound by AFRINIC’s adopted policies such that these policies have no direct effect on legacy resource holders, and it is up to those legacy-holders to adhere to AFRINIC’s policies. Thus, the authors should bear in mind that obligations impacting legacy resource holders may not necessarily achieve the intended results if the legacy resource holders refuse to opt for voluntary registration of the transfer with AFRINIC.
3. In regard to outbound transfers of resources. It is understood that the intended transfers will be channelled through AFRINIC. Therefore, other than simply setting out the conditions for transfers, AFRINIC’s role in the whole process must also be adequately defined. In particular, it is unclear as to whether AFRINIC’s role in the whole process would be limited to facilitating the administrative aspect of the intended transfers only with or without such legal responsibilities attached thereto, more so that AFRINIC will be relying on representation made to it when attending to similar requests. Accordingly, it is proposed that the burden of conducting such adequate due diligence with respect to the source holder or the concerned IPv4 number resources be borne by the intended recipient, and that AFRINIC’s role should be limited to act as a facilitator only without bearing any legal responsibility whatsoever in that process.
4. It is not clear in the proposed policy whether the concerned IPv4 legacy resource will lose its legacy status upon transfer into the AFRINIC’s service region in as much as the current RSA is not presently tailored for that purpose.
5. it is also important to clarify whether, in case of inbound transfers of legacy resources, AFRINIC will be able to execute its RSA with the obvious risk of the concerned IP number resources being reclaimed by AFRINIC in case of a subsequent breach of the RSA, despite that the recipient organisation would have most probably paid good consideration (financial value) for such transfers.

On the topic of reciprocity, APNIC just gave an update ... they are okay in terms of reciprocity with this version of the proposal

Since IPv4 & ASN resources from the AFRINIC Pool can only be transferred in-region(Intra), AFRINIC will not lose its current resource members to other RIRs in outgoing transfers. This proposal will therefore have a minimal financial impact on AFRINIC’s revenue.

The implementation will take about 12 months because of the complexity of inter-RIR transfers.

12.4) PDWG Chairs Summary

PDWG Chair Vincent Ngundi said there were no discussions on the mailing list and therefore there are no contentious issues. He then asked the authors to address the issues raised in the Impact assessment, in the interest of time.

12.5) Author response to Staff Assessment

The authors responded as follows:-

Rewording of the definitions was accepted as follows:-

The author also mentioned that:-

* Others is well defined -mentioned 3.1.5 and 3.4
* Section 2.0 cover concern in regard to the reclamation of resources
* Most of legal aspect are comments and they think that they already covered them in Sections 3.4 and 3.6 of the proposal. Members need to sign the RSA and resources will be covered by AFRINIC policy.
* On the topic of "hold down" time: Section 3.6 which addresses that. We don't want to put concrete to justify the need for people building network in the region .

12.6) Open Microphone Discussion on the Proposal

1. Oluwabunmi EGBEYEMI mentioned that she does not support the policy as non members would have to be member. Vincent Ngundi, PDWG Chair asked her why she does not support the proposal and there was no response. He said there is a part (of her comment) which is not a valid concern.
2. Jordi mentioned that he also sent his questions on the mailing list and he doesn’t think this proposal is reciprocal with ARIN and that it becomes null. ARIN is the major donor to the other regions. It is utopic to believe that if many addresses are coming from LACNIC to AFRINIC, LACNIC can take action and can even change the policy before even the one in AFRINIC becomes implemented. In this case, he agrees with the legal assessment and believes there are several points to mention and to take into consideration the staff assessment which he does not think are just editorial. Taking care of the editorial changes in this part of PDP is against the PDP.
3. Paul Hjul from Crystal Web commented on the fact that it is indicated that there are no discussions on the mailing list. The issue with this policy is the other conflicting policies -one is under appeal and another one in discussion and that it is problematic to take the view that nobody raised valid objections to the proposal. We therefore need to be quite careful on how we handle the situation.

4. PDWG Chair Vincent Ngundi clarified that there were discussions on the previous version. There was an update to the impact assessment on this. It's been discussed extensively but not for that version.
5. Gregoire Ehoumi mentioned that the authors prefer that the staff provide the update on reciprocity from ARIN as they even updated us on the position from APNIC today. RIPE is the big donor of IPv4 resources.
6. Arnaud mentioned that this proposal has been discussed plenty and is more adapted to the AFRINIC region and he supports this proposal as it protects the resources of AFRINIC as compared to the proposal that was discussed yesterday.
7. Prof Nii Quaynor Ghana Dot Com made some comments :-

* RIRs are the rightful person to determine reciprocity.
* Since this is not a global policy and a subset of RIRs find it reciprocal, then we can start.
* any policy have different lifecycle , the right way to go is to think hard about the purpose of the policy
* the net gainers in numbers is RIPE and if RIPE is open, we should go and might get the benefit

In response to the above queries, Noah Maina, co-author mentioned that:-
* They are trying to be extremely conservative and that resources are allowed to flow in the region.
* This policy has been out for so long and the second version of this policy is being discussed. The feedback we are seeing today has already been addressed pretty much. As a start, we can work with RIRs that are reciprocal and as times change, improvements can be bought at a later stage by the working group.

Alain Aina as co-author also mentioned that:-
* In the AFRINIC context, there are two types of entities- the members and the legacy holders. Legacy Holders are allowed to transfer their resources without becoming a member of AFRINIC
* If legacy resources are transferred inbound, you must become a member
* ARIN is the biggest donor but its policy is based on the need.
* If other RIRs can change their policy to protect their region, then we can also protect AFRICA.

Madhvi Gokool from AFRINIC responded on reciprocity with ARIN, the latter has confirmed that this proposal is not reciprocal with them.

12.7) PDWG Chairs decision

After deliberation, the PDWG Co-chairs made the following decision and announced that having considered the current version of the policy proposal, the deliberations during the current PPM and the editorial changes that the authors have committed to, they have determined that rough consensus has been reached. The updated version of the policy proposal will now move to Last Call and the community can further engage in that period.

13) Open Microphone on the PPM

Vincent Ngundi started the open microphone session and mentioned that he would like to first make a presentation based on the co-chairs’ experiences.

The following elements were raised :-
* The need to have clarity in the problem statement, else the proposal will not align
* Avoid proposals that seek to address administrative functions
* Need to have arguments based on facts and if statistical in nature are based on empirical data.
Number of proposals under discussion
* Impacts both the PDWG and Secretariat
* PDP is a bottom up process and is based on volunteers
* Stretches the resources devoted to produce impact assessments. A lot of work goes behind the preparation of these public policy meetings
A dramatic increase in policy discussions on the mailing list as the PPM nears :-
* Summaries are prepared and some discussions may inadvertently be left out.

Code of conduct

* Unfriendly to new entrants and it is a hostile Community & Environment
* Loss of contributive culture
Participation in policy discussions
* Few participations are noted
* Community input that does not consider the legal part of Staff assessment, AFRINIC Bylaws, RSA as these form part of the AFRINIC working environment
* There is a view that voting is equivalent to community participation which is not the same. The work is done by rough consensus.
* Further issues:
* Competing policies
* Authors do not want to come together
* Brainstorming are required
* The culture of the community needs to change so that it becomes more collaborative
* Authors of the “Co-chair recall proposal” never presented at any PPM and when they were approached, they did not respond to the request to present in this PPM during AF34. Authors need to allocate time and resources when they submit proposals.

The co-chairs thanked the audience for their participation and support and opened the microphone for the participants to intervene :-

Intervenants
1. Jordi mentioned that :
* In regard to RSA and Bylaws , he always read them and if they are wrong or are broken, the resource members need to fix it.
* Opportunity for improvement is there and we need to make the most use of it
* There is a different interpretation of the CPM which needs to be addressed
* The problem statement should not be mandatory, it can be a problem statement or Opportunity of improvement.
* He thanked staff and co-chair for the good job done
* Requested that the impact analysis be well in advance(many weeks) in order to have proper discussions to allow authors more time to update their proposals.
* He suggested that Co-chairs should take their time to take the discussion even after the meeting. No need to stress during the meeting itself by not adjourning the meeting
* Asked a question to the PDWG Chairs - What part of the PDP is telling the chairs that they can hold the last call if there is a pending appeal. He says this is an illegal act against the community and it helps those who are looking for opportunities to collapse the PDP.
* SInce the Chairs and Appeal Committee are different bodies, they do not need to synchronise. What is acceptable is that the Board may not ratify the proposal if there is a pending appeal. This gives a weapon for people to initiate a recall.

Vincent Ngundi Co chair responded as follows :-
* The policy proposal is not the only way to improve things. There could be ways to improve the legal documents, bylaws and RSA. In regard to appeal, legal opinion is sought. The appeal committee can also be expedited to make the ecosystem work much better.

2. Eddy Kayihura, CEO of AFRINIC intervened and mentioned that:-
* When a policy is under appeal, the Board cannot ratify
* Legal team can assess what happens if a policy reaches consensus while another one is under appeal
* IMpact assessments - a reasonable and clear service level agreement to be worked on
3. Ashok Radhakissoon AFRINIC Legal advisor
* To answer the question about Policy lying in appeal and another policy reaches
* Legal term Conflicting judgement is used.
* When there an appeal, we wait for final disposal of this appeal before we proceed
* This is based on the legal framework and is used everywhere.
* Best legal judicial course if to wait for final response of the appeal

4. Alain Aina
* Commended the PDWG Chairs for work accomplished
* ICP-2, bylaws and RSA are the founding documents.
* Changes to bylaw are possible and have been done in the past. We should stay in the remit of what is doable
* Conflicting policies do not need to be based on the title only
* If we accept proposals based on the merit of problem statement, the conflicting policies will go away as the proposals may not be solving the same problem
* Asked the CEO that the Consolidated Policy Manual work has not been completed well and the CPM is not in compliance with the policy text. He appeals that the CPM be completed.

5. Prof NII Quaynor
* Thanked the co-chairs since a clear process is being followed
* Proposals with single author to be reduced
* Number of proposals authored by same author to be reduced
* We should reduce pressure on PDP and move to more fruitful thinking
* Addressed Jordi and said there is a culture in each region
* Leave the company alone
* Use of words as weapons or collapse
* In our culture, community is not on the top of the organisation
* Community is clearly defined by the bylaws of the organisation
* You may have the best idea but avoid telling co-chairs what to do
* If the problem statement is not important, we want to ask you to please understand that problem statement is important
* Be more familiar in the community and the culture
________________

6. Noah Maina
Thanked the co-chairs and working group
Questions:
1. Follow-up with impact assessment with CEO, how can we better improve this? He Suggested a discussion on the impact assessments with authors, staff and PDWG chairs one month before the PPM.
2. Co-authors could also present the proposals
3. Authors , if they have an idea engage with the Working Group and the latter can help to have more engagement and work towards the problem statement
4. Believes we are in the right trajectory
5. If people are open to discussing, the community can better work together

7. Madhvi Gokool from AFRINIC mentioned that :-
* The CPM review is already in progress
* As soon as this is completed or significant progress is made, we will provide feedback to the community.

8. Sylvain ABOKA BAYA mentioned that
* Policy should focus on clearly defined problem as a good practice
* Problem statement that is well defined can help to define a solution and to reach agreement
* Anonymity is perfect, with responsibilities
* It is perfect to address the concern
* Thanked the Co-chairs for the good work done, policy team and participants

9. Jordi mentioned that
* He was pointing to Abuse policy in his question to the PDWG Chairs
* The policy reached consensus and moved to Last Call. Because there is an appeal , the website mentioned that it has status “held by co-chair decision”
* No way we should hold a policy if there is an appeal. PDWG Chairs should move on with the process.
* If board cannot process,

10. Alain Aina also mentioned that :-
* Consensus is already definedRFC 7282 and IETF we should also make more analysis on this
* RFC clearly mentions how Objections should be managed
* We have commitments to global community w.r.t ICP 2, technical committee,
* Check proposal against IETF standards
* Assess the impact the policy will have on the networks and membership and NOT on the individuals
* PDP is backed by the Working Group, there is room to hold discussions, design teams, meetings with the engineering team before we go policy way

11. Anthony Ubah commented as follows:-

* The PDWG should encourage similar policy authors to sit together to work on the idea together and Come to an agreement to have fewer policies
* Narrow down the number of proposals
* Spend more time to discuss on roadmaps

12. Prof Nii Quaynor mentioned that:-
* PDWG chairs to devote some attention to building a community and groups, we need to cultivate the culture
* Avoid RPD list being used to make elections even if the covid has moved us a bit
* PDP to be used for consensus.

13. Noah Maina mentioned that:
* Instead to focus on being an author, we could channel to the Working Group(WG)
* WG can then guide the authors

Darwin Da Costa mentioned that we are in the process of learning from, guide & supporting each other. The chairs commit to taking into consideration all the feedback received and improving.
Vincent Ngundi mentioned that collaboration (essence of the community), the culture of working together and clarity of purpose(problem statement) are key elements

Close of the AFRINIC PPM

Vincent Ngundi and Darwin da Costa, PDWG Co-chairs, thanked the AFRINIC Board, Policy Liaison Team and everyone who contributed to making the PPM a success, before closing the AFRINIC-34 Public Policy Meeting.

Last Modified on - 15 July 2024